Small leaf breakdown in a Savannah headwater stream

Page 1

R

S

ERa

b

a

ARRAA

KBCRDSF

C

h0

Limnologica 51 (2015) 131–138

Contents lists available at ScienceDirect

Limnologica

jo ur nal ho me page: www.elsev ier .com/ locate / l imno

eview

mall leaf breakdown in a Savannah headwater stream

lisa Araújo Cunha Carvalho Alvima,∗, Adriana de Oliveira Medeirosb,enan Souza Rezendea, José Francisco Gonc alves Jr. a

Universidade de Brasília, Departamento de Ecologia, IB, Campus Darcy Ribeiro, Asa Norte, Brasília CEP 70.910-900, DF, BrazilUniversidade Federal da Bahia, Instituto de Biologia, Campus Universitário de Ondina, Salvador CEP 40.170-115, BA, Brazil

r t i c l e i n f o

rticle history:eceived 5 March 2014eceived in revised form 20 October 2014ccepted 24 October 2014vailable online 11 November 2014

eywords:acchariserradoocky field

a b s t r a c t

The chemical nature and nutritional quality of leaves influence microbial colonization, microbial activ-ity and consequently leaf breakdown rates. In the present study, we compared the decomposition ofBaccharis concinna and Baccharis dracunculifolia leaves and the influence of leaf quality on the micro-bial activity during the decomposition process. This investigation was conducted in a Brazilian savannaheadwater stream with a riparian zone composed predominantly of herbaceous and shrubs. The break-down coefficient was higher in B. dracunculifolia than in B. concinna; for both species, increases in leafmass were observed after the 60th day. The secondary compounds were quickly leached in the firstseven days, but the structural compounds persisted longer and served as the main carbon source for thedetritus-associated microorganisms. The highest values of ergosterol were observed in the final stages

ecompositionecondary compoundsungi

of leaf breakdown and indicated the difficulty of colonization on the detritus; these values were relatedto the increase in leaf mass. The ATP content increased without corresponding increase in ergosterolcontent, suggesting a biofilm formation during leaf breakdown. These results indicated that the totalmicrobial biomass can assimilate organic compounds released from detritus by the enzymatic action of

fungi, demonstrating the importance of this group for releasing the energy stored in small leaves.

© 2014 Elsevier GmbH. All rights reserved.

ontents

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

Study site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132Water parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132Experimental procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132Characteristics of detritus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133Microbial activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133Data analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133Physical and chemical stream characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133Leaf breakdown. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133Physical and chemical detritus characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134Leaf-associated microorganisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135Breakdown rates and characteristics of leaf detritus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

Contribution of microorganisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

∗ Corresponding author. Tel.: +55 61 3107 2987.E-mail address: [email protected] (E.A.C.C. Alvim).

ttp://dx.doi.org/10.1016/j.limno.2014.10.005075-9511/© 2014 Elsevier GmbH. All rights reserved.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137

Page 2

1 nolog

I

tlImt

wcaettcsa

opbcoedppde2sa

dosacmCwwwoae

svlsdhWafwhdbtsrtl

32 E.A.C.C. Alvim et al. / Lim

ntroduction

The functioning of fluvial ecosystems is determined by the rela-ionships between organisms and the environment, as well as byocal physical and chemical processes (Allan and Castillo, 2007).n headwaters and small streams, the functioning depends on the

etabolic activity of organisms and the flow of energy between theerrestrial and aquatic systems (Gomi et al., 2002).

Several investigators have noted that in headwater streams,here light can be a limiting factor for primary producers (as a

onsequence of the dense forest cover), the main source of energynd organic matter is leaves from riparian vegetation (Vannotet al., 1980; Pozo et al., 1997; Franc a et al., 2009). In streams wherehe riparian canopy is poorly developed, light is not a limiting fac-or for the growth and development of the algal community, andonsequently, high autochthonous production becomes the mainource of energy in aquatic food webs (Bunn et al., 2003; Estevesnd Gonc alves, 2011).

Decomposition is an ecological process essential for the flowf energy and organic matter cycling in headwater streams. Thisrocess is responsible for mineralization, aiding mineral remo-ilization to the trophic web (Allan and Castillo, 2007), and theolonization and conditioning of the detritus by microorganisms isne of the most important factors in this remobilization (Gessnert al., 1999). Microorganisms (fungi and bacteria) colonize theetritus, producing extracellular enzymes that are able to degradelant structural polymers such as cellulose, hemicelluloses andectin, making detritus more palatable for consumption by shred-ers (Canhoto and Grac a, 1999). Bacteria usually colonize in thearly stages, using easily assimilated molecules (Komínková et al.,000). Fungi reach a higher biomass and act during the advancedtages of decomposition (Gulis and Suberkropp, 2003b; Nikolchevand Bärlocher, 2005).

According to Gessner and Chauvet (1994), studies of leaf break-own must consider the chemical nature and nutritional qualityf leaves, which directly influence the breakdown rates. Plantpecies with low contents of structural macromolecules (lignin)nd defense compounds (polyphenols and tannins) and high con-entrations of nitrogen and phosphorus are more susceptible toicrobial colonization, leading to higher decay rates (Gessner and

hauvet, 1994; Ostrofsky, 1997; Das et al., 2008). Leaf toughness,hich may be related to high concentrations of lignin (a compoundith a complex molecular structure and low nutrient content) andaxes, which provide rigidity to the plant tissue and form layers

f preventive barriers to water loss, can also influence microbialctivity, reducing leaf breakdown (Canhoto and Grac a, 1999; Ravent al., 2007; Li et al., 2009).

Plants of the Cerrado present higher concentrations ofecondary and structural compounds, which protect against herbi-ores and pathogens in these ecosystems and may directly affecteaf breakdown. In addition, the riparian zone of the rocky fieldystem is composed predominantly of herbaceous and shrubs andoes not form a true canopy over the water body, allowing theigh light incidence on the stream (Oliveira-Filho and Ratter, 2002).e propose the following hypotheses: (i) the small leaves associ-

ted with the open canopy increase the leaf breakdown due to theacilitation of microbial colonization due to its negative correlationith leaf size when compared to other Cerrado streams; (ii) theigher concentrations of secondary and structural compounds inetritus will negatively affect microbial colonization and thus leafreakdown rates. Our objectives were as follows: (1) to comparehe decomposition rates of leaves from two typical and abundant

pecies in high-altitude areas of the Cerrado biome, including theiparian zones of headwater streams; (2) to compare changes inhe concentrations of secondary and structural compounds duringeaf breakdown; and (3) to analyze changes in the reproduction and

ica 51 (2015) 131–138

biomass of fungi and total microbial community correlating withthe dynamics of chemical compounds of detritus. We chose Baccha-ris dracunculifolia DC. and Baccharis concinna G.M. Barroso becausethey are common in the riparian zone, and the Baccharis genus isimportant for honey production and produces many compoundsof pharmacological interest. The study of these species may helpto demonstrate the importance of the conservation of this genus;thus, any human exploitation should be minimal and consider thatthese species contribute to the energy stability of ecosystems.

Methods

Study site

This study was conducted in a 2nd order stream (GeraldinhoStream) in Serra do Cipó, Minas Gerais, Brazil, within the SãoFrancisco Basin (19◦16′55.51′′S, 43◦35′34.46′′W; altitude 1135 m).The riparian zone is composed predominantly of herbaceous andshrubs. The experiment was conducted in the dry season from Mayto September 2009. The mean annual temperature in the regionvaries between 17.0 and 18.5 ◦C, and the mean annual precipitationvaries between 1450 and 1800 mm (Gonc alves et al., 2006).

Water parameters

In each sampling period, we measured current speed with aFluxometer (model SWOFFER 2100 series), temperature and dis-solved oxygen (YSI 55 dissolved oxygen meter), pH (Digimed MD20), and electrical conductivity (Minipa MCD-2000) in the field. Wecollected 1 L of water in each period to analyze total alkalinity usingthe Gran method (Carmouze, 1994). Nitrate and soluble reactivephosphorus concentrations were measured using the cadmium col-umn reduction method and the ascorbic acid method, respectively(APHA, 2005).

Experimental procedure

Senescent leaves of the shrubs B. dracunculifolia and B. concinnawere collected from various individuals around the stream withnets before they fell to the ground; samples were collected 6months before the experiment. Leaves were air-dried and incu-bated separately in fine-mesh litterbags (0.5 mm mesh size), witheach bag containing 1.5 ± 0.1 g of dried leaves. Sixty-four litterbagswere placed in rows in the stream just above the streambed andtied to steel rods and submerged rocks.

The decomposition rate was measured by the weight loss of thedetritus incubated in the stream for a period of 120 days (withpartial removal after 3, 7, 15, 21, 30, 60, 90, 120 d). On each samp-ling date, four litterbags with each leaf species were removed,numbered sequentially, placed in individual plastic bags, and trans-ported to the laboratory on ice in an insulated container. In addition,four replicates per species were prepared, corresponding to dayzero, which were used to assess the weight loss resulting frompreparation, handling, and transporting the sample to the field, i.e.,to correct for weight loss that did not result from decomposition.These replicates were also used to determine the initial chemicalcomposition (structural and secondary compounds).

In the laboratory, the leaf material was carefully rinsed with dis-tilled water, and 20 leaves of the same size were removed fromeach litterbag and divided into four groups of five leaves each.Each group of leaves was used to determine: (1) ash free dry mass(AFDM) of leaves; (2) fungal biomass, estimated from the ergoste-

rol concentration; (3) sporulation rate of aquatic hyphomycetes;and (4) microbial biomass, estimated from the ATP concentration.For AFDM analysis, groups of leaves were dried at 60 ◦C for 72 h andthen placed in the muffle furnace for 4 h at 550 ◦C to combust the

Page 3

nologica 51 (2015) 131–138 133

oaAai

C

tt

batrcr

M

tofa(aa

ai2fTa0

iwtT−pel

D

uW(fir

spmtcTaait

Table 1Chemical and physical characteristics (mean ± standard error, n = 9) of Geraldinhostream during the experiment (May–September 2009).

Parameter Mean (SE) Range

Temperature (◦C) 21.0 (±0.5) 19.1–23.9Electrical conductivity

(�S/cm)59.6 (±16.9) 5.41–132.0

pH 7.0 (±0.3) 6.1–8.9Dissolved oxygen

(mg/L)8.1 (±0.2) 7.6–8.7

Alkalinity (mEq/L) 12.5 (±3.9) 0.99–32.3Flow velocity (m/s) 0.06 (±0.009) 0.04–0.13Discharge (m3/s) 0.003 (±0.001) 0.001–0.007Ammonia (mg/L N) <0.05 –Nitrate (mg/L N) <0.10 –

B. concinna leaves broke down more slowly than those of B.dracunculifolia (k = 0.0023 d−1 and k = 0.0064 d−1, respectively). Todecompose 50% of the mass of B. concinna, 301 days would berequired, whereas for B. dracunculifolia 108 days would be required.

E.A.C.C. Alvim et al. / Lim

rganic compounds. For fungal and microbial biomass (ergosterolnd ATP, respectively), groups of leaves were frozen until analysis.fter washing, the remaining leaves were placed on trays and driedt 60 ◦C for 72 h to determine the dry weight and further trituratedn a mill to analyze the chemical composition.

haracteristics of detritus

The leaves were measured to obtain the mean lengths and werehen scanned using ImageJ software (Rasband, 2006) to determinehe mean leaf area.

The concentration of total phenolic compounds was measuredy extraction of the polyphenols in 5 mL of 70% acetone for 1 ht 4 ◦C (Bärlocher and Grac a, 2005). The condensed tannins con-ent was estimated as protein-precipitating potential using theadial diffusion assay described by Grac a and Bärlocher (2005). Theoncentrations of cellulose and lignin were determined gravimet-ically, following the procedure of Gessner (2005b).

icrobial activity

The biomass of aquatic fungi on the detritus was determinedhrough extraction of ergosterol, a sterol restricted to membranesf fungi as described by Gessner (2005a). The group of leaves wasrozen at −20 ◦C for extraction, the leaves were placed in methanol,nd the extraction of lipids and hydrolysis was performed in boiling60 ◦C) KOH/methanol. The extract was purified by passage through

SPE cartridge. Next, the ergosterol was eluted with isopropanolnd quantified by HPLC (Waters Co.).

For each sample period, each group of leaves was incubated onn orbital shaker (100 rpm) at 18 ◦C in Erlenmeyer flasks contain-ng 30 mL of filtered stream water to induce sporulation (Bärlocher,005). After 48 h, the suspension containing the spores was trans-erred to vials, and the samples were fixed with 2 mL of formalin.o count spores under the microscope, 0.1 mL of Triton (5%) wasdded to the samples, which were then filtered and stained with.1% cotton blue in lactophenol.

The total microbial biomass was evaluated by measuring ATPn the detritus, as described by Abelho (2005). Five frozen leaves

ere placed in 5 mL of 0.05 M HEPES and 1.2 N sulfuric acid con-aining 8 g/L of oxalic acid and were then milled and centrifuged.he supernatant was filtered, neutralized by NaOH, and frozen at20 ◦C. For the quantification of ATP, an aliquot of 20 �L of sam-le was removed, 130 �L of buffer by HEPES and 50 �L of Fireflynzyme (Sigma) were added, and the sample was measured in auminometer (Turner Biosystems Co.).

ata analysis

The decomposition rates were determined by adjusting the val-es for percentage mass loss by the negative exponential modelt = Wo × e−kt, where Wt is the mass loss remaining at time t

days), Wo is the initial weight and k is the decomposition coef-cient (Olson, 1963). These values were estimated by exponentialegression analysis.

We used a generalized linear models (GLMs) that were con-tructed and adjusted to a normal error distribution (function glm,ackage stats for R) to evaluate the differences in mass loss in theicrobial community and the chemical characteristics of the detri-

us. The models were subjected to residual analysis to verify modelompliance with the chosen error distribution (package RRJ for R).he data for mass loss and the concentrations of ergosterol, ATP

nd secondary and structural compounds (response variable) werenalyzed in terms of sampling period (time), leaf species, and thenteraction between these two factors (explanatory variables). Allhe models were analyzed using the normal distribution (link = log;

Soluble reactivephosphorus (mg/L P)

<0.015 –

test = F), and an ANOVA of these GLMs was performed. A paired t testwas performed to verify differences between the mean areas andmean lengths in the two leaf species. The analyses were performedin R v2.6.2 (R Development Core Team, 2008).

Results

Physical and chemical stream characteristics

During the study period, the pH values were slightly acid to alka-line, ranging from 6.1 to 8.9; the dissolved oxygen content was 8.1(±0.2) mg/L and the water velocity was 0.06 (±0.009) m/s. Gerald-inho stream showed high electrical conductivity and low nutrientconcentrations (Table 1).

Leaf breakdown

Both leaf species showed rapid mass loss during the first 3days of incubation (18% for B. concinna and 16% for B. dracun-culifolia; Fig. 1). A slight increase in mass was observed after the60th day of incubation for both species; therefore data after day60 were not included in decay rate estimates. The percentages ofmass remaining on the 60th day were 78.3% for B. concinna and62.3% for B. dracunculifolia. These results were significantly differ-ent between the species and periods of the experiment (Table 2).

Fig. 1. Mass loss (mean % ± standard error) in Baccharis concinna and B. dracunculi-folia leaves during breakdown in Geraldinho stream.

Page 4

134 E.A.C.C. Alvim et al. / Limnologica 51 (2015) 131–138

Table 2Generalized linear models to evaluate whether the mass loss, the microbial community, and chemical characteristics of detritus varied between the species Baccharis concinnaand B. dracunculifolia, as a function of incubation time. Deviance: sum of squares.

Response variable Explanatory variable DF Deviance Resid. DF Resid. deviance F p

Mass loss Null 66 5279Detritus 1 826 65 4452 58.3 <0.001Time 8 2453 57 1999 21.6 <0.001Detritus × time 8 1304 49 694 11.5 <0.001

Total polyphenols Null 65 4696Detritus 1 1729 64 2967 62.8 <0.001Time 8 1131 56 1835 5.1 <0.001Detritus × time 8 512 48 1322 2.3 0.034

Condensed tannin Null 64 34Detritus 1 14 63 19 115.9 <0.001Time 8 7 55 11 7.7 <0.001Detritus × time 8 6 47 5 6.1 <0.001

Cellulose Null 61 1159Detritus 1 135 60 1024 16.8 <0.001Time 8 512 52 511 7.9 <0.001Detritus × time 8 157 44 354 2.4 0.028

Lignin Null 61 1688Detritus 1 240 60 1448 15.9 <0.001Time 8 458 52 990 3.8 0.002Detritus × time 8 326 44 664 2.7 0.017

Ergosterol Null 61 4682035Detritus 1 177325 60 4504710 3.2 0.079Time 8 1756106 52 2748604 3.9 0.001Detritus × time 8 330759 44 2417845 0.7 0.645

Sporulation Null 62 4544Detritus 1 46. 61 4498 0.7 0.412Time 7 712 54 3786 1.5 0.185Detritus × time 7 630 47 3155 1.3 0.253

ATP Null 62 265139071926439388255710

P

fmh(d

twdcAw(5bo

TIac

Detritus 1 12Time 7 23Detritus × time 7 51

hysical and chemical detritus characteristics

The incubated leaves had a mean leaf area of 0.46 cm2 (±0.02)or B. concinna and 1.30 cm2 (±0.09) for B. dracunculifolia; these

eans were significantly different (t = −8.7; p < 0.001). The leavesad a mean length of 1.44 cm (±0.03) in B. concinna and 3.05 cm±0.09) in B. dracunculifolia; these means were also significantlyifferent (t = −14.4; p < 0.001).

The initial concentrations of total polyphenols and condensedannins differed significantly between the two species (Table 2) andere approximately 3 and 14 times higher in B. concinna than in B.

racunculifolia, respectively (Table 3). The loss of these secondaryompounds over time differed in the two detritus types (Table 2).fter only seven days of incubation, the loss of total polyphenolsas estimated at 59% for B. concinna and 37% for B. dracunculifolia

Fig. 2a). The loss of condensed tannins in the first three days was4% for B. concinna, and concentrations for B. dracunculifolia wereelow the detection limit after 3 days (Fig. 2b). The concentrationsf structural compounds also differed between the two types of

able 3nitial and final content (mean % ± standard error) of secondary (total polyphenolsnd condensed tannins) and structural (cellulose and lignin) compounds of Baccharisoncinna and B. dracunculifolia.

Baccharis concinna Baccharis dracunculifolia

Initial Final Initial Final

Total polyphenols 34.0 (9.4) 12.1 (1.7) 10.1 (1.9) 3.3 (0.5)Condensed tannins 2.3 (0.4) 0.1 (0.1) 0.1 (0.1) –Cellulose 22.9 (0.6) 22.4 (1.1) 15.8 (2.6) 13.0 (1.9)Lignin 24.0 (0.2) 31.2 (2.1) 34.9 (1.8) 24.8 (4.7)

1 61 252212640 3.4 0.0704 54 228273815 0.9 0.5079 47 176716706 1.9 0.081

detritus (Table 2). Initially, B. concinna contained more celluloseand less lignin than B. dracunculifolia (Fig. 2c and d). B. concinnacontained a higher proportion of both these compounds after 120days of decomposition, whereas B. dracunculifolia only had a higherproportion of cellulose (Table 3).

Leaf-associated microorganisms

The initial values of ergosterol were 689.9 �g/g AFDM for B.concinna and 774.9 �g/g AFDM for B. dracunculifolia, indicating con-siderable fungal colonization of the detritus before incubation inthe stream. The concentrations of ergosterol did not differ betweenthe leaf types, but there was a significant difference between theincubation times (Table 2).

During incubation, the initial concentration of ergosteroldecreased after seven days of incubation in both B. concinna andB. dracunculifolia (349.5 �g/g AFDM and 392.7 �g/g AFDM, respec-tively). By 60 days, there were variations in the increase anddecrease in the concentrations, followed by an increase in the con-centration of ergosterol, reaching 908.5 �g/g AFDM for B. concinnaand 908.4 �g/g AFDM for B. dracunculifolia at 120 days (Fig. 3A).The lowest measured ergosterol concentrations were 349.5 �g/gAFDM on day 7 for B. concinna and 371.4 �g/g AFDM on day 21 forB. dracunculifolia.

The peak of sporulation of aquatic hyphomycetes occurredon day 30 in B. concinna (17 spores mg/AFDM) and on day 21 in

B. dracunculifolia (19 spores mg/AFDM; Fig. 3B). The lowest valuefound for B. concinna occurred on day 7 (2 spores mg/AFDM), andthe lowest value found for B. dracunculifolia occurred on day 3(3 spores mg/AFDM). The total number of aquatic hyphomycete

Page 5

E.A.C.C. Alvim et al. / Limnologica 51 (2015) 131–138 135

F ean %i nd (d)

st

btttsd

D

B

aBabfet2ect

ig. 2. Chemical changes in concentrations of secondary and structural compounds (mn Geraldinho stream. (a) Total polyphenols; (b) condensed tannins; (c) cellulose; a

pores did not differ between the leaf types or among incubationimes (Table 2).

The ATP concentration (indicating of the total microbialiomass) ranged from 1054 to 4023 nmol/g in B. concinna and 774o 4278 nmol/g in B. dracunculifolia (Fig. 3C). The ATP content ofhe debris did not differ between the species or among the incuba-ion times (Table 2). The highest values were observed in the finaltages for both B. concinna and B. dracunculifolia (60th and 90thays, respectively).

iscussion

reakdown rates and characteristics of leaf detritus

The coefficients of decomposition found in this study indicated relatively slow decay rate for B. concinna (k < 0.0041), whereas. dracunculifolia decayed at a moderate rate (0.0041 > k < 0.0173)ccording to the classification of Gonc alves et al. (2013). The leafreakdown rates for the two species were similar to those foundor other species in tropical regions (Gonc alves et al., 2006; Charat al., 2007; Moretti et al., 2007) but lower compared to studies inemperate zones (Braatne et al., 2007; Baudoin et al., 2008; Abelho,

009). The highest breakdown rates found in streams of temperatecosystems may be related to lower concentrations of recalcitrantompounds in the detritus (Ardón et al., 2009). This shows thathe smaller leaf sizes associated with higher open canopies do

± standard error) in decomposing leaves of Baccharis concinna and B. dracunculifolia lignin.

not change the leaf breakdown rates compared to other Cerradosystems. However, the significant increase in mass after 60th dayshould be noted and will be discussed below.

The results of this study suggest that the breakdown rates wereaffected by differences in leaf detritus characteristics and chemicalcomposition. The decay rate of B. dracunculifolia may be related tolower concentrations of inhibitory compounds, which would facil-itate microbial colonization (Mathuriau and Chauvet, 2002). Thelower decay rate of B. concinna was related to higher concentrationsof secondary and structural compounds, which inhibit microbialcolonization and, consequently, retard the decomposition of thedetritus (Ostrofsky, 1997). However, the inhibitory effect of the sec-ondary compounds appeared to influence the decomposition untilday 15 in B. concinna; after this period, its effect tended to decreasedue to rapid leaching.

The concentrations of condensed tannins, which are anti-herbivory compounds that affect microbial colonization, of B.concinna and B. dracunculifolia were lower than in other tropicalspecies (Ardón and Pringle, 2008; Ardón et al., 2009). Accordingto studies on tropical species, secondary compounds, which arecompounds produced by plants to fight herbivory, do not stronglyaffect detritus decomposition due to rapid leaching (Ardón et al.,

2006; Ardón and Pringle, 2008). In this study, polyphenols wereleached in the first seven days and tannins in the first three days;these results support the findings of Ardón et al. (2006) and Ardónand Pringle (2008) and contrast the hypothesis of Stout (1989), who

Page 6

136 E.A.C.C. Alvim et al. / Limnolog

Fig. 3. Contribution of microorganisms in the detritus of Baccharis concinna andB. dracunculifolia during the leaf breakdown in Geraldinho stream. (A) Concentra-tion of ergosterol (mean and standard error); (B) total number of spores of aquatichyphomycetes (mean and standard error); and (C) concentration of ATP (mean andstandard error).

ica 51 (2015) 131–138

stated that secondary compounds decrease the decomposition rateof tropical species.

The leaves of B. dracunculifolia showed smaller initial propor-tions of cellulose and larger proportions of lignin compared to B.concinna. However, these high levels of lignin did not result in aslower decay rate of B. dracunculifolia, although the concentrationof this structural compound is considered a limiting factor in leafdecomposition (Ardón and Pringle, 2008). Despite their high initiallignin content, leaves of B. dracunculifolia have thin cuticles and asofter texture, which favored a more rapid breakdown when com-pared to B. concinna. In leaves of B. dracunculifolia, the proportion ofcellulose and lignin increased during the first month of the exper-iment, indicating that molecules other than structural compoundswere degrading.

In B. concinna, there were decreases in the proportion of struc-tural compounds from the 90th day of incubation; however, at theend of the experiment, this ratio increased, indicating an accu-mulation of these compounds or the use of other less-refractorycompounds, corroborating the results of Suberkropp et al. (1976).As lignin is resistant to enzymatic degradation, the higher propor-tion of this component in the leaf lowers the relative amount ofavailable carbon compounds (Gessner and Chauvet, 1994). Thus,lignin may be a reverse index of the availability of carbon forthe decomposers. Although cellulose is the major constituent ofcell walls and confers rigidity to plant tissues, certain species ofmicroorganisms can degrade it more easily (Tomme et al., 1995).

Contribution of microorganisms

The microbial community has an important role in the releaseof stored energy in detritus, mainly in tropical regions where inver-tebrate shredders are rare (Boyero et al., 2012). In this study,both species had higher initial concentrations of ergosterol, indi-cating natural leaf colonization by terrestrial fungi (endophyticand epiphytic fungi) (Nikolcheva and Bärlocher, 2005). The higherergosterol content found in the present study is attributed tothe associated terrestrial taxa, which are present in the phyllo-sphere or as endophytes, and the newly arrived conidia of aquatichyphomycetes that had settled on the leaf surface. In addition,the rates of decay for B. dracunculifolia and B. concinna were low.Traditional techniques for studying the fungal community compo-sition in streams favor the detection and identification of aquatichyphomycetes, and the use of molecular techniques can determinethe presence and contributions of other fungal groups (Nikolchevaand Bärlocher, 2004). The results presented in the aforementionedstudy raise the possibility that during leaf decomposition, fungiother than aquatic hyphomycetes may substantially contribute toleaf decomposition in streams, such as Chytridiomycota, Oomycotaand Zygomycota. According to Leroy et al. (2011), endophytic fungican sequester nutrients in leaf tissue for their own use, and thisfungal mycelial tissue may have an inhibitory effect on bacterial orfungal decomposers in the stream and could explain the reduceddecomposition rates found in our study. During the leaf litter incu-bation, the ergosterol concentration oscillated, most likely due tothe succession of the fungi community. After 60 days, the concen-tration of ergosterol increased, which it can be explained by thecolonization of aquatic hyphomycetes (Bärlocher and Grac a, 2002;Grac a and Canhoto, 2006). Moreover, we observed the increasein leaf mass which it has been discussed in the literature as aconsequence of microbial biofilms growing on the leaf surface andmasking the leaf mass loss (Sales et al., 2014).

The production of spores by the aquatic hyphomycetes was

also high when compared to other studies performed in tropi-cal (Gonc alves et al., 2007) and neotropical streams (Mathuriauand Chauvet, 2002) and mainly occurred at the beginning ofthe breakdown process when resources tended to be more

Page 7

nolog

atclpoVhdoodrth

spcwtimmsufftmgbtn

das2cptp

c(tfltl

dSftciaardtgs2

E.A.C.C. Alvim et al. / Lim

vailable. In contrast, Bärlocher et al. (2010) found values of upo 500 spores/mg/d on leaves exposed in the Republic of Panama,ompared to a maximum of 17 spores/mg/d in our study. Duringeaf decomposition, sporulation declines, leading to a temporalattern of the production of spores, with peaks at the beginningf the process related to the quality of the detritus (Gessner andan Ryckegem, 2003; Pérez et al., 2012). This suggests that aquaticyphomycetes dominate the fungal community early in the break-own process and are a more reliable indicator of the performancef this group in the decomposition process than the concentrationf ergosterol, which represents all groups of fungi colonizing theetritus. Furthermore, in this study, the concentration of ergoste-ol was more stable and then increased between 30 and 90 days;his increase could be associated with fungi other than aquaticyphomycetes.

Our results for the fungal biomass dynamics agree with othertudies and confirmed that fungi play a key role in the decom-osition of leaf litter (Gonc alves et al., 2007; Li et al., 2009). Thisonclusion is based on the high estimated fungal biomass, whichas higher than reports in the literature to both temperate and

ropical streams (Abelho, 2001; Ardón et al., 2006). These resultsndicate that the production of fungi is higher in tropical environ-

ents. The growth of fungal biomass is similar to the sequence ofass loss. These high values can be related to the size of the leaf

pecies utilized in the present study because the entire leaf wassed for the analysis of ergosterol, allowing us to estimate the entireungal community present in the leaves. This procedure differsrom the sampling of an aggregate distribution of the communityhat results when leaf discs are taken from leaves (the standard

ethodology for this type of study). Furthermore, due to their rapidrowth in branched filaments, fungi can rapidly absorb nutrientsy increasing the surface/volume ratio of the debris, facilitating theranslocation of internal substances such as organic substances andutrients (Gessner and Van Ryckegem, 2003).

In our study, the proportion of lignin decreased after the 90thay, suggesting that this fungal community may include speciesble to produce enzymes that degrade this compound. As shown ineveral studies (Gessner and Van Ryckegem, 2003; Krauss et al.,011; Sridhar and Sudheep, 2011), the changes in physical andhemical characteristics of the detritus during the decompositionrocess may have changed the abundance of fungal species inhe community, with fungi capable of degrading recalcitrant com-ounds becoming more abundant.

The maximum ATP concentration found in this study was highompared to other studies in streams located in the same areaGonc alves et al., 2006, 2007). The leaf mass loss was proportionalo the microbial colonization, and this relationship was observedor B. dracunculifolia, particularly on the 60th day. The use of smalleaves most likely increased the possibility of microbial coloniza-ion, facilitating bacterial access to the resources available in theeaves.

Fungi and bacteria interact synergistically and antagonisticallyuring leaf breakdown (Gulis and Suberkropp, 2003b; Sridhar andudheep, 2011). In the present study, the highest values of ATP wereound in the final stages of decomposition, although with fluctua-ions over time and higher concentrations of ergosterol. This ATPoncentration could indicate that the microbial community assim-lated organic compounds released from leaves by the enzymaticction of fungi, as found by Gulis and Suberkropp (2003a). Gulisnd Suberkropp (2003b) suggested that fungi may control bacte-ial activity because the competition for food and bacteria is partlyependent on the fungi, as fungi increase the range of coloniza-

ion and resources available for the bacteria. Therefore, bacteriarow best in conjunction with fungi rather than in their absence,howing a high activity in the presence of fungi (Romaní et al.,006).

ica 51 (2015) 131–138 137

ATP concentration increases without corresponding increasein the concentration of ergosterol, this suggests that microorgan-isms other (we suggest that could be biofilm because mucilage andgreen colors observed during washing leaves) than fungi are accu-mulating during leaf breakdown (Gonc alves et al., 2006, 2007).The importance of biofilm growth on submerged leaves in tropi-cal streams without vegetation cover has been neglected, but theseorganisms could significantly enrich the nutritional value of leaves.The microbial community present in the biofilm varied in relationto resource availability, and its enzymatic activity allows the opti-mal allocation of resources, increasing the microbial growth rate(Sinsabaugh et al., 2010). Weight gain during leaf breakdown maybe caused by the growth of biofilm on the surface of the detri-tus, as observed for Syzygium cordatum (Myrtaceae) by Mathookoet al. (2000). In a laboratory study in a temperate region, the expo-sure to light increased the quality and quantity of biofilm (Frankenet al., 2005). Despite the observed high values of ergosterol andATP, decomposition did not occur as expected, most likely due tothe inactivity of fungi that may have benefitted from the biofilmby taking up some of their nutrients. This would indicate that aperiod of biofilm production occurs and that the leaves are only asubstrate; after a certain amount of time this relationship would bedisrupted, and the leaves would resume their decomposition.

Conclusions

Our results partly confirmed our hypothesis that small leafsizes increase the litter decomposition rates and rejected ourhypothesis that the chemical composition of the detritus affectedmicrobial colonization because the secondary compounds wereleached in the first days of incubation and did not directly influ-ence decomposition in this tropical stream. However, we foundthat the concentration of structural compounds controlled thisprocess, and these compounds may be the main carbon sourcefor the detritus-associated microorganisms. The relationship of anincrease in fungal biomass and a decrease in the proportion of ligninindicates that the fungal community associated with leaf detritusconsisted of species that produced extracellular enzymes able todegrade lignin. The increase in weight after the 60th day of incu-bation may be related to biofilm growth, which was favored bythe direct entry of light into the stream. These data are the firstfor these two plant species, which are key species in the structureof rocky field vegetation. Furthermore, this study was fundamen-tal to understanding the flow of energy and the maintenance ofmetabolism in high-altitudinal streams.

Acknowledgements

We thank the members of the UFMG Benthos Ecology Lab-oratory for help in the field and some laboratory analyses.The first author received a scholarship from CNPq during herMaster’s research. This research was financially supported byFAPEMIG/PRONEX (Edital 20/2006; proc. 465/07) through theproject “Ecological and Climatic Dimensions of Biodiversity in Bac-charis: from molecules to organisms” coordinated by Dr. G.W.Fernandes.

References

Abelho, M., 2001. From litterfall to breakdown in stream: a review. Sci. World 1,658–680.

Abelho, M., 2005. Extraction and quantification of ATP as a measure of microbialbiomass. In: Grac a, M.A.S., Bärlocher, F., Gessner, M.O. (Eds.), Methods to Study

Litter Decomposition: A Practical Guide. Springer, Dordrecht, pp. 223–229.

Abelho, M., 2009. Leaf-litter mixtures affect breakdown and macroinvertebrate col-onization rates in a stream ecosystem. Int. Rev. Hydrobiol. 94, 436–451.

Allan, J.D., Castillo, M.M., 2007. Stream Ecology: Structure and Function of RunningWaters. Springer, Dordrecht.

Page 8

1 nolog

A

A

A

A

B

B

B

B

B

B

B

B

C

C

C

D

E

F

F

G

G

G

G

G

G

G

G

38 E.A.C.C. Alvim et al. / Lim

PHA, 2005. Standard Methods for the Examination of Water and Wastewater, 18thed. American Public Health Association, Washington, DC.

rdón, M., Pringle, C.M., 2008. Do secondary compounds inhibit microbial- andinsect-mediated leaf breakdown in a tropical rainforest stream, Costa Rica?Oecologia 155, 311–323.

rdón, M., Pringle, C.M., Eggert, S.L., 2009. Does leaf chemistry differentially affectbreakdown in tropical vs temperate streams? Importance of standardized ana-lytical techniques to measure leaf chemistry. J. North Am. Benthol. Soc. 28,440–453.

rdón, M., Stallcup, L.A., Pringle, C.M., 2006. Does leaf quality mediate the stimula-tion of leaf breakdown by phosphorus in Neotropical streams? Freshwater Biol.51, 618–633.

ärlocher, F., 2005. Sporulation by aquatic hyphomycetes. In: Grac a, M.A.S., Bär-locher, F., Gessner, M.O. (Eds.), Methods to Study Litter Decomposition: APractical Guide. Springer, Dordrecht, pp. 185–187.

ärlocher, F., Graca, M.A.S., 2002. Exotic riparian vegetation lowers fungal diver-sity but not leaf decomposition in Portuguese streams. Freshwater Biol. 47,1123–1135.

ärlocher, F., Grac a, M.A.S., 2005. Total phenolics. In: Grac a, M.A.S., Bärlocher, F.,Gessner, M.O. (Eds.), Methods to Study Litter Decomposition: A Practical Guide.Springer, Dordrecht, pp. 97–100.

ärlocher, F., Helson, J.E., Williams, D.D., 2010. Aquatic hyphomycete communitiesacross a land-use gradient of Panamanian streams. Fundam. Appl. Limnol. 177/3,209–221.

audoin, J.M., Guérold, F., Feltren, V., Chauvet, E., Wagner, P., Rousselle, P., 2008. Ele-vated aluminium concentration in acidified headwater streams lower aquatichyphomycete diversity and impairs leaf-litter breakdown. Microb. Ecol. 56,260–269.

oyero, L., Pearson, R.G., Dudgeon, D., Ferreira, V., Grac a, M.A.S., Gessner, M.O., Boul-ton, A.J., Chauvet, E., Yule, C.M., Albarino, R.J., Ramírez, A., Helson, J.E., Callisto,M., Arunachalam, M., Chará, J., Figueroa, R., Mathooko, J.M., Gonc alves Jr., J.F.,Moretti, M.S., Chará-Serna, A.M., Davies, J.N., Encalada, A., Lamothe, S., Buria,L.M., Castela, J., Comejo, A., Li, A.O.Y., M’Erimba, C., Villanueva, V.D., Zúniga,M.C., Swan, C.M., Barmuta, L.A., 2012. Global patterns of stream detritivore dis-tribution: implications for biodiversity loss in changing climates. Global Ecol.Biogeogr. 21, 134–141.

raatne, J.H., Sullivan, S.M.P., Chamberlain, R., 2007. Leaf decomposition andstream macroinvertebrate colonization of Japanese Knotweed, an invasive plantspecies. Int. Rev. Hydrobiol. 92, 656–665.

unn, S.E., Davies, P.M., Winning, M., 2003. Sources of organic carbon supportingthe food web of an arid zone floodplain river. Freshwater Biol. 48, 1–17.

anhoto, C., Grac a, M.A.S., 1999. Leaf barriers to fungal colonization and shredders(Tipula lateralis) consumption of decomposing Eucalyptus globules. Microb. Ecol.37, 163–172.

armouze, J.P., 1994. O Metabolismo dos Ecossistemas Aquáticos: Fundamen-tos Teóricos, Métodos de Estudo e Análises Químicas. Editora EdgardBlücher/FAPESP, São Paulo.

hara, J., Baird, D., Telfer, T., Giraldo, L., 2007. A comparative study of leaf break-down of three native tree species in a slowly-flowing headwater stream in theColombian Andes. Int. Rev. Hydrobiol. 92, 183–198.

as, M., Royer, T.V., Leff, L.G., 2008. Fungal communities on decaying leaves instreams: a comparison of two leaf species. Mycol. Prog. 7, 267–275.

steves, F.A., Gonc alves Jr., J.F., 2011. Etapas do metabolismo aquático. In:Esteves, F.A. (Ed.), Fundamentos de Limnologia. Interciência, Rio de Janeiro,pp. 119–124.

ranc a, J.S., Gregório, R.S., Paula, J.D., Gonc alves Jr., J.F., Ferreira, F.A., Callisto, M.,2009. Composition and dynamics of allochthonous organic matter inputs andbenthic stock in a Brazilian stream. Mar. Freshwater Res. 60, 990–998.

ranken, R.J.M., Peeters, E.T.H.M., Gardeniers, J.J.P., Beijer, J.A.J., Scheffer, M., 2005.Growth of shredders on leaf litter biofilms: the effect of light intensity. Fresh-water Biol. 50, 459–466.

essner, M.O., 2005a. Ergosterol as a measure of fungal biomass. In: Grac a, M.A.S.,Bärlocher, F., Gessner, M.O. (Eds.), Methods to Study Litter Decomposition: APractical Guide. Springer, Dordrecht, pp. 189–195.

essner, M.O., 2005b. Proximate lignin and cellulose. In: Grac a, M.A.S., Bärlocher, F.,Gessner, M.O. (Eds.), Methods to Study Litter Decomposition: A Practical Guide.Springer, Dordrecht, pp. 115–120.

essner, M.O., Chauvet, E., 1994. Importance of stream microfungi in controllingbreakdown rates of leaf litter. Ecology 75, 1807–1817.

essner, M.O., Chauvet, E., Dobson, M., 1999. A perspective on leaf litter breakdownin streams. Oikos 85, 377–384.

essner, M.O., Van Ryckegem, C., 2003. Water fungi as decomposers in freshwaterecosystems. In: Bitton, G. (Ed.), Encyclopedia of Environmental Microbiology.Wiley, New York, NY, http://dx.doi.org/10.1002/0471263397.env314 (onlineed.).

omi, T., Sidle, R.C., Richardson, J.S., 2002. Understanding processes and downstreamlinkages of headwater systems. BioScience 52, 905–916.

onc alves J.F.JJr., Franc a, J.S., Medeiros, A.O., Rosa, C.A., Callisto, M., 2006. Leaf break-down in a tropical stream. Int. Rev. Hydrobiol. 91, 164–177.

onc alves Jr., J.F., Grac a, M.A.S., Callisto, M., 2007. Litter decomposition in a Cerradosavannah stream is retarded by leaf toughness, low dissolved nutrients and alow density of shredders. Freshwater Biol. 52, 1440–1451.

ica 51 (2015) 131–138

Gonc alves Jr., J.F., Martins, R.T., Ottoni, B.M.P., Couceiro, S.E.M., 2013. Uma visãosobre a decomposic ão foliar em sistemas aquáticos brasileiros. In: Hamada, N.,Nessimian, J.L., Querino, R.B. (Eds.), Insetos Aquáticos na Amazônia brasileira:taxonomia, biologia e ecologia. Editora do INPA, Manaus, pp. 89–116.

Grac a, M.A.S., Bärlocher, F., 2005. Radial diffusion assay for tannins. In: Grac a, M.A.S.,Bärlocher, F., Gessner, M.O. (Eds.), Methods to Study Litter Decomposition: APractical Guide. Springer, Dordrecht, pp. 101–108.

Graca, M.A.S., Canhoto, C., 2006. Leaf litter processing in low orders streams. Lim-nética 25, 1–10.

Gulis, V., Suberkropp, K., 2003a. Effect of inorganic nutrients on relative contrib-utions of fungi and bacteria to carbon flow from submerged decomposing leaflitter. Microb. Ecol. 45, 11–19.

Gulis, V., Suberkropp, K., 2003b. Interactions between stream fungi and bacteriaassociated with decomposing leaf litter ad different levels of nutrient availabil-ity. Aquat. Microb. Ecol. 30, 149–157.

Komínková, D., Kuehn, K.A., Büsing, N., Steiner, D., Gessner, M.O., 2000. Microbialbiomass, growth, and respiration associated with submerged litter of Phragmitesaustralis decomposing in a littoral reed stand of a large lake. Aquat. Microb. Ecol.22, 271–282.

Krauss, G.-J., Solé, M., Krauss, G., Schlosser, D., Wesenberg, D., Bärlocher, F., 2011.Fungi in freshwaters: ecology, physiology and biochemical potential. FEMSMicrobiol. Rev. 35, 620–651.

Leroy, C.J., Fischer, D.G., Halstead, K., Pryor, M., Bailey, J.K., Schweitzer, J.A., 2011.A fungal endophyte slows litter decomposition in streams. Freshwater Biol. 56,1426–1433.

Li, A.O.Y., Lily, C.Y., Dudgeon, D., 2009. Effects of leaf toughness and nitrogen contenton litter breakdown and macroinvertebrates in a tropical stream. Aquat. Sci. 71,80–93.

Mathooko, J.M., Magana, A.M., Nyang’au, I.M., 2000. Decomposition of Syzygiumcordatum in a Rift Valley stream ecosystem. Afr. J. Ecol. 38, 365–368.

Mathuriau, C., Chauvet, E., 2002. Breakdown of leaf litter in a neotropical stream. J.North Am. Benthol. Soc. 21, 384–396.

Moretti, M., Gonc alves Jr., J.F., Callisto, M., 2007. Leaf breakdown in two tropicalstreams: differences between single and mixed species packs. Limnologica 37,250–258.

Nikolcheva, L.G., Bärlocher, F., 2004. Taxon-specific fungal primers reveal unex-pectedly high diversity during leaf decomposition in a stream. Mycol. Prog. 3,41–49.

Nikolcheva, L.G., Bärlocher, F., 2005. Seasonal and substrate preferences of fungicolonizing leaves in streams: traditional versus molecular evidence. Environ.Microbiol. 7, 270–280.

Oliveira-Filho, A.T., Ratter, J.A., 2002. Vegetation physiognomies and woody flora ofthe Cerrado Biome. In: Oliveira, P.S., Marquis, R.J. (Eds.), The Cerrados of Brazil:Ecology and Natural History of a Neotropical Savanna. Columbia UniversityPress, New York, NY, pp. 91–120.

Olson, J.S., 1963. Energy storage and the balance of producers and decomposers inecological systems. Ecology 44, 321–331.

Ostrofsky, M.L., 1997. Relationship between chemical characteristics of autumn-shed leaves and aquatic processing rates. J. North Am. Benthol. Soc. 16, 750–759.

Pérez, J., Descals, E., Pozo, J., 2012. Aquatic hyphomycete communities associatedwith decomposing alder leaf litter in reference headwater streams of the BasqueCountry (northern Spain). Microb. Ecol. 64, 279–290.

Pozo, J., González, E., Díez, J.R., Molinero, J., Elósegi, A., 1997. Inputs of particu-late organic matter to streams with different Riparian Vegetation. J. North Am.Benthol. Soc. 16, 602–611.

R Development Core Team, 2008. R: A Language and Environment for StatisticalComputing. R Foundation for Statistical Computing, Vienna, Austria, ISBN 3-900051-07-0, Available at 〈http://www.r-project.org〉.

Rasband, W.S., 2006. ImageJ. U.S. National Institutes of Health, Bethesda, MD,〈http://rsb.info.nih.gov/ij〉.

Raven, P.H., Evert, R.F., Eichhorn, S.E., 2007. Biologia Vegetal. Guanabara Koogan, Riode Janeiro.

Romaní, A.M., Fischer, H., Mille-Lindblom, C., Tranvik, L.J., 2006. Interactions of bacte-ria and fungi on decomposing litter: differential extracellular enzyme activities.Ecology 87, 2559–2569.

Sales, M.A., Gonc alves Jr., J.F., Dahora, J.S., Medeiros, A.O., 2014. Influence of leaf qual-ity in microbial decomposition in a headwater stream in the Brazilian Cerrado:a one-year study. Microb. Ecol., http://dx.doi.org/10.1007/s00248-014-0467-5.

Sinsabaugh, R.L., Van Horn, D.J., Shah, J.J.F., Findlay, S., 2010. Ecoenzymatic stoi-chiometry in relation to productivity for freshwater biofilm and Planktoncommunities. Microb. Ecol. 60, 885–893.

Sridhar, K.R., Sudheep, N.M., 2011. The spatial distribution of fungi on decomposingwoody litter in a freshwater stream, Western Ghats, India. Microb. Ecol. 61,635–645.

Stout, T.J., 1989. Effects of condensed tannins on leaf processing in mid-latitude andtropical streams: a theoretical approach. Can. J. Fish. Aquat. Sci. 46, 1097–1106.

Suberkropp, K., Godshalk, G.L., Klug, M.J., 1976. Changes in the chemical composition

of leaves during processing in a woodland stream. Ecology 57, 720–727.

Tomme, P., Warren, R.A.J., Gilkes, N.R., 1995. Cellulose hydrolysis by bacteria andfungi. Adv. Microb. Physiol. 37, 1–81.

Vannote, R.L., Minshall, G.W., Cummins, K.W., Sedell, J.R., Cushing, C.E., 1980. Theriver continuum concept. Can. J. Fish. Aquat. Sci. 37, 130–137.