natural environments is still poorly understood. Macromo- lecular carbon degradation is a synergistic result of many
populations interacting in consortia. In addition, fungal interactions with bacteria can have profound effects on
lignocellulase activities Lang et al., 1997. Thus, a broad approach that considers the components and activities of the
entire soil community may be productive. The analysis of microbial phospholipid fatty acids PLFAs is a broad
measure of microbial community composition. PLFA analysis can be used to estimate the relative sizes of fungal,
actinomycete, anaerobe, Gram 1 and Gram 2 commu- nities. We attempted to link PLFA biomarker data with
two functional measures, lignocellulase activities and substrate utilization capacities BIOLOG, of several soil
microbial communities under changing land use condi- tions.
An understanding of the relationships between the composition of the microbial community and its physiolo-
gical capacity is particularly important in agroecoystems. In these ecosystems, frequent soil disturbance, carbon and
nutrient alterations, and changes in the water and light regime can alter the composition of the microbial commu-
nity Kennedy and Smith, 1995; Bossio et al., 1998. We analyzed soils from five tropical forest sites surrounding
pineapple plantations of differing ages and stages within the management cycle. Our objectives were to assess the
composition of the soil microbial community and its func- tional capabilities across a conversion in land use from
tropical forest to cultivated soils and to answer two ques- tions. In what ways do these soils differ in microbial
biomass,
community composition,
and physiological
capabilities? Can lignocellulase activities be related to changes in community composition which occur with chan-
ging land use?
2. Materials and methods
2.1. Field sites The research was conducted in Tahiti 17830
S, 149850
W on the Society Islands of French Polynesia. The dominant plant species in our forest sites were Hibiscus
tillaceus , Nephrolepis and Gleichenia linearis, Psidium
guajava , Pistache, and Eugenia cumini. After clearcutting
the forest, the forest litter was burned and the plantations planted within one year. The same owner used similar
management practices on each plantation. Thirty thousand pineapple plants Bromilacea ananas comelsus var quentii
per hectare were grown for three to four years after which time the plants became unproductive and were chipped and
incorporated into the surface soil. Each plant was given 80 g fertilizer per year 30:10:30:10 — N:P:K:S. Precipitation is
1728 mm annually, mostly falling in winter, and mean annual temperature is 268C. The soil is a Tropeptic Haplus-
tol and the soil texture is a silt loam to sandy loam with ,5 clay.
Five tropical forest sites were selected randomly from the perimeter of the pineapple plantations. One 30 m transect
was placed in each forest site. Four plantations were selected that differed in age or stage within the management
cycle. The youngest was a six-year old plantation with three-year old pineapple plants. The second, a nine-year
old plantation was bare: the previous rotation of pineapple plant biomass was chipped and incorporated into the soil
within the preceding three months. The third, a twelve-year old plantation, was at the same stage within the management
cycle as the nine-year old plantation. The fourth, a fourteen- year old plantation had three-year old pineapple plants on it.
Thus, the six- and nine-year old plantations were considered the ‘young’ treatment. The twelve- and fourteen-year old
plantations were considered as the ‘old’ treatment. The six- and fourteen-year old plantations were considered as
the ‘mature plant’ treatment and the nine- and twelve-year old plantations were considered the ‘bare’ treatment. Three
30-m transects were established from the center of each plantation along a randomly determined compass heading.
Every three meters along each transect, soil cores 12.5 cm long × 4.5 cm diameter were driven into the soil with a
mallet. The following day, cores were pulled from the soil, wrapped in plastic, and placed in a cooler at t48C.
Soils were transported to the laboratory within 24 h of sampling. All soils were sampled in July 1998. Three
cores were randomly selected from each transect, bulked, and homogenized by mixing and removing roots and coarse
organic material. Analyses were performed on three composited samples per transect, except where indicated
n 15 for forest; n 9 per plantation.
2.2. Enzyme activities The enzymes assayed were b-1,4-glucosidase EC
3.2.1.21, cellobiohydrolase EC 3.2.1.91, b-xylosidase EC 3.2.1.37, phenol oxidase EC 1.10.3.2, peroxidase
EC 1.11.1.7 phosphomonoesterase EC 3.1.3.2 and sulphatase EC 3.1.6.1. The substrates for the b-glucosi-
dase, b-xylosidase, cellobiohydrolase, phosphatase, and sulphatase assays were p-nitrophenol pNP b-d-glucopyr-
anoside, pNP-b-xylopyranoside, pNP-cellobioside, pNP- phosphate, bis-pNP-phosphate, and pNP-sulphate, respec-
tively. Phenol oxidase substrate was 10 mM l-dihydroxy- phenyalanine DOPA. Peroxidase substrate was 10 mM l-
DOPA solution and 0.3 H
2
O
2
. All substrates were made in 50 mM pH 5.0 acetate buffer. Five grams of soil were added
to 50 ml 50 mM acetate buffer solution pH 5.0, briefly shaken by hand and stirred on a stir plate for 30 s. Fifty
m l of the mixture was added to each lane of a 96-well
microplate that contained 150 ml of 50 mM pNP-substrate solution or acetate buffer sample control. There were five
analytical replicates and eight sample controls of each extract. Plates were incubated at 278C for 2 h after
which time 50 ml of the solution in each well was trans- ferred to another 96-well plate containing 50 ml of 1.0 N
M.P. Waldrop et al. Soil Biology Biochemistry 32 2000 1837–1846 1838
NaOH to terminate the reaction. In the case of phenol oxidase and peroxidase, however, aliquots were transferred to empty
wells. Absorbance was measured at 410 nm for pNP-subtrates and at 469 nm for l-DOPA assays using a Spectramax plus
spectrophotometer Molecular Devices. Peroxidase activity was calculated as the difference between samples reacted with
and without H
2
O
2
Sinsabaugh et al., 1993. Units for phenol oxidase and peroxidase are in absorbance units h
21
g
21
dry soil. All other enzyme activities were expressed as mmol
substrate converted h
21
g
21
dry soil. The specific activities of the enzymes were calculated by dividing total enzyme
activities by the microbial biomass C determined by CFDE. 2.3. Microbial community composition
Phospholipid fatty acid analysis was conducted on 5 g samples of freeze dried soil using the procedure by White
and Ringelberg 1998. Extracted phospholipid samples were analyzed using a Hewlett Packard 6890 Gas Chroma-
tograph with a 25 m × 0.2 mm × 0.33 mm Ultra 2 5- phenyl-methylpolysiloxane column Hewlett Packard
using hydrogen as the carrier, nitrogen as the make up gas, and air to support the flame. The GC analyzed a 1 ml
injection with a 1:100 split, at an initial temperature of 1708C, ramped to 2608C at 28C min
21
at a constant flow rate of 0.4 ml per min, with a rapid increase to 3108C for
2 min at the end of each run to bake out the column. Run time was approximately 27 min per sample. Peaks were
identified using bacterial fatty acid standards and MIDI peak identification software MIDI, Inc., Newark, DE.
2.4. Substrate utilization patterns Substrate utilization patterns were measured using the
BIOLOGe Gram-negative microtiter plates BIOLOG, Inc.. Methods were similar to those of Bossio and Scow
1995. Twenty gram samples of field moist soil were shaken for 1 h in 200 ml 0.7 mm NaCl and aliquots
from a 10
23
dilution were pipetted into plates using an 8- channel micropipetter set to 150 ml
23
. Plates were incubated at 288C and read on a BIOLOGe Microplate Reader every
12 h. Color development in the control well due to utiliza- tion of background dissolved organic carbon was
subtracted from absorbance readings in all other wells. Negative values were set to zero. A time point was chosen
based on its average well color development AWCD to analyze for each plate Garland, 1996. Time points chosen
had AWCD values between 0.75 and 1.0. Prior to statistical analysis, individual well absorbance values were normal-
ized by total plate color to account for possible differences in inoculation densities between samples. These processed
data were used for principal components analysis.
2.5. Microbial biomass, nutrients, and pH Microbial biomass C and N were measured using the
chloroform fumigation and direct extraction CFDE tech- nique Vance et al., 1987. A conversion factor of 2.64 was
used to convert extracted C to biomass C. Available nutri- ents were determined using the Mehlich 3 extraction method
Mehlich, 1984. Extracts were then analyzed for P, S, Ca, Mg, and Mn on an Inductively Coupled Plasma Spectro-
photometer Thermo Jarrell Ash. Total C and N of oven dried soils were measured by combustion on a Carlo-Erba
NA 2100 carbon and nitrogen analyzer. For pH measure- ments, soil samples were mixed with an adequate amount of
deionized water to create a flowing paste. The paste was allowed to equilibrate for several minutes and the pH of
the paste was determined using an Accumet Basic pH meter Fischer Scientific.
2.6. Statistics Analysis of Variance ANOVA was performed using
Super ANOVA software Abacus Concepts. Tukey–Kramer
M.P. Waldrop et al. Soil Biology Biochemistry 32 2000 1837–1846 1839
Table 1 Effect of land use treatments on physical, chemical, and biological properties of soils
Variable Forest
Young plantation Old plantation
Bare Mature plant
u
m
‡
0.32
ax
0.29
b
0.27
c
0.28
y
0.28
y
pH
‡
4.57
ax
4.29
a
3.97
b
4.21
y
4.00
y
C
†
7.81
ax
3.52
b
3.60
b
3.52
y
3.60
y
N
†
0.40
ax
0.20
b
0.19
b
0.17
y
0.21
z
Ca
†
mg 10 g
21
7.0
ax
1.0
b
0.4
c
0.7
y
0.8
y
K
†
mg 10 g
21
1.2
ax
0.6
a
0.6
a
0.8
x
0.3
y
Mg
†
mg 10 g
21
5.3
ax
0.3
b
0.1
c
0.2
y
0.1
z
Mn
†
mg kg
21
6.5
ax
2.7
b
0.9
c
2.0
y
1.6
y
P
†
mg 100 g
21
4.8
ax
8.8
a
3.8
b
4.3
x
9.7
x
S
†
mg 100 g
21
1.3
ax
1.8
a
2.3
a
1.7
x
2.6
y
Zn
†
mg 10 kg
21
4.3
ax
9.8
a
5.6
a
9.1
x
6.8
x
Biomass
‡
mg g
21
855
ax
222
b
209
b
193
y
238
y
Labile C
‡
mg g
21
258
ax
198
a
188
b
171
y
216
x
u
m is the mass water content of soil. The ‘Bare’ treatment recently had mature plant biomass chipped and incorporated into the soil. The ‘Mature plant’ treatment had mature pineapple plants growing on the soil. Letters denote differences among treatments a,b,c among forest, young, and old; x,y,z among
forest, bare, and mature plant. Values are means
†
n 3 for forest and 6–9 for plantation treatments;
‡
n 14–15 for forest and plantation treatments.
highly significant difference HSD tests and Scheffe’s S tests in case of nonparametric data were performed as post-hoc
tests. Principal Components Analysis PCA was performed using JMP software SAS Institute. The first and second prin-
cipal components PC1 and PC2, respectively were used for further correlation analysis. The Spearman’s Rho correlation
analysis was used in case of nonparametric data. All effects noted were significant at the P , 0.05 level.
3. Results
