The environmental impact of anthropogenic disturbances such as agriculture, on the
soil ecosystem, and particularly on soil microbial structural and functional diversity,
is of great importance to soil health, conservation and remediation. Therefore, this
study assessed the effects of various land use and management practices on both the
structural (genetic) and functional (catabolic) diversity of the soil bacterial and fungal
communities, at two long-term sites in KwaZulu-Natal. The first site is situated at
Baynesfield Estate, and the second at Mount Edgecombe Sugarcane Research
Institute. At site 1, the land uses investigated included soils under pre-harvest burnt
sugarcane (Saccharum officinarum, Linn.) (SC); maize (Zea mays, Linn.) under
conventional tillage (M); permanent kikuyu (Pennisetum clandestinum, Chiov)
pasture (KIK); pine (Pinus patula, Schiede) plantation (PF); and wattle (Acacia
mearnsii, De Wild) plantation (W), all fertilized; and undisturbed native grassland
(NAT) that had never been cultivated or fertilized. At site 2, a sugarcane (Saccharum
officinarum × S. spontaneum var. N27) pre-harvest burning and crop residue retention
trial was investigated. The treatments studied included conventional pre-harvest
burning of sugarcane with the tops removed (Bto), and green cane harvesting with
retention of crop residues on the soil surface as a trash blanket (T). Each of these
treatments was either fertilized (F) or unfertilized (Fo).
The polymerase chain reaction (PCR), followed by denaturing gradient gel
electrophoresis (DGGE) were used to determine the structural diversity, and
community level physiological profiling (CLPP) using BIOLOG plates, the catabolic
diversity. In addition, the soils were analysed with respect to selected
physicochemical variables, and the effects of these on the soil microbial communities
were determined. Replicate soil samples (0–5 cm) were randomly collected from three
independent locations within each land use and management, at both sites. Soil
suspensions for the CLPP analyses were prepared from fresh soil subsamples (within
24 h of collection) for the bacterial community analyses, and from 8-day-old soil
subsamples (incubated at 4°C to allow for spore germination) for the fungal
community analyses. BIOLOG EcoPlates™ were used for the bacterial CLPP study
and SF-N2 MicroPlates™ for the fungal analysis, the protocols being adapted and optimized for local conditions. This data was log [X+1]-transformed and analysed by
principal component analysis (PCA) and redundancy analysis (RDA). For PCRDGGE,
total genomic DNA was isolated directly from each soil subsample, and
purified using the MO BIO UltraClean™ soil DNA Isolation kit. Protocols were
developed and optimized, and fragments of 16S rDNA from soil bacterial
communities were PCR-amplified, using the universal bacterial primer pair
341fGC/534r. Different size 18S rDNA sequences were amplified from soil fungal
communities, using the universal fungus-specific primer pairs NS1/FR1GC and
FF390/FR1GC. Amplicons from both the bacterial and fungal communities were
fingerprinted by DGGE, and bands in the fungal DGGE gels were excised and
sequenced. The DGGE profiles were analysed by Bio-Rad Quantity One™ Image
analysis software, with respect to band number, position, and relative intensity.
Statistical analyses of this data then followed.
Soil properties [organic C; pH (KCl); exchangeable acidity; total cations (ECEC);
exchangeable K, Ca and Mg; and extractable P] were determined by PCA and were
shown to have affected the structural and catabolic diversity of the resident microbial
communities. At Baynesfield, canonical correspondence analysis (CCA) relating the
selected soil variables to bacterial community structural diversity, indicated that
ECEC, K, P and acidity were correlated with CCA1, accounting for 33.3% of the
variance, whereas Mg and organic C were correlated with CCA2 and accounted for
22.9% of the variance. In the fungal structural diversity study, pH was correlated with
CCA1, accounting for 43.8% of the variance, whereas P, ECEC and organic C were
correlated with CCA2, and accounted for 30.4% of the variance. The RDA of the
catabolic diversity data showed that the same soil variables affecting fungal structural
diversity (organic C, P, ECEC and pH) had influenced both the bacterial and fungal
catabolic diversity. In both the bacterial and fungal RDAs, organic C, P and ECEC
were aligned with RDA1, and pH with RDA2. However in the bacterial analysis,
RDA1 accounted for 46.0%, and RDA2 for 27.5% of the variance, whereas in the
fungal RDA, RDA1 accounted for only 21.7%, and RDA2 for only 15.0% of the
variance.
The higher extractable P and exchangeable K concentrations under SC and M, were
important in differentiating the structural diversity of these soil bacterial and fungal communities from those under the other land uses. High P concentrations under M
were also associated with bacterial catabolic diversity and to a lesser extent with that
of the soil fungal communities under M. Similarly, the higher organic C and
exchangeable Mg concentrations under KIK and NAT, possibly contributed to the
differentiation of these soil bacterial and fungal communities from those under the
other land uses, whereas under PF, the high exchangeable acidity and low pH were
possibly influencing factors. Under W, low concentrations of P and K were noted.
Other factors, such as the presence/absence and frequency of tillage and irrigation,
and the diversity of organic inputs due to the diversity of the above-ground plant
community, (in NAT, for example) were considered potentially important influences
on the nature and diversity of the various land use bacterial and fungal communities.
At Mount Edgecombe, CCA showed that organic C and Mg had a significant effect
on soil bacterial structural diversity. Organic C was closely correlated with CCA1,
accounting for 58.7% of the variance, whereas Mg was associated with CCA2, and
accounted for 41.3% of the variance. In the fungal structural diversity study, ECEC
and pH were strongly correlated with CCA1 and accounted for 49.1% of the variance,
while organic C was associated with CCA2, accounting for 29.6% of the variance. In
the functional diversity studies, RDA showed that both bacterial and fungal
community catabolic diversity was influenced by soil organic C, pH, and ECEC. In
the bacterial analysis, RDA1 was associated with organic C and pH, and accounted
for 43.1% of the variance, whereas ECEC was correlated with RDA2, accounting for
36.9% of the variance. In the fungal analysis, RDA1 was correlated with ECEC and
accounted for 47.1% of the variance, while RDA2 was associated with pH and
organic C, accounting for 35.8% of the variance. The retention of sugarcane harvest
residues on the soil surface in the trashed treatments caused an accumulation of
organic matter in the surface soil, which did not occur in the pre-harvest burnt
sugarcane. This difference in organic C content was a factor in differentiating both
bacterial and fungal communities between the trashed and the burnt treatments. Soil
acidification under long-term N fertilizer applications caused an increase in
exchangeable acidity and a loss of exchangeable Mg and Ca. Thus, as shown by CCA,
a considerably lower exchangeable Mg concentration under F compared to Fo plots
resulted, which was influential in differentiating the bacterial and fungal communities
under these two treatments. In the structural diversity study at Baynesfield, differences were found in bacterial
community species richness and diversity but not in evenness, whereas in the fungal
analysis, differences in community species richness, evenness and diversity were
shown. The soil bacterial and fungal communities associated with each land use were
clearly differentiated. Trends for bacterial and fungal diversity followed the same
order, namely: M < SC < KIK < NAT < PF < W. At Mount Edgecombe, no
significant difference (p > 0.05) in bacterial structural diversity was found with oneway
analysis of variance (ANOVA), but two-way ANOVA showed a slight
significant difference in bacterial community species richness (p = 0.05), as an effect
of fertilizer applications. A significant difference in fungal species richness (p = 0.02)
as a result of management effects was detected, with the highest values recorded for
the burnt/fertilized plots and the lowest for the burnt/unfertilized treatments. No
significant difference was shown in species evenness, or diversity (p > 0.05), in either
the bacterial or the fungal communities.
In the catabolic diversity study at site 1, the non-parametric Kruskal-Wallis ANOVA
showed that land use had not affected bacterial catabolic richness, evenness, or
diversity. In contrast, while fungal catabolic richness had not been affected by land
use, the soil fungal community catabolic evenness and diversity had. At site 2, the
land treatments had a significant effect on soil bacterial community catabolic richness
(p = 0.046), but not on evenness (p = 0.74) or diversity (p = 0.135). In the fungal
study, land management had no significant effect on the catabolic richness (p =
0.706), evenness (p = 0.536) or diversity (p = 0.826).
It was concluded, that the microbial communities under the different land use and
trash management regimes had been successfully differentiated, using the optimized
protocols for the PCR-DGGE of 16S rDNA (bacteria) and 18S rDNA (fungi).
Sequencing bands produced in the 18S rDNA DGGE, enabled some of the soil fungal
communities to be identified. CLPP of the soil microbial communities using BIOLOG
plates showed that, on the basis of C substrate utilization, the soil bacterial and fungal
communities’ catabolic profiles differed markedly. Thus, it was shown that the
different land use and management practices had indeed influenced the structural and
catabolic diversity of both the bacterial and fungal populations in the soil. / Thesis (Ph.D.)-University of KwaZulu-Natal, Pietermaritzburg, 2011.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:ukzn/oai:http://researchspace.ukzn.ac.za:10413/7974 |
| Date | January 2011 |
| Creators | Wallis, Patricia Dawn. |
| Contributors | Titshall, Louis William., Hunter, Charles H., Morris, Craig Duncan. |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0032 seconds