Soil is an important ecosystem that supports a wide variety of organisms such as bacteria, fungi, arthropods and nematodes. This sensitive ecosystem may be influenced by various factors, including agricultural management practices. With the introduction of genetically modified (GM) glyphosate-tolerant (RoundUp ® Ready: RR) crops, herbicides such as glyphosate have been increasingly used. However, little is known about the effect of glyphosate on the biological communities in these herbicide-sprayed soils. With the intimate proximity that microorganisms and nematodes have with the roots of plants, these organisms can be used to assess changes that may occur in the soil surrounding roots of RR crops. The aim of this study was to determine microbial community structure and nematode diversity, with emphasis on that of non-parasitic nematodes, in soil samples from conventional soybean (CS) - and RR- soybean fields compared to that in adjacent natural veld (NV) areas. Samples were collected from twenty three sites at six localities that are situated within the soybean-production areas of South Africa. These sites represented fields where RR and CS soybean grew, as well as surrounding NV. All RR fields have been treated with glyphosate for no less than five years. Microbial community structures of the twenty three sites in the RR, CS and NV ecosystems were determined by phospholipid fatty acid (PLFA) analyses. Nematode diversity was determined by extracting the nematodes from soil samples and conducting a faunal analysis. Soil physical and chemical properties were determined by an independent laboratory, Eco-Analytica (North West University, Potchefstroom) according to standard procedures. Results from this study indicated differences in microbial community structure between the various localities. However, there were no significant (p ≤ 0.05) differences in microbial community structures between RR- and CS ecosystems. Soils of both RR- and CS crops were primarily dominated by bacteria. Nematode identification and faunal analysis also indicated no significant (p ≤ 0.05) differences between the different non-parasitic/beneficial nematodes that were present in soils of these two ecosystems during the time of sampling. Non-parasitic nematode communities were primarily dominated by bacterivores. A faunal analysis indicated that most of the sites contained enriched, but unstructured soil food-webs. However, four of the sites showed enriched and structured food webs due to the presence of non-parasitic nematodes with high coloniser-persister (cp) values. Relationships between non-parasitic nematode – and microbial communities showed that there was a positive relationship between nematode functional groups and their corresponding microbial prey. From the results obtained in this study, it can be concluded that the community structures of both non-parasitic nematodes and microorganisms shared similarities. These community structures showed no long-term detrimental effects of glyphosate application in the soils surrounding roots of RR soybean crops. Relationships existed between non-parasitic nematode and microbial communities in the rhizosphere of soybean crops and natural veld. For example, bacterivore nematodes had a strong positive relationship with gram-negative bacteria. Similar but weaker relationships also existed between carnivores, omnivores, plantparasitic nematodes and gram-negative bacteria. A positive relationship also existed between fungivores and fungal fatty acids. This emphasises the value of these organisms as indicators of soil health and also the impact that agricultural practices can have on soils. / MSc (Environmental Sciences), North-West University, Potchefstroom Campus, 2014

Soil is an important ecosystem that supports a wide variety of organisms such as bacteria, fungi, arthropods and nematodes. This sensitive ecosystem may be influenced by various factors, including agricultural management practices. With the introduction of genetically modified (GM) glyphosate-tolerant (RoundUp ® Ready: RR) crops, herbicides such as glyphosate have been increasingly used. However, little is known about the effect of glyphosate on the biological communities in these herbicide-sprayed soils. With the intimate proximity that microorganisms and nematodes have with the roots of plants, these organisms can be used to assess changes that may occur in the soil surrounding roots of RR crops. The aim of this study was to determine microbial community structure and nematode diversity, with emphasis on that of non-parasitic nematodes, in soil samples from conventional soybean (CS) - and RR- soybean fields compared to that in adjacent natural veld (NV) areas. Samples were collected from twenty three sites at six localities that are situated within the soybean-production areas of South Africa. These sites represented fields where RR and CS soybean grew, as well as surrounding NV. All RR fields have been treated with glyphosate for no less than five years. Microbial community structures of the twenty three sites in the RR, CS and NV ecosystems were determined by phospholipid fatty acid (PLFA) analyses. Nematode diversity was determined by extracting the nematodes from soil samples and conducting a faunal analysis. Soil physical and chemical properties were determined by an independent laboratory, Eco-Analytica (North West University, Potchefstroom) according to standard procedures. Results from this study indicated differences in microbial community structure between the various localities. However, there were no significant (p ≤ 0.05) differences in microbial community structures between RR- and CS ecosystems. Soils of both RR- and CS crops were primarily dominated by bacteria. Nematode identification and faunal analysis also indicated no significant (p ≤ 0.05) differences between the different non-parasitic/beneficial nematodes that were present in soils of these two ecosystems during the time of sampling. Non-parasitic nematode communities were primarily dominated by bacterivores. A faunal analysis indicated that most of the sites contained enriched, but unstructured soil food-webs. However, four of the sites showed enriched and structured food webs due to the presence of non-parasitic nematodes with high coloniser-persister (cp) values. Relationships between non-parasitic nematode – and microbial communities showed that there was a positive relationship between nematode functional groups and their corresponding microbial prey. From the results obtained in this study, it can be concluded that the community structures of both non-parasitic nematodes and microorganisms shared similarities. These community structures showed no long-term detrimental effects of glyphosate application in the soils surrounding roots of RR soybean crops. Relationships existed between non-parasitic nematode and microbial communities in the rhizosphere of soybean crops and natural veld. For example, bacterivore nematodes had a strong positive relationship with gram-negative bacteria. Similar but weaker relationships also existed between carnivores, omnivores, plantparasitic nematodes and gram-negative bacteria. A positive relationship also existed between fungivores and fungal fatty acids. This emphasises the value of these organisms as indicators of soil health and also the impact that agricultural practices can have on soils. / MSc (Environmental Sciences), North-West University, Potchefstroom Campus, 2014

This work expands the knowledge on phospholipid metabolism in the kinetoplastid parasites: T. brucei, T. cruzi, Leishmania spp. that cause neglected tropical diseases and the related non-human pathogenic Crithidia fasiculata. As a close relative of parasitic kinetoplasts, specifically Leishmania, it is hypothesised that Crithidia fasiculata possesses a similar lipid biosynthetic capability and therefore represent an attractive model organism. Database mining the Crithidia genome revealed the ability to biosynthesise all of the main phospholipid species. Utilising various lipidomic techniques, a high level of an ω-6 18:3 fatty acid was observed, alongside an uncommon Δ19:0 fatty acid that was later identified to be exclusive attributed to PE species. Sphingolipid metabolism was shown to resemble that of Leishmania and T. cruzi, given the exclusive production of inositol-phosphoceramide species and no sphingomyelin species being observed. Using labelled precursors, Crithidia were seen to uptake and incorporate extracellular inositol into both phosphatidylinositol and inositol-phosphoceramide species. Crithidia were also shown to utilise both the Kennedy pathway and methylation of phosphatidylethanolamine to form phosphatidylcholine. The phospholipidome of T. cruzi revealed several phosphatidylserine species for the first time, suggesting a functional phosphatidylserine synthase. Current knowledge of T.cruzi sphingolipid biosynthesis was also confirmed as only inositol xxxi phosphoceramide species were observed. The identification and subsequent characterisation of novel phosphonolipid species are reported for the first time. Utilising lipidomic methodologies and labelled precursors, the relative contribution of the intracellular inositol pools within bloodstream and procyclic T. brucei towards PI biosynthesis was examined. This highlighted that the synthesis/turnover rates for specific phosphatidylinositol and inositol-phosphoceramide species are unequal. Efforts to optimise media conditions highlighted that under reduced levels of serum/glucose/inositol, bloodstream T. brucei unexpectedly adjusts its inositol metabolism. The procyclic parasite exemplifies this fact, as under inositol/glucose deficient media conditions they appear to have adapted to utilising glucogenesis and inositol de-novo synthesis. This work highlights that these parasites are rapidly dividing, their unique features of lipid metabolism may be exploitable for drug discovery purposes.

Recent declines in returns from primary forest products in New Zealand and projected increases in world food prices have led to the land-use conversion from plantation forest to pastoral farming in many lowland areas. After decades of forest cover the soils are in many cases less than adequate for pastoral farming, as they are acidic, with toxic levels of exchangeable aluminum, and contain low levels of available nitrogen (N), very high carbon (C):N ratio, and are devoid of earthworms and structural integrity. Overcoming the major site limitations of low soil pH and available N was a major priority and a field experiment was established in April 2005 to determine the impact of various rates of lime and N in relation to pasture establishment and production. Concerns about the short and long-term effects of these inputs on biological soil quality gave rise to the present study. The effects of land-use change and establishment inputs were assessed by comparison of selected treatment plots with two adjacent reference sites (long-term pasture and a 60–year Pinus radiata forest) on the same soil type. The effects of lime and N on soil biological quality were investigated under field and controlled environment conditions by determination of: microbial community structure (phospholipid fatty acids - PLFA), microbial biomass (total PLFA), and microbial activity (dehydrogenase activity). Soil physical (percentage water-stable aggregates) and chemical (pH, and total C and N) properties were also determined. Similarly, the effects of earthworm addition on soil biological properties were explored in a short-term glasshouse pot experiment. The role of earthworms as indicators of soil biological quality in the field was assumed by nematodes and these were assessed in field trial plots and the reference sites mentioned above. Land-use change and applications of lime and N contributed to changing the microbial community structure determined by principal component analysis of transformed PLFA data. However, the effect of lime was more pronounced in the field, while N contributed most to changing microbial community structure in the glasshouse. Mean microbial activity in the field increased from 4 µg dwt/hr without lime to 16 and 21 µg dwt/hr where lime was applied at 5 and 10 tons/hectare (t/ha), respectively. Mean microbial activity in the field was markedly higher (7-fold) than in the glasshouse at similar rates of lime. Lime application also increased soil moisture retention in the field, mean gravimetric soil moisture increased from 0.33 in control plots to 0.38 and 0.39 in plots treated with 5 and 10 t/ha lime, respectively. Lime application was associated with greater soil aggregate stability. Soils from test plots treated with 5 and 10 tons/ha lime had 45-50% water-stable aggregates compared to 34% in treatments without lime. After 16 weeks in pots, earthworm treatments increased mean plant dry matter (DM)/pot by at least 19% above the control. The increase was attributed primarily to greater N mineralization in the presence of earthworms. For the duration of the trial the earthworm species tested (Apporectodea caliginosa and Lumbricus rubellus, individually or combined) did not affect any of the measured soil microbial properties. However, the survival rate of A. caliginosa was 83% compared to 25% for L. rubellus. The control not receiving any lime or N and plots treated with 10t/ha lime and 200 kgN/ha had similar nematodes species composition, comprising 40% each of bacterial and fungal feeding nematodes. They differed markedly from the reference sites as the forest soil was dominated by plant associated species (38%) and the long-term pasture had 44% plant parasitic nematodes. Accordingly, the soil food web condition inferred from nematode faunal analysis characterized all test plots as basal, stressed and depleted, while the forest soil was categorized as highly structured and fungal dominated. The findings of this thesis demonstrated that land-use change from forest to pasture can have significant impacts on soil biological properties, earthworms can contribute to pasture productivity even in the short term, and nematode faunal analysis is a robust and reliable indicator of soil biological quality.

Recent declines in returns from primary forest products in New Zealand and projected increases in world food prices have led to the land-use conversion from plantation forest to pastoral farming in many lowland areas. After decades of forest cover the soils are in many cases less than adequate for pastoral farming, as they are acidic, with toxic levels of exchangeable aluminum, and contain low levels of available nitrogen (N), very high carbon (C):N ratio, and are devoid of earthworms and structural integrity. Overcoming the major site limitations of low soil pH and available N was a major priority and a field experiment was established in April 2005 to determine the impact of various rates of lime and N in relation to pasture establishment and production. Concerns about the short and long-term effects of these inputs on biological soil quality gave rise to the present study. The effects of land-use change and establishment inputs were assessed by comparison of selected treatment plots with two adjacent reference sites (long-term pasture and a 60–year Pinus radiata forest) on the same soil type. The effects of lime and N on soil biological quality were investigated under field and controlled environment conditions by determination of: microbial community structure (phospholipid fatty acids - PLFA), microbial biomass (total PLFA), and microbial activity (dehydrogenase activity). Soil physical (percentage water-stable aggregates) and chemical (pH, and total C and N) properties were also determined. Similarly, the effects of earthworm addition on soil biological properties were explored in a short-term glasshouse pot experiment. The role of earthworms as indicators of soil biological quality in the field was assumed by nematodes and these were assessed in field trial plots and the reference sites mentioned above. Land-use change and applications of lime and N contributed to changing the microbial community structure determined by principal component analysis of transformed PLFA data. However, the effect of lime was more pronounced in the field, while N contributed most to changing microbial community structure in the glasshouse. Mean microbial activity in the field increased from 4 µg dwt/hr without lime to 16 and 21 µg dwt/hr where lime was applied at 5 and 10 tons/hectare (t/ha), respectively. Mean microbial activity in the field was markedly higher (7-fold) than in the glasshouse at similar rates of lime. Lime application also increased soil moisture retention in the field, mean gravimetric soil moisture increased from 0.33 in control plots to 0.38 and 0.39 in plots treated with 5 and 10 t/ha lime, respectively. Lime application was associated with greater soil aggregate stability. Soils from test plots treated with 5 and 10 tons/ha lime had 45-50% water-stable aggregates compared to 34% in treatments without lime. After 16 weeks in pots, earthworm treatments increased mean plant dry matter (DM)/pot by at least 19% above the control. The increase was attributed primarily to greater N mineralization in the presence of earthworms. For the duration of the trial the earthworm species tested (Apporectodea caliginosa and Lumbricus rubellus, individually or combined) did not affect any of the measured soil microbial properties. However, the survival rate of A. caliginosa was 83% compared to 25% for L. rubellus. The control not receiving any lime or N and plots treated with 10t/ha lime and 200 kgN/ha had similar nematodes species composition, comprising 40% each of bacterial and fungal feeding nematodes. They differed markedly from the reference sites as the forest soil was dominated by plant associated species (38%) and the long-term pasture had 44% plant parasitic nematodes. Accordingly, the soil food web condition inferred from nematode faunal analysis characterized all test plots as basal, stressed and depleted, while the forest soil was categorized as highly structured and fungal dominated. The findings of this thesis demonstrated that land-use change from forest to pasture can have significant impacts on soil biological properties, earthworms can contribute to pasture productivity even in the short term, and nematode faunal analysis is a robust and reliable indicator of soil biological quality.

New Zealand’s 5.3 million strong dairy herd returns approximately 106 million litres of urine to pasture soils daily. The urea in that urine is rapidly hydrolysed to ammonium (NH₄⁺), which is then nitrified, with denitrification of nitrate (NO₃⁻) ensuing. Nitrous oxide (N₂O), a potent greenhouse gas (GHG), is produced via nitrification and denitrification, which are enzyme-catalysed processes mediated by soil microbes. Thus microbes are linked intrinsically to urine patch chemistry. However, few previous studies have investigated microbial dynamics in urine patches. Therefore the objective of these four experiments was to investigate the effects on soil microbial communities of cow urine deposition. Methods used included phospholipid fatty acid (PLFA) analyses of microbial community structure and microbial stress, dehydrogenase activity (DHA) assays measuring microbial activity, and headspace gas sampling of N₂O, ammonia (NH₃) and carbon dioxide (CO₂) fluxes. Experiment 1, a laboratory study, examined the influence of soil moisture and urinary salt content on the microbial community. Both urine application and high soil moisture increased microbial stress, as evidenced by significant changes in PLFA trans/cis and iso/anteiso ratios. Total PLFAs and DHA showed a short-term (< 1 week) stimulatory effect on microbes after urine application. Mean cumulative N₂O-N fluxes were 2.75% and 0.05% of the nitrogen (N) applied, from the wet (70% WFPS) and dry (35% WFPS) soils, respectively. Experiment 2, a field trial, investigated nutrient dynamics and microbial stress with plants present. Concentrations of the micronutrients, copper, iron and molybdenum, increased up to 20-fold after urine application, while soil phosphorus (P) concentrations decreased from 0.87 mg kg ⁻¹ to 0.48 mg kg⁻¹. Plant P was also lower in urine patches, but total PLFAs were higher, suggesting that microbes had utilised the available nutrients. Microbial stress again resulted from urine application but, in contrast to experiment 1, the fungal biomass recovered after its initial inhibition. Studies published during the course of this thesis reported that hippuric acid (HA) and its hydrolysis product benzoic acid (BA) significantly reduced N₂O-N emissions from synthetic cow urine, thus experiment 3 investigated this effect using real cow urine. Cumulative N₂O-N fluxes were 16.8, 5.9 and 4.7% of N applied for urine (U) alone, U+HA and U+BA, respectively. Since NH₃-N volatilisation remained unchanged, net gaseous N emissions were reduced. Trends in total PLFAs and microbial stress were comparable to experiment 1 results. Experiment 4 studied HA effects at different temperatures and found no inhibition of N₂O-N fluxes from HA-amended urine. However, mean cumulative N₂O-N fluxes were reduced from 7.6% of N applied at 15–20°C to 0.2% at 5–10°C. Total cumulative N emissions (N₂O-N + NH₃-N) were highest at 20°C (17.5% of N applied) and lowest at 10°C (9.8% of N applied). Microbial activity, measured as potential DHA, increased with increasing temperature. This work has clearly shown that the stimulation and inhibition of the soil microbial community by urine application are closely linked to soil chemistry and have significant impacts not only on soil nutrient dynamics but also on N₂O-N emissions and their possible mitigation.