BACTERIAL GROWTH EFFICIENCY: ASSESSMENT IN TERRESTRIAL ECOSYSTEMS By Zarraz M. Lee A DISSERTATION Submitted to Michigan State University In partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Microbiology and Molecular Genetics 2011 ABSTRACT BACTERIAL GROWTH EFFICIENCY: ASSESSMENT IN TERRESTRIAL ECOSYSTEM AND INFLUENCING FACTORS By Zarraz M. Lee Bacteria in soil are responsible for converting labile root exudates into microbial biomass that is more stable. This transformation is important for stabilizing soil organic carbon (SOC) and maintaining soil fertility. Bacterial partitioning of SOC into biosynthesis of new biomass or mineralization to carbon dioxide is defined as bacterial growth efficiency (BGE). BGE is also an integral component of models that simulate carbon dynamics in soil. However, variation in BGE in terrestrial ecosystems is not well understood, nor are environmental factors that influence it. In fact, BGE is often assumed to be constant. This dissertation explores BGE in four terrestrial biomes at the Kellogg Biological Station Long Term Ecological Research site and assesses factors that may influence BGE. BGE is calculated from bacterial production (BP) and respiration (BR) using the formula BGE = BP/(BP+BR). In terrestrial ecosystems, these parameters are often measured using a single radiolabeled substrate, which masks any influence of BGE specific to resources available in the soil. We developed a method that permits BGE measurements to reflect the nutrient status 3 of the soil. BP was measured as H-leucine incorporation rate into protein and BR as oxygen consumption rate, both without exogenously added substrates. Using this method, variation of BGE was assessed for soils collected from deciduous forest and three different croplands. We showed that BGE was not constant, but varied from 0.23 to 0.63. Bacterial communities in soils ii from soybean monoculture cropland tended to have a higher BGE than those in deciduous forests or rotational cropland soils. BGE in cropland soils exhibited a large seasonal variation not observed in forest soils. BGE can also be influenced by the composition of bacterial communities, as different bacterial species have different energy and growth requirements. However, it is challenging to link a general function such as carbon transformation to the structure of the bacterial community because of high functional redundancy. Using 16S rRNA gene surveys of rotation cropland and forest soils, it is proposed that the efficiency of bacterial communities can be predicted. In addition to identification of the types of bacteria in the soil sample, 16S rRNA gene surveys also provide a glimpse of the lifestyle of the bacteria. The identity of the bacteria provides an estimate of the number of 16S rRNA genes in the genome, which can then indicate the ecological strategy of the bacteria. Based on the ecological strategy, the growth efficiency can be predicted. The rotation cropland and forest soils have different community composition but the overall 16S rRNA gene copy number is the same, consistent with the observation that these two communities have similar growth efficiencies. Inclusion of measured values of BGE into DAYCENT, a model simulating nutrient cycling in soils, revealed the sensitivity of the model to changes in BGE. When simulating carbon dynamics in a mock ecosystem, decreasing the default BGE value of 0.45 to the average BGE in forest soils, 0.35, reduced the active carbon fraction by 22%. This led to a 5% reduction in the predicted total soil carbon at equilibrium. Therefore, site-specific BGE is important for improving the predictive capacity of SOC models, especially when investigating the effects of changes in climate, soil edaphic properties and land management practices on labile SOC transformations. iii ACKNOWLEDGEMENTS First of all, I am grateful for the guidance and support provided by my advisor, Dr. Tom Schmidt. He has provided me endless opportunities to mature as a scientist and continuously encouraged me to challenge my scientific creativity. He introduced me to a whole new area of microbial ecology and I am truly thankful to be part of his team. I would like to thank my committee members, Dr. Michael Bagdasarian, Dr. Rob Britton, Dr. Jay Lennon and Dr. Vince Young, for their support and advice throughout my graduate career. I am especially thankful to them for the encouragement they provided when I thought I had come to a dead end. I would like to extend a thank you to the staff members in the Microbiology and Molecular Genetics program for their assistance in making graduate school free of bureaucratic hassle. I would like to acknowledge the staffs at Kellogg Biological Station, especially Cathy McMinn and Stacey Vanderwulp, for their help in collecting site data. I would also like to acknowledge the Michigan State University’s Graduate School, the College of Natural Sciences and the Department of Microbiology and Molecular Genetics for travel funding. The Schmidt lab had made my graduate career a pleasurable and memorable experience. I am thankful to Jorge Rodriguez and Dion Antonopoulous for mentoring me when I rotated in the Schmidt lab. They are the best teachers in molecular techniques and were great role models. I owe my excitement in microbial physiology to Stephanie Eichorst and Kwi Kim. Stephanie is the greatest mentor and the best friend I have ever known. She was like a family member to me; she was there for me whenever I missed home and always inviting me to her home to celebrate American holidays. Kwi was not only a lab technician but she made a day of failed experiments much better with her mom-like hugs and encouragements. Kevin Theis was always willing to share his expertise in statistics while Tracy Teal and Vicente Gomez were only a few steps away iv when I was confused about molecular analysis. I am thankful to John Dover for all the help he provided and for enjoyable conversations about everything, from science to photography. Uri Levine, Ben Roller and Keara Towery are great friends who brought fun into the lab and Ben for his shared interest in BGE. I would like to thank Dr. Clegg Waldron for all the advice he provided. I also owe him for all the inspirations that came to me during our discussions. The long process of obtaining a Ph. D. would not have been easy without the friendship and occasional distraction from friends I made. Lin Pei, Theng Theng, Jo, Pei Hua, Deckson, May and Alan all made being away from home less painful. The memories we made will always bring a smile and brighten gloomy days. Kristin Huizinga, Kristi Whitehead, Jennifer Gray, Fan Yang, Dena Cologgi, Allison Speers and Mike Manzella were always around whenever I need to take a break or even ask for favors. I hope to continue to keep in touch with all of you and will be excited hear more about where science takes each of us. Lastly, I am indebted to my parents, Dalwi Lee and Khoo Lay Im, and grandparents for their endless love and continuous encouragement. They are always a phone call away whenever I need them. I would not have had the opportunity to make it here if it wasn’t for the sacrifice my parents made. Dad, thank you for making sure I got through with this challenge. This degree and experience comes with an expensive price to pay. I am guilty for the loneliness by grandma, Khoo Guat Sin, felt and for all the birthdays I missed. I owe her a career that she can relate to, but I know she is proud that I chased after my own dreams. Ahmah, you will always hold a special place in my heart. I would also like to acknowledge the love, support and challenge from a special person, John Quick. I am excited that you comprehend the enthusiasm I have for microbial ecology. I am looking forward to all the possible ways we can integrate our expertise to help play our role in making the world a better place. v TABLE OF CONTENTS LIST OF TABLES……………………………………………………………………………...viii LIST OF FIGURES………………………………………………………………………………ix CHAPTER 1: BACTERIAL DRIVEN SOIL CARBON SEQUESTRATION BACTERIAL DRIVEN CARBON TRANSFORMATIONS…………………………….1 BACTERIAL GROWTH EFFICIENCY…………………………………………………4 SOIL BACTERIAL COMMUNITIES……………………………………………………8 MICROBIAL ACTIVITY IN SOIL CARBON MODELS……………………………...12 REFERENCES…………………………………………………………………………..14 CHAPTER 2: MEASURING BACTERIAL GROWTH EFFICIENCY IN SOIL INTRODUCTION……………………………………………………………………….21 MATERIALS AND METHODS………………………………………………………...24 RESULTS………………………………………………………………………………..28 DISCUSSION……………………………………………………………………………30 CONCLUSION…………………………………………………………………………..37 REFERENCES…………………………………………………………………………..38 CHAPTER 3: BACTERIAL GROWTH EFFICIENCY IN SOILS UNDER DIFFERENT LAND MANAGEMENT PRACTICES INTRODUCTION……………………………………………………………………….43 MATERIALS AND METHODS………………………………………………………..45 RESULTS………………………………………………………………………………..47 DISCUSSION……………………………………………………………………………51 CONCLUSION…………………………………………………………………………..55 ACKNOWLEDGEMENTS……………………………………………………………...55 REFERENCES…………………………………………………………………………..56 CHAPTER 4: rRNA OPERON COPY NUMBER AS A GENETIC MARKER FOR ECOLOGICAL STRATEGIES INTRODUCTION……………………………………………………………………….61 ABUNDANCE OF rrn OPERONS IN BACTERIA AND ARCHAEA………………...63 ECOLOGICAL STRATEGIES OF BACTERIA WITH DIFFERENT rrn OPERON COPY NUMBER………………………………………………………………………...66 BACTERIAL COMMUNITIES IN SOIL……………………………………………….71 CONCLUSION…………………………………………………………………………..74 MATERIALS AND METHODS………………………………………………………...75 ACKNOWLEDGEMENTS……………………………………………………………...76 vi REFERENCES…………………………………………………………………………..77 CHAPTER 5: SIGNIFICANCE AND FUTURE DIRECTIONS……………………………….82 ACKNOWLEDGEMENTS……………………………………………………………...90 REFERENCES…………………………………………………………………………..91 APPENDIX A: STREPTOMYCIN-RESISTANT E. COLI MUTANT EXHIBITS INCREASED BACTERIAL GROWTH EFFICIENCY INTRODUCTION………..……………………………………………………………...95 MATERIALS AND METHODS………………………………………………………...96 RESULTS………………………………………………………....................................101 DISCUSSION………………….……………………………………………………….106 FUTURE DIRECTIONS………………….……………………………………………107 ACKNOWLEDGEMENTS…………………………………………………………….108 REFERENCES…………………………………………………………………………109 vii LIST OF TABLES Table 1-1: Microbial growth efficiencies (MGE) in agricultural soils………………….……….9 Table 1-2: Microbial growth efficiencies in grassland soils……………………………..……..10 Table 1-3: Microbial growth efficiencies in forest soils…………………………………..……11 Table 2-1: Methods for calculating growth efficiency and the corresponding assumptions. SC = Substrate consumed, MR = microbial respiration, MP = microbial production, ST = total substrate………………………………………………………………………………………….22 Table 4-1: Genera with more than two rrs copy number.……………………………………….64 Table 4-2: Sequence divergence of the rrs gene and rrn operon copy number of bacteria with intragenomic heterogeneity of more than 3%. *Sequence divergence was calculated from neighbor-joining distance matrix produced from the Arb software. The maximum divergence is reported………………………………………………….……………………………………….66 Table 4-3: Characteristics of the lifestyle of bacteria that are correlated with its rrn operon copy a number. The descriptions in bold have yet to be tested. This characteristic is further discussed in b Appendix A. This has been shown by Roller, B. in the Schmidt lab. ………………………….70 Table A-1: The R value for E. coli XAC and E. coli SmR strains carrying plasmid pYC or pYLC and their R values. The values shown are the averages of three independent experiments with standard deviations in parentheses. Values in bold are values reported by Dong and Kurland..103 viii LIST OF FIGURES Figure 1-1: Bacteria play a central role in soil carbon transformations. Plant-derived carbon (litter material and root exudates) serves as substrates for bacterial growth. Solid black arrows represent decomposition that immobilizes carbon. Dashed black arrows represent degradation by extracellular enzymes. Solid grey arrows represent chemical and physical interactions between lignin and bacterial products with soil particles to form stable passive carbon. Dotted black arrow represents the decreased protection of passive carbon that releases labile carbon for decomposition by bacteria………………………………………………..……………………….3 Figure 1-2: Soil organic carbon serves as both a carbon and energy source for aerobic heterotrophs in soil………………………………………………………………………………...4 3 Figure 2-1: Workflow of measuring BGE for soil samples using H-leucine. TCA: trichloroacetic acid……………………………………………………………………………………………….25 Figure 2-2: a) Leucine incorporation rate (µg C/g soil/hr) by bacterial communities in grassland soil for leucine concentrations between 0.3 nM and 110 µM. b) Leucine incorporation rate for leucine concentrations below 5 µM fitted to Michaelis-Menton kinetics (Vmax = 1.1 µg C/g soil/hr, Km = 0.97 µM). Each point represents the average of three experimental replicate with one standard deviation as the error bars. Grassland soil was used in this experiment as the trials performed during method development were tested using grassland soils………………………29 Figure 2-3: Incorporation of leucine over time for grassland soil at KBS LTER when leucine concentration is 660 nM. Each symbol represents average of three biological replicates. The 2 error bar is smaller than the symbols. The linear regression has an r of 0.995…………………30 Figure 2-4: Bacterial growth efficiency in rotation and forest soils measured using leucine/oxygen (striped) and 14 3 H- C-acetate (solid) in February and March……………………….31 Figure 3-1: a) The linear relationship between bacterial production and respiration for soil samples from four different biomes and at different times of the year. The solid line represents 2 the linear fit, BR = 0.51*BP + 1.19, r = 0.20, p-value < 0.001. b) Boxplot showing the variation of BGE in the four different biomes from June through September. Different letter indicates significant difference from post-hoc Tukey HSD test with a p-value ≤ 0.1 (n=9)………………49 Figure 3-2: Seasonal variation of BGE in rotation cropland (dashed line) and deciduous forest (solid line). Each point represents three experimental replicates and error bars are standard error………………………………………………………………………………………………49 Figure 3-3: Relative a) Bacterial production, bacterial respiration and b) BGE for microcosms amended with carbon, inorganic nitrogen, carbon and inorganic nitrogen mixture or R2 broth. All measurements are normalized to the unamended microcosms by percentage difference. Asterisks ix represent significant difference from the unamended microcosm, where ‘*’ has p-value ≤ 0.1 and ‘**’ has p-value ≤ 0.05 (Student’s t-test)………………………………………………………...50 Figure 3-4: a) Seasonal variation of C:N ratio in forest and rotation cropland soils. Each point represents three experimental replicates and error bars represent standard error. b) The linear correlation between C:N ratio and BGE in forest and rotation cropland soils collected in March through September. The solid line represents the linear regression, BGE = - 0.04(C:N ratio) + 2 0.68, r = 0.27, p-value = 0.001………………………………………………………………….54 Figure 4-1: Distribution of 16S rRNA gene copy number in bacteria. The analysis was performed on 652 unique bacterial species (gray bars)……………………………………………………...65 Figure 4-2: The relative abundance of different phyla for each 16S rRNA gene copy number…67 Figure 4-3: Correlation between 16S rRNA gene copy number and tRNA genes in bacterial genomes. The data are gathered from 867 bacterial genomes. The solid line represents the linear 2 fit, y = 31.8x + 6.3, R = 0.78, p-value < 0.001…..……………………………………………...68 Figure 4-4: Relative abundance of bacterial phyla in 16S rDNA libraries constructed from rotation cropland and forest soils in September 2009. Each bar represents average relative abundance from two replicate plots and error bar is one standard deviation. ‘*’ represents significant differences from Student’s t-test with p-value < 0.05. ……………………………....71 Figure 4-5: Relative abundance of the 20 most abundant OTUs in forest and rotation cropland based on the order level of the OTU. The possible rrn operon copy number is provided in P brackets. ‘ ’ represents taxonomic classification above the order level. ‘*’ represents significant difference based on Student’s t-test with p-value < 0.05. ………………………….....................73 Figure 5-1: Soil organic carbon submodel in the DAYCENT ecosystem model. Ovals represent the application of growth efficiency in carbon transformations. Blue ovals represent growth efficiency values that were adjusted in the sensitivity analysis. This figure is adapted from the CENTURY User’s Manual………………………………………………………………………85 Figure 5-2. Sensitivity of the active carbon pool (a) and the annual passive carbon pool (b) to variations in growth efficiency. The active carbon pool at different growth efficiencies is presented as a percentage relative to the active carbon pool when growth efficiency is 0.45…...86 Figure 5-3. Framework relating changes in climate, land management practices and edaphic factors on BGE. The dotted arrows represent processes that can influence BGE……...………..87 Figure 5-4. Seasonal variation of CO2 flux (a, b) and BGE (c, d) in the conventional agriculture (a, c) and deciduous forest (b, d) sites. Each point represents the average of three experimental replicates and the error bars are standard error. Flux data was obtained from the KBS LTER dataset (retrieved on March 21, 2011).……...…………………………………………………...89 x Figure A-1: Plasmid pYLC. bla: ampicillin resistance gene, ori: origin of replication, Plac: lac promoter, eyfp: yellow fluorescent protein gene, lacZ: ß-galactosidase gene, and ecfp: cyan fluorescent protein gene………………………………………………………………………….98 Figure A-2: Ribosome drop-off frequency for wild type (WT) and SmR mutant (SmR) in M9Gly54 medium during log-phase growth. R-value is calculated as C:Y expression from pYLC plasmid/C:Y expression for pYC plasmid when induced with 2 mM IPTG. Each point represents the average of three measurements with the standard error..…………………………………...102 Figure A-3: Bacterial growth efficiency of E. coli wild-type and streptomycin resistant strains in three different conditions. GlyAA: M9 medium supplemented with 19 amino acids and 0.4% glycerol. GlyPA: M9 medium supplemented with 0.4 mM proline and arginine, and 0.2% 14 glycerol. BGE in GlyPA medium was measured using two different methods; C-glycerol and 3 3 H-leucine/oxygen. The H-leucine/oxygen measurement was provided by Roller, B. from Schmidt lab. Each bar represents the average of two replicates with standard deviation as error bars……………………………………………………………………………………………...104 Figure A-4: Relative doubling time of E. coli MG1655 wild-type and rpsL mutant strain (rpsL6K42N) in M9 minimal medium with 0.2% glucose, 0.2% glycerol or 0.4% acetate as carbon source. The doubling time was normalized to growth of wild-type strain in M9-Glucose. Each bar represents average of three biological replicates and the error bar represents standard error……………………………………………………………………………………………..105 Figure A-5: Relative BGE of E. coli MG1655 and rpsL mutant in M9 minimal medium with 0.2% glycerol as carbon source. BGE was measured with two different methods and normalized to the WT strain for each method. Each bar represents average of two replicates and the error bar represents standard deviation..………………………………………………………………….105 xi CHAPTER 1 BACTERIAL-DRIVEN SOIL CARBON SEQUESTRATION Bacterial Driven Carbon Transformations Soil organic carbon (SOC) makes up the largest terrestrial reservoir of carbon with 1550 Pg carbon, almost three times more than the amount of carbon in vegetation (36). Some of the SOC is vulnerable to changes in climate, particularly moisture and temperature, and especially to land use change. The conversion of forest to intensively managed cropland can result in the loss of up to 50% of the stored carbon (72). Due to the large amount of carbon stored in soil, a small change in soil carbon stock can have a significant impact on the amount of carbon in the atmosphere, which is currently 760 Pg (36). Historical change in land use during the years 1850 to 2000 was estimated to release up to 156 Pg CO2-C, second only to industrial sources of CO2 (31). Not only does the loss of soil carbon increase atmospheric CO2, it also decreases soil fertility (71). Replenishing the carbon lost from soil following deforestation would offset some of the CO2 released to the atmosphere (37). With the decreasing capability of the ocean to serve as a carbon sink, more attention is being focused on carbon sequestration in soil (30). Promoting carbon sequestration in soil is also important for improving soil fertility and preventing soil degradation. Soil degradation is one of the most challenging problems associated with terrestrial carbon loss because degraded soils lead to decreases in primary productivity and are difficult to rejuvenate (23). Cropland that was converted back into grassland takes decades to accumulate SOC and does not necessarily return to the pre-agriculture SOC content (47). The amount of 1 carbon that can be stored in soil is determined by the amount of carbon input, primarily from photosynthesis, and the amount of carbon lost during decomposition of SOC by microbes. Therefore to increase soil carbon there has to be an increase in primary productivity, a decrease in carbon loss during decomposition, or both. Plant-derived carbon (litter material and root exudates) is the major input of organic carbon into soils. Some of this carbon is directly assimilated by microbes and incorporated into biomass (Figure 1-1, solid arrows). More recalcitrant plant polymers such as lignin are degraded with extracellular enzymes into labile, low-molecular weight substances that are subsequently consumed by microbes (Figure 1-1, dashed arrows). Simultaneously, some compounds undergo chemical interactions to form humic acids and become physically protected in soil aggregates (Figure 1-1, grey arrows). This carbon pool is known as passive carbon and it is not typically accessible for decomposition by bacteria. The structure of bacterial biomass and extracellular polysaccharides that allow strong interactions with soil particles make bacterial products a crucial component of SOC formation, especially in assisting with humic formation (41, 42). In fact, recalcitrant carbon in the passive carbon pool is dominated with carbon of microbial origin (32). The central role of bacterial activity in soil organic carbon formation is further supported by the observation that molecular structure of soil carbon is similar across all land management practices (21). Carbon in the passive carbon pool can become accessible to active microbes. This often happens when cropland is tilled or there is an increase in soil moisture (22). Tillage homogenizes aggregates while moisture increases diffusion rate of low molecular weight substances, making the carbon accessible to bacteria (Figure 1-1, dotted arrow). Tillage also increases oxygen availability, which can increase decomposition rate (4). No-tillage practices 2 can lead to the recovery of more than 10% of soil carbon in the surface layer (45). Additionally, when carbon input increases but the bacteria are still limited for nutrients such as nitrogen, extracellular enzymes can be produced to mine for the limiting nutrient and enhance decomposition (6). Therefore, bacterial communities in soil are also responsible for the loss of passive soil carbon. Figure 1-1: Bacteria play a central role in soil carbon transformations. Plant-derived carbon (litter material and root exudates) serves as substrates for bacterial growth. Solid black arrows represent decomposition that immobilizes carbon. Dashed black arrows represent degradation by extracellular enzymes. Solid grey arrows represent chemical and physical interactions between lignin and bacterial products with soil particles to form stable passive carbon. Dotted black arrow represents the decreased protection of passive carbon that releases labile carbon for decomposition by bacteria. The efficiency with which bacteria transform carbon substrates into biomass can be measured as bacterial growth efficiency (BGE). BGE is defined as the fraction of the total carbon consumed that is incorporated into biomass (13). Aerobic heterotrophic bacteria assimilate plantderived carbon both to synthesize biomass and to respire to CO2 for energy (Figure 1-2). A more efficient bacterial community will be able to incorporate a greater percentage of carbon consumed into biomass material and so increase the abundance of material for passive carbon formation. Therefore, soils with more efficient bacterial communities have higher potential to 3 become a carbon sink and are more capable of maintaining soil fertility. More efficient carbon transformations by the bacterial communities will also preserve more energy in the ecosystem for higher trophic level organisms. Figure 1-2: Soil organic carbon serves as both a carbon and energy source for aerobic heterotrophs in soil. Bacterial Growth Efficiency During aerobic growth, BGE is determined by two parameters of bacterial activity – the synthesis of biomass, also known as bacterial production (BP), and bacterial respiration (BR) (Figure 1-2). Hence, BGE is calculated as BP/(BP+BR) (13). In pure cultures of bacteria, BGE is found to be influenced by environmental conditions and intrinsic factors (61). At ranges of resource availability typically provided in laboratory media, BP and BR are strongly coupled. This means that BGE remains constant as substrate concentration varies (51). For example, BGE 4 for Aerobacter cloacae stays within the range of 0.56 to 0.6 when grown in a medium containing glucose as the sole carbon and energy source (26). However, at very low substrate concentrations, BP and BR become uncoupled, causing BGE to decrease with substrate availability (69). This uncoupling of BP and BR is due to the constant requirement for maintenance energy within a bacterium. Maintenance energy is the energy used to support non-growth processes such as cellular integrity, ion gradients and endogenous metabolism (52). Uncoupling of BP and BR can also occur at high carbon substrate concentrations, when growth becomes limited by nutrients such as nitrogen or phosphorus (43). Under these conditions, bacteria frequently respire the excess carbon in a futile cycle known as energy spilling (56). In addition to resource availability, BGE is also dependent on the composition of the growth medium because this influences the energy required for biomass synthesis. For example, Streptococcus bovis is found to exhibit a 40% increase in growth efficiency when amino acids are provided in the growth medium (55). Amino acid biosynthesis requires between 12 and 74 molecules of ATP for each amino acid synthesized, therefore more of the substrate carbon is required for respiration (BR) leading to a decrease in BGE. BGE is also influenced by the amount of free energy available in the growth substrate. A low energy substrate such as methane will provide less carbon for growth after the energy demand is fulfilled (51). Intrinsic factors are determined by the genetic makeup of an organism. There are two ways in which differences in genetic capability can influence BGE. One is the growth-related energy requirement and the other is the maintenance energy requirement. An organism that uses a catabolic pathway that yields more energy will be able to allocate more carbon for growth instead of ATP synthesis. Streptococcus faecalis, utilizing the Embden-Meyerhof pathway that yields 2 moles ATP/mole glucose, has a higher BGE than Leuconostoc mesenteroides using the 5 Entner-Doudoroff pathway that yields 1 mole ATP/mole glucose (5). It has been proposed that different organisms have different maintenance energy requirements, but more research is required to determine the interspecies variability (57, 73). Growth-related energy requirements can also be decreased by reducing the cost of biomass synthesis. The most energetically expensive cellular process is translation, the process of protein synthesis. With four ATP molecules required for the formation of each peptide-bond, protein synthesis consumes up to 60% of cellular ATP during exponential growth (66). Different bacteria can have ribosomes with different translational performance. Fast-growing bacteria have higher translational power, which is the rate of functional protein synthesized per unit translational machinery (15). In E. coli, different isolates can have ribosomes with different efficiencies, as measured by the rate of amino acid incorporation per ribosome (54). As protein synthesis is such an energy demanding process, bacteria with ribosomes that can synthesize more functional proteins per unit energy are predicted to be more efficient, especially in environments with limited resources (15). This hypothesis is briefly explored in Appendix A using E. coli strains having ribosomes with different translational performances. The sensitivity of growth efficiency to both environmental factors and genomic make-up can result in a large variation in BGE of bacterial communities in the environment. Additionally, the quality and concentration of resources often fluctuate in the environment, which can limit growth and cause uncoupling of BP and BR. It is therefore not surprising that where BGE has been intensively explored, such as in the aquatic environment, it is found to range from 0.05 to 0.6 (14). Aquatic BGE is mainly driven by changes in quality and quantity of dissolved organic substrates. Using chlorophyll as an indicator of system productivity, Lopez-Urrutia showed that BP is sensitive to resource availability but BR remains the same. Hence, the variation of BGE is 6 driven by the capability of bacterioplankton to incorporate substrate into biomass (44). Consequently, low BGE is a characteristic of oligotrophic lakes with low primary productivity (2, 10). The influence of substrate quality becomes more important in eutrophic environments where more labile carbon promotes increased efficiency (3). Due to the central role of bacterioplankton in carbon cycling in the aquatic system, high BR becomes a significant contribution to oceanic CO2 flux into the atmosphere (11). Growth efficiency in terrestrial ecosystems is not as well explored as in aquatic ecosystems because the heterogenous nature of soil makes measuring growth efficiency difficult. Additionally, terrestrial growth efficiency measurements are commonly based on a single substrate, typically applied at concentrations not commonly found in soil (Tables 1-1 to 1-3). Early measurements of microbial growth efficiency (MGE) using radiolabeled glucose found that MGE varies with the substrate concentration applied to the soil and the incubation time (9). This results in MGE measurements that are more reflective of the experimental perturbation than the nutrient status of the soil. Furthermore, MGE measurements using a single substrate represent the efficiency of microbes that can assimilate and mineralize the added substrate, rather than the entire active community. I adapted a method from aquatic ecology that does not require addition of exogenous substrate (Chapter 2). This method of measuring growth efficiency reflects influences from the soil environment and the overall activity of the bacterial community. Several measurements of MGE have been reported for terrestrial biomes. However, the sensitivity of MGE to the substrate chosen by the investigators makes cross-site comparisons less meaningful (68). Using a substrate that can be utilized by most bacteria, such as ammonium nitrate, Schimel calculated MGEs for grasslands and croplands using three different C:N ratio. The grassland soils tended to have microbes that are more efficient (59). However, without more 7 growth efficiency measurements for a wider range of terrestrial biomes, factors that regulate growth efficiency in the terrestrial ecosystem cannot be determined. For example, the influence of land management practices on microbial-driven carbon transformations requires cross-site comparisons. Focusing on the dynamics of bacterial growth, we measured BGE in four terrestrial biomes at the Kellogg Biological Station Long Term Ecological Research (KBS LTER) site to identify factors that can regulate BGE in terrestrial ecosystems (Chapter 3). Soil Bacterial Communities Transformations of soil carbon are common functions for all heterotrophic microbes in soil. Unlike specialized functions such as denitrification or ammonia oxidation, there is no single functional gene that links CO2 production by heterotrophs to the structure of microbial communities (24). In fact, while the diversity of methane-oxidizing bacteria is correlated with the level of methane oxidation (a specialized metabolic function), there is no correlation between heterotroph diversity and CO2 production (a broadly distributed metabolic pathway) (40). An extensive study of bacterial community diversity in 98 soils across America found that pH was the only explanatory factor for differences in bacterial diversity (18). Yet it is well documented that community compositions change in response to different substrate types and, in some cases, the change influences carbon mineralization rate (17, 67). This suggests that in order to study the influence of changes in community structure on carbon transformations, it is important to first identify bacterial groups that respond to different substrate availability and type (33). Changes in community structure indicate that when labile substrate becomes available carbon mineralization rate increases; this is often correlated with increased abundance of fast-responding bacteria like Gammaproteobacteria and Firmicutes (12). 8 Soil Texture Substrate / Nutrients Factors Silt loam (23% clay, 0.5% orgC) Glucose (800 g C/g soil) Low grazing Fine sandy loam (18% clay, 1.8% orgC) 14 C-Glucose (1g C/g soil) Low grazing Sandy loam (1.6% orgC) Silt loam (2.2% orgC) 14 Silt loam (1.3% totC) High grazing Method Ref. 0.57 - 0.58 YS/YB/YC (S) High grazing C-Glucose, N, P (36 - 2304 g C/g soil) MGE 0.24 - 0.65 0.51 - 0.59 (20) 0.27 - 0.68 Glucose concentration 14 YB (S) 0.9 YB (L) 0.57 - 0.75 YR (L) 0.63 YS (S) 0.91 - 0.92 YS (L) 0.67 - 0.72 Glucose concentration C-Glucose (30-300 g C/g soil) C:N:P = 150:10:1 0.56 - 0.72 Glucose concentration, Time 14 YR (L) (8) YR (S) 0.92 - 0.98 (28) (9) 9 Black soil (49% clay, 2.1% orgC) C-glucose (937 g C/g soil), C:N = 25:1 Clay loam 26% clay, 1.3% orgC) 14 C-Glucose (0.01 g C/g soil) Eutric Cambisols (3% totC) 14 Silt loam (26% clay, 1.6% orgC) 13 C-Glucose (800 g C/g soil) F:B ratio, C:N ratio YS (S) 0.59 - 0.62 (70) Silt loam Cellobiose (420 g C/g soil) Temperature, Time YR (L) 0.60 - 0.80 (65) Clay loam 14 Straw application YC (L) 0.05 – 0.43 (29) C:N ratio, oxic status, temperature YC (L) 0.30 - 0.61 (16) Soil texture YB(N) (L) 0.25 - 0.59 (59) Fine clay (51% clay, 2.1% totC) Sandstone, loamy silt, fine shale * None Plants, Time C-Glucose (1 M - 10 mM) C-Wheat straw Rice straw 15 NH4SO4 (1 g N/g soil) Table 1-1:Microbial growth efficiencies (MGE) in agricultural soils. 9 (64) (48) Soil Texture Substrate / Nutrients Soddy-podzolic (1.9% orgC) Gray forest (2.3% orgC) Glucose (4 mg C/g soil), N, P (C:N:P = 1:10:1 to 1:30:1) Silt loam (2.4% totC) 14 C-Glucose (0.0009 - 260 g Glucose concentration C/g soil), N, P Sandy loam (12% Clay) 14 Sterilized soil C-Glucose (300 g C/g soil), 15 N (20 g N/g soil), 32 Sandy silt (1% totC) 13 10 Clay (42% Clay) Course loam (2.89% orgC) Sandstone, Loamy silt, Fine shale 15 15 Method MGE Ref. Substrate, Soil type YR (S) 0.80 - 0.95 (7) YB (S) 0.5 - 0.85 YB (L) 0.4 - 0.61 YS (S) 0.78 - 0.86 YS (L) 0.61 - 0.67 YS (S) 0.80 - 0.88 YS (L) 0.64 - 0.71 YB (L) 0.21 - 0.36 YB (L) 0.46 - 0.62 YB(N) (L) 0.24 - 0.58 YB(N) (L) 0.29 - 0.46 Soil texture, Moisture, Time P C Cellulose (495 mg C/g soil) 13 Factors C-Acetate, N-NH3, vapor NH4SO4 (1 g N/g soil) Nutrient (C:N:P = 80:8:1) Moisture, Method Soil texture Table 1-2: Microbial growth efficiencies in grassland soils. 10 (62) (74) (19) (27) (59) Soil Texture Substrate / Nutrients Factors Soddy-podzolic (1.9% orgC) Glucose (4 mg C/g soil), N, P (C:N:P = 1:10:1/1:30:1) Gray forest (2.3% orgC) Sterilized soil Umbrisol (28% Clay, 16% totC) Glucose (300 - 4000 g C/g soil), C:N:P = 10:1:1 Upland taiga, Floodplains 14 Loamy (8.7% totC) 15 C-Glucose, 14 C-Phenolics 15 NH4 NO3 MGE Ref. Substrate, Soil type YR (S) 0.80 - 0.95 (7) Glucose concentration YB (S) 0.76 - 0.83 (58) Glucose concentration YB (L) 0.47 - 0.66 (63) Substrate, Soil type YC (S) 0.21 - 0.85 (68) Time Sandy loam (29% clay, 5% 14C-Glucose organic matter) (0.004 - 400 g C/g soil) Method YB(N) (L) 0.15 - 0.60 (25) 11 Table 1-3: Microbial growth efficiencies in forest soils. Methods: (S) and (L) indicates incubation duration YS = (Sc-MR)/Sc, 1 day and > 1 day, respectively. YB = MP/(MP+MR), YB(N) indicates that MP was calculated from N immobilization YC = MB/Sc YR = 1-(MR/Sc) where Sc is the amount of substrate consumed, MR = microbial respiration, and MP = microbial production. *: No comparisons were made. MGE was measured to calculate microbial biomass turnover and maintenance energy. 11 One genetic marker that might be used to gauge the efficiency with which bacteria incorporate carbon into biomass is rrn operon copy number. Bacteria can have 1 to 15 copies of rrn operons per genome (35). Bacteria with high rrn operon copy number tend to response faster and are more competitive during high substrate availability. Bacteria with fewer rrn operons have ecological strategy characteristics of a K-strategist, with enhanced capability to grow at low substrate concentrations (34). The copy number of rrn operons can be predicted from molecular surveys of 16S rDNA. Based on phylogenetic association, we can assign rrn operon copy numbers to bacterial groups that respond to environmental factors such as substrate availability. The distribution of rrn operons in bacteria and archaea was explored to determine the confidence with which rrn operons can be assigned to important phylotypes (39) (Chapter 4). Microbial Activity in Soil Carbon Models One of the motivations for understanding how microbes respond to changes in land management practices is the desire to better predict how the practices influence the dynamics of soil organic carbon. Soil nutrient models have been developed to make quantitative predictions that could guide decision-makers on sustainable land management practices (53). Our understanding of the factors that influence carbon transformations must be expressible in mathematical equations to be useful for incorporation into such models. Microbially driven carbon transformations are typically described using first-order rate kinetics as a function of the carbon concentration (50). Two environmental factors that consistently influence this rate are temperature and moisture, both of which are well documented to correlate with rates of carbon mineralization (1). Growth efficiency is then applied to determine the fraction of the consumed carbon that is transformed into microbial biomass. Due 12 to a lack of direct measurements of growth efficiency in soil, a fixed value between 0.35 and 0.5 has been commonly assumed in soil carbon models (38, 50, 60). It is currently treated as a global parameter, where the same growth efficiency is used for simulations of all types of terrestrial biomes. This implies that all microbial communities are equally efficient at transforming carbon into microbial biomass. As discussed earlier, growth efficiency is influenced by environmental factors and potentially by changes in community structure. Therefore, understanding regulation of growth efficiency in different biomes is crucial to improve accuracy of site-specific models. The incorporation of accurate MGE into models will acknowledge the versatility of microbial communities in responding to different substrate type and nutrient status, rather than treating soil microbial communities as a single entity. This is especially important for simulations with short timescales or models with more robust microbial mechanism such as the inclusion of carbon allocation for extracellular enzymes (38, 49). Site-specific MGE measures will also become more important as improved models are able to take into consideration ecosystem changes that result from shifts in microbial communities (46). The application of MGE in SOC models and its significance in understanding microbial-driven soil carbon dynamics are further discussed in Chapter 5. 13 REFERENCES 14 REFERENCES 1. Amundson, R. 2001. The carbon budget in soils. Annu Rev Earth Pl Sc 29:535-562. 2. Anderson, T. R., and C. M. Turley. 2003. Low bacterial growth efficiency in the oligotrophic eastern Mediterranean Sea: a modelling analysis. J Plankton Res 25:10111019. 3. Apple, J. K., and P. del Giorgio. 2007. Organic substrate quality as the link between bacterioplankton carbon demand and growth efficiency in a temperate salt-marsh estuary. Isme J 1:729-742. 4. Balesdent, J., C. Chenu, and M. Balabane. 2000. Relationship of soil organic matter dynamics to physical protection and tillage. Soil Till Res 53:215-230. 5. Bauchop, T., and S. R. Elsden. 1960. The Growth of Micro-Organisms in Relation to Their Energy Supply. J Gen Microbiol 23:457-469. 6. Billings, S. A., J. Lichter, S. E. Ziegler, B. A. Hungate, and D. D. Richter. 2010. A call to investigate drivers of soil organic matter retention vs. mineralization in a high CO2 world. Soil Biol Biochem 42:665-668. 7. Blagodatsky, S. A., E. G. Demyanova, E. I. Kobzeva, and V. N. Kudeyarov. 2002. Changes in the efficiency of microbial growth upon soil amendment with available substrates. Eurasian Soil Sci+ 35:874-880. 8. Bremer, E., and P. Kuikman. 1994. Microbial Utilization of C- [U]Glucose in Soil Is Affected by the Amount and Timing of Glucose Additions. Soil Biol Biochem 26:511517. 9. Bremer, E., and C. Vankessel. 1990. Extractability of Microbial C-14 and N-15 Following Addition of Variable Rates of Labeled Glucose and (NH4)2SO4 to Soil. Soil Biol Biochem 22:707-713. 10. Carlson, C. A., N. R. Bates, H. W. Ducklow, and D. A. Hansell. 1999. Estimation of bacterial respiration and growth efficiency in the Ross Sea, Antarctica. Aquat Microb Ecol 19:229-244. 11. Carlson, C. A., P. A. Del Giorgio, and G. J. Herndl. 2007. Microbes and the Dissipation of Energy and Respiration: From Cells to Ecosystems. Oceanography 20:89100. 14 15 12. Cleveland, C. C., D. R. Nemergut, S. K. Schmidt, and A. R. Townsend. 2007. Increases in soil respiration following labile carbon additions linked to rapid shifts in soil microbial community composition. Biogeochemistry 82:229-240. 13. del Giorgio, P. A., and J. J. Cole. 2000. Bacterial energetics and growth efficiency., p. 289-325. In D. L. Kirchman (ed.), Microbial ecology of the oceans. Wiley, New York. 14. del Giorgio, P. A., and J. J. Cole. 1998. Bacterial growth efficiency in natural aquatic systems. Annu Rev Ecol Syst 29:503-541. 15. Dethlefsen, L., and T. M. Schmidt. 2007. Performance of the translational apparatus varies with the ecological strategies of bacteria. J Bacteriol 189:3237-3245. 16. Devevre, O. C., and W. R. Horwath. 2000. Decomposition of rice straw and microbial carbon use efficiency under different soil temperatures and moistures. Soil Biol Biochem 32:1773-1785. 17. Eilers, K. G., C. L. Lauber, R. Knight, and N. Fierer. 2010. Shifts in bacterial community structure associated with inputs of low molecular weight carbon compounds to soil. Soil Biol Biochem 42:896-903. 18. Fierer, N., and R. B. Jackson. 2006. The diversity and biogeography of soil bacterial communities. P Natl Acad Sci USA 103:626-631. 19. Fontaine, S., G. Bardoux, L. Abbadie, and A. Mariotti. 2004. Carbon input to soil may decrease soil carbon content. Ecol Lett 7:314-320. 20. Frey, S. D., V. Gupta, E. T. Elliott, and K. Paustian. 2001. Protozoan grazing affects estimates of carbon utilization efficiency of the soil microbial community. Soil Biol Biochem 33:1759-1768. 21. Grandy, A. S., and J. C. Neff. 2008. Molecular C dynamics downstream: The biochemical decomposition sequence and its impact on soil organic matter structure and function. Sci Total Environ 404:297-307. 22. Grandy, A. S., and G. P. Robertson. 2006. Aggregation and organic matter protection following tillage of a previously uncultivated soil. Soil Sci Soc Am J 70:1398-1406. 23. Gregorich, E. G., K. J. Greer, D. W. Anderson, and B. C. Liang. 1998. Carbon distribution and losses: erosion and deposition effects. Soil Till Res 47:291-302. 24. Hallin, S., C. M. Jones, M. Schloter, and L. Philippot. 2009. Relationship between Ncycling communities and ecosystem functioning in a 50-year-old fertilization experiment. Isme J 3:597-605. 16 25. Hart, S. C., G. E. Nason, D. D. Myrold, and D. A. Perry. 1994. Dynamics of Gross Nitrogen Transformations in an Old-Growth Forest - the Carbon Connection. Ecology 75:880-891. 26. Herbert, D., R. Elsworth, and R. C. Telling. 1956. The Continuous Culture of Bacteria - a Theoretical and Experimental Study. J Gen Microbiol 14:601-622. 27. Herron, P. M., J. M. Stark, C. Holt, T. Hooker, and Z. G. Cardon. 2009. Microbial growth efficiencies across a soil moisture gradient assessed using C-13-acetic acid vapor and N-15-ammonia gas. Soil Biol Biochem 41:1262-1269. 28. Hill, P. W., J. F. Farrar, and D. L. Jones. 2008. Decoupling of microbial glucose uptake and mineralization in soil. Soil Biol Biochem 40:616-624. 29. Holland, E. A., and D. C. Coleman. 1987. Litter placement effects on microbial and organic matter dynamics in an agroecosystem. Ecology 68:425-433. 30. Houghton, R. A. 2007. Balancing the global carbon budget. Annu Rev Earth Pl Sc 35:313-347. 31. Houghton, R. A. 2003. Revised estimates of the annual net flux of carbon to the atmosphere from changes in land use and land management 1850-2000. Tellus B 55:378390. 32. Kiem, R., and I. Kogel-Knabner. 2003. Contribution of lignin and polysaccharides to the refractory carbon pool in C-depleted arable soils. Soil Biol Biochem 35:101-118. 33. King, G. M. 2011. Enhancing soil carbon storage for carbon remediation: potential contributions and constraints by microbes. Trends Microbiol 19:75-84. 34. Klappenbach, J. A., J. M. Dunbar, and T. M. Schmidt. 2000. rRNA operon copy number reflects ecological strategies of bacteria. Appl Environ Microbiol 66:1328-1333. 35. Klappenbach, J. A., P. R. Saxman, J. R. Cole, and T. M. Schmidt. 2001. rrndb: the Ribosomal RNA Operon Copy Number Database. Nucleic Acids Res 29:181-184. 36. Lal, R. 2004. Agricultural activities and the global carbon cycle. Nutr Cycl Agroecosys 70:103-116. 37. Lal, R., F. Follett, B. A. Stewart, and J. M. Kimble. 2007. Soil carbon sequestration to mitigate climate change and advance food security. Soil Sci 172:943-956. 38. Lawrence, C. R., J. C. Neff, and J. P. Schimel. 2009. Does adding microbial mechanisms of decomposition improve soil organic matter models? A comparison of four models using data from a pulsed rewetting experiment. Soil Biol Biochem 41:1923-1934. 17 39. Lee, Z. M. P., C. Bussema, and T. M. Schmidt. 2009. rrnDB: documenting the number of rRNA and tRNA genes in bacteria and archaea. Nucleic Acids Res 37:D489-D493. 40. Levine, U. Y., T. K. Teal, G. P. Robertson, and T. M. Schmidt. 2011. Agriculture's impact on microbial diversity and associated fluxes of carbon dioxide and methane. Isme J. 41. Liang, C., and T. C. Balser. 2008. Preferential sequestration of microbial carbon in subsoils of a glacial-landscape toposequence, Dane County, WI, USA. Geoderma 148:113-119. 42. Liang, C., G. Cheng, D. Wixon, and T. Balser. 2010. An Absorbing Markov Chain approach to understanding the microbial role in soil carbon stabilization. Biogeochemistry:1-7. 43. Liu, Y. 1998. Energy uncoupling in microbial growth under substrate-sufficient conditions. Appl Microbiol Biot 49:500-505. 44. Lopez-Urrutia, A., and X. A. G. Moran. 2007. Resource limitation of bacterial production distorts the temperature dependence of oceanic carbon cycling. Ecology 88:817-822. 45. McConkey, B. G., B. C. Liang, C. A. Campbell, D. Curtin, A. Moulin, S. A. Brandt, and G. P. Lafond. 2003. Crop rotation and tillage impact on carbon sequestration in Canadian prairie soils. Soil Till Res 74:81-90. 46. McGuire, K. L., and K. K. Treseder. 2010. Microbial communities and their relevance for ecosystem models: Decomposition as a case study. Soil Biol Biochem 42:529-535. 47. McLauchlan, K. K., S. E. Hobbie, and W. M. Post. 2006. Conversion from agriculture to grassland builds soil organic matter on decadal timescales. Ecol Appl 16:143-153. 48. Nguyen, C., and A. Guckert. 2001. Short-term utilisation of C-14-[U]glucose by soil microorganisms in relation to carbon availability. Soil Biol Biochem 33:53-60. 49. Parton, W. J., M. Hartman, D. Ojima, and D. Schimel. 1998. DAYCENT and its land surface submodel: description and testing. Global Planet Change 19:35-48. 50. Parton, W. J., D. S. Schimel, C. V. Cole, and D. S. Ojima. 1987. Analysis of Factors Controlling Soil Organic-Matter Levels in Great-Plains Grasslands. Soil Sci Soc Am J 51:1173-1179. 51. Payne, W. J. 1970. Energy Yields and Growth of Heterotrophs. Annu Rev Microbiol 24:17-52. 18 52. Pirt, S. J. 1965. Maintenance Energy of Bacteria in Growing Cultures. P Roy Soc Lond B Bio 163:224-231. 53. Priess, J. A., C. Schweitzer, F. Wimmer, O. Batkhishig, and M. Mimler. 2011. The consequences of land-use change and water demands in Central Mongolia. Land Use Policy 28:4-10. 54. Rang, C. U., R. Mikkola, S. Molin, and P. L. Conway. 1997. Ribosomal efficiency and growth rates of freshly isolated Escherichia coli strains originating from the gastrointestinal tract. Febs Lett 418:27-29. 55. Russell, J. B. 1993. Effect of Amino-Acids on the Heat-Production and Growth Efficiency of Streptococcus-Bovis - Balance of Anabolic and Catabolic Rates. Appl Environ Microbiol 59:1747-1751. 56. Russell, J. B. 2007. The energy spilling reactions of bacteria and other organisms. J Mol Microb Biotech 13:1-11. 57. Russell, J. B., and G. M. Cook. 1995. Energetics of Bacterial-Growth - Balance of Anabolic and Catabolic Reactions. Microbiol Rev 59:48-62. 58. Sawada, K., S. Funakawa, and T. Kosaki. 2009. Threshold concentrations of glucose to increase the ratio of respiration to assimilation in a Japanese arable soil and a strongly acidic Japanese forest soil. Soil Sci Plant Nutr 55:634-642. 59. Schimel, D. 1988. Calculation of microbial growth efficiency from 15 N immobilization. Biogeochemistry. 60. Schimel, J. P., and M. N. Weintraub. 2003. The implications of exoenzyme activity on microbial carbon and nitrogen limitation in soil: a theoretical model. Soil Biol Biochem 35:549-563. 61. Schmidt, T., and A. Konopka. 2009. Physiological and Ecological Adaptations of SlowGrowing, Heterotrophic Microbes and Consequences for Cultivation, p. 101-120. In S. S. Epstein (ed.), Uncultivated Microorganisms, vol. 10. Springer Berlin / Heidelberg. 62. Schneckenberger, K., D. Demin, K. Stahr, and Y. Kuzyakov. 2008. Microbial utilization and mineralization of [C-14]glucose added in six orders of concentration to soil. Soil Biol Biochem 40:1981-1988. 63. Shen, J., and R. Bartha. 1996. Metabolic efficiency and turnover of soil microbial communities in biodegradation tests. Appl Environ Microbiol 62:2411-2415. 64. Shields, J. A., E. A. Paul, W. E. Lowe, and D. Parkinson. 1973. Turnover of microbial tissue in soil under field conditions. Soil Biology and Biochemistry 5:753-764. 19 65. Steinweg, J. M., A. F. Plante, R. T. Conant, E. A. Paul, and D. L. Tanaka. 2008. Patterns of substrate utilization during long-term incubations at different temperatures. Soil Biol Biochem 40:2722-2728. 66. Stouthamer, A. H. 1973. Theoretical Study on Amount of Atp Required for Synthesis of Microbial Cell Material. A Van Leeuw J Microb 39:545-565. 67. Strickland, M. S., C. Lauber, N. Fierer, and M. A. Bradford. 2009. Testing the functional significance of microbial community composition. Ecology 90:441-451. 68. Sugai, S. F., and J. P. Schimel. 1993. Decomposition and Biomass Incorporation of C14-Labeled Glucose and Phenolics in Taiga Forest Floor - Effect of Substrate Quality, Successional State, and Season. Soil Biol Biochem 25:1379-1389. 69. Tempest, D. W., and O. M. Neijssel. 1984. The Status of YATP and Maintenance Energy as Biologically Interpretable Phenomena. Annu Rev Microbiol 38:459-486. 70. Thiet, R. K., S. D. Frey, and J. Six. 2006. Do growth yield efficiencies differ between soil microbial communities differing in fungal: bacterial ratios? Reality check and methodological issues. Soil Biol Biochem 38:837-844. 71. Tiessen, H., E. Cuevas, and P. Chacon. 1994. The role of soil organic matter in sustaining soil fertility. Nature 371:783-785. 72. Tilman, D. 1998. The greening of the green revolution. Nature 396:211-212. 73. van Bodegom, P. 2007. Microbial maintenance: A critical review on its quantification. Microb Ecol 53:513-523. 74. Vanveen, J. A., J. N. Ladd, and M. Amato. 1985. Turnover of Carbon and Nitrogen through the Microbial Biomass in a Sandy Loam and a Clay Soil Incubated with [C14(U)] Glucose and [N-15] (NH4)2SO4 under Different Moisture Regimes. Soil Biol Biochem 17:747-756. 20 CHAPTER 2 MEASURING BACTERIAL GROWTH EFFICIENCY IN SOIL Introduction Soil carbon transformations are driven primarily by microbial communities in soil. Bacteria, which can make up to 80% of the microbial biomass, play an important role in decomposing plant-derived organic compounds and consequently in forming soil organic carbon (8). The process of decomposition contributes to soil respiration and immobilizes carbon into microbial biomass, which plays an important role in soil carbon sequestration. The amount of plant-derived carbon that is respired or incorporated into bacterial biomass is determined by growth efficiencies of the bacterial communities. Bacterial growth efficiency (BGE) is defined as the fraction of the total carbon substrate consumed that is incorporated into new biomass (9). It is calculated from bacterial respiration (BR) and bacterial production (BP). In terrestrial ecosystems, growth efficiency is typically measured for the composite microbial community (which will be referred to as MGE in this dissertation). There are four major methods by which growth efficiency is determined. Each method is based on different assumptions so they are not readily comparable (Table 1)(16). In aquatic environments, the amount of substrate consumed is typically measured as the depletion of dissolved organic carbon (DOC) while microbial biomass is measured as the accumulation of particulate organic carbon (POC) (10). In soil, organic carbon available to microbes is difficult to quantify and is highly dependent on the extraction method (24). Therefore, the amount of substrate consumed in soil cannot be measured directly. Bacteria are known to have strong interactions with soil particles so the heterogeneous matrix of soil also 21 makes measuring bacterial biomass difficult (6). Due to these challenges, investigators often use a single radiolabeled compound to determine the use of substrates (Chapter 1, Tables 1-1 to 1-3). MGE Calculation Parameters measured Substrate consumed CO2 produced Assumptions All substrate consumed that is not respired is YS = (SC – MR)/SC incorporated into biomass. Extracellular compounds are considered as incorporated carbon. Microbial biomass The amount of substrate consumed is equal to YB = MP/(MP+MR) the sum of biomass carbon and respired CO2 produced carbon. Extracellular compounds are not considered as incorporated carbon. Substrate consumed Carbon consumed that is not incorporated into YC = MB/SC Microbial biomass biomass material is respired. Extracellular compounds are not considered as incorporated carbon. Substrate added Added carbon that is not respired is YR = 1-(MR/ST) incorporated into biomass material. CO2 produced Extracellular compounds are considered as incorporated carbon. Table 2-1: Methods for calculating growth efficiency and the corresponding assumptions. SC = Substrate consumed, MR = microbial respiration, MP = microbial production, ST = total substrate. One drawback to using a single radiolabeled compound is that the growth efficiency measured will not be representative of all carbon transformations by the overall bacterial community unless every member is able to transport and metabolize the added substrate. The organic compounds in soil that serve as carbon or energy sources for heterotrophs are a mixture of saccharides, organic acids, amino acids and phenolics (7). The relative abundance of these different compounds can vary with the plant species growing on the soil (29). If the carbon substrate used to measure BGE represents only a small fraction of the total available carbon, the growth efficiency measurement will be a poor measure of overall bacterial activity. An example of the problems associated with using single substrates is provided by observations using radiolabeled glucose. MGE varies with the concentration of glucose applied 22 to the soil (4). At low concentrations of substrate MGE is very high - in some cases even higher than the growth efficiency of pure cultures during unconstrained balanced growth (2, 32). Bremen and Kuikman suggested that MGE was overestimated when the amount of substrate provided was too low because it could be assimilated even though the microbes were unable to initiate growth (3). Later studies demonstrated that when carbon is provided at low concentrations it is stored in the cytoplasm (32). In addition to the affects of varying substrate concentration, the influence of extended incubations on MGE has also been measured. Over a period of 100 days, substrate availability in the soil microcosm decreased, as did the concentrations of other nutrients such as nitrogen and phosphorus. Bacterial community will also change in both physiological state and composition. As a consequence growth efficiency was observed to decrease with incubation time (19). Hill et al. suggest that MGE should be measured shortly after the soil is collected because root exudates such as glucose have extremely short half-lives (20). It is therefore important to have growth efficiency measurements that are reflective of the nutrient status in the soil rather than the added substrate. This is particularly significant when comparing the efficiencies of microbial communities across different sites. It is also important to distinguish the properties of bacteria and fungi as they contribute differently to soil carbon sequestration and respond differently to environmental factors (26, 35). This chapter introduces a method to measure BGE in soil without the addition of exogenous carbon substrate. It is adapted from bacterial productivity measurements in aquatic ecosystems (27). The method measures BP by incorporation of trace amounts of leucine and simultaneously measures BR by the consumption of oxygen (Figure 2-1). 23 Materials and Methods Soil Sampling Soil samples were collected from the W. K. Kellogg Biological Station Long Term Ecological Research main site (KBS LTER, Hickory Corners, MI) in August 2009 and Spring 2010. The site description and soil type are available at http://kbs.msu.edu. Five soil cores of 2.5 cm diameter were collected from the top 10 cm at the conventional agriculture sites with a corn/soybean/wheat crop rotation (T1), deciduous forests (DF), and native grassland (T8). The crop at the conventional agriculture sites was soybean in 2009 and wheat in 2010. The litter layer in the deciduous forest soil was removed prior to sampling. Soil cores from each experimental site were pooled and brought back to the laboratory on ice. The soil was homogenized with a 4 o mm sieve and stored at 12 C until used for experiments. All measurements were made within 48 hours of sampling. Soil pH and moisture were determined prior to the start of an experiment (34). Soil Incubation Field moist soil samples were resuspended in 10 mM MES buffer (1:1 w:v). The buffer was adjusted to the same pH as the soil and supplemented with 50 µg/ml cycloheximide to inhibit fungal growth and 2.3 mM sodium pyrophosphate to assist with dispersion of bacteria. The soil suspension was stirred for 30 minutes at 200 rpm then filtered through 8 layers of cheesecloth followed by a 100 µm cell strainer (BD, Franklin Lakes, NJ) to remove large soil particles. Unlabeled L-leucine (Sigma-Aldrich, MO) was added to 5 ml of the soil slurry to a final concentration of 2 µM. A sample of 550 µl was removed for measuring oxygen 3 consumption prior to addition of radiolabeled leucine. Immediately after H-leucine (250 nCi, Sigma-Aldrich, MO) was added, 50 µl slurry was removed and added to 5 ml scintillation 24 cocktail (Biosafe II, RPI Corp., Mt. Prospect, IL) to determine total radioactivity. To determine leucine saturation kinetics, the concentration of added leucine was varied with different amounts of unlabeled L-leucine. Final concentrations ranged from 0.3 nM to 110 µM. The soil slurry was o incubated for 30 minutes at 25 C while shaking at 100 rpm. To determine the linearity of leucine incorporation, soil slurries with 0.66 µM leucine were sampled over a 1 hour incubation. 3 Figure 2-1: Workflow of measuring BGE for soil samples using H-leucine. TCA: trichloroacetic acid. When BGE was determined using 14 C-acetate, the soil slurry was incubated in a 60 ml serum bottle with a magnetic stir bar. The serum bottle was sealed and 50 µl was sampled for determining total radioactivity immediately following the addition of 14 C-acetate. This was a modification of the method developed by Eichorst for measuring BGE in pure cultures (13). 25 Bacterial Production 3 250 µl of the soil slurry spiked with H-leucine was collected in chilled 2 ml screw-cap tube containing 100% trichloroacetic acid (TCA, final concentration 5% (v/v)) at time zero (T0) and after 30-minute incubation (T30). For measuring the linearity of incorporation the slurry was o also sampled 60 minutes after addition of labeled leucine. The acid mixtures were stored at 5 C 3 for at least 30 minutes to precipitate macromolecules. The method for determining H-leucine incorporation rate was adapted from Bååth, E. et al. (1). Briefly, the TCA-precipitated macromolecules were collected by centrifugation at 11,000 g for 15 minutes and washed with 1 ml 5% TCA followed by 80% ethanol. After the last washing step, the pellets were dissolved in o 0.4 ml 1M sodium hydroxide and incubated in a 90 C water bath for 1 hour with vortexing every 15 minutes. The dissolved pellets were cooled to room temperature before mixing with scintillation cocktail. All supernatant fractions from a sample were pooled and 1 ml was used to determine the amount of radioactivity that had not been incorporated into biomass. Samples were kept in scintillation cocktail overnight in the dark before quantifying the radioactivity with a Beckman Coulter LS 6000TA Liquid Scintillation Counter (Brea, CA), courtesy of Dr. YongHui Zheng. All counts were corrected for background and quenching. Quenching factors were 3 determined by internal-standard method (30). The average recovery of the radiolabeled Hleucine was 85.6% ± 1%. 3 The amount of leucine incorporated into biomass was determined as the H-leucine in 3 TCA precipitate at T30 minus H-leucine in TCA precipitate at T0. The incorporated leucine was 26 converted into the total amount of protein synthesized and total carbon incorporated using the conversion factors suggested by Simon and Azam (38). All leucine incorporated during the 30 minute incubation was assumed to be biological since a formaldehyde-treated soil slurry had no significant incorporation of leucine into TCA precipitate (data not shown). When measured using 14 C-acetate incorporation, TCA (final concentration 5% (v/v)) was added after 30 minutes incubation to stop growth and to acidify the slurry so that the CO2 produced was released into the headspace. The serum bottle was stirred in an ice-bath on a magnetic stir plate to allow simultaneous precipitation of macromolecules. TCA-precipitated samples were collected into chilled 2 ml screw cap tubes after 2 hours of venting with nitrogen gas. The TCA-precipitated slurry was processed as described above. It is assumed that all 14 C label in the TCA-precipitate is carbon incorporated into biomass. Bacterial Respiration Oxygen consumption during the 30-minute incubation was determined by monitoring the oxygen concentration of a 0.5 ml sample in an oxygen chamber with a micro-respirometer (Unisense, Denmark). The amount of oxygen consumed was converted to carbon units by assuming a respiratory quotient (RQ) of 1. Basal RQ for bulk soil incubated in a static condition averages 0.78 and does not exceed 1 when incubated for a short period (11). In this experiment the soil slurries were shaking in ambient oxygen condition as they were incubated to promote aerobic heterotrophy as the major metabolic pathway. Carbon dioxide produced in the soil slurry spiked with 14 C-acetate was released into the headspace of the serum bottle when the slurry was acidified with 5% TCA. The 27 14 CO2 was vented for two hours into three phenethlyamine:methanol (1:1 v/v) base traps arranged in tandem. Triplicate 1 ml trap solution was collected and each added to 5 ml scintillation cocktail to measure 14 CO2 captured. Data Analysis 3 Bacterial production from the incorporation of H-leucine into macromolecules and respiration from oxygen consumption were expressed as the mass of carbon and normalized to soil dry weight and per hour (µg C/g dry soil/hr). BGE was calculated from these two parameters using the formula BGE = BP/(BP+BR). Measurements from soils collected from three experimental sites for each biome were treated as biological replicates except when noted. In the 14 C-acetate method BP and BR were the radioactivity in macromolecule fraction and in base trap, respectively. Differences between methods and samples were assessed using the Student’s t-test. All statistical analyses were performed with R: A Language and Environment for Statistical Computing (39). Results Leucine Incorporation The rate of leucine incorporation increased with leucine concentrations ranging from 0.3 nM to 110 µM (Figure 2-2a). No saturation of leucine incorporation was observed for the range of concentration tested. However, two different patterns of incorporation were observed. At concentrations below 1 µM, the amount of leucine incorporated increased linearly with leucine 28 concentration (slope = 1.23 ± 0.01 µg C/g dry soil/hr/µM leucine). At higher concentrations, the change in leucine incorporation rate per µM leucine concentration decreased by half (slope = 0.06 ± 0.004 µg C/g dry soil/hr/µM leucine) (Figure 2-2a). Michaelis-Menten kinetics fitted to leucine incorporation rates at concentrations below 5 µM suggests that saturation of some leucine transporters can be achieved at concentrations above 1 µM (Figure 2-2b). Leucine incorporation is linear for at least one hour of incubation (Figure 2-3). b) 1.2 10 1 0.1 0.01 0.001 10-­2 10-­1 0.1 1 10 100 Leucine  Concentration  (µM) Leucine  Incorporation  Rate Leucine  Incorporation  Rate a) 1.0 0.8 0.6 0.4 0.2 0.0 0 1 2 3 4 5 Leucine  Concentration  (µM) Figure 2-2. a) Leucine incorporation rate (µg C/g soil/hr) by bacterial communities in grassland soil for leucine concentrations between 0.3 nM and 110 µM. b) Leucine incorporation rate for leucine concentrations below 5 µM fitted to Michaelis-Menton kinetics (Vmax = 1.1 µg C/g soil/hr, Km = 0.97 µM). Each point represents the average of three experimental replicate with one standard deviation as the error bars. Grassland soil was used in this experiment as the trials performed during method development were tested using grassland soils. Bacterial Growth Efficiency BGE was calculated for soil slurries from measurements of leucine incorporated into TCA-precipitable material and oxygen consumption (BGEH). This method was compared to the method using 14 C-acetate (BGEC) for soil slurries prepared from crop rotation and forest soils 29 collected in February and March. BGEC is on average two-fold higher than BGEH from the same sample. BGEC was consistently higher for bacterial communities in forest soils than the communities from the rotation soils (Figure 2-4). The BGEC of rotation soil bacterial communities increased significantly in March, although still lower than in forest soils. On the other hand, BGEH was not significantly different between the two biomes or sampling times. Figure 2-3. Incorporation of leucine over time for grassland soil at KBS LTER when leucine concentration is 660 nM. Each symbol represents average of three biological replicates. The 2 error bar is smaller than the symbols. The linear regression has an r of 0.995. Discussion Measuring growth efficiency Growth efficiency for microbial communities in soil is typically measured using a tracer substrate, such as glucose or acetate (Chapter 1, Tables 1-1 to 1-3). These are substrates common in soil but the total organic carbon available to bacterial soil communities is a much more complicated mixture of saccharides, carboxylic acids, amino acids and the more resistant phenolic compounds (31). Radiolabeled tracer substrate is often favored because it is a direct 30 measure of carbon allocation since no conversion factors are required. Additionally, the use of stable isotopes instead of radioactive compounds for this method can be used to identify bacterial groups that are involved in transforming the particular carbon substrate (18). 0.9   0.8   3H-­Feb   3H-­Mar   14C-­Feb   14C-­Mar   0.7   BGE   0.6   0.5   0.4   0.3   0.2   0.1   0.0   Rotation     Forest   Figure 2-4: Bacterial growth efficiency in rotation and forest soils measured using leucine/oxygen (striped) and 14 3 H- C-acetate (solid) in February and March. Understanding bacterial growth on a single substrate can be important in describing the dynamics of that particular substrate in different environments. For example, forest soils appear to have acetate oxidizers that are more efficient than the ones in the rotation soils (Figure 2-4). However, it is impossible to determine how much a single carbon substrate, when incorporated into bacterial biomass, contributes to overall soil organic carbon stabilization. Our interest is how soil bacteria in different soil types or under different land management influence carbon cycling. Therefore, it was crucial to develop a method that would allow the measurement of growth efficiency on the mixed carbon substrates present in the soils. 31 Bacterial Production 3 The use of H-leucine as a measure of bacterial production was introduced for aquatic systems by Kirchman et al. and later developed by Simon and Azam (27, 38). It is a sensitive measure of bacterial production because bacteria make protein (and hence, incorporating leucine) throughout cell division. It is also advantageous that protein makes up a large fraction of total cell biomass and this fraction is relatively constant. Therefore leucine incorporation can be reliably converted into biomass production (38). Using protein synthesis as a measure of bacterial production avoids the uncertainty of the molecular fate when carbon is assimilated. Leucine specifically reflects carbon that is incorporated into biomass and not carbon assimilated and stored (as when using glucose). Nguyen and Guckert showed that when MGE is calculated from 14 C-glucose, MGE decreased by 24% when only 14 C incorporated into structural biomass (non-fumigation extractable biomass) is taken into account (32). 3 The H-leucine incorporation method was later modified by Baath et al. to measure production of bacteria extracted from soil (1). The extraction step to prepare the bacteria for leucine incorporation is likely to change the physiology of the bacteria. Recovery of bacterial cells might be variable, depending on the interaction between bacterial cells present and their interactions with soil particles. To reduce potential stress on the bacterial community, we eliminated the centrifugation step of Baath et al. and simply made a soil slurry. This further allows the BP measurement to be reflective of the substrates typically available in the soil sample, rather than exogenously provided substrate. 32 The drawback of measuring BP using radiolabeled leucine is that incorporation is dependent on leucine concentration. A saturating concentration is required to assure that the BP measured is not limited by leucine. However, this is challenging as no saturating asymptote was observed for the grassland soil with leucine concentrations up to 110 µM (Figure 2-2). Saturating concentrations for other amino acids such as glycine have been reported to be in the mM range (23). The difficulty in saturating leucine uptake is not surprising because the high diversity of bacteria in soil can include transporters with a wide range of kinetics. The different kinetics observed could be due to changes in the type of leucine transporters that are active. Thus when leucine concentration increases, the high affinity transporters become saturated while the low affinity transporters are activated. E. coli and Pseudomonas aeruginosa have high affinity transporters with Km’s ranging from 0.2 to 0.34 µM (22, 33). Although no complete saturation was observed, biphasic leucine incorporation kinetics indicates that some transporters, such as the high affinity transporters are being saturated. This is consistent with the finding that leucine incorporation rate at low leucine concentrations can be fitted to the Michaelis-Menten function. Biphasic leucine saturation kinetics was previously observed for microbial communities associated with decaying wetland litter and for pure cultures (12, 17). Buesing and Gessner suggested avoiding leucine concentrations in the µM range as they are more likely to promote uptake by fungi (5). Therefore in our study the concentration of leucine used for BP measurement was 2 µM. It was selected to saturate the high affinity transporters and minimize assimilation by fungal communities. It was also expected that most bacteria in soil would be expressing the high affinity transporters because amino acids in soil that are available for bacteria are present in extremely low concentrations (21). Additionally, it was important to use lower concentrations to minimize the 33 use of leucine as carbon and energy source preventing the measured BGE from representing bacterial growth on leucine. To identify appropriate incubation times over which to measure productivity, the kinetics of radiolabeled leucine incorporation were monitored over the course of one hour and shown to increase linearly over this period. However a shorter incubation time of 30 minutes was selected to minimize changes in the composition of the bacterial community (Figure 2-3). This short incubation time also reduces the risk of the community becoming carbon-limited because the carbon made available from homogenizing the soil aggregates can be rapidly decomposed (20). Precipitation of macromolecules using trichloroacetic acid is required to determine the amount of leucine incorporated because up to 50% of the leucine is taken up within minutes of addition but not incorporated into protein (data not shown, (21)). An incubation of 30 minutes is also suitable for measuring oxygen consumption because oxygen was not depleted. Therefore BP and BR measurements could be made simultaneously. Bacterial Respiration Bacterial respiration has typically been measured as CO2 production, but the need for short incubation times (30 minutes) prohibited accurate measures of CO2 production in the soil slurry. Fortunately recent technological developments in oxygen sensors have made direct measurements of oxygen concentration more practical. The Unisense respirometer is highly sensitive and allows measurements to be made in as short as 10 minutes. The drawback to oxygen consumption as a measure of bacterial respiration is the need to know the respiratory quotient (RQ) to convert the amount of oxygen consumed to the amount of carbon respired. 34 RQ for agriculture or forest soils are found to range between 0.77 and 0.8 (11). It approaches 1.0 only when substrate is added in concentrations higher than 100 µg C/g dry soil (11). In this study, the amount of water-soluble organic carbon in the slurry ranged between 223 µg C/g dry soil and 623 µg C/g dry soil. The disruption of soil aggregates during slurry preparation is expected to increase substrate availability when compared to bulk soil. In addition, the soil slurry was incubated shaking, maintaining a well-aerated condition to promote aerobic respiration. Hence, an RQ = 1 was assumed to be reasonable. If the RQ had been assumed to be 0.77 then there would be approximately 6% underestimation of BGE, which is not a significant change (data not shown). Bacterial Growth Efficiency BGEC and BGEH were both calculated using the YC method (Table 2-1). A high BGE close to the theoretical maximum of 0.8 was obtained when using radiolabeled acetate to measure BGE in the forest soil (Figure 2-1). This is unexpected as acetate has less free energy available than glucose or glycerol. BGE measurements for E. coli growing on acetate are approximately half of the BGE when glycerol is supplied as the sole carbon and energy source (data not shown). The correlation between substrate quality and growth efficiency was also observed for soil samples (14). Based on previous short-term soil MGE measurements using radiolabeled glucose, it was expected that BGEC using acetate would be below 0.5 (Chapter 1, Tables 1-1 to 1-3). The high BGEC-acetate obtained in this study is consistent with our hypothesis that added substrate will be taken up and stored instead of mineralized if other nutrients are limiting or the conditions do not support growth (3, 32, 36). 35 3 On the other hand, the lower BGE obtained using the H-leucine/oxygen method is consistent with BGE measurements calculated using ammonium nitrate. Schimel introduced a C:N-dependent method where microbial production is calculated from the immobilization of ammonium nitrate (37). Yet, the BGEH measurements for KBS LTER soils are at the lower range of Schimel’s agriculture soils MGE measurement (Chapter 1, Table 1-1). This is consistent with the prediction that KBS LTER soils are not conducive for efficient growth. Another reason that BGEH is lower than BGEC is that BGEC does not take into consideration use of substrate by bacteria that are not actively synthesizing new biomass. This subgroup of bacteria does not contribute to carbon sequestration but can still be respiring to generate maintenance energy. It is unclear how much microbial respiration is from such metabolically arrested cells, but a large fraction of the bacterial community in soil is expected to 3 be in a non-growing state (28). The H-leucine/oxygen method includes respiration from all bacterial cells in the soil slurry whereas BGEC only accounts for carbon respired from the added acetate. Respiration from non-acetate derived carbon by metabolically arrested cells could explain some of the difference between BGEH and BGEC. As discussed in Chapter 1, growth efficiency is highly dependent on substrate availability and nutrient status. BGEH was measured for soil slurries prepared from bulk soil. This means that the BGE measured is more likely to represent the potential growth efficiency with all the substrates and nutrients available in the soil slurry. In the field, carbon availability is sparse and more heterogeneous. A large fraction of soil carbon is found within small aggregates and is not available to majority of the bacterial community in situ (25). However, if all the soil samples 36 were treated the same way during slurry preparation, differences in BGE measurements are not due to differences in heterogeneity but are due to different active bacterial communities or environmental factors such as nutrient status. Conclusion The use of single radiolabeled substrates is useful for determining the fate of carbon in the substrate or to identify a particular group of decomposers (15, 18). However, it is not a suitable method to measure the overall bacterial growth efficiency in soil. Cross-site comparisons to determine the influence of land management practices require the understanding of the function performed by the composite bacterial communities and in the presence of substrate typically available for growth. Therefore, we suggest that cross-site comparisons of 3 bacterial growth efficiency should be made using the H-leucine/oxygen combination introduced here. 37 REFERENCES 38 REFERENCES 1. Bååth, E., M. Pettersson, and K. H. Sˆderberg. 2001. Adaptation of a rapid and economical microcentrifugation method to measure thymidine and leucine incorporation by soil bacteria. Soil Biology and Biochemistry 33:1571-1574. 2. Blagodatsky, S. A., E. G. Demyanova, E. I. Kobzeva, and V. N. Kudeyarov. 2002. Changes in the efficiency of microbial growth upon soil amendment with available substrates. Eurasian Soil Sci+ 35:874-880. 3. Bremer, E., and P. Kuikman. 1994. Microbial Utilization of C-14[U]Glucose in Soil Is Affected by the Amount and Timing of Glucose Additions. Soil Biol Biochem 26:511517. 4. Bremer, E., and C. Vankessel. 1990. Extractability of Microbial C-14 and N-15 Following Addition of Variable Rates of Labeled Glucose and (NH4)2SO4 to Soil. Soil Biol Biochem 22:707-713. 5. Buesing, N., and M. O. Gessner. 2003. Incorporation of radiolabeled leucine into protein to estimate bacterial production in plant litter, sediment, epiphytic biofilms, and water samples. Microb Ecol 45:291-301. 6. Chenu, C., and G. Stotzky. 2002. Interactions between microorganisms and soil particles: an overview. , p. xvi, 566 p. In P. M. Huang, J. M. Bollag, and N. Senesi (ed.), Interactions between soil particles and microorganisms : impact on the terrestrial ecosystem. John Wiley & Sons, Chichester, West Sussex, England. 7. Dakora, F. D., and D. A. Phillips. 2002. Root exudates as mediators of mineral acquisition in low-nutrient environments. Plant Soil 245:35-47. 8. de Vries, F. T., E. Hoffland, N. van Eekeren, L. Brussaard, and J. Bloem. 2006. Fungal/bacterial ratios in grasslands with contrasting nitrogen management. Soil Biol Biochem 38:2092-2103. 9. del Giorgio, P. A., and J. J. Cole. 2000. Bacterial energetics and growth efficiency., p. 289-325. In D. L. Kirchman (ed.), Microbial ecology of the oceans. Wiley, New York. 10. del Giorgio, P. A., and J. J. Cole. 1998. Bacterial growth efficiency in natural aquatic systems. Annu Rev Ecol Syst 29:503-541. 11. Dilly, O. 2003. Regulation of the respiratory quotient of soil microbiota by availability of nutrients. FEMS Microbiol Ecol 43:375-381. 39 12. Driessen, A. J., S. de Jong, and W. N. Konings. 1987. Transport of branched-chain amino acids in membrane vesicles of Streptococcus cremoris. J Bacteriol 169:5193-5200. 13. Eichorst, S. A. 2007. Isolation and characterization of members of the phylum Acidobacteria from soils. . Michigan State University, East Lansing. 14. Fischer, H., J. Ingwersen, and Y. Kuzyakov. 2010. Microbial uptake of low-molecularweight organic substances out-competes sorption in soil. Eur J Soil Sci 61:504-513. 15. Fischer, H., and Y. Kuzyakov. 2010. Sorption, microbial uptake and decomposition of acetate in soil: Transformations revealed by position-specific C-14 labeling. Soil Biol Biochem 42:186-192. 16. Frey, S. D., V. Gupta, E. T. Elliott, and K. Paustian. 2001. Protozoan grazing affects estimates of carbon utilization efficiency of the soil microbial community. Soil Biol Biochem 33:1759-1768. 17. Gillies, J. E., K. A. Kuehn, S. N. Francoeur, and R. K. Neely. 2006. Application of the [H-3]leucine incorporation technique for quantification of bacterial secondary production associated with decaying wetland plant litter. Appl Environ Microbiol 72:5948-5956. 18. Haichar, F. E., C. Marol, O. Berge, J. I. Rangel-Castro, J. I. Prosser, J. Balesdent, T. Heulin, and W. Achouak. 2008. Plant host habitat and root exudates shape soil bacterial community structure. Isme J 2:1221-1230. 19. Hart, S. C., G. E. Nason, D. D. Myrold, and D. A. Perry. 1994. Dynamics of Gross Nitrogen Transformations in an Old-Growth Forest - the Carbon Connection. Ecology 75:880-891. 20. Hill, P. W., J. F. Farrar, and D. L. Jones. 2008. Decoupling of microbial glucose uptake and mineralization in soil. Soil Biol Biochem 40:616-624. 21. Hobbie, J., and E. Hobbie. 2010. Amino acid cycling in plankton and soil microbes studied with radioisotopes: measured amino acids in soil do not reflect bioavailability. Biogeochemistry:1-22. 22. Hoshino, T. 1979. Transport-Systems for Branched-Chain Pseudomonas-Aeruginosa. J Bacteriol 139:705-712. 23. Jones, D. L., and D. V. Murphy. 2007. Microbial response time to sugar and amino acid additions to soil. Soil Biol Biochem 39:2178-2182. 24. Jones, D. L., and V. B. Willett. 2006. Experimental evaluation of methods to quantify dissolved organic nitrogen (DON) and dissolved organic carbon (DOC) in soil. Soil Biol Biochem 38:991-999. 40 Amino-Acids in 25. Killham, K., M. Amato, and J. N. Ladd. 1993. Effect of Substrate Location in Soil and Soil Pore-Water Regime on Carbon Turnover. Soil Biol Biochem 25:57-62. 26. King, G. M. 2011. Enhancing soil carbon storage for carbon remediation: potential contributions and constraints by microbes. Trends Microbiol 19:75-84. 27. Kirchman, D., E. Knees, and R. Hodson. 1985. Leucine Incorporation and Its Potential as a Measure of Protein-Synthesis by Bacteria in Natural Aquatic Systems. Appl Environ Microbiol 49:599-607. 28. Lennon, J. T., and S. E. Jones. 2011. Microbial seed banks: the ecological and evolutionary implications of dormancy. Nat Rev Microbiol 9:119-130. 29. Meier, C. L., and W. D. Bowman. 2008. Links between plant litter chemistry, species diversity, and below-ground ecosystem function. P Natl Acad Sci USA 105:19780-19785. 30. Mulrooney, S. B., W. A. Wood, and J. R. Paterek. 2007. Physical analysis and purification methods. , p. xix, 1069 p. In C. A. Reddy, T. J. Beveridge, J. A. Breznak, G. A. Marzluf, T. M. Schmidt, and L. R. Snyder (ed.), Methods for general and molecular microbiology, 3rd ed. ASM Press, Washington, D.C. 31. Neumann, G. 2007. Root Exudates and Nutrient Cycling, p. 123-157. In P. Marschner and Z. Rengel (ed.), Nutrient Cycling in Terrestrial Ecosystems, vol. 10. Springer Berlin Heidelberg. 32. Nguyen, C., and A. Guckert. 2001. Short-term utilisation of C-14-[U]glucose by soil microorganisms in relation to carbon availability. Soil Biol Biochem 33:53-60. 33. Rahmania.M, D. R. Claus, and D. L. Oxender. 1973. Multiplicity of Leucine Transport-Systems in Escherichia-Coli K-12. J Bacteriol 116:1258-1266. 34. Robertson, G. P., D. C. Coleman, C. S. Bledsoe, and P. Sollins. 1999. Standard soil methods for long-term ecological research. Oxford University Press, New York. 35. Rousk, J., and E. Baath. 2007. Fungal and bacterial growth in soil with plant materials of different C/N ratios. Fems Microbiology Ecology 62:258-267. 36. Sawada, K., S. Funakawa, and T. Kosaki. 2009. Threshold concentrations of glucose to increase the ratio of respiration to assimilation in a Japanese arable soil and a strongly acidic Japanese forest soil. Soil Sci Plant Nutr 55:634-642. 37. Schimel, D. 1988. Calculation of microbial growth efficiency from 15 N immobilization. Biogeochemistry. 38. Simon, M., and F. Azam. 1989. Protein-Content and Protein-Synthesis Rates of Planktonic Marine-Bacteria. Mar Ecol-Prog Ser 51:201-213. 41 39. Team, R. D. C. 2008. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. 42 CHAPTER 3 BACTERIAL GROWTH EFFICIENCY IN SOILS UNDER DIFFERENT LAND MANAGEMENT PRACTICES Introduction Soil bacteria play a central role in transformations of soil organic carbon (SOC) and nutrient cycling. Carbon inputs, in the form of root exudates and litter, are decomposed and transformed into bacterial biomass. Bacteria produce metabolites and extracellular enzymes that assist with humification and aggregation, both of which increase the protection of soil organic matter from further decomposition (35). Bacterial biomass continues to contribute to the stabilization of soil organic matter even after the cells die, as biomass can serve as a substrate for humification (24). Understanding the involvement of microbes in soil carbon sequestration is becoming more important as there is increasing evidence that biological factors strongly influence the dynamics of dissolved organic carbon (18, 23). Soil heterotrophic bacteria depend on carbon input from plants as both carbon and energy sources. Therefore transformation of carbon from plants into bacterial biomass is not complete because some of the carbon is respired into the atmosphere as CO2. When a change in CO2 production is observed, it could represent a change in the production of plant-derived carbon, the rate of decomposition of plant compounds or a change in growth efficiency. Bacterial growth efficiency (BGE) is a measure of the fraction of consumed carbon that is being incorporated into new biomass (Chapter 2, (9)). It is an informative measure of the physiological state of the bacterial community and its ability to stabilize labile organic carbon. In pure culture studies, BGE was found to be sensitive to both substrate quality and availability, and can rapidly decrease at low substrate concentrations (38). BGE can also be influenced by innate processes of 43 the bacterial community, which are dictated by the genetic makeup of its constituents (32). Different bacterial species can have different energy requirements for both growth and nongrowth purposes, and can use catabolic pathways that yield different amounts of energy. Even though BGE provides valuable information, measurements in terrestrial ecosystems have been sparse. Comparisons across studies are challenging because many use different methods and assumptions (2, 17, 25, 30). One particularly common complication is that growth efficiency measurements vary with the concentrations of labeled substrates used in the experiment. Additionally, growth efficiency in soil is often measured for the total microbial community consisting of both bacteria and fungi. Yet, it is essential to elucidate the function of these communities separately because bacteria and fungi respond differently in soils under different land managements (21, 28). Fungi are typically assumed to be more efficient than bacteria, but agricultural soils with different fungal:bacterial biomass ratios were found to have similar growth efficiency (40). It is also important to distinguish the role of bacteria and fungi if the objective is to link community dynamics to the changes in function of bacteria and fungi in carbon transformations. A survey of microbial growth efficiency in terrestrial ecosystems suggests that it is sensitive to temperature, soil type, substrate quality and availability, but not fungal:bacterial ratio (7, 11, 14, 34, 40). However, growth efficiency is still widely assumed to be constant for terrestrial biomes (26, 31). Focusing on bacterial communities, we tested the assumption that BGE is constant for soils under different land managements. Bacterial biomass production and respiration were measured without the addition of exogenous substrate to calculate the potential growth efficiency of bacterial communities in soil collected from a forest and three differently managed croplands at the Kellogg Biological Station Long Term Ecological Research (KBS 44 LTER) site. The sensitivity of BGE to resource availability was determined by laboratory incubation of soil from tilled, corn/soybean/wheat rotation cropland with different nutrient amendments. Materials and Methods Soil Sampling Soil samples were collected from the W. K. Kellogg Biological Station Long Term Ecological Research main site (KBS LTER, Hickory Corners, MI) in 2010. The site description and soil characteristics are available at http://kbs.msu.edu. Five soil cores of 2.5 cm diameter were collected from the top 10cm at the conventional agriculture sites with corn/soybean/wheat crop rotation (T1), deciduous forests (DF), soybean monoculture cropland (B19) and a cultivated land that was left fallow (B21). The crop at the conventional agriculture sites when the soil was sampled was wheat. The litter layer in the deciduous forest was removed prior to sampling. Soil cores collected from each experimental site were pooled and brought back to lab on ice. The o soils were homogenized through a 4mm sieve and stored at 12 C until used for experiments. All measurements were made within 48 hours of soil sampling. Soil pH and moisture were o determined prior to the start of experiments (27). Soil samples were also frozen at -80 C for chemical analysis. Soil Chemistry Water-soluble dissolved organic carbon (DOC) and total dissolved nitrogen (DN) were extracted using a protocol modified from Jones and Willett (19). Briefly, frozen soil samples were suspended in sterile water (1:2 w/v) supplemented with sodium pyrophosphate. The soil 45 slurry was shaken at 200 rpm for 30 minutes followed by centrifugation at 8000g for 10 minutes. o All steps were performed at 4 C to reduce decomposition. The supernatant was filtered through 0.45µm syringe filter to remove particulate organic matter, including most bacteria. DOC and DN were measured using a Shimadzu TOC-TN analyzer (Shimadzu, Columbia, MD), courtesy of Dr. Steve Hamilton at KBS. Bacterial Growth Efficiency 3 BGE was measured according to the H-leucine/oxygen method described in Chapter 2. Microcosm Experiments Field-moist soil samples from rotational croplands were mixed with talc that was saturated with buffer or amendment solutions (10 µl solution/50 mg talc/g soil). The amendment solutions were labile carbon mixture, inorganic nitrogen, carbon and nitrogen mixture, or R2 broth (Difco, Detroit, MI). The labile carbon solution contains a combination of glucose, fructose, sucrose, lactic acid, citric acid, succinic acid, and acetic acid. It is a modification of the artificial root exudate solution used by Baudoin et al. (3). The saccharides provided twice the amount of carbon supplied by the carboxylic acids, providing a total of 250 µg C/g dry soil. Nitrogen was added at a final concentration of 25 µg N/g dry soil. The carbon and nitrogen mixture has a C:N ratio of 10. R2 broth was applied at 100 µl/g dry soil, providing 342 µg C and o 19 µg N per g of dry soil. The microcosms were incubated in the dark at 25 C for 24 hours. Soil slurries were prepared from these microcosms and BP and BR measurements were made as 46 described in Chapter 2. Soils were collected at three different experimental sites of rotation cropland to prepare three microcosms for each treatment. Data Analysis 3 The incorporation rate of H-leucine into macromolecules was used as a measure of bacterial production (BP), while bacterial respiration (BR) was based on measurement of oxygen consumption rate. Both rate measurements were converted to mass of carbon and normalized to soil dry weight and time of incubation (µg C/g dry soil/hr). Bacterial Growth Efficiency (BGE) was calculated from these two parameters using the formula BGE = BP/(BP+BR). Soils collected from three different experimental sites for each biome were treated as independent biological replicates of each management practice. ANOVA was used to assess treatment effects while the effect of seasonality was assessed by repeated measures ANOVA. Pair-wise comparisons between treatments were evaluated by Tukey’s HSD. For the treatment effect, only measurements that were performed on the same dates (June, August and September) were included in the analysis. Statistical significance is recognized at p-value ≤ 0.1. All statistical analyses were performed with R: A Language and Environment for Statistical Computing (37). Results BGE in field samples Bacterial production (BP) and respiration (BR) were measured for soil slurries prepared from soil sampled from four different land managements. Although archaea could be contributing to the measured production and respiration, archaea contributes only 1.4% of the 47 total rRNA sequences in KBS LTER soils (33). Generally, a positive correlation was observed 2 between BR and BP (Fig. 3-1a, r = 0.20, p-value < 0.001). However, only 20% of the variation in BR could be explained by variation in BP. Forest and soybean monoculture cropland soils have similar respiration rates but the bacterial community from soybean monoculture soils synthesized more biomass (data not shown). The measured BP and BR were used to calculate bacterial growth efficiency. BGE varies almost 3-fold, from 0.23 to 0.63. Soils under different land managements have bacterial communities that exhibit different BGE (Figure. 3-1b, ANOVA p-value = 0.001). With a median BGE of 0.52, the bacterial communities in soybean monoculture cropland soils are the most efficient, allocating up to 63% of substrate carbon into biomass. The median BGE is at least 9% more than the other three biomes. Bacterial communities in fallow soils have the lowest bacterial production (0.84 µg C/g dry soil/hr) but with a median BGE of 0.43. These deciduous forest soils have bacterial communities with a BGE of 0.37, which is below the commonly assumed growth efficiency of 0.45 in soil (26). BGE in rotation cropland soils is the most variable, ranging from 0.24 to 0.59 (Figure. 31b). Some of this variability is due to a seasonal effect (Figure. 3-2, RM-ANOVA p-value < 0.001). All three experimental sites of rotation cropland soils exhibit a similar trend, where two peaks of BGE were observed. The first peak occurred in May, during active wheat growth, and the second peak was observed in September, one and half months after the wheat crop was harvested and the field was left bare. This seasonal variation was not observed for forest soils, which stayed constant throughout the year (Figure. 3-2). 48 b) 0.7 3.5 b 3.0 a 0.6 2.5 2.0 BGE Bacterial  Respiration  (µg  C/g  dry  soil/hr) a) 1.5 0.5 ab a 0.4 1.0 0.3 0.5 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Bacterial  Production  (µg  C/g  dry  soil/hr) 0.2 Forest Rotation Fallow Soybean Figure 3-1. a) The linear relationship between bacterial production and respiration for soil samples from four different biomes and at different times of the year. The solid line represents 2 the linear fit, BR = 0.51*BP + 1.19, r = 0.20, p-value < 0.001. b) Boxplot showing the variation of BGE in the four different biomes from June through September. Different letter indicates significant difference from post-hoc Tukey HSD test with a p-value ≤ 0.1 (n=9). 0.6 BGE 0.5 0.4 0.3 0.2 0.1 Apr Jun Aug Oct Dec Figure 3-2. Seasonal variation of BGE in rotation cropland (dashed line) and deciduous forest (solid line). Each point represents three experimental replicates and error bars are standard error. BGE in microcosm amendments To ascertain the influence of substrate availability on BGE, measurements were made of BP and BR for 24-hour microcosms prepared from rotation cropland soil. Carbon amendments of 49 easily oxidizable saccharides and carboxylic acids commonly found in root exudates did not stimulate BP. Instead, it increased respiration rate almost two-fold, which led to a significant decrease in BGE from 0.49 in the unamended control to 0.40 (Figure 3-3). Nitrogen addition by itself had no influence on BP, BR or BGE (Figure 3-3). When carbon was added in combination with inorganic nitrogen there was a simultaneous increase in both BP and BR such that BGE remained the same as in the unamended control. A microcosm with R2 broth, a rich medium commonly used to isolate heterotrophs from the environment, was the only microcosm that exhibited increased BGE compared to the control (Figure 3-3b). Bacterial  Activity (%  Unamended  Control) a) 200 BGE (%  Unamended  Control) * 150 100 ** ** ** 50 0 b) BP BR 30 20 10 0 -­10 -­20 -­30 Carbon Nitrogen C:N R2B * * Carbon Nitrogen C:N R2B Figure 3-3. Relative a) Bacterial production, bacterial respiration and b) BGE for microcosms amended with carbon, inorganic nitrogen, carbon and inorganic nitrogen mixture or R2 broth. All measurements are normalized to the unamended microcosms by percentage difference. Asterisks represent significant difference from the unamended microcosm, where ‘*’ has p-value ≤ 0.1 and ‘**’ has p-value ≤ 0.05 (Student’s t-test). 50 Discussion 3 Using the H-leucine/oxygen method, BGE in terrestrial ecosystems varied three-fold, from 0.23 to 0.63 (Figure 3-1). It is acknowledged that the BGE measurements more likely represent potential growth efficiency of the bacterial community than efficiency in situ because they were made in a soil slurry instead of bulk soil. In the field, carbon availability is lower and less homogeneous; a large fraction is found within aggregates and is not available to a majority of the bacterial community (20). Slurry preparation will disrupt soil aggregates and allow some of the resources in the bulk soil to become more accessible. However, all the soil samples were treated the same way during slurry preparation and therefore differences in BGE are due to differences either in the bacterial community from the different soil treatments (further discussed in Chapter 4) or in environmental factors between soil samples, such as total available carbon and nutrients. Measurements of BGE in KBS LTER soils are within a narrower range than in aquatic environments, where BGE ranges from 0.025 to 0.81, (9). All four biomes in this study are rich in carbon compared to oligotrophic lakes with their very dilute concentration of dissolved organic carbon. Even for the relatively low carbon input fallow soils, BGE is maintained in the range of 0.26 to 0.46 (Figure 3-1). In the oligotrophic aquatic environment where substrate availability is low, bacteria express a plethora of catabolic genes to scavenge any substrate as soon as it becomes available. This is the “generalist” approach to surviving in low substrate environments (13). This can be costly because more energy must be available for synthesis of proteins and maintaining an active, responsive state. In soil, substrate diffusion is slower, allowing time for catabolic enzymes to be synthesized only when the substrate is made available. 51 Additionally, a large fraction of bacteria in soil is likely to be in a metabolically arrested state, which requires less maintenance energy than an active cell (22). More than half of the soil samples analyzed exhibited median growth efficiencies lower than the commonly assumed microbial growth efficiency of 0.41 to 0.45 (42). In fact, all but one sample in the forest soils had BGE less than 0.45. Unless the growth efficiency of the fungal community is much higher than the bacterial community, an assumption of 0.45 overestimates the efficiency of carbon transformation by microbes in forest soil. The fungal community is generally assumed to be more efficient than the bacterial community, due to higher structural biomass in fungi, but there is insufficient evidence to suggest that this is always the case (40). Furthermore, actual BGEs in bulk soil are expected to be lower than the potential BGEs measured in soil slurries in these experiments. The large variation in BGE across all samples indicates that biomass synthesis and energy synthesis activity are not tightly coupled in soil bacteria. The community is likely to be either energy or nutrient starved, as expected for environmental samples where growth substrates are often limiting. When carbon substrates do become available, the stoichiometry of nutrients might not be ideal for increased productivity but it can still be assimilated and mineralized. This is often observed when high C:N ratio substrates, such as root exudates, are released into the soil (15). The mineralization of carbon without increased biomass synthesis can be explained by maintenance energy requirement or by energy spilling reactions. Energy spilling reactions are more likely to explain the short-lived increase in respiration observed for soil bacteria flushed with high concentration of carbon substrates. Without nutrients such as nitrogen and phosphorus, the ability to make biomass is still limited. But to maintain an active state, the cells will have to recycle reducing agents such as NADH and hence consume 52 oxygen and produce CO2 (29). This is especially important for those soil bacteria that are unable to store carbon sources (43). On the other hand, a large fraction of bacteria in the soil are in a metabolically arrested state (22). With the exception of spore-formers, bacteria that are in this state still require energy sources for non-growth purposes (maintenance energy) (1, 41). The low growth efficiency for forest soil is in agreement with the stoichiometry of microbial biomass in forest soil. Forest microbial biomass tend to have high C:N and N:P ratio, indicating phosphorus limitation, which will require the highest maintenance energy (8, 39). The microcosm study with rotation cropland soil suggests that biomass synthesis is not carbon limited because carbon amendment does not increase productivity (Figure 3-3a). In fact, we propose that the C:N ratio is a more important driver of BGE in terrestrial ecosystems than carbon availability, which is a major driver of BGE in aquatic ecosystems (10). The increased BGE observed for soybean monoculture cropland is correlated with increased nitrogen content in soil during growth of soybean (36). When BGE was measured for the rotation cropland during soybean growth, a BGE of 0.46 ± 0.05 was observed. BGE of this magnitude was observed during wheat growth only in May, when it was 0.48 ± 0.02. This peak of BGE in wheat coincides with two applications of nitrogen fertilizer during the week before the May sample was collected. Our hypothesis is supported by the observation that the seasonal variation in BGE is driven by changes in C:N ratio (Figure 3-4a). The variation in C:N ratio explains 27% of BGE variation in forest and rotation cropland soils (Figure 3-4b). The lack of response when nitrogen is added individually to microcosms indicates that the bacterial community is not nitrogen limited. However evidence for synergistic effects of nitrogen supplementation and carbon addition was found in the microcosm study. The addition of nitrogen with carbon at a C:N ratio of 10 increases bacterial production by two-fold, even 53 though the same amount of carbon was mineralized as in the microcosm amended with carbon only (Figure 3-3). It appears that the microcosms are nitrogen and carbon co-limited. This means that increased sequestration of carbon into bacterial biomass can be promoted only if increased carbon availability is accompanied by supplementation with other nutrients required for biomass synthesis. a)12 11 Rotation Forest b) 0.7 0.6 9 8 0.5 BGE C:N  Ratio 10 0.4 0.3 7 6 0.2 5 Mar AprMay Jun Jul Aug Sep Oct 0.1 5 6 7 8 9 C:N  Ratio 10 11 Figure 3-4. a) Seasonal variation of C:N ratio in forest and rotation cropland soils. Each point represents three experimental replicates and error bars represent standard error. b) The linear correlation between C:N ratio and BGE in forest and rotation cropland soils collected in March through September. The solid line represents the linear regression, BGE = - 0.04(C:N ratio) + 2 0.68, r = 0.27, p-value = 0.001. Soil bacteria are often described as carbon starved because carbon addition often leads to increased CO2 production (12). The extra mineralized carbon comes not only from the added substrate but also from older SOC (16). This phenomenon is also known as the priming effect (6). The use of CO2 production as a measure of bacterial activity does not differentiate between increased growth (activity) and decreased efficiency. Our BGE measurements demonstrate that 54 the increased CO2 production is not due to increased growth but is instead caused by a decreased efficiency of carbon transformation (Figure 3-3). A priming effect associated with decreased growth efficiency will eventually lead to loss of soil carbon, as there is no replenishing of the older organic carbon by increasing biomass synthesis. Growth efficiency will decrease if the added carbon stimulates the bacterial community to tap into older soil organic matter for nitrogen or an energy source to synthesize biomass from the added carbon (5). Therefore, carbon addition by itself, either through increased plants per unit area, or by CO2 fertilization, will not necessarily lead to increased carbon sequestration in soil. Conclusion The assessment of BGE in four terrestrial biomes at KBS LTER sites showed that it is not constant. Bacterial communities in soils under different land managements have different growth efficiencies. This work suggests that C:N ratio is the major regulator of BGE in soil. Understanding BGE dynamics can help clarify how soil carbon transformation is being affected by changes in land management practices. Further investigation into carbon availability and nutrient requirements will have to be performed to determine what C:N ratio will support high growth efficiency, promoting bacterial-driven carbon sequestration in soil (4). Acknowledgements I would like to acknowledge Dr. Clegg Waldron and Ben Roller for helpful discussions on BGE in soil. I would also like to thank David Weed for helping with DOC and DN measurements. 55 REFERENCES 56 REFERENCES 1. Anderson, T. H., and K. H. Domsch. 2010. Soil microbial biomass: The ecophysiological approach. Soil Biol Biochem 42:2039-2043. 2. Barros, N., S. Feijoo, J. A. Simoni, C. Airoldi, B. Ramajo, A. Espina, and J. R. Garcia. 2008. A mass and energy balance to provide microbial growth yield efficiency in soil - Sensitivity to metal layering phosphates. J Therm Anal Calorim 93:657-665. 3. Baudoin, E., E. Benizri, and A. Guckert. 2003. Impact of artificial root exudates on the bacterial community structure in bulk soil and maize rhizosphere. Soil Biol Biochem 35:1183-1192. 4. Billings, S. A., J. Lichter, S. E. Ziegler, B. A. Hungate, and D. D. Richter. 2010. A call to investigate drivers of soil organic matter retention vs. mineralization in a high CO2 world. Soil Biol Biochem 42:665-668. 5. Billings, S. A., and S. E. Ziegler. 2008. Altered patterns of soil carbon substrate usage and heterotrophic respiration in a pine forest with elevated CO2 and N fertilization. Global Change Biol 14:1025-1036. 6. Bird, J. A., D. J. Herman, and M. K. Firestone. 2011. Rhizosphere priming of soil organic matter by bacterial groups in a grassland soil. Soil Biology and Biochemistry 43:718-725. 7. Blagodatsky, S. A., E. G. Demyanova, E. I. Kobzeva, and V. N. Kudeyarov. 2002. Changes in the efficiency of microbial growth upon soil amendment with available substrates. Eurasian Soil Sci+ 35:874-880. 8. Cleveland, C., and D. Liptzin. 2007. C:N:P stoichiometry in soil: is there a “Redfield ratio” for the microbial biomass? Biogeochemistry 85:235-252. 9. del Giorgio, P. A., and J. J. Cole. 1998. Bacterial growth efficiency in natural aquatic systems. Annu Rev Ecol Syst 29:503-541. 10. del Giorgio, P. A., J. J. Cole, and A. Cimbleris. 1997. Respiration rates in bacteria exceed phytoplankton production in unproductive aquatic systems. Nature 385:148-151. 11. Devevre, O. C., and W. R. Horwath. 2000. Decomposition of rice straw and microbial carbon use efficiency under different soil temperatures and moistures. Soil Biol Biochem 32:1773-1785. 57 12. Duah-Yentumi, S., R. Ronn, and S. Christensen. 1998. Nutrients limiting microbial growth in a tropical forest soil of Ghana under different management. Appl Soil Ecol 8:19-24. 13. Egli, T. 2010. How to live at very low substrate concentration. Water Res 44:4826-4837. 14. Fischer, H., J. Ingwersen, and Y. Kuzyakov. 2010. Microbial uptake of low-molecularweight organic substances out-competes sorption in soil. Eur J Soil Sci 61:504-513. 15. Fontaine, S., G. Bardoux, L. Abbadie, and A. Mariotti. 2004. Carbon input to soil may decrease soil carbon content. Ecol Lett 7:314-320. 16. Fontaine, S., G. Bardoux, D. Benest, B. Verdier, A. Mariotti, and L. Abbadie. 2004. Mechanisms of the priming effect in a savannah soil amended with cellulose. Soil Sci Soc Am J 68:125-131. 17. Herron, P. M., J. M. Stark, C. Holt, T. Hooker, and Z. G. Cardon. 2009. Microbial growth efficiencies across a soil moisture gradient assessed using C-13-acetic acid vapor and N-15-ammonia gas. Soil Biol Biochem 41:1262-1269. 18. Jiao, N., G. J. Herndl, D. A. Hansell, R. Benner, G. Kattner, S. W. Wilhelm, D. L. Kirchman, M. G. Weinbauer, T. W. Luo, F. Chen, and F. Azam. 2010. Microbial production of recalcitrant dissolved organic matter: long-term carbon storage in the global ocean. Nat Rev Microbiol 8:593-599. 19. Jones, D. L., and V. B. Willett. 2006. Experimental evaluation of methods to quantify dissolved organic nitrogen (DON) and dissolved organic carbon (DOC) in soil. Soil Biol Biochem 38:991-999. 20. Killham, K., M. Amato, and J. N. Ladd. 1993. Effect of substrate location in soil and soil pore-water regime on carbon turnover. Soil Biology and Biochemistry 25:57-62. 21. King, G. M. 2011. Enhancing soil carbon storage for carbon remediation: potential contributions and constraints by microbes. Trends Microbiol 19:75-84. 22. Lennon, J. T., and S. E. Jones. 2011. Microbial seed banks: the ecological and evolutionary implications of dormancy. Nat Rev Microbiol 9:119-130. 23. Liang, C., and T. C. Balser. 2011. Microbial production of recalcitrant organic matter in global soils: implications for productivity and climate policy. Nat Rev Microbiol 9:-. 24. Liang, C., G. Cheng, D. Wixon, and T. Balser. 2010. An Absorbing Markov Chain approach to understanding the microbial role in soil carbon stabilization. Biogeochemistry:1-7. 58 25. Nguyen, C., and A. Guckert. 2001. Short-term utilisation of C-14-[U]glucose by soil microorganisms in relation to carbon availability. Soil Biol Biochem 33:53-60. 26. Parton, W. J., D. S. Schimel, C. V. Cole, and D. S. Ojima. 1987. Analysis of Factors Controlling Soil Organic-Matter Levels in Great-Plains Grasslands. Soil Sci Soc Am J 51:1173-1179. 27. Robertson, G. P., D. C. Coleman, C. S. Bledsoe, and P. Sollins. 1999. Standard soil methods for long-term ecological research. Oxford University Press, New York. 28. Rousk, J., and E. Baath. 2007. Fungal and bacterial growth in soil with plant materials of different C/N ratios. Fems Microbiology Ecology 62:258-267. 29. Russell, J. B. 2007. The energy spilling reactions of bacteria and other organisms. J Mol Microb Biotech 13:1-11. 30. Schimel, D. 1988. Calculation of microbial growth efficiency from 15 N immobilization. Biogeochemistry. 31. Schimel, J. P., and M. N. Weintraub. 2003. The implications of exoenzyme activity on microbial carbon and nitrogen limitation in soil: a theoretical model. Soil Biol Biochem 35:549-563. 32. Schmidt, T., and A. Konopka. 2009. Physiological and Ecological Adaptations of SlowGrowing, Heterotrophic Microbes and Consequences for Cultivation, p. 101-120. In S. S. Epstein (ed.), Uncultivated Microorganisms, vol. 10. Springer Berlin / Heidelberg. 33. Schmidt, T. M., D. H. Buckley, and J. R. Graber. 1998. Phylogenetic analysis of nonthermophilic members of the kingdom Crenarchaeota and their diversity and abundance in soils. Appl Environ Microbiol 64:4333-4339. 34. Schneckenberger, K., D. Demin, K. Stahr, and Y. Kuzyakov. 2008. Microbial utilization and mineralization of [C-14]glucose added in six orders of concentration to soil. Soil Biol Biochem 40:1981-1988. 35. Six, J., S. D. Frey, R. K. Thiet, and K. M. Batten. 2006. Bacterial and fungal contributions to carbon sequestration in agroecosystems. Soil Sci Soc Am J 70:555-569. 36. Smith, R. G., and K. L. Gross. 2007. Assembly of weed communities along a crop diversity gradient. J Appl Ecol 44:1046-1056. 37. Team, R. D. C. 2008. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. 38. Tempest, D. W. 1978. Biochemical Significance of Microbial-Growth Yields Reassessment. Trends Biochem Sci 3:180-184. 59 39. Tempest, D. W., and O. M. Neijssel. 1984. The Status of Yatp and Maintenance Energy as Biologically Interpretable Phenomena. Annu Rev Microbiol 38:459-486. 40. Thiet, R. K., S. D. Frey, and J. Six. 2006. Do growth yield efficiencies differ between soil microbial communities differing in fungal: bacterial ratios? Reality check and methodological issues. Soil Biol Biochem 38:837-844. 41. van Bodegom, P. 2007. Microbial maintenance: A critical review on its quantification. Microb Ecol 53:513-523. 42. Voroney, R. P., and E. A. Paul. 1984. Determination of Kc and Kn Insitu for Calibration of the Chloroform Fumigation Incubation Method. Soil Biol Biochem 16:914. 43. Wang, J. G., and L. R. Bakken. 1998. Screening of Soil Bacteria for Poly-betaHydroxybutyric Acid Production and Its Role in the Survival of Starvation. Microb Ecol 35:94-101. 60 CHAPTER 4 rRNA OPERON COPY NUMBER AS A GENETIC MARKER FOR ECOLOGICAL STRATEGIES Some of these results have been published in the article: Lee, Z. M. P., C. Bussema, and T. M. Schmidt. 2009. rrnDB: documenting the number of rRNA and tRNA genes in bacteria and archaea. Nucleic Acids Res 37:D489-D493. Introduction Soil harbors a high diversity of bacteria with more than 10,000 species in a gram of soil (35). The structure of bacterial communities in soil has been well characterized using cultureindependent molecular surveys. Molecular analyses have allowed scientists to glimpse into the diversity of bacteria in various environments and to observe how the community responds to perturbations (17). This is especially important because we have yet to culture representative strains from more than 90% of bacterial species in soil (18). A particular molecule that has been exploited as a phylogenetic marker to identify bacteria in the environment is the 16S rRNA (31). Along with 21 ribosomal proteins, the 16S rRNA makes the 30S small subunit of the ribosome. Due to the central role of ribosomes in protein synthesis, the rRNA gene sequence has regions that are highly conserved interspersed with variable region. A variable region of the rRNA gene can be amplified using primers annealing to the flanking conserved regions. Sequence analysis of the variable region based on databases of known bacterial species, such as Ribosomal Database Project and SILVA, will allow identification of the bacterial genome that harbors the particular sequence (8, 33). Surveys of genes encoding for central enzyme in specialized functions like denitrification and methane oxidation have allowed the functions of various groups of bacteria to be defined 61 (23). Soil carbon transformation or CO2 production on the other hand, is a general function with a high degree of functional redundancy (3). This makes it challenging to link the structure of the bacterial community to general function and it is often found that changes in the function (ie. CO2 production) is not correlated with shifts in the community composition (15). There is no single gene that can be targeted as a marker for CO2 production because different bacteria can process substrates using multiple different pathways. Consequently, perturbations of soil microcosms, or survey of the terrestrial ecosystem is often based on community composition determined by 16S rDNA sequence analysis. One aspect of 16S rDNA surveys that is often overlooked is the presence of multiple genes per genome that encode the rRNAs. The 16S rRNA genes, along with the 23S and 5S rRNA genes, have a unique characteristic relative to other bacterial genes. It can be present from one to 15 copies per genome (21). The multiplicity of rRNA genes has been recognized as a caveat in molecular analyses targeting rRNA genes (19). It is a hindrance in quantifying relative abundance of different sequences. A bacterium with seven copies of 16S rRNA genes per genome will appear to be more abundant than a bacterium with two copies of the gene, even if the two species were present in equal abundance. The copy number effect becomes more influential for environments dominated by bacteria with different rRNA gene copy numbers. However, with increasing characterization of the rRNA operon copy number of ecologically relevant bacteria, their abundance can be better described. This has led to the compilation of rrnDB, an online database cataloging rRNA genes copy number in bacteria and archaea (http://rrndb.mmg.msu.edu/, (22)). 62 Albeit a concern, rRNA genes copy number is a potential genetic marker for understanding the ecological strategies of bacteria. This chapter explores the distribution of rRNA genes in bacteria and discusses the characteristics of bacteria with different rRNA genes copy number. Using molecular survey of bacterial community in rotation cropland and forest soils at the KBS LTER site, we used rRNA genes copy number to make hypothesis regarding the ecological strategies of the bacterial communities. Abundance of rrn operons in Bacteria Ribosomal RNA genes in bacteria are typically arranged in an operon, the rrn operon. The first gene is the 16S rDNA (rrs), followed by 23S rDNA (rrl) and 5S rDNA (rrf) with an internal transcribed spacer (ITS) region between rrs and rrl (6). There are five different groups of ITS, some of which can encode up to four tRNA genes (41). The intragenomic heterogeneity of the ITS region also makes it a suitable target for culture-independent identification when the rDNA heterogeneity does not provide sufficient resolution such as at the subspecies level (43). While 5’-rrs-rrl-rrf-3’ is the general operon structure, there are various exceptions. The second most commonly found operon structure is the separation of the rrs gene from the rrl-rrf genes cluster. This structure is common in intracellular bacteria, such as members of the class Rickettsiales, and other host-associated bacteria such as species in the genus Helicobacter (25, 28). Members of the family Brachyspiraceae and Leptospiraceae in the phylum Spirochaetes have three separate promoters for each of the rRNA genes. All of the species with unusual rrn operon structure has either one or two sets of rRNA genes. Bacterial species with one or two copies of rrn operon dominate the current catalogue of rrn operon copy number (22). Bacteria with eight copies or more only makes up 10.3% of unique 63 species in the catalog (Figure 4-1). It consists of members of the phylum Firmicutes and class Gammaproteobacteria with only one member from the Betaproteobacteria class (Figure 4-2). Two well-defined genera that belong exclusively in this group are Shewanella and Vibrio. There are currently only two species with 15 rrn operons – Clostridium paradoxum and Photobacterium profundum (34, 45). Most genera have members with the same rrn operon copy number or differ by one. However, some genera such as Lactobacillus and Clostridium can have a wider range of rrn operon copy number (Table 4-1). It is important to note that the rrn operon copy number distribution is highly biased towards bacteria with fully sequenced genomes. It does not necessarily mean that the domain Bacteria is dominated with low rrn operon copy number members. Genus rrs copies Lactobacillus 1-9 Geobacter, Gluconacetobacter, Pelobacter 2-4 Bifidobacterium 2-5 Aquifex 2/6 Rhodobacter 3-4 Ruegeria 3/5 Corynebacterium, Ralstonia 3-6 Clostridium 3-15 Burkholderia 4-6 Bacteroides, Pseudomonas, Streptococcus 4-7 Acinetobacter, Methylobacterium 5-7 Pseudoalteromonas 5/9 Bacillus 7-14 Alkaliphilus, Desulfotomaculum 8-10 Shewanella 8-12 Paenibacillus, Vibrio 8-13 Table 4-1: Genera with more than two rrs copy number. Archaeal genomes have a smaller range of rrn operon copy number, with only 1 to 4 copies per genome. Members with one rrn operon dominate the catalogue, constituting 60% of the archaea in the rrnDB (22). Archaea with more than one rrn operon are all from the phylum 64 Euryarchaeota. Archaeal genomes are also often found to have the 5’-rrs-rrl-3’ genes cluster encoded separately from rrf. This unusual operon is found in all known Thermoplasma genomes. 160   Number  of  unique  species   140   120   100   80   60   40   20   0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16S  rRNA  gene  copy  number   Figure 4-1: Distribution of 16S rRNA gene copy number in bacteria. The analysis was performed on 652 unique bacterial species (gray bars). In molecular surveys of 16S rRNA gene, two sequences with a distance of more than 3% difference are often defined to be in separate operational taxonomic units (OTUs). This is often seen as equivalent to different species although the exact definition of species is still debatable (39). When multiple rrn operons are present in a genome, it is common to find some sequence differences (1). However, intragenomic heterogeneity of the 16S rRNA gene is typically below 1% (1). The highest level of divergence within a bacterial genome was reported for Thermomonospora chromogena at 6.6% (Table 4-2). The archaeon Halosimplex carlsbadense has operons with the largest sequence divergence at 7.4% (Table 4-2, (5)). Acinas et al. raised a concern where intragenomic heterogeneity could lead to overestimation in diversity of microbial communities (1). With the exceptions of bacteria and archaea listed in Table 4-2, diversity 65 measures based on the well-accepted OTU cutoff of 3% will not be influenced by intragenomic heterogeneity (7). Organism Bacteria Thermobispora bispora Thermomonospora chromogena Thermoanaerobacter tengcongensis Thermoanaerobacter pseudethanolicus Desulfotomaculum kuznetsovii Desulfitobacter hafniense Photobacterium profundum rrn operon (#) Divergence (%) Ref. 4 6 4 4 Unknown 6 15 6.4% 6.6%* 6.5% 3.7%* 8.3% 4.8%* 5.1% (47) (48) (1) (44) (45) Archaea Haloarcula marismortui 2 5.6%* (29) Halosimplex carlsbadense 2 7.4%* (5) Natrinema sp. XA3-1 4 5.4%* (5) Table 4-2: Sequence divergence of the rrs gene and rrn operon copy number of bacteria with intragenomic heterogeneity of more than 3%. *Sequence divergence was calculated from neighbor-joining distance matrix produced from the Arb software. The maximum divergence is reported. Ecological strategies of bacteria with different rrn operon copy number Ecological strategies of organisms were originally characterized as either r- or K-selected by MacArthur and Wilson, based on a set of traits observed in organisms colonizing different stages of succession (26). Andrews and Harris then adopted the r/K concept for microbial ecology, proposing that bacteria can be characterized along an r/K spectrum based on growth phenotype (4). Bacteria that are r-selected have high maximum specific growth rate (µmax) while K-selected bacteria are more competitive in low nutrient environments, utilizing resources more efficiently for biomass synthesis (4). To determine where along the r/K spectrum a bacterial species will be placed, the growth characteristics of the bacteria must be determined. This means that a representative of the species must be cultivable. 66 Others   Verrucomicrobia   Thermotogae   Tenericutes   Spirochetes   Proteobacteria   Planctomycetes   Firmicutes   Deinococcus   Cyanobacteria   Chloroflexi   Chlorobi   Chlamydiae   Bacteroidetes   Aquificae   Actinobacteria   Acidobacteria   100   90   80   70   60   50   40   30   20   10   0   1   2   3   4   5   6   7   8   9   10   11   12   13   14   15   16S  rRNA  gene  copy  number   Figure 4-2: The relative abundance of different phyla for each 16S rRNA gene copy number. For the interpretation of the references to color in this and all other figures, the reader is referred to the electronic version of this dissertation. The combination of genome analysis and bacterial physiology found that high growth rate in bacteria is attributed to increased gene dosage. This can be achieved by having multiple copies per genome or positioning of the gene near the origin of replication (10). Increased numbers of rrn operons allow for more ribosome to be synthesized simultaneously and hence, more proteins can be synthesized per unit time when the environmental conditions permit fast growth. In an enzymatic reaction, both enzyme and substrate can be limiting factors. Increasing the amount of ribosomes (the enzyme) can only influence protein synthesis rate if it is accompanied by increased availability of tRNA (the substrate). There is a significant positive correlation between rRNA gene copy number and the abundance of tRNA genes in the genome (Figure 4-3)(12, 22). This observation provides evidence for a selection pressure to maintain high rrn operon copy number. It is generally observed that bacteria with more rrn operons have a higher growth rate than bacteria with fewer rrn operons, supporting Couturier and Rocha’s observation of a 67 correlation between growth rate and gene dosage (10). Dethlefsen and Schmidt showed that in addition to gene dosage, bacteria with more rrn operons have ribosomes that are faster than ribosomes in bacteria with fewer rrn operons (12). In the environment, where resource availability is more variable, the benefit of having high rrn operon copy number is to be able to respond faster to favorable growth conditions. Klappenbach et al. found that soil bacteria that form early colonies on agar plate have on average, three-fold more rrn operons than late-forming colonies (20). In the same study, the addition of a herbicide changed the relative abundance of herbicide degraders where herbicide amended soils select for high rrn operon copy number degraders (20). Using E. coli as the model organism, others have provided further evidence that multiple rrn operons is crucial for fast response time (9, 40). 180 tRNA  gene  copy  number 160 140 120 100 80 60 40 20 0 0 2 4 6 8 10 12 14 16S  rRNA  gene  copy  number 16 Figure 4-3: Correlation between 16S rRNA gene copy number and tRNA genes in bacterial genomes. The data are gathered from 867 bacterial genomes. The solid line represents the linear 2 fit, y = 31.8x + 6.3, R = 0.78, p-value < 0.001. A faster response time will allow bacteria with high rrn operon copy number to be more competitive in environments experiencing high nutrient pulses. Members of the genera 68 Shewanella and Vibrio with eight or more rrn operons are often found to be in nutrient rich environments. Shewanella are versatile in their respiratory capability and are often found in association with fermenters, exploiting almost any fermentative carbon substrates and electron acceptors made available to them (30). The genus Vibrio consists of one of the fastest growing bacteria, Vibrio natriegens, with a doubling time of less than 10 minutes (2, 13). However, there is a trade-off to maintaining multiple rrn operons. Although ribosome synthesis is a highly regulated process, the rrn operon promoters are constitutive promoters (24, 37). This means that a basal amount of rRNA will always be synthesized. When conditions become unfavorable for growth, the cost of rRNA synthesis will become a burden to the cells. Some high rrn operon copy number bacteria such as E. coli adapt to nutrient scarcity by dimerizing the ribosomes (46). This inactivates the ribosomes so that protein synthesis can be downregulated. Under this condition, bacteria with fewer rrn operons will become more competitive. Model oligotrophic organisms such as caulobacters and acidobacteria spp. that are abundant in low nutrient environments have one or two rrn operon(s) (14, 32). Based on the correlations of rrn operon copy number and various physiological characteristics of bacteria, it is suggested that rrn operon copy number is a potential indicator of the ecological strategy of the bacteria, including the rate and efficiency of substrate utilization. Bacteria with high rrn operon copy number resemble r-strategists, where it is selected for high specific growth rate, with ribosomes that have higher translational power, respond quickly to increased resource availability and are more competitive in environments that experience frequent pulses of nutrient (Table 4-3). On the other hand, bacteria with few rrn operons can be considered as K-strategists that are more competitive in low nutrient environments and more efficient at resource allocation to biomass (Table 4-3). 69 1 7 15 rRNA gene copy number per genome Few rrn operons Less tRNA encoding genes (11) Low translational power (11) a Low ribosome drop-off frequency Low growth rates Form late colonies on agar plates (20) Late responders to nutrient input (20) Oligotrophs Low Kmin Genome Translation machinery High translational power High ribosome drop-off frequency Physiology High growth rates Form early colonies on agar plates Early responders to nutrient Copiotrophs High Kmin Low BGE b High BGE K-selected phenotype Widely distributed Highly abundant in environment with constant low nutrient supply Multiple rrn operons More tRNA encoding genes r-selected phenotype Ecology More abundant in environment with frequent pulse of high nutrient. Table 4-3. Characteristics of the lifestyle of bacteria that are correlated with its rrn operon copy a number. The descriptions in bold have yet to be tested. This characteristic is further discussed in b Appendix A. This has been shown by Roller, B. in the Schmidt lab. The rrn operon copy number of environmental 16S rDNA sequences can be determined based on phylogenetic association to close relatives with known rrn operon copy number. The rrn operon copy numbers can be found in the rrnDB database (22). This will allow hypotheses to be made regarding the growth efficiency of the bacteria because we can infer ecological strategies based on rrn operon copy number. It can also assist in setting up a microbial community component in ecosystem models (27). 70 Bacterial communities in soil Others Gemmatimonadetes Rotation Forest Unknown Nitrospirae TM6 * Chloroflexi Planctomycetes * Bacteroidetes Actinobacteria Verrucomicrobia * Acidobacteria Proteobacteria 0.0 0.1 0.2 0.3 Relative Abundance 0.4 Figure 4-4: Relative abundance of bacterial phyla in 16S rDNA libraries constructed from rotation cropland and forest soils in September 2009. Each bar represents average relative abundance from two replicate plots and error bar is one standard deviation. ‘*’ represents significant differences from Student’s t-test with p-value < 0.05. The bacterial communities in forest and rotation cropland soils from the KBS LTER main site were characterized by surveying the 16S rRNA gene sequences. At the phylum level, bacterial communities from forest and rotation cropland have similar composition. Only the phyla Verrucomicrobia, Planctomycetes and TM6 exhibited significant differences, where they are more abundant in forest soil (Figure 4-4). The phyla Proteobacteria, Acidobacteria and Verrucomicrobia describe more than 70% of the bacterial communities. This is not surprising as these three phyla are often found in soil. When compared to other terrestrial biomes, the KBS 71 LTER forest and rotation soils have more Verrucomicrobia and Acidobacteria than grassland soils (42). Sequence analysis at a finer level of resolution, the order level, revealed more differences between the two bacterial communities (data not shown). In addition to orders in the phyla Verrucomicrobia and Planctomyces, higher relative abundance of the order Rhizobiales was also observed in the forest soil. On the other hand, Sphingomonadales was found to be more abundant in the rotation cropland. The differences between forest and rotation soil bacterial communities at the order level can be observed even when the analysis was performed on the 20 most abundant OTUs (cutoff = 0.03). An exception is orders within the phylum Acidobacteria. There are more sequences clustering with Acidobacteria from the 20 most abundant OTUs at the order level in the rotation cropland soil. These OTUs make up more than 10% of the total bacterial 16S rDNA sequences from rotation soil (Figure 4-5). To analyze the ecological strategies of the bacterial community from these two biomes, rrn operon copy number was assigned to each OTU based on taxonomic classification at the order level. The predicted average rrn operon copy number for bacterial communities from both sites is 2.0 ± 0.05. This is slightly lower than the rRNA gene:single copy housekeeping gene ratio calculated from metagenomic library of the forest and rotation soils (Forest = 2.6 ± 0.1, Rotation = 3.1 ± 0.3, Teal, Gomez and Schmidt, 2011, manuscript in preparation). This discrepancy is due to several reasons. Metagenomic libraries have a lower coverage than libraries generated from pyrosequencing of 16S rDNA amplicons. It represents only up to 85 genomes, an extremely small fraction of the overall bacterial community, even if it is only compared to the 20 most abundant OTUs (Teal et al., 2011, manuscript in preparation). Additionally, rrn operon 72 copy number cannot be assigned to 1% of the 20 most abundant OTUs due to lack of information from close relatives (classified as Bacteria in Figure 4-5). Planctomycetales   Flavobacteriales   Forest Rotation (2) (6) (2,4) Xanthomonadales   (6) Sphingobacteriales   (2,4) Myxococcales   ProteobacteriaP Rhizobiales   BacteriaP   * (1-­3)   Sphingomonadales   VerrucomicrobiaP   * (2)   * AcidobacteriaP   0   (1,2)   * (1,2)   0.02   0.04   0.06   0.08   0.10   0.12 0.14   Relative  Abundance Figure 4-5: Relative abundance of the 20 most abundant OTUs in forest and rotation cropland based on the order level of the OTU. The possible rrn operon copy number is provided in P brackets. ‘ ’ represents taxonomic classification above the order level. ‘*’ represents significant difference based on Student’s t-test with p-value < 0.05. However, the similarity of average rrn operon copy number in the forest and rotation soil suggests that the dominant bacteria have K-selected phenotypes. This suggests low concentrations of resources in both the forest and rotation soils, selecting for bacteria that are more efficient at resource use. Based on the average rrn operon copy number, it is predicted that the bacterial community in forest and rotation soils have similar growth efficiency when exposed to the same growth conditions. Therefore, variation in growth efficiency between the forest and rotation soil is more likely to be driven by environmental factors. Most soils are often seen as a nutrient limiting environment but not all bacterial communities are similar in terms of rrn operon copy number. A survey of 71 soil samples from 73 biomes with different carbon content found that acidobacteria are more prevalent in low resource availability biomes while the abundance of copiotrophic Bacteroidetes and Betaproteobacteria is positively correlated with increased resource availability (16). Furthermore, it is often observed that when low molecular weight substrates such as glucose and citric acid are added to soil, Acidobacteria becomes outcompeted by fast-growing bacteria with high rrn operon copy number such as Burkholderia spp. (4 – 6 copies) and Flavobacterium spp. (6 copies) (15). On the other hand, the abundance of members of the phylum Actinobacteria does not correlate with carbon availability. This phylum has a wider range of rrn operon copy numbers, with the Corynebacterium and Streptomyces having six copies and the cellulose degraders Cellulomonas having only two rrn operons. These observations provide further support for characterizing bacteria based on ecological strategies using rrn operon copy number. Conclusion The application of molecular surveys has provided considerable insight into the structure of bacterial communities in soil. Efforts to link general function such as CO2 production to community composition has been difficult because all metabolically active heterotrophs in soil produce CO2, but more specialized functions like methane oxidation can been linked to the diversity of methane-oxidizing bacteria. The capacity to identify ecological strategies of bacteria in soil could help accelerate our understanding on how aboveground activity influences bacterial activity. We suggest that rrn operon copy number is a suitable candidate as an indicator for the ecological strategies of bacteria. Bacteria with high rrn operon copy number tend to have the phenotype of an r-strategist while low rrn operon copy number bacteria will be more competitive in low nutrient environments. Therefore, the inclusion of rrn operon copy number classification 74 in molecular analyses can provide information regarding the ecology of the bacterial community and make predictions on the influence of community composition on bacterial growth efficiency. Materials and Methods Soil Sampling Soil samples were collected from the W. K. Kellogg Biological Station Long Term Ecological Research main site (KBS LTER, Hickory Corners, MI) in September 2009. The site description and soil type is available at http://kbs.msu.edu. Five soil cores of 2.5 cm diameter were collected from the top 10cm at the rotation cropland sites with corn/soybean/wheat crop rotation and deciduous forests. The rotation cropland was planted with soybean at the time of collection. The litter layer in the deciduous forest soil was removed prior to sampling. Soil cores from each site were pooled, sieved and flash frozen with liquid nitrogen in the field, and transported back to the o laboratory on ice. Soils were stored at -80 C upon arrival at the laboratory. DNA Extraction DNA extraction from two replicate plots of each biome was performed by Dr. Tracy Teal according to the protocol in Teal et al. (Teal et al., manuscript in preparation). Briefly, DNA was extracted using a direct soil extraction method followed by cesium-chloride gradient purification (36). DNA samples were sent to the Josephine Bay Paul Center in Comparative Molecular Biology and Evolution at Marine Biological Laboratory, Woods Hole, MA., for amplification and sequencing of the V6 region in the rDNA sequence. 75 Data Analysis The OTU assignments were downloaded from the Visualization and Analysis of Microbial Population Structures (VAMPS) project at the Josephine Bay Paul Center (38). rrn operon copy number was assigned based on the most specific taxonomy level of the sequence. If the rrn operon copy number for that taxonomy level is not available, the next broader taxonomy level is used to obtain a range of rrn operon copy number. The copy number for the rrn operon is obtained from the rrnDB database (22). Acknowledgement I would like to thank the Schmidt lab September 2009 soil sampling team for collecting the soil samples, Dr. Tracy Teal for DNA extraction work, and Dr. Kevin Theis for allowing me to use his computer for sequence analysis. I would also like to acknowledge Drs. Kevin Theis and Tracy Teal for their assistance in sequence analysis and helpful discussions. 76 REFERENCES 77 REFERENCES 1. Acinas, S. G., L. A. Marcelino, V. Klepac-Ceraj, and M. F. Polz. 2004. Divergence and redundancy of 16S rRNA sequences in genomes with multiple rrn operons. J Bacteriol 186:2629-2635. 2. Aiyar, S. E., T. Gaal, and R. L. Gourse. 2002. rRNA promoter activity in the fastgrowing bacterium Vibrio natriegens. J Bacteriol 184:1349-1358. 3. Allison, S. D., and J. B. H. Martiny. 2008. Resistance, resilience, and redundancy in microbial communities. P Natl Acad Sci USA 105:11512-11519. 4. Andrews, J. H., and R. F. Harris. 1986. R-Selection and K-Selection and Microbial Ecology. Adv Microb Ecol 9:99-147. 5. Boucher, Y., C. J. Douady, A. K. Sharma, M. Kamekura, and W. F. Doolittle. 2004. Intragenomic heterogeneity and intergenomic recombination among haloarchaeal rRNA genes. J Bacteriol 186:3980-3990. 6. Brosius, J., T. J. Dull, D. D. Sleeter, and H. F. Noller. 1981. Gene Organization and Primary Structure of a Ribosomal-RNA Operon from Escherichia coli. J Mol Biol 148:107-127. 7. Coenye, T., and P. Vandamme. 2003. Intragenomic heterogeneity between multiple 16S ribosomal RNA operons in sequenced bacterial genomes. Fems Microbiol Lett 228:45-49. 8. Cole, J. R., Q. Wang, E. Cardenas, J. Fish, B. Chai, R. J. Farris, A. S. Kulam-SyedMohideen, D. M. McGarrell, T. Marsh, G. M. Garrity, and J. M. Tiedje. 2009. The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. Nucleic Acids Res 37:D141-145. 9. Condon, C., D. Liveris, C. Squires, I. Schwartz, and C. L. Squires. 1995. rRNA operon multiplicity in Escherichia coli and the physiological implications of rrn inactivation. J Bacteriol 177:4152-4156. 10. Couturier, E., and E. P. Rocha. 2006. Replication-associated gene dosage effects shape the genomes of fast-growing bacteria but only for transcription and translation genes. Mol Microbiol 59:1506-1518. 11. Dethlefsen, L. 2004. Translational power differs between bacteria pursuing different ecological strategies. Michigan State University, East Lansing. 12. Dethlefsen, L., and T. M. Schmidt. 2007. Performance of the translational apparatus varies with the ecological strategies of bacteria. J Bacteriol 189:3237-3245. 78 13. Eagon, R. G. 1962. Pseudomonas natriegens, a Marine Bacterium with a Generation Time of Less Than 10 Minutes. J Bacteriol 83:736-&. 14. Eichorst, S. A., C. R. Kuske, and T. M. Schmidt. 2011. Influence of Plant Polymers on the Distribution and Cultivation of Bacteria in the Phylum Acidobacteria. Appl Environ Microbiol 77:586-596. 15. Eilers, K. G., C. L. Lauber, R. Knight, and N. Fierer. 2010. Shifts in bacterial community structure associated with inputs of low molecular weight carbon compounds to soil. Soil Biol Biochem 42:896-903. 16. Fierer, N., M. A. Bradford, and R. B. Jackson. 2007. Toward an ecological classification of soil bacteria. Ecology 88:1354-1364. 17. Hugenholtz, P., and N. R. Pace. 1996. Identifying microbial diversity in the natural environment: a molecular phylogenetic approach. Trends Biotechnol 14:190-197. 18. Keller, M., and K. Zengler. 2004. Tapping into microbial diversity. Nat Rev Microbiol 2:141-150. 19. Kjelleberg, S., R. J. Case, Y. Boucher, I. Dahllof, C. Holmstrom, and W. F. Doolittle. 2007. Use of 16S rRNA and rpoB genes as molecular markers for microbial ecology studies. Appl Environ Microbiol 73:278-288. 20. Klappenbach, J. A., J. M. Dunbar, and T. M. Schmidt. 2000. rRNA operon copy number reflects ecological strategies of bacteria. Appl Environ Microbiol 66:1328-1333. 21. Klappenbach, J. A., P. R. Saxman, J. R. Cole, and T. M. Schmidt. 2001. rrndb: the Ribosomal RNA Operon Copy Number Database. Nucleic Acids Res 29:181-184. 22. Lee, Z. M., C. Bussema, 3rd, and T. M. Schmidt. 2009. rrnDB: documenting the number of rRNA and tRNA genes in bacteria and archaea. Nucleic Acids Res 37:D489493. 23. Levine, U. Y., T. K. Teal, G. P. Robertson, and T. M. Schmidt. 2011. Agriculture's impact on microbial diversity and associated fluxes of carbon dioxide and methane. Isme J. 24. Liang, S. T., M. Bipatnath, Y. C. Xu, S. L. Chen, P. Dennis, M. Ehrenberg, and H. Bremer. 1999. Activities of constitutive promoters in Escherichia coli. J Mol Biol 292:19-37. 25. Liao, D. Q. 2000. Gene conversion drives within genic sequences: Concerted evolution of ribosomal RNA genes in bacteria and archaea. J Mol Evol 51:305-317. 79 26. MacArthur, R. H., and E. O. Wilson. 2001. The theory of island biogeography. Princeton University Press, Princeton. 27. McGuire, K. L., and K. K. Treseder. 2010. Microbial communities and their relevance for ecosystem models: Decomposition as a case study. Soil Biol Biochem 42:529-535. 28. Merhej, V., M. Royer-Carenzi, P. Pontarotti, and D. Raoult. 2009. Massive comparative genomic analysis reveals convergent evolution of specialized bacteria. Biol Direct 4:-. 29. Mylvaganam, S., and P. P. Dennis. 1992. Sequence heterogeneity between the two genes encoding 16S rRNA from the halophilic archaebacterium Haloarcula marismortui. Genetics 130:399-410. 30. Nealson, K., and J. Scott. 2006. Ecophysiology of the Genus Shewanella, p. 1133-1151. In M. Dworkin, S. Falkow, E. Rosenberg, K.-H. Schleifer, and E. Stackebrandt (ed.), The Prokaryotes. Springer New York. 31. Pace, N. R., D. A. Stahl, D. J. Lane, and G. J. Olsen. 1986. The analysis of natural microbial populations by ribosomal RNA sequences. Adv Microb Ecol 9:1-55. 32. Poindexter, J. S. 1981. The caulobacters: ubiquitous unusual bacteria. Microbiol Rev 45:123-179. 33. Pruesse, E., C. Quast, K. Knittel, B. M. Fuchs, W. Ludwig, J. Peplies, and F. O. Glockner. 2007. SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids Res 35:7188-7196. 34. Rainey, F. A., N. L. Ward-Rainey, P. H. Janssen, H. Hippe, and E. Stackebrandt. 1996. Clostridium paradoxum DSM 7308T contains multiple 16S rRNA genes with heterogeneous intervening sequences. Microbiology 142 ( Pt 8):2087-2095. 35. Roesch, L. F., R. R. Fulthorpe, A. Riva, G. Casella, A. K. M. Hadwin, A. D. Kent, S. H. Daroub, F. A. O. Camargo, W. G. Farmerie, and E. W. Triplett. 2007. Pyrosequencing enumerates and contrasts soil microbial diversity. Isme J 1:283-290. 36. Sambrook, J., and D. W. Russell. 2006. The condensed protocols from Molecular cloning : a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 37. Schneider, D. A., W. Ross, and R. L. Gourse. 2003. Control of rRNA expression in Escherichia coli. Curr Opin Microbiol 6:151-156. 80 38. Sogin, M. L., H. G. Morrison, J. A. Huber, D. Mark Welch, S. M. Huse, P. R. Neal, J. M. Arrieta, and G. J. Herndl. 2006. Microbial diversity in the deep sea and the underexplored "rare biosphere". P Natl Acad Sci USA 103:12115-12120. 39. Stackebrandt, E., and B. M. Goebel. 1994. Taxonomic Note: A Place for DNA-DNA Reassociation and 16S rRNA Sequence Analysis in the Present Species Definition in Bacteriology. International Journal of Systematic Bacteriology 44:846-849. 40. Stevenson, B. S., and T. M. Schmidt. 2004. Life history implications of rRNA gene copy number in Escherichia coli. Appl Environ Microbiol 70:6670-6677. 41. Stewart, F. J., and C. M. Cavanaugh. 2007. Intragenomic variation and evolution of the internal transcribed spacer of the rRNA operon in bacteria. J Mol Evol 65:44-67. 42. Strickland, M. S., C. Lauber, N. Fierer, and M. A. Bradford. 2009. Testing the functional significance of microbial community composition. Ecology 90:441-451. 43. Trcek, J., and M. Teuber. 2002. Genetic and restriction analysis of the 16S-23S rDNA internal transcribed spacer regions of the acetic acid bacteria. Fems Microbiol Lett 208:69-75. 44. Turova, T. P., B. B. Kuznetsov, E. V. Novikova, A. B. Poltaraus, and T. N. Nazina. 2001. [Heterogeneity of nucleotide sequences of 16S ribosomal RNA genes from the Desulfotomaculum kuznetsovii type strain]. Mikrobiologiia 70:788-795. 45. Vezzi, A., S. Campanaro, M. D'Angelo, F. Simonato, N. Vitulo, F. M. Lauro, A. Cestaro, G. Malacrida, B. Simionati, N. Cannata, C. Romualdi, D. H. Bartlett, and G. Valle. 2005. Life at depth: Photobacterium profundum genome sequence and expression analysis. Science 307:1459-1461. 46. Wada, A., K. Igarashi, S. Yoshimura, S. Aimoto, and A. Ishihama. 1995. Ribosome Modulation Factor - Stationary Growth Phase-Specific Inhibitor of Ribosome Functions from Escherichia coli. Biochem Bioph Res Co 214:410-417. 47. Wang, Y., Z. Zhang, and N. Ramanan. 1997. The actinomycete Thermobispora bispora contains two distinct types of transcriptionally active 16S rRNA genes. J Bacteriol 179:3270-3276. 48. Yap, W. H., Z. Zhang, and Y. Wang. 1999. Distinct types of rRNA operons exist in the genome of the actinomycete Thermomonospora chromogena and evidence for horizontal transfer of an entire rRNA operon. J Bacteriol 181:5201-5209. 81 CHAPTER 5 SIGNIFICANCE AND FUTURE DIRECTIONS Bacterial growth efficiency (BGE) is a measure of how bacteria allocate resources for biomass or energy synthesis. Growth efficiency is a central parameter in soil organic carbon (SOC) models that simulate microbial-driven organic carbon transformations. Soil carbon models are important for determining sustainable land management practices. One such model is the DAYCENT model, the daily version of the CENTURY model. DAYCENT is a biogeochemistry model that simulates plant-soil nutrient cycling, specifically carbon, nitrogen, phosphorus and sulfur dynamics in terrestrial ecosystems (12, 14). In the model, growth efficiency is applied to determine the fractionation of carbon at various steps of microbially driven carbon transformations. Microbial growth efficiency (MGE) is currently treated as a global parameter, where a single value is used for simulating carbon cycling and carbon dioxide flux in various biomes, from forests to grasslands and conventional agricultural soils (2, 15). In the DAYCENT model, most growth efficiencies are currently fixed at 0.45. This means that for every microbial-driven carbon transformation, 55% of the transformed carbon is lost into the atmosphere as CO2, while the remaining 45% is incorporated into microbial biomass (14). By using a constant growth efficiency value, the DAYCENT model assumes that organic compounds with different stabilities are converted with equal efficiency into microbial biomass, and that microbes from different biomes have equivalent growth efficiencies. In Chapter 3, it was shown that bacterial communities, which can constitute up to 80% of total microbial biomass in soils, in land under different management practices have different growth efficiencies (Chapter 3, (3)). BGE ranged from 0.23 to 0.63, with soybean monoculture 82 cropland soil having the most efficient bacterial communities (Chapter 3). More than 70% of the soil samples have BGE less than 0.45. This means that a growth efficiency of 0.45 will be overestimating the efficiency of carbon immobilized into microbial biomass when simulating some of the biomes at KBS LTER. Schimel and Weintraub introduced a model with a growth efficiency value of 0.35 to take into account the allocation of substrate for extracellular enzyme production (17). This model would be a better representation of microbially driven carbon dynamics in KBS LTER soils. A sensitivity analysis was performed using the DAYCENT model to determine if the variation in BGE observed for field samples was sufficient to change soil carbon dynamics. An executable version of the DAYCENT model was obtained from Cindy Keough at the National Resource Ecology Laboratory at Colorado State University. A mock ecosystem was simulated using the site-specific parameters for Kellogg Biological Station and vegetation parameters for a temperate deciduous forest. Both input parameters were obtained from the Vegetation Ecosystem Modeling and Analysis Project (VEMAP) simulation, which is available through Oak Ridge National Laboratory Distributed Active Archive Center (http://daac.ornl.gov/MODELS/guides/century_vemap.html, retrieved on February 12, 2011, (10)). The weather input for DAYCENT model was obtained from the National Weather Service Station at KBS (retrieved from KBS LTER Data Catalog on October 31, 2010). The mock ecosystem was simulated for 1990 years to initiate soil organic carbon pools and plant biomass, and to allow the system to reach equilibrium. The output from the initial simulation was then used to run a simulation for 70 years with different global values for microbial growth efficiency. There are ten fixed growth efficiency parameters in the DAYCENT soil organic carbon submodel (Figure 5-1). By default, seven of these growth efficiency parameters are fixed at a 83 value of 0.45. Surface structural carbon is composed of crop residue with high lignin content. Lignin is more recalcitrant to microbial degradation, thus the high lignin component of crop residue enters the slow carbon pool without being completely decomposed by the microbial community. The fungal community, which is often assumed to be more efficient than the bacterial community, is primarily thought to be the major decomposer of surface structural carbon (6, 14). Hence, the growth efficiency applied for decomposition of surface structural carbon is 0.55. The incorporation of surface microbial carbon into slow carbon is thought to have the highest efficiency at 0.6. On the other hand, bacterial community is responsible for decomposition of belowground carbon such as root exudates or dead root biomass. The incorporation of soil active carbon into the slow carbon pool is a function of soil texture, where the efficiency is lowest in soil with high sand content (Figure 5-1, (14)). In the sensitivity analysis, the seven growth efficiency parameters originally fixed at 0.45 were varied. The active carbon pool was most sensitive to changes in growth efficiency. This carbon pool consists of root exudates, deposition from dead root cells, active microbial biomass and microbial metabolites. A 22% decrease in the carbon pool was observed when growth efficiency was decreased from 0.45 to 0.35 (Figure 5-2a). This was expected since growth efficiency is a direct measure of the capability of the microbial community to incorporate resources into biomass material. Additionally, the decrease in the active carbon pool was sufficient to influence the rate of soil passive carbon accumulation (Figure 5-2b). When the default growth efficiency (0.45) was used in the simulation, the mock ecosystem accumulated 2 passive carbon at a rate of 0.35 g/m /year. A decrease of growth efficiency by 5% stabilizes passive carbon (0.4, Figure 5-2b). A further decrease to 0.35 led to a loss in passive carbon (0.35, Figure 5-2b). 84 SURFACE Structural  C BELOW-­GROUND Metabolic  C Structural  C Metabolic  C 0.55 CO2 Microbial  C CO2 CO2 0.60 CO2 CO2 Lignin Slow  C (10-­50  yrs) CO2 Lignin Passive  C Active  C CO2 sand CO2 Slow  C (10-­50  yrs) CO2 CO2 Figure 5-1: Soil organic carbon submodel in the DAYCENT ecosystem model. Ovals represent the application of growth efficiency in carbon transformations. Blue ovals represent growth efficiency values that were adjusted in the sensitivity analysis. This figure is adapted from the CENTURY User’s Manual (13). Passive carbon formation is a combination of the biological, chemical and physical transformations of carbon. Decreased growth efficiency will reduce the biologically mediated carbon transformations, but the reduction in active carbon pool will decrease the amount of substrate (microbial biomass and extracellular products) available for chemical and physical transformations (9). Stable carbon is often found to have more microbially derived carbon (4, 7, 8). Therefore, to supply more microbially derived carbon for stable carbon formation, it is important to practice land management that will support the growth of microbial communities with high growth efficiency. 85 Figure 5-2. Sensitivity of the active carbon pool (a) and the annual passive carbon pool (b) to variations in growth efficiency. The active carbon pool at different growth efficiencies is presented as a percentage relative to the active carbon pool when growth efficiency is 0.45. The sensitivity of ecosystem models to changes in growth efficiency emphasizes the importance of using site-specific values of growth efficiency. This is crucial when the model is used to make predictions about how changes in land management practices could influence SOC dynamics (16). However, to effectively apply growth efficiency in biogeochemistry models, we must better understand how growth efficiency is regulated in the field. Focusing on bacterial growth efficiency, the work from this dissertation presents a framework for how BGE can be 86 influenced by climate change, soil edaphic properties and land management practices (Figure 53). Figure 5-3. Framework relating changes in climate, land management practices and edaphic factors on BGE. The dotted arrows represent processes that can influence BGE. The two major sources of perturbation in terrestrial ecosystems are climate change and land management practices. These environmental factors can influence the function of bacterial communities in soil, where the function in this context is defined as soil carbon transformation. Bacteria respond to changes in environmental factors as soon as immediately after the perturbation to several hours or days, where the response is typically measured as CO2 production (11). The response of bacteria to perturbation is also influenced by edaphic factors (20). Some of the major factors that have been shown to exert a change in function are moisture, litter type, and resource availability (19). Microbial-driven soil carbon transformations are often measured as CO2 production. While CO2 production is a good indicator of microbial activity, it is not a good measure of the efficiency at which carbon is transformed into microbial biomass. For example, we found no 87 correlation between soil CO2 flux and BGE (Figure 5-4). Field CO2 flux is a combination of root respiration and microbial respiration. Microbial CO2 production can constitute up to 50% of soil CO2 efflux (5). However, there are two possible explanations for changes in microbial CO2 production. First, it can represent an overall change in bacterial growth. This can be measured as change in decomposition rate, either because more bacteria become active or the new condition allows for increased decomposition rate. Change in CO2 production can also be caused by changes in energy expenditure for biomass synthesis. BGE can distinguish between these two scenarios. If the change in CO2 production is due to the first reason, one will observe no difference in BGE, while the second situation will lead to a change in BGE. In laboratory incubation, a response was observed as soon as 24 hours after incubation (Chapter 3). It is this immediate response of bacterial communities to changes in aboveground activity that leads to within site variation in BGE (Figure 5-3: short-term response). Over multiple generations, the changes in the reproduction of individual populations will lead to changes in community structure. This will occur if the perturbation has differential effect on the community members. If the perturbation is not removed, bacteria that can survive better in the new environment will subsequently increase in abundance. If the change in community composition leads to change in ecological strategy, it is expected that growth efficiency will be affected (Chapter 4). This change in growth efficiency is mediated by the potential of the new bacterial community to efficiently allocate substrate for growth. This potential is dictated by the genetic capability of the community, which will influence the type of metabolic pathways used, energy required for cellular maintenance, investment of energy for production of extracellular 88 enzymes to access substrates, efficiency of translational machineries and various energy expenditure processes (18). a) 80 Flux  (kg  CO2-­C/ha/day) Flux  (kg  CO2-­C/ha/day) 60 60 40 40 20 0 c) b) 80 20 Apr Jun Aug Oct Dec 0.6 0 d) Apr Jun Aug Oct Dec Apr Jun Aug Oct Dec 0.6 0.5 0.4 0.4 BGE BGE 0.5 0.3 0.3 0.2 0.2 0.1 Apr Jun Aug Oct 0.1 Dec Figure 5-4. Seasonal variation of CO2 flux (a, b) and BGE (c, d) in the conventional agriculture (a, c) and deciduous forest (b, d) sites. Each point represents the average of three experimental replicates and the error bars are standard error. Flux data was obtained from the KBS LTER dataset (retrieved on March 21, 2011). In conclusion, I have shown that BGE in the terrestrial ecosystem is not static and can be influenced by climate, edaphic properties and land management practices via two pathways. The short-term response is mediated by changes in the function of bacterial communities while the delayed effect is due to changes in community composition. The variability of BGE in the terrestrial ecosystem and sensitivity of carbon models to changes in BGE highlights the need to apply site-specific BGE in SOC models. Future research on the growth efficiency of bacterial 89 communities in soil should be focused on a wider range of biomes. Attention should be placed on biomes that have resource availability with different C:N:P stoichiometry. Research on terrestrial BGE variation should be conducted in conjunction with the assessment of bacterial communities and soil properties. This will allow variation in BGE to be incorporated into SOC models, where it is a function of community composition and environmental factors. Finally, the hypothesis that BGE is a suitable indicator of biological carbon sequestration in soil should continue to be tested. The contribution of this dissertation work is consistent with the goal put forth by the scientific community to better understand how changes in aboveground activity influence the stability of soil organic carbon (1). Acknowledgements I would like to thank Cindy Keough from the National Resource Ecology Laboratory at Colorado State University for her patience and assistance in preparing inputs for the DAYCENT model. 90 REFERENCES 91 REFERENCES 1. Billings, S. A., J. Lichter, S. E. Ziegler, B. A. Hungate, and D. D. Richter. 2010. A call to investigate drivers of soil organic matter retention vs. mineralization in a high CO2 world. Soil Biol Biochem 42:665-668. 2. De Gryze, S., A. Wolf, S. R. Kaffka, J. Mitchell, D. E. Rolston, S. R. Temple, J. Lee, and J. Six. 2010. Simulating greenhouse gas budgets of four California cropping systems under conventional and alternative management. Ecol Appl 20:1805-1819. 3. de Vries, F. T., E. Hoffland, N. van Eekeren, L. Brussaard, and J. Bloem. 2006. Fungal/bacterial ratios in grasslands with contrasting nitrogen management. Soil Biol Biochem 38:2092-2103. 4. Grandy, A. S., and J. C. Neff. 2008. Molecular C dynamics downstream: The biochemical decomposition sequence and its impact on soil organic matter structure and function. Sci Total Environ 404:297-307. 5. Hanson, P. J., N. T. Edwards, C. T. Garten, and J. A. Andrews. 2000. Separating root and soil microbial contributions to soil respiration: A review of methods and observations. Biogeochemistry 48:115-146. 6. Holland, E. A., and D. C. Coleman. 1987. Litter placement effects on microbial and organic matter dynamics in an agroecosystem. Ecology 68:425-433. 7. Kiem, R., and I. Kogel-Knabner. 2003. Contribution of lignin and polysaccharides to the refractory carbon pool in C-depleted arable soils. Soil Biol Biochem 35:101-118. 8. Liang, C., and T. C. Balser. 2008. Preferential sequestration of microbial carbon in subsoils of a glacial-landscape toposequence, Dane County, WI, USA. Geoderma 148:113-119. 9. Liang, C., G. Cheng, D. Wixon, and T. Balser. 2010. An Absorbing Markov Chain approach to understanding the microbial role in soil carbon stabilization. Biogeochemistry:1-7. 10. Melillo, J. M., J. Borchers, J. Chaney, H. Fisher, S. Fox, A. Haxeltine, A. Janetos, D. W. Kicklighter, T. G. F. Kittel, A. D. Mcguire, R. Mckeown, R. Neilson, R. Nemani, D. S. Ojima, T. Painter, Y. Pan, W. J. Parton, L. Pierce, L. Pitelka, C. Prentice, B. Rizzo, N. A. Rosenbloom, S. Running, D. S. Schimel, S. Sitch, T. Smith, and I. Woodward. 1995. Vegetation Ecosystem Modeling and Analysis Project - Comparing Biogeography and Biogeochemistry Models in a Continental-Scale Study of Terrestrial Ecosystem Responses to Climate-Change and Co2 Doubling. Global Biogeochem Cy 9:407-437. 92 11. Oburger, E., and D. L. Jones. 2009. Substrate mineralization studies in the laboratory show different microbial C partitioning dynamics than in the field. Soil Biol Biochem 41:1951-1956. 12. Parton, W. J., M. Hartman, D. Ojima, and D. Schimel. 1998. DAYCENT and its land surface submodel: description and testing. Global Planet Change 19:35-48. 13. Parton, W. J., B. McKeown, V. Kirchner, and D. S. Ojima. 1992. CENTURY User's Manual. Colorado State University. 14. Parton, W. J., D. S. Schimel, C. V. Cole, and D. S. Ojima. 1987. Analysis of Factors Controlling Soil Organic-Matter Levels in Great-Plains Grasslands. Soil Sci Soc Am J 51:1173-1179. 15. Pepper, D. A., S. J. Del Grosso, R. E. McMurtrie, and W. J. Parton. 2005. Simulated carbon sink response of shortgrass steppe, tallgrass prairie and forest ecosystems to rising [CO2], temperature and nitrogen input. Global Biogeochem Cy 19:GB1004. 16. Priess, J. A., C. Schweitzer, F. Wimmer, O. Batkhishig, and M. Mimler. 2011. The consequences of land-use change and water demands in Central Mongolia. Land Use Policy 28:4-10. 17. Schimel, J. P., and M. N. Weintraub. 2003. The implications of exoenzyme activity on microbial carbon and nitrogen limitation in soil: a theoretical model. Soil Biol Biochem 35:549-563. 18. Schmidt, T., and A. Konopka. 2009. Physiological and Ecological Adaptations of SlowGrowing, Heterotrophic Microbes and Consequences for Cultivation, p. 101-120. In S. S. Epstein (ed.), Uncultivated Microorganisms, vol. 10. Springer Berlin / Heidelberg. 19. Strickland, M. S., C. Lauber, N. Fierer, and M. A. Bradford. 2009. Testing the functional significance of microbial community composition. Ecology 90:441-451. 20. van Veen, J., L. van Overbeek, and J. van Elsas. 1997. Fate and activity of microorganisms introduced into soil. Microbiol. Mol. Biol. Rev. 61:121-135. 93 APPENDICES 94 APPENDIX A STREPTOMYCIN-RESISTANT E. COLI MUTANT EXHIBITS INCREASED BACTERIAL GROWTH EFFICIENCY Introduction Protein synthesis, also known as translation, is one of the most expensive cellular processes. It is catalyzed by ribosome, a ribonucleoprotein complex. The catalytic center of the ribosome consists primarily of rRNA while ribosomal proteins play a structural role to increase the efficiency and to stabilize the structure of the ribosome (10). One such protein that plays an important role in determining the efficiency of translation is ribosomal protein S12. It is encoded by the gene rpsL and interacts with 16S rRNA as part of the 30S small subunit. Ribosomal protein S12 is a central ribosomal protein in assisting polypeptide elongation and is involved in determining translational accuracy (9). Some mutation in rpsL in E. coli can render the bacteria resistant to streptomycin. One very well studied mutation is a point mutation in codon 42 that changes a lysine codon to asparagine. This mutation does not only renders the mutant streptomycin resistant; it also causes the ribosome to be slower at peptide elongation. The decreased elongation rate is correlated with decreased growth rate and increased fidelity (2). Additionally, the rpsL mutant also exhibits increased ribosome drop-off frequency (5). Ribosome drop-off is a phenomenon where the peptidyl-tRNA dissociates from the translational machinery prior to reaching the stop codon. When compared to other translational errors such as missense substitutions and frameshift errors, ribosome drop-off contributes most significantly to inefficient translation (5). Ribosome drop-off event results in formation of free peptidyl-tRNA that arrests protein synthesis if not hydrolyzed by peptidyl-tRNA hydrolase (3). 95 In wild type E. coli K12 strains, ribosome drop-off occurs at a rate of once in 4000 codons (8). During the translation of ß-galactosidase mRNA, the 1:4000 codons frequency is equivalent to one drop-off occurrence for every four times translation is initiated. The significance of drop-off occurrence is also supported by the findings that peptidyl-tRNA hydrolase is essential for growth in E. coli and B. subtilis (7, 12). E. coli rpsL mutant with high fidelity ribosomes was shown to have a drop-off frequency of once every two translation initiation (5). During exponential growth, protein synthesis can consume up to 60% of cellular ATP (17). The ATP spent on making truncated polypeptides is considered wasteful because the ATP is not invested in synthesizing proteins for building biomass. This means that to synthesize the same amount of biomass, the rpsL mutant will require more ATP. Increased demand for ATP due to increased ribosome drop-off is expected to decreased bacterial growth efficiency. It is expected that the mutant will allocate less carbon substrate assimilated for biomass because it is mainly used for energy synthesis. Here I tested the hypothesis that ribosome drop-off frequency is correlated with BGE. I hypothesized that the high fidelity mutant with lower translational yield will need more energy for biomass synthesis. A theoretical calculation predicts that the high fidelity mutant will have a 10% decrease in BGE when compared to wild type. This hypothesis was tested by measuring ribosome drop-off frequency and BGE for E. coli K12 wild type and an rpsL(K42N) mutant. Materials and Methods Strains and growth characterization The wild-type and streptomycin resistant mutant for ribosome drop-off assay were E. coli XAC and E. coli SmR obtained from Dr. Albert Dahlberg (11). It is the same strains that were 96 used in Dong and Kurland’s study on ribosome drop-off (5). Cultures were maintained in LB medium, which was also used to grow cultures for recombineering procedures. Ribosome dropoff assay was performed in M9 minimal medium supplemented with 19 amino acids (no methionine) and the carbon source is 0.4% glycerol (M9-Gly54) (8). Another E. coli strain used in this study is E. coli MG1655 provided by Dr. Robert Britton at Michigan State University. Growth characterization was made in M9 minimal medium and the carbon sources are 0.2% glucose, 0.2% glycerol or 0.4% acetate. E. coli mutant rpsL6 was constructed by recombineering as described below, by introducing a single point mutation. Plasmid construction All plasmids were constructed by restriction digestion and ligation reactions. The E. coli specific plasmid carrying the gene ylc (pYLC), which encodes the reporter protein YLC (a chimeric protein from the fusion of enhanced yellow fluorescent protein (EYFP), ß-galactosidase and enhanced cyan fluorescent protein (ECFP)) and pYC were constructed by Drs. Uri Levine and Mike Weigand (Figure A-1). Briefly, the gene encoding EYFP was amplified from pEYFP with primers containing HindIII and BamHI restriction sites and cloned into the multiple cloning sites (MCS) upstream of gene encoding ECFP in pECFP (BD Biosciences, Palo Alto, CA). This construction is name pYC. The lacZ gene from pSV-ß-galactosidase control vector (Promega, Madison, WI) was amplified using primers engineered with BamHI and AgeI restriction sites. Plasmid pYC was digested with restriction enzymes BamHI and AgeI to insert the lacZ amplicon between eyfp and ecfp. This final product is the pYLC plasmid. In both pYC and pYLC plasmids, the genes yc and ylc are expressed under the control of lac promoter, which can be induced by 97 isopropyl-ß-D-1-thiogalactopyranoside (IPTG). The negative control plasmid, pBla, was constructed by digested pECFP with StuI and PvuII followed by blunt-end ligation. Figure A-1. Plasmid pYLC. bla: ampicillin resistance gene, ori: origin of replication, Plac: lac promoter, eyfp: yellow fluorescent protein gene, lacZ: ß-galactosidase gene, and ecfp: cyan fluorescent protein gene. Ribosome drop-off Ribosome drop-off assay requires quantification of independent expression of two reporter proteins, YC and YLC. YC is a fusion of the enhanced yellow fluorescent protein and enhanced cyan fluorescent protein, abbreviated EYFP:ECFP while the YLC chimeric protein has a ß-galactosidase inserted between the two fluorescent proteins. Non-specific fluorescence was measured from a culture carrying pBla plasmid, which is the background plasmid of pYC and pYLC. 98 Overnight cultures of E. coli XAC and E. coli SmR expressing pBla, pYC and pYLC were inoculated into M9 medium supplemented with 18 amino acids and 100 µg/ml ampicillin, and 0.4% glycerol as carbon source to an OD600 of 0.2 and induced with IPTG to a final concentration of 2 mM. Following induction, OD600 and relative fluorescence units (RFU) were measured at every increment of 0.1 OD600 until OD600 reaches 0.8. Optical density was measured using Spectronic 20D+ while fluorescence were measured using Spectramax M5 spectrofluorometer at excitation 435 nm and emission 475 nm with a cutoff filter at 455 nm for ECFP expression, and excitation 520 nm and emission 575 nm with a cutoff filter at 550 nm for EYFP expression. The background fluorescence from pBla was substracted from pYC and pYLC expression to correct for any non-specific fluorescence. The corrected fluorescence signals were then divided by the corresponding OD600 to normalize to biomass. The ratio of ECFP expression to EYFP expression for each chimeric protein was calculated as RFU(C)/OD divided by RFU(Y)/OD. Ribosome processivity, which is completion of translation, is expressed as, R = average C/Y ratio from YLC average C/Y ratio from YC An R value equals to 1 means that the ribosome always complete translation. If drop-off occurs, there will be less ECFP from YLC than YC and hence, R will be less than 1. Drop-off frequency can be calculated as 1-R. Construction of K42N mutant by recombineering Plasmid pSIM6 carrying phage genes required for recombination was obtained from Dr. Robert Britton at Michigan State University. It was introduced into E. coli MG1655 by 99 electroporation according to the protocol from Molecular Cloning (4, 16). Successful transformants were selected by plating on LB agar plates supplemented with 100 µg/ml ampicilin. The transformants were confirmed by PCR amplification of the ori region in plasmid pSIM6 using SD3 (5’-CTGAAAGCACAGCGGAGATGTGTATAAGAGACAGCTGGACAG TAAGA-3’)-SD4 (5’-TTGTATGGAACAACGAGATGTGTATAAGAGACAGCTGACGGGT TTTG-3’) primer set (4). The expression of recombination genes in the transformants was o induced for 15 minutes at 42 C. The induced cells were then made electrocompetent after induction by washing with ice-cold deionized water according to the protocol in Current Protocols in Molecular Biology (18). An 80 bp oligo, RPSL3 (5’-ACG AAC ACG GCA TAC TTT ACG CAG CGC GGA GTT CGG TTT GTT AGG AGT GGT AGT ATA TAC ACG AGT ACA TAC GCC ACG TT-3’, corresponding to lagging strand that complements position 208 to 287 in the rpsL gene was introduced into the electrocompetent cells expressing recombination genes by electroporation. The electroporated cells were grown in SOC medium for 2 hours at o 32 C. Selection of positive recombinants was made by plating the culture on LB agar plate supplemented with 100 µg/ml streptomycin. The pSIM6 plasmid was cured by growing the recombinants at 37oC followed by selection of colonies that do not result in products when amplified with SD3-SD4 primer set. The base pair change in rpsL gene was confirmed by sequencing product from PCR amplification with RPSL1-RPSL2 primer set (1). BGE measurements using 14 C-labeled substrate BGE measurements for all strains were made in M9 medium with glycerol as the carbon source and supplements were as described in the figures. The strains were grown in 750 ml side- 100 o arm flasks with magnetic stir bars, at 37 C and shaking at 200 rpm. Optical density at 600 nm for each culture was monitored with a Thermo Spectronic 20D+. The cultures were added with uniformly labeled 14 C-glycerol (0.25 µCi, Sigma-Aldrich, St. Louis, MO) at early log-phase, when OD600 is approximately 0.1. Immediately after 14 C-glycerol addition, the flasks were sealed and triplicates of 100 µl sample were removed into 5 ml scintillate (Biosafe II, RPI) to determine total label added. The culture was incubated for two doublings and then added with 5% trichloroacetic acid (TCA) to stop growth and to acidify the culture to release any CO2 produced into the headspace. The headspace was vented with nitrogen gas into three phenethylamine:methanol (1:1 v/v) base trap arranged in tandem for two hours. During these two hours, the acidified culture was placed on ice while stirring to promote precipitation of macromolecules. At the end of the two hours, triplicate of 1 ml samples were collected from the base traps into 5 ml scintillate to measure the activity of 14 C-CO2. This was the measurement for bacterial respiration (BR). Bacterial production (BP) was measured as the amount of 14 C activity in the TCA precipitated fraction, which was collected by filtration through 0.7 µm glass fiber filter (Whatman GF/F, Piscataway, NJ). Bacterial growth efficiency (BGE) was calculated from BP and BR using the formula BGE = BP/(BP+BR). Results Chimeric protein YLC was developed as a reporter protein for quantification of ribosome drop-off. It is a fusion of enhanced yellow fluorescent protein (EYFP), ß-galactosidase and enhanced cyan fluorescent protein (ECFP). Ribosome drop-off will result in a ECFP:EYFP ratio 101 of less than 1 with the assumption that drop-off occurring prior to the completion of EYFP mRNA is negligible. To correct for differential expression between different strains of bacteria, the ECFP:EYFP ratio from YLC expression was compared to the ratio from YC expression. It is expected that more drop-off will occur during the translation of the longer YLC mRNA and hence, less ECFP will be synthesized from YLC mRNA translation than from YC mRNA translation. Throughout exponential phase, R in both wild type and streptomycin resistant mutant, strain SmR, exhibit a constant processivity (Figure A-2). The SmR strain has significantly lower ribosome processivity than the wild type strain (Table A-1). This means that the mutant has a higher ribosome drop-off frequency because it completes translation less often than the wild type strain. This result is consistent with Dong and Kurland’s measurement for the same strains of E. coli (5). Figure A-2. Ribosome drop-off frequency for wild type (WT) and SmR mutant (SmR) in M9Gly54 medium during log-phase growth. R-value is calculated as ECFP:EYFP expression from pYLC plasmid/ECFP:EYFP expression for pYC plasmid when induced with 2 mM IPTG. Each point represents the average of three measurements with the standard error. Growth efficiencies of wild type and SmR mutant were measured in the same medium, M9 medium supplemented with 19 amino acids and 0.4% glycerol as the carbon source (M9102 Gly54). Bacterial growth efficiency was measured using 14 C-glycerol. Unexpectedly, no difference in BGE was detected between the wild type and SmR mutant strains (Figure A-3). However, the amount of glycerol consumed in both wild type and SmR mutant cultures were only 23 and 38%, respectively (data not shown). This indicates that the bacteria are using amino acids as the major carbon and energy source. Therefore, BGE measurement using 14 C-glycerol in this medium is not a suitable representation of the growth efficiencies for these two strains. BGE measurement using 14 C-glycerol was repeated in M9 medium with 0.2% glycerol as carbon source and supplemented with only proline and arginine. Both strains are auxotrophic for proline and arginine therefore these two amino acids could not be eliminated from the growth medium. In this medium, the mutant was observed to be more efficient than the wild type strain 3 (Figure A-3). BGE measured using H-leucine/oxygen consumption method in the same medium yield the same observation, where the mutant is more efficient than the wild type strain (Figure A-3). Strain E. coli XAC C/Y of pYLC 0.796 (0.021) C/Y of pYC 1.136 (0.041) E. coli SmR 0.692 (0.033) 1.257 (0.051) R value 0.701 (0.009) 0.75 0.550 (0.011) 0.53 Table A-1. The R value for E. coli XAC and E. coli SmR strains carrying plasmid pYC or pYLC and their R values. The values shown are the averages of three independent experiments with standard deviations in parentheses. Values in bold are values reported by Dong and Kurland (5). The capability of streptomycin resistant mutant to be more efficient at resource allocation for growth was confirmed with a different strain of E. coli, E. coli MG1655 and an rpsL mutant. The mutant, rpsL6-K42N, was constructed by introducing a base pair change into the rpsL gene by recombineering. The mutant is resistant to the antibiotic streptomycin and has increased 103 doubling time compared to wild-type strain when grown on glucose as a carbon source (Figure A-4). The doubling time for wild-type increases as the carbon source is changed from glucose to glycerol or acetate. In M9-acetate, the wild type strain had a doubling time four times longer than growth in glucose while the rpsL6-K42N mutant only increased doubling time by 2.8-fold. This resulted in the mutant growing at a faster rate than the wild type strain in M9-acetate (Figure A4). 0.9 0.8 0.7 WT SmR BGE 0.6 0.5 0.4 0.3 0.2 0.1 0 GlyAA GlyPA(14C) GlyPA(3H) Figure A-3. Bacterial growth efficiency of E. coli wild-type and streptomycin resistant strains in three different conditions. GlyAA: M9 medium supplemented with 19 amino acids and 0.4% glycerol. GlyPA: M9 medium supplemented with 0.4 mM proline and arginine, and 0.2% 14 glycerol. BGE in GlyPA medium was measured using two different methods; C-glycerol and 3 3 H-leucine/oxygen. The H-leucine/oxygen measurement was provided by Roller, B. from Schmidt lab. Each bar represents the average of two replicates with standard deviation as error bars. Similar to the observations for E. coli XAC and E. coli SmR, the streptomycin resistant mutant, rpsL6-K42N, is more efficient. It allocates at least 5% more carbon into biomass synthesis when compared to the wild-type strain (Figure A-5). The distinction is more evident 3 when BGE was measured using H-leucine/oxygen consumption, which is a more sensitive measure of carbon allocation specifically into biomass. 104 Relative  Doubling  Time   5   4   WT   rpsL6   3   2   1   0   Glucose Glycerol Acetate Figure A-4. Relative doubling time of E. coli MG1655 wild-type and rpsL mutant strain (rpsL6K42N) in M9 minimal medium with 0.2% glucose, 0.2% glycerol or 0.4% acetate as carbon source. The doubling time was normalized to growth of wild-type strain in M9-Glucose. Each bar represents average of three biological replicates and the error bar represents standard error. 2.0   Relative  BGE   1.6   14C-­Gly   3H-­Leu   1.2   0.8   0.4   0.0   WT   rpsL6   Figure A-5. Relative BGE of E. coli MG1655 and rpsL mutant in M9 minimal medium with 0.2% glycerol as carbon source. BGE was measured with two different methods and normalized to the WT strain for each method. Each bar represents average of two replicates and the error bar represents standard deviation. 105 Discussion Streptomycin resistant mutant with a K42N mutation in ribosomal protein S12 have increased ribosome drop-off frequency when compared to wild-type strain. This is consistent with Dong and Kurland’s measurement for these two strains of E. coli (5). Premature dissociation of ribosome leads to the formation of truncated proteins that may not be properly folded into its functional conformation. Therefore, it is expected that the mutant will require more ATP per unit functional protein. The high cost of protein synthesis is expected to decrease growth efficiency of the mutant. Contrary to the theoretical expectation, the streptomycin resistant mutant exhibit increased BGE in all the growth conditions tested except when the medium was supplemented with 19 amino acids. Therefore, I was unable to provide evidence that ribosome drop-off can lead to a decrease in growth efficiency. An alternative explanation for the increased growth efficiency exhibited by the mutant is that missense error, which is the incorporation of the non-cognate tRNA, is more costly than the synthesis of truncated polypeptide. Missense error occurs at a rate of 1 in 1000 to 10000 codons (14) but can increase during nutrient starvation (13). It is hypothesized that missense error can sequester the ClpX protease away from hydrolyzing sigma factor 38 (6). Sigma factor 38, also known as RpoS, is a sigma factor that is being induced during carbon starvation. It is important for stress protection and the mechanism is by arresting metabolic activity while upregulating stress protective enzymes such as enzymes involved in oxidative stress. In the streptomycin resistant mutant, the absence of aberrant proteins due to low missense error allows ClpX protease to hydrolyze RpoS, even when the cells have entered stationary phase (6). This allows the cells to continue incorporating substrate into biomass while in the wild-type strain, increased missense 106 errors induced an RpoS-dependent growth retardation, allocating more resources to synthesizing protective enzymes. It was previously reported in Salmonella enterica, that K42N mutants had a higher growth rate in medium with poor carbon source (15). This increased growth rate is correlated with decreased RpoS in the cells compared to the wild type strain. In this study, we suggest that the increased growth rate observed in poor medium is due to increased growth efficiency. When grown in the presence of 19 amino acids, the culture does not experience amino acid starvation until it reaches stationary phase, which will not be detected because BGE was measured during exponential growth. However, when majority of the amino acids are removed, the wild type strain may experience amino acid starvation due to inefficient translation. The dependency of RpoS stability on ClpX protease activity suggests that missense error in wild-type strain is more costly than ribosome drop-off. Therefore, we hypothesized that the increased growth efficiency exhibited by the mutant is due to increased translational accuracy. Future Directions To determine if the mutant is truly more efficient at protein synthesis or the increased BGE is due to decreased missense error, we first have to measure the missense error rate for the wild-type and mutant strains under the growth conditions that BGE was measured. BGE measurements and missense error rates must also be measured in the same medium with and without amino acid supplement to determine if the correlation between growth efficiency and ribosome fidelity is dependent on amino acid requirement. 107 Acknowledgements I like to acknowledge Drs. Uri Levine and Mike Weigand for constructing the pYLC plasmid, Dr. Dion Antonopoulos for assistance in plasmid constructions and Dr. Stephanie Eichorst for teaching me how to measure BGE and helpful discussions. I like to thank Ben Roller 3 for sharing the BGE measurements he made using H-leucine. I am also thankful for the help from Dr. Scott Mulrooney in the protein purification work. I am also grateful to Dr. Thomas Schmidt for his understanding, support and patience throughout the course of this entire project. I especially thank Dr. Robert Britton for all the helpful discussions and advice he provided to me. 108 REFERENCES 109 REFERENCES 1. Amoako, K. K., T. Prysliak, A. A. Potter, S. K. Collinson, W. W. Kay, and B. J. Allan. 2004. Attenuation of an avian pathogenic Escherichia coli strain due to a mutation in the rpsL gene. Avian Dis 48:19-25. 2. Bohman, K., T. Ruusala, P. C. Jelenc, and C. G. Kurland. 1984. Kinetic Impairment of Restrictive Streptomycin-Resistant Ribosomes. Mol Gen Genet 198:90-99. 3. Das, G., and U. Varshney. 2006. Peptidyl-tRNA hydrolase and its critical role in protein biosynthesis. Microbiology 152:2191-2195. 4. Datta, S., N. Costantino, and D. L. Court. 2006. A set of recombineering plasmids for gram-negative bacteria. Gene 379:109-115. 5. Dong, H. J., and C. G. Kurland. 1995. Ribosome Mutants with Altered Accuracy Translate with Reduced Processivity. J Mol Biol 248:551-561. 6. Fredriksson, A., M. Ballesteros, C. N. Peterson, O. Persson, T. J. Silhavy, and T. Nystrom. 2007. Decline in ribosomal fidelity contributes to the accumulation and stabilization of the master stress response regulator sigma(S) upon carbon starvation. Gene Dev 21:862-874. 7. Heurgue-Hamard, V., L. Mora, G. Guarneros, and R. H. Buckingham. 1996. The growth defect in Escherichia coli deficient in peptidyl-tRNA hydrolase is due to starvation for Lys-tRNA(Lys). The EMBO journal 15:2826-2833. 8. Jorgensen, F., and C. G. Kurland. 1990. Processivity Errors of Gene-Expression in Escherichia coli. J Mol Biol 215:511-521. 9. Liljas, A. 1990. Some structural aspects of elongation, p. 309-317. In W. E. Hill, A. E. Dahlberg, R. A. Garrett, P. B. Moore, D. Schlessinger, and J. R. Warner (ed.), The Ribosome: Structure, Function and Evolution. American Society for Microbiology, Washington, D. C. 10. Lilley, D. M. 2001. The ribosome functions as a ribozyme. Chembiochem 2:31-35. 11. Lodmell, J. S., and A. E. Dahlberg. 1997. A conformational switch in Escherichia coli 16S ribosomal RNA during decoding of messenger RNA. Science 277:1262-1267. 12. Menez, J., R. H. Buckingham, M. de Zamaroczy, and C. K. Campelli. 2002. PeptidyltRNA hydrolase in Bacillus subtilis, encoded by spoVC, is essential to vegetative growth, whereas the homologous enzyme in Saccharomyces cerevisiae is dispensable. Mol Microbiol 45:123-129. 110 13. Ofarrell, P. H. 1978. Suppression of Defective Translation by Ppgpp and Its Role in Stringent Response. Cell 14:545-557. 14. Ogle, J. M., and V. Ramakrishnan. 2005. Structural insights into translational fidelity. Annu Rev Biochem 74:129-177. 15. Paulander, W., S. Maisnier-Patin, and D. I. Andersson. 2009. The Fitness Cost of Streptomycin Resistance Depends on rpsL Mutation, Carbon Source and RpoS (sigma(s)). Genetics 183:539-546. 16. Sambrook, J., and D. W. Russell. 2001. Molecular cloning : a laboratory manual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 17. Stouthamer, A. H. 1973. Theoretical Study on Amount of ATP Required for Synthesis of Microbial Cell Material. A Van Leeuw J Microb 39:545-565. 18. Thomason, L., D. L. Court, M. Bubunenko, N. Costantino, H. Wilson, S. Datta, and A. Oppenheim. 2001. Recombineering: Genetic Engineering in Bacteria Using Homologous Recombination, Current Protocols in Molecular Biology. John Wiley & Sons, Inc. 111