A Green Technology for the Production of Biofuels Dan Whalen
A Green Technology for the Production of Biofuels
Dan Whalen
Abstract
. The past several decades have been demarcated by growing concerns about
climate change, a global condition attributed to the emission of green house gases
(GHG). The mitigation of this global threat has been met by the development of green
technologies and alternative fuel sources in an effort to diminish GHG production. One
of the most promising technologies currently being developed is the conversion of
agricultural biomass into alternative fuel sources. Indeed, the conversion of cellulose into
alternative energy sources has become a cornerstone of green technologies. However, the
development of efficient methods for the use of hemicellulose, a major component of
woody materials, in the production of alternative fuels remains elusive. In the present
study, we developed a microbial system capable of converting xylose, the dominant
structural component of hemicellulose, into valueadded products. A microbial
consortium grown under various nutrient conditions allowed the efficient metabolic
conversion of xylose into hexoses. Microbes cultured in a standard mineral medium were
found to have the highest conversion rates of xylose to glucose. In contrast, microbes
exposed to nutrient stress preferentially generated mannose. The proficient metabolism
of xylose was attributed to an NADPdependent xylose dehydrogenase. Intriguingly,
NADPdependent xylose dehydrogenase displayed a lower activity profile in the nutrient
stress. Thus, we have developed a microbial system capable of converting xylose into
fermentable hexoses. The optimization of this process will enable the production of fuels
in a sustainable manner.
Acknowledgements
I would like to thank Dr Vasu Appanna for his guidance and support throughout
the duration of my project. His willingness to address questions and concerns allowed
me to be confident in my analysis and approach to the project.
My appreciation goes out to Dr. Ryan Mailloux for his patience in guiding me
through a vast array of new procedures. Despite my constant harassment, he continued to
answer my questions with his full attention. Without him , this project would not have
been possible.
Joe Lemire for helping me refine some of my new found lab techniques as well as
providing general support.
My lab mates Steph Leclair, Rami Darwich and Catehrine Kossar for making this
experience as enjoyable and smooth rolling as it was.
My girlfriend, Cara Blasutti, and my parents, Oscar and Kate Whalen for support
through this year which was personally very trying. Without them I would have had to
abandon the project before it started.
Table of Contents ` Page
1.Introduction 5 Global Warming 5 Alternative Fuels 5 Carbohydrates in Lignocellulosic Biomass 6 General Bacteria Metabolism 8 Xylose Metabolism in Soil Bacteria 11
2. Materials 13 2.1 High Phosphate Media 13 2.2 Peroxide Stress 13 2.3 Zinc Stress 14 2.4 Low Phosphate Stress 14 2.5 Bacterial Storage 14 2.6Growing Cultures 14 2.7 Cell Storage Buffers 15 2.8 HPLC Mobile Phase 15
3. Methods 16 3.1 Bradford Biomass Assay 16 3.2 Dubois Carbohydrate Assay 16 3.3 HPLC Analysis 16 3.4 Isolation of Cellular Fractions 17 3.5 Blue Native PolyarylamideGel Electrophoresis 17 3.6 InGel Staining for XDH Characterization 19 3.7 Bial’s Test for Pentose Concentration 20
4. Results 20 4.1 Bacterial Growth Curve 21 4.2 HPLC Analysis 21 4.3 Carbohydrate Content 23 4.4 Dubois Assay 23 4.5 Bials Test 23 4.6 Gas Chromatography 24 4.7XDH Activity Staining 25 4.8 Zinc and Low Phosphate Cultures 26 4.9 Pentose Concentration in Zinc and Low Phosphate Cultures 27
5. Discussion 27 5.1 Biomass 28 5.2 Spent Fluid Analysis 28 5.3 Dubois Assay 29 5.4 Pentose Concentration 29 5.5 Gas Chromatography 30 5.6 Xylose Dehydrogenase Characterization 31
6. Optimization and Further Research 34
List of Figures Page
Figure 1: Glycolysis………………………………………………………………….9
Figure 2: The Citric Acid Cycle……………………………………………………..10
Figure 3 : Proposed Pathway for DXylose Metabolism……………………………12
Figure 4 : Bacterial Growth Curve in 20mM Xylose………………………………...21
Figure 5: Chromatogram of Spent Fluid HPLC……………………………………...22
Figure 6: Results of Dubois Assay…………………………………………………...23
Figure 7 : Bial’s Test Results………………………………………………………….24
Figure 8: XDH Activity Stains………………………………………………………..26
Figure 9: Bacterial Growth Curve for Zinc and Low Phosphate Cultures……………27
Figure 10: Pentose Concentration in Zinc and Low Phosphate Cultures……………..28
Figure 11: Metabolic Pathway for DXylose Metabolism In Soil Bacteria…………...29
List of Tables Page
Table 1: Conversion Rates of Xylose into Hexosans…………………………………25
1. Introduction:
Global warming has become one of the most severe environmental issues in recent
history. The impact of climate change will be severe if measures are not taken to reverse
and or slow the mechanisms contributing to the global increase in temperature.
Greenhouse gas production is the greatest contributor to the rise in global temperature.
The accumulation of greenhouse gases (GHG) in our atmosphere has caused the heat
waves produced as the suns radiation strikes the earth to become trapped inside the
earth’s atmosphere (Crowley 2000). CO2 emissions produced from hydrocarbon
combustion in automobiles, coal power stations, and aircraft account for the damaging
effects of GHG, Thus, cleaner fuels and alternative energy sources are required to
diminish GHG levels. While nuclear and wind power generators are replacing the carbon
dioxide producing coal power stations, transportation is in need of a major overhaul in
order to meet the environmental needs of the future.
Alternative Fuels
Current projects underway which are capable of replacing hydrocarbon burning
vehicles include hydrogen fuel cells, as well as alcohol fueled engines. Hydrogen is
possibly the cleanest initiative proposed as it uses the energy stored in hydrogen to power
the vehicle with H20 as the major waste product. However, issues such as safe
transportation due to hydrogen’s highly combustible nature as well as sufficient supply
are hurdles in its acceptance into the mainstream infrastructure (Ashley 2005). Ethanol
on the other hand is perhaps the leading liquid fuel replacement for hydrocarbons as it
produces very little carbon dioxide as a biproduct. The benefits of switching to ethanol
based fuel supplies are that very little needs to be done to either current automotives or
the infrastructure which supplies them to make the switch to ethanol ultimately happen.
This is because ethanol is both less volatile than hydrogen and can increase the efficiency
of hydrocarbon combustion (Lee 1997). The major argument against ethanol
development as a major fuel source is that there exists insufficient supply resources for its
production. Indeed, current methods for ethanol production involve using the starch from
corn as the glucose source for ethanol fermentation. The problem with this process is
that the yield of corn required for mass scale production is simply far too high. Corn has
a relatively low energy per mass ratio meaning that in order to supply the world’s
growing energy needs; corn would have to be grown over the entire earth’s available
surface (Sun and Cheng 2002). However, alternative sources of fermentable sugars are
now being employed to fill the energy gap. Other fermentable sugar sources include
cellulose and hemicellulose. Butanol has also become an interesting prospect due to its
higher energy density than ethanol and lower hydroscopicity. Butanol can be produced
using metabolic and genetic engineering techniques on bacteria such as Escherisia coli in
order increase the formation of Butanol (Atsumi et al 2008). Ethanol has been criticized
somewhat because a method for storage which nullifies its higher hydroscopicity has not
yet been found. Some techniques have been devised as of late which are approaching
viability, involving creating butanol and other high carbon alcohols using Escherechia
coli along with some non native enzymes (Atsumi et al 2008).
Carbohydrates from Lignocellulosic Biomass
Cellulose is the most abundant carbohydrate on the planet and is therefore an
attractive prospect for the production of biofuels. Cellulose is simply a polymer of D
glucose which can be harnessed to generate ethanol efficiently using glucose fermenting
bacteria (Lin and Tanaka 2005). Cellulose is the major structural component of plant cell
walls, wood, and bark making this Dglucose polymer the greatest renewable energy
resource on the planet. This makes biomass resources easily available in many rural
areas, and consequently may also help play an important role in providing jobs in rural
communities if it can be efficiently utilized as a fuel source (Lin and Tanaka 2005). The
first step in the preparation of lignocellulosic biomass for ethanol production is to
delignate it to free the polymers of cellulose and hemicellulose from the binding agent
lignin. Treatment of wood with chlorous acid yields a holocellulose mixture which upon
treatment with potassium hydroxide will produce a soluble hemicellulose/cellulose
mixture. The cellulose must then be broken apart with either chemical hydrolysis or
biological means using the enzyme cellulase (HahnHagerdahl et al. 1994). The
liberated Dglucose can then be fermented using Sachromyces cerevisae or other
fermentative bacteria to produce ethanol (Sedlack and Ho 2004). The fermentation of D
glucose has been performed efficiently for many hundreds of years as this fermentation
process is nearly identical to that used by beer and wine makers. The ethanol can then be
further distilled in order to be made properly combustible. Lignocellulosic biomass also
contains hemicellulose, a complex polymer of hexosans and pentosans. Hemicellulose is
closely associated with lignin forming a major structural moiety in higher plant tissues.
The structure of Hemicellulose is composed of a Dxylose backbone with side chains of
arabinose, hexoses and modified carbs. Roughly 30% of the dry mass of lignocellulosic
biomass is actually the pentose sugar containing polymer Hemicellulose. 30% of
hemicelluloses sugars in hardwood trees is DXylose, a number which can reach 70% in
many types of straw, sugarcane bagasse and soybean stalks (Lin and Tanaka 2006).
Current methods are being studied using xylose fermenting bacteria to maximize ethanol
production and efficiency. The presence of Dxylose along with Dglucose in the
hydrolyzed product presents biochemical problem when trying to directly ferment both
simultaneously. This is because the glucose inhibits the xylose fermentation and the
xylose inhibits the fermentation of glucose when using recombinant fermentative bacteria
(Sedlack and Ho 2004). Indeed, the inability o fermenting microbes to properly ferment
xylose and other pentosans makes the use of agricultural biomass for biofuel production
inefficient. Thus, it is important to find more economically viable methods of converting
xylose into value added products in order to increase the overall efficiency of
lignocellulotic biomass fermentation.
General Bacterial Metabolism
Bacteria are the key to finding cheap efficient ways to metabolize wood
carbohydrates because evolution has allowed for the broad development of their
metabolic pathways over millions of years. Bacteria rely on various metabolic pathways
in order to convert various carbon sources into building blocks and ATP. Carbohydrate
metabolism in bacteria involves a complex network of metabolic reactions which include
parts of glycolysis, gluconeogenesis, the citric acid cycle, as well as the Pentose
Phosphate Pathway and other intermediate shunts. Glycolysis is the initial phase in the
production of energy from glucose and involves a series of 10 enzyme catalyzed steps
which conclude with the production of the 3 carbon pyruvate molecules. The first stage
of glycolysis involves the conversion of D glucose into the trioses, glyceraldehyde3
phosphate and DHAP. This conversion requires the phosphorilation, isomerization, and
aldol cleavage of Dglucose. Stage 2 of glycolysis involves the NADdependent
oxidation of the trioses, a process coupled to the generation of 4 ATP. In order to
maintain glycolytic flux the NADH is oxidized by the conversion of pyruvate to ethanol,
a process called alcoholic fermentation. Alcoholic fermentation is required to maintain
glycolytic efficiency when O2 is scarce. However, when O2 is available, NADH is
oxidized by the electron transport chain. (Horton et al).
Fig 1 Glycolysis
Gluconeogenesis is basically the reverse of glycolysis. When ATP production is
no longer required pyruvate is used to fuel the production of nucleotide and carbohydrate
precursors needed for biosynthetic reactions. The 3 enzymes which are not homologous
between both glycolysis and gluconeogenesis are the enzymes which modulate the rate of
the reactions in either direction and are controlled indirectly by the ATP levels inside the
cell. If ATP reserves in the cell are lower than required however, the pyruvate produced
during glycolysis is brought into the mitocondria where it is decarboxylated by pyruvate
dehydrogenase to form Acetyl CoA. The Citric acid cycle then combines a molecule of
oxaloacetate which is formed as a product of the previous cycle to form citrate. The citric
acid is then systematically oxidized by the concerted action of TCA cycle enzymes
generating NADH, FADH2, and evolving CO2.
Fig 2: The Citric Acid Cycle
The NADH and FADH2 donate their electrons to the respiratory complexes. The
electrons are then passed down their electrochemical gradient to the terminal electron
acceptor O2. This generates the H+ gradient required to drive ATP production. Bacteria
can also produce acetyl CoA from lipids by cleaving the fatty acids from the glycerol in
stored lipids with phosphodiesterases to produce 3 fatty acid chains which are essentially
chains of Acetyl groups which are easily converted to acetyl CoA(Horton et al 2006).
These two pathways are key in the production of energy in all non photorespiring cells,
both eukaryotic and prokaryotic.
Xylose Metabolism in Soil Microbes
Soil microbes have been shown to exhibit some variation in the metabolism of
both xylose and arabinose. Since Xylose is a major constituent of agricultural biomass,
the aim of this project is to develop a microbial system capable of converting Dxylose
into value added products. If the bacteria are metabolizing bacteria for energy
production, xylose tends to follow either the proposed Weinberg or Dahms pathway. The
intial step in both the Dahms and the Weinberg proposed pathways is the oxidation of
xylose into xylonolactone by xylose dehydrogenase, while reducing one molecule of
NAD+ in the process. In the next step D Xylonate is produced through the use of the
enzyme xylonolactase. Subsequent removal of a molecule catalyzed by Xylonate
Dehydratase yields 2 keto 3 deoxyxylonate and it is here where the two proposed
pathways begin to differ (Stephens et al 2006). In the weinbeg pathway the xylonite is
dehydrated and then oxidized before becoming aKG which enters the TCA cycle for
either energy production or so it can exit as malate to be further transformed for storage
purposes as in fig 1. The Dahms pathway involves utilizing the 2 keto 3 deoxypentose as
a substrate to produce one molecule of pyruvate and one glycoaldehyde molecule.
Fig 3 Proposed pathway fro DXylose Metabolism
If the bacteria cell determines that ATP levels inside the cell are adequate it can
then shift its metabolism to produce carbohydrates such as glucose and mannose in a
more direct manner by utilizing reactions of the Pentose phosphate pathway (PPP). This
is the common mechanism utilized by yeasts to metabolize xylose into ethanol. In this
proposed mechanism the xylose is transformed into xylitol by using nadp+ dependent
xylsoe reductase. Xylitol is then converted to xylulose 5 phosphate, a substrate for PPP
degredation through subsequent oxidation and phosphorylation reactions . The xylose 5
phosphate then enters the Pentose phosphate pathway to yield pyruvate which can either
be used for anaerobic energy production yielding ethanol, or sequestered for glucose
production through the gluconeogenic pathway. The conversion of xylose to glucose
would also be an economically viable process as glucose can easily be fermented into
ethanol as well. By producing additional ethanol or glucose from the hemicellulosic
components of lignocellulose the overall viability of utilizing biomass would increase
considerably.
The media in which any organism is grown accounts for the specificity of the
metabolic pathways utilized by that organism both for energy production and energy
storage purposes. By determining experimentally the preference of specific bacteria for
producing a desired product it is possible to maximize the production of that product.
The reason for this is that the buildup of metabolites, reactive oxygen species, and other
chemicals can change the ability of various enzymes in the cell to deal with their
substrate. In this experiment I will be growing cultures of soil bacteria in a variety of
control and nutrient stress environments in order to analyze and maximize the production
of value added products such as keto acids, pyruvate, butanol, ethanol and glucose. I
will be monitoring the growth rate and protein concentration at confluency of the bacteria
in each media using Spectrofluorometric techniques. The metabolic components of the
media at various timepoints will also be analyzed using various methods. Some enzyme
analysis will also be performed in order to try and determine the metabolic direction of
both control and stress media cultures in an effort to steer the metabolism of the organism
towards whichever value added metabolite is being produced. My goal is to produce in
an efficient manner a metabolite from xylose which could offset the cost of ethanol or
butanol production, both considered to be viable alternatives for the dwindling oil
reserves and increased GHGs of our world.
2. Materials
Microbial growth conditions and media
High phosphate control media was prepared by adding 12.0g Na2HPO4, 6.0 g KH2PO4,
0.4 g MgSO4∙7H2O, 1.6 g NH4Cl, and 1 mL of a trace element solution to 600 mL of
ddH2O. Trace element solution contained 2 μM FeCl3∙6H2O, 1 μM MgCl2∙4H2O, 0.05
μM Zn(NO3)2∙6H2O, 1 μM CaCl2, 0.25 μM CoSO4∙7H2O, 0.1 μM CuCl2∙2H2O, 0.1 μM
NaMoO4∙2H2O. The pH of the trace element solution was adjusted to 2.75 with dilute
HCl, and the solution was stored at 4 o C. The pH of the trace element solution was
adjusted to 6.8 with dilute NaOH and the volume was brought to 360 mL with ddH2O.
After stirring and bringing the pH of the phosphate media to 6.8 with 2M HCL, the final
volume was brought to 2L by adding ddH20. The media was subsequently autoclaved at
121 o C for 60 min and allowed to cool to room temperature. Xylose Stock was prepared
by adding 6.0g of DXylose to 200ml of ddH20 and then stirred. The Xylose stock was
sterilized using a disposable vacuum unit with a pore size of 0.2 microns to sterilize.
20ml of 20mm Xylose and 180 mL of High Phosphate media were then added to a
500mL Erlenmeyer flask. The final concentration of Xylose in the control growth media
was 20mm.
Peroxide Stress
The peroxide media was prepared as above however 200ul of 100mM peroxide was
added to the 500ml Erlenmeyer flask along with 180ml of High Phosphate media and
20ml of Xylose stock.
Zinc Stress
Zinc media was prepared by adding 400ul of 10mM Zinc to 180ml of High Phosphate
Media and 20ml of Xylose Stock.
Low Phosphate Stress
Low phosphate media was prepared by adding100x less Na2HPO4. The media therefore
included 0.12g Na2HPO4, 6.0 g KH2PO4, 0.4 g MgSO4∙7H2O, 1.6 g NH4Cl, and 1 mL of
a trace element solution. The volume of the media was adjusted to 2L and sterilized as
above. The Xylose Phosphate media contained 180ml of low Phosphate media and 20ml
of Xylose stock.
Bacterial Storage
Xylose slants were prepared by 270ml of High Phosphate Media and 4.8g of Agar to a
500ml beaker and was stirred over heat until it dissolved. The solution was then
autoclaved for 60 min at 121 o C in order to sterilize. 30ml of Xylose stock was then
added to the warm solution and 10ml portions of the agar solution was poured into sterile
test tubes. The slants were allowed to harden with the test tubes lying at an acute angle
inorder to maximize the surface area of the hardened gel. After the slants were cooled,
they were then refrigerated at 4 degrees Celcius. Slants were then inoculated every time
a new culture was started and allowed to grow for 4 days until visible growth was
observed. The inoculated slants were then refrigerated at 4 degrees Celcius.
Growing Cultures
Precultures were made adding 90ml of Phosphate media and 10ml of Xylose stock into a
250ml Erlenmeyer flask. The media was then inoculated with a streak from a fresh
Xylose slant incubated for 48 h in a water bath shaker (model G76, New Brunswick
Scientific) at 26 o C and 140rpm. 1mL of preculture was then isolated and added to
200ml of each of the 4 different stress media in a 500mL Erlenmeyer flask. These
cultures were again allowed to grow in the water bath shaker and 10ml samples were
isolated at 8h post inoculation and then 24h intervals up to 96h and refrigerated at 4
degrees Celsius. Biomass isolation was performed by subsequent centrifugation of the
samples at 4000 rpm for 30 min and removal of the supernatant. The pellet was then
resuspended in 1ml of NaOH.
Cell Storage Buffer (CSB) with DTT
Cell storage buffer was made by adding0.03084 g Dithiothreitol (1 mM), 1.211 g
Trizma base (50 mM), 0.035 g Phenylmethylsulfonylfluoride (1 mM). DTT and Trizma
base to 80 mL of ddH2O and heated until fully dissolved. PMSF was then added and the
solution was cooled and brought to pH = 7.4 with 2 M HCl. Total volume of 200 mL was
achieved by adding of ddH20. The buffer was then covered and refrigerated at 4 º C.
Tris Reaction Buffer
Reaction buffer was made using (5 mM) MgCl2 and 25 mM Tris in a final volume of 1 L.
Tris and MgCl2 were added to 500 mL ddH2O. A pH of 7.4 was then achieved using 2M
HCl. Final volume was then brought to 1 L using ddH2O.
HPLC Mobile Phase
HPLC mobile phase was prepared by adding 2.72 g of 20mM (≥99.9%) KH2PO4 to 500
mL ddH2O. Mobile phase pH was adjusted to 2.9 with 2 M HCl.
3. Methods
Determination of Biomass
Biomass was monitored using the Bradford Assay. 10mL isolated samples were
centrifuged at 4000rpm for 30min. The supernatant was removed and saved for HPLC
analysis. The pellets were resuspended in 1ml of NaOH, placed in microcetrifuge tubes,
and boiled for 5 minutes in order to degrade the cell membranes and denature proteins.
1.5 mL cuvettes were used for the assay. Samples were diluted in the cuvette with 200uL
of Bradford reagent. The volume was then brought to 1mL with ddH2O and allowed to
stand for 5 minutes. The absorbance was then measured at 595nm using a
spectrophotometer. BSA, at a final concentration of 10ug/mL was used as the standard.
Dubois Carbohydrate Assay
Total carbohydrate concentrations were determined using the Dubois
carbohydrate assay. 800 µL of ddH2O, 200 µL of sample supernatant, 1 mL of phenol
(5% w/v) and 5 mL of H2SO4 (90% v/v) were mixed in 10 mL test tubes. Absorbance
was measured at 488 nm by transferring 1 mL of each reaction mixture into 1.5 mL
cuvettes. Standard curves were developed using Dglucose varying from 0.1 to 1mg/mL.
Bials Test for Pentose Concentration
In order to measure the consumption of Xylose and formation of Pentose sugars
in the culture media, a Bial’s Assay was performed. Bial’s reagent was prepared by
adding 0.3g of orcinol and 0.05 g of ferric chloride to 100mL of 12M HCl. A standard
curve was created adding 0.01ml, 0.02 ml, 0.03ml, 0.04ml, and 0.05ml of 1mg/ml xylose
stock to 5 different test tubes. 1ml of Bial’s reagent was then added to each test tube and
boiled for 5 minutes in a fume hood. The contents of each test tube were then transferred
to 1.5 ml cuvettes and their absorbance measured at 660nm to measure the formation
furfurol which forms a blue green product in the presence of orcinol and ferric chloride.
Samples were prepared in triplicate by adding 150ul of the spent fluid of 8, 24, 48, 72, 96
and 120 hour isolations to small test tubes. 3ml of Bial’s reagent was then added to each
test tube and the tubes were heated in a boiling water bath for 5 minutes. The absorbance
was again measured at 660nm. Both control and stress media were tested.
Metabolic analysis and HPLC
The supernatant of all samples were separated using a Phenomex Reverse Phase
C18 column with an injection volume of 10 µL and a flow rate of 0.7 mL/min. The
separations module is coupled with a UV/VIS spectrophotometer that was calibrated at an
absorbance of 240 nm in order to pick up organic acids. Samples were prepared for
HPLC by running 2mL of spent fluid through 1mm of cotton in a Pasteur pipette .
Samples were then injected into the HPLC column.
Subcellular fractionation and preparation for Blue Native PAGE
Cell cultures were grown for 30h and then poured into centrifuge tubes. After
centrifuging for 30 min at 4000rpm the supernatant fractions were removed and the
pellets were resuspended in 0.85% NaCl and then centrifuged again at 4000rpm for
another 30min. The supernatant fractions were discarded and the pellets were
resuspended in 500ul of CSB. Low intensity sonication was performed (Brunswick
Ultrasonic Processor) with four cycles of 15 s at power level ≤4 on ice. This was done to
lyse the cell membranes and liberate to separate cytoplasmic and membrane bound
proteins. The lysate was then separated into membrane and soluble fractions by
centrifuging at 50 000 rpm in 40º C for 2h. The soluble fraction was decanted and placed
in a micro centrifuge tube. The pellet was resuspended in minimal of CSB. Protein
concentrations of both fractions were determined using the Bradford method. 133ul of 3x
BN buffer, 50 ul of 10% malticide, sufficient fraction to give a final protein concentration
of 125 ul/ml were put into a microcentrifuge tube and the final volume brought up to
500uL with ddH20. Membrane fractions and 500ul of soluble fractions were then frozen
at 21ºC until needed fro Blue Native Page electrophoresis.
Blue Native Polyacrylamide Gel Electrophoresis (BNPAGE)
The BioRad MiniProtean 2 electrophoresis system was used for running all of the gels.
Table 2 outlines the contents of all polyacrylamide gels. The separating gel was
composed of 2.9 mL of 4% acrylamide solution and 2.9 ml of 16% acrylamide solution.
416% linear gradient gels were generated using a gradient former (BioRad) and a
peristaltic pump. This type of BioRad gradient allows for broad range separation, with
1mm spacers being used in all cases. After all the gel was in the form the gel was
overlaid with isopropanol to allow for proper polymerization. After waiting 20 minutes
for solidification of the separating gel, the isopropanol was absorbed from on top. The
stacking gel ingredients were then mixed together and inserted on top of the stacking gel
with a pasture pipette. While the stacking gel was still liquid, a plastic comb was inserted
to allow for the formation of wells. After polymerization of the stacking gel the combs
were removed and 60ug of protein was applied to the wells and subsequently overlaid
with cathode buffer (see table 3 for BNPage buffers). The gels were then placed in a
electrophoresis tank with the inner portion of tank being filled with blue cathode buffer
and the outer tank was filled with anode buffer. The tank was then placed in the
refrigerator at 4ºC and a voltage of 80V was applied between ends of the gel until the
protein had migrated into the separating gel. Upon the proteins reaching the separating
gel the voltage was increased to 200V. Once the proteins had migrated halfway through
the separating gel the blue cathode buffer was replaced with colorless cathode buffer.
Once the running front had reached the bottom of the gel the electrophoresis was halted.
Reagent 4% Gel 16% Gel Stacking Gel 49.5% Acrylamide 243 937 273 3x BN Buffer 967 967 1136
ddH2O 1699 223 2000 75% Glycerol 0 773 0 10% APS 9.7 7.6 30 TEMED 1 0.8 2.5
Ingredients for 4%16% BNPAGE Gels. All volumes are in µL.
Solutions for BNPAGE
Blue Cathode Buffer (1L) 8.96 g Tricine (50 mM) 3.138 g BisTris (15 mM) 0.2 g Coomassie blue G 250 pH 7.0 at 4 o C
3x BN Buffer 9.84 g Aminocaproic acid (1.5 M) 1.567 g BisTris (150 mM) pH 7.0 at 4 o C
Coomassie Brilliant Blue Staining Solution 50% Methanol 10% Glacial acetic acid 0.2% Coomassie Blue R 250
Colourless Cathode Buffer (1L) 8.96 g Tricine (50 mM) 3.138 g BisTris (15 mM) pH 7.0 at 4 o C
Anode Buffer 10.45 g BisTris (50 mM) pH 7.0 at 4 o C
Destaining Solution 50% Methanol 10% Glacial acetic acid
InGel Stain for Xylose Dehydrogenase Characterization
Activity of Xylose Dehydrogenase (XDH) was visualized by separating
membrane cell fractions on a 4%16% BNPAGE gel and incubating the gel in 4 different
reaction mixtures. All reaction mixtures contained 250ul phenozine methyl sulfate
(PMS), 500ul of Iodonitrotetrazolium (INT), and 75ul of 200mM xylose. In order to
characterize the enzyme XDH, I then varied the concentration of two cofactors, NAD+
and NADP+. One mixture was 0.5mM NAD, one was 0.1 mM NAD, one was 0.5mM
NADP and one contained 0.1mM NADP. Gel slabs were incubated in 3mL of reaction
mixture per dual lanes of control and stress. Formazan precipitate forms directly over the
site of enzyme catalysis with the amount of precipitate being proportional to enzyme
activity. Upon visualization, the reaction is quenched by transferring the gel slab into
destaining solution.
4. Results
As shown in figure 4 both the peroxide stressed and control cell cultures
reached confluency in roughly 30 hours. However, the biomass of peroxide stress was
less than control. The biomass measurements were for total protein content in the cell
fraction of the media and were measured using the Bradford method. Upon reaching
confluency, the biomass in both cultures remained relatively stable.
Bacterial Growth Curve in 20mm Xylose Media
0.05
0
0.05
0.1
0.15
0 24 46 72 96 120
Time Of Isolation(hrs)
Bio
mas
s (T
otal
Prot
ein,
mg/
ml)
Control
Peroxide Stress
Fig 4 :Bacterial Growth in 20mM Xylose Media
HPLC Analysis
Keto Acids are a valuable commodity to the medical community thus we
deciphered if xylose was being converted into either pyruvate or aKG.
Figure 5: Chromatogram of 20mM Xylose Control (left) and Peroxide stress media c
(right) cultures spent fluid
HPLC analysis of the media revealed the presence of pyruvate from the peroxide stressed
cultures. Indeed a small peak at 5 minutes was recorded in the peroxide stressed media.
No peak was observed in the control. However, the spike was not significant; in fact the
absorbance spike of the pyruvate did not even match the peak in absorbance due to the
mobile phase. After calculating the integral of the area underneath the curve at 5 minutes
the concentration of pyruvate in the stress culture was found to be in the microgram per
milliliter range. The results from the HPLC spent fluid analysis demonstrated an overall
lack of buildup with regards to metabolic intermediates.
Measurement of Carbohydrate Levels
pyruvate
0.008
0.001
0.004
2 4 6 8 10
A
Minutes
pyruvate
0.008
0.001
0.004
2 4 6 8 10
A
Minutes 4 10
0.008
A
minutes
0.001
4 10
0.008
A
minutes
0.001
A Dubois assay was performed to measure the total carbohydrate content in the
spent fluid of both control and peroxide stressed cells. Both cultures showed a slight rise
in carbohydrate concentration after the low point at 24hr, followed by a stable period of
little to no change in carbohydrate levels. The control media however showed a sharp
rise in carbohydrates after 120 hours.
Total Carbohydrate Content in Spent Fluid of 20mM Xylose Cultures
0
0.1
0.2
0.3
0.4
24 48 72 96 120
Time of Isolation
Carbo
hydrate
Con
centratio
n (m
g/mL)
control stress
Figure 6: Results from Dubois assay showing relatively stable carbohydrate
concentrations with low points at 24hr isolation. Control media had an increase in
total carbohydrates at the 120 hour mark.
A Bial’s test was then performed in order to detect the levels of pentose sugars in
the mixture. Results show that the low point in pentosans occurs around the 24 hours of
growth point. This is coincides with a time to confluency for both cultures of around 30
hours. After the 24 hour point there was some increase in pentose concentration as
xylose is evidently converted into other pentoses. The concentration of pentose sugars
goes from 3g per Liter at the time of inoculation, all of which is xylose, to a low
concentration of 1g per Liter after 24hours in both culture media. These data indicate
that xylose is consumed quite rapidly since the concentration of xylose at time 0 was
3g/mL. Furthermore, these data suggest that xylose is converted into hexose sugars.
Thus, we setout to identify which hexoses were being produced.
Pentose Concentration
0
0.005
0.01
0.015
0.02
0.025
8 24 48 72 96 120
Time of Isolation (Hours)
Pen
tose Con
centratio
n (m
g/ml)
control stress
Figure 7: Bial’s test results showing a depletion of pentosans after 24 hours
and a slight increase after 120 hours.
Isolations from both cultures were taken after 24 hours and the spent fluid was
sent to Paprican labs for carbohydrate analysis. The lab used gas chromatography to
distinguish between carbohydrates in the mixture. Results indicated that almost all of the
xylose in both control and stress media was consumed after 24 hours. In the control
media the xylose concentration went from 20mM to 1.6mM at the 24 hour isolation and
the concentration after 24 hours in the stress media was 1.7mM. The overall
carbohydrate content in the mixture was still quite high remaining at 10mM. The
conversion of Xylose into other sugars was apparent with glucose concentration of 6mM
in the control culture and 4.5mM in the stress media after 24 hours of incubation. The
remainder of the carbohydrates in the spent fluid was mannose with concentrations of
2.6mM for the control and 2.9mM in the peroxide sample. The level of glucose and
mannose produced are significant given the market value and ethanol yielding properties
of these two carbohydrates. Table 1 provides a summary of xylose to hexose conversion
rates.
Type of hexosan/Culture Xylsoe to Mannose Xylose to Glucose
Control 10% 29%
Stress 14% 18%
Table 1: Conversion rates from xylose into hexose sugars.
Xylose Dehydrogenase Activity Staining
BlueNative PAGE was performed on membrane fractions form both control and
stress cultures. After separating the protein, activity stains were performed in order to
test fro activity of XDH under different concentrations of either NAD + or NADP+
cofactors. This was performed in order to better characterize xylose dehydrogenase. The
consumption of NAD+ would indicate that xylose dehydrogenase could directly impact
the aerobic respiration of the cell. A strict dependence on NADP+ could play an
important role in the ability of the bacteria to control reactive oxygen species as NADP+
is essential not only in biosynthetic reactions but in antioxidant defense.
Fig 8 Xylose dehydrogenase activity stains with 0.1mM nadp+ (top left), 0.5
mM nadp (top right), 0.1mM nad+ bottom left, and 0.5 mM nad+ (bottom right).
Control membrane fraction in left lane, stress membrane fraction in right.
The activity of XDH was greatest in the reaction mixtures containing NADP+ as a
cofactor. The reaction mixture containing 0.5mM NADP+ showed the most substantial
activity bands, however even at concentrations of 0.1mM nadp+ the activity of XDH was
quite evident. The 0.1 mM NAD+ showed very little activity whereas the 0.5mM NAD+
reaction mixture showed some moderate activity. In all reaction mixtures the activity
bands were more evident for the control fractions then for the stress fractions.
Zinc and Low Phosphate Cultures
Zinc and low phosphate cultures were also started and their biomass observed to
determine if these cultures would be viable in the production of value added products
especially monosacharides due to the tendency of these stresses to increase
gluconeogenesis. Results showed a slower time to confluency than the other two cultures
and consequently I focused mainly on the control and peroxide stress cultures from that
point on.
Bacterial Growth in 20mM Xylose
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18
24 48 72 96
Time of Isolation (hrs)
Biomass (m
g/ml o
f total
protein)
zinc low phosphate
Fig 9 Bacterial Growth Curve for Zinc and Low Phosphate Stress cultures
Near the end of the project I did however have time to perform a Bial’s test on the
zinc and low phosphate cultures. Results showed a low level of pentose production later
on in the growth phase which may indicate production of hexosans or keto acids. Further
investigation should be performed in order to better analyze the metabolism of the
bacteria in these cultures.
Pentose Concentration in Spent Fluid
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
24 48 72 96
Time of Isolation
Pen
tose Con
centratio
n (m
g/ml)
low phosphate zinc
Figure 10 Pentose Concentration in low phosphate and zinc stressed cultures
Discussion
The formation of economically viable products from xylose is an important step
in bringing lignocellosic biomass to the forefront of alternative fuel technology. The aim
of this project was to try to find an efficient method for converting xylose into value
added products such as ethanol, butanol, glucose or keto acids by manipulating the xylose
containing media. 20mM xylose cultures were grown in control, peroxide, zinc and low
phosphate mineral mediums, although much of the focus was placed on the control and
peroxide stress cultures. It was identified that much of the xylose was converted to
hexoses in both control (39 percent) and peroxide (32 percent) cultures. Production of
organic acids in the spent fluid was insignificant.
Biomass
The biomass was measured using the Bradford method. Growth rates for control
and peroxide cultures were relatively similar as both reached confluency in just under 30
hours. Protein concentration for peroxide cultures at confluency was slightly lower than
in the control media with protein concentrations of 1.2 and 1.3mg/ml respectively. After
reaching confluency the biomass in each culture remained quite stable. The oxidative
stress caused by the peroxide in the media was the most likely contributor to the lower
cellular levels at confluency as high levels of peroxide is know to be caustic.
Hydrogen Peroxide was chosen as a stressor due to its ability to increase cellular levels of
ROS. These toxic moieties have been shown to perturb various cellular structures and
functions. For instance, ROS are capable of altering various enzymes within the TCA
cycle and possibly causing the buildup of keto acids in the culture media. Thus,
hydrogen peroxide is an excellent candidate for modifying the molecular pathways of
xylose metabolism.
Spent Fluid Analysis
HPLC was used to analyze the spent fluid of both cultures at 24 hour increments.
The absorbance of the eluate was measured at 210nm in order to detect the presence of
organic acids such as pyruvate and alpha keto glutarate. I hypothesized the buildup of
organic acids in the peroxide stressed media as ROS have been shown to inhibit TCA
cycle enzymes such as akG dehydrogenase. Results from the HPLC analysis showed that
there was however very little buildup of either pyruvate or akG, suggesting that either the
TCA was functioning optimally or the metabolism of the bacteria was shifted mostly
towards gluconeogenesis. In the peroxide stressed cells there was a small peak after 5
minutes, however the peak did not even manage to eclipse that of the mobile phase which
eluted from the stationary phase after 4 minutes. It was clear that in both control and
stress cultures the metabolism of the bacteria was not geared to produce organic acids in
high concentrations.
Dubois Assay
The next step in the metabolic analysis of microbes in the two cultures was to
measure the carbohydrate levels in the spent fluid. The Dubois assay itself measures only
the presence of aldehyde groups which are present in both pentose and hexose sugars.
Both control and stress showed relatively stable carbohydrate concentrations in their
media from inoculation to 120 hours post inoculation time points. Since it was obvious
that xylose was being consumed as the bacteria proliferated, it was clear that there was a
production of another carbohydrate in the growth media. The production of glucose,
mannose or other carbohydrates in the growth media would account for the overall levels
of carbohydrates in the growth media being stagnant. At the 120 hour time point, a spike
in carbohydrate levels in the control media was observed which may have been due to a
gluconeogenic shift in the overall metabolism in the cell causing a buildup of glucose in
the spent fluid.
Pentose Concentrations
In order to better understand the rate at which pentoses were being consumed as
well as produced, a Bial’s test was performed. The production of furfurol from pentose
which produces a blue green color in the presence of iron and orcinol in high molar HCl
was monitored at 660nm. Results showed that the lowest concentrations of pentosans
were present 24 hours after bacterial inoculation in both stress and control media. Low
levels in the stress media were 0.011mg/ml and 0.012mg/ml in the control cultures spent
fluid. This suggests that the highest concentrations of non pentose carbohydrates would
occur around this time since results form the Dubois assay show a stable concentration of
carbohydrates in the media. After 24 hours, pentose concentration increased slightly in
the peroxide cultures back up to levels 0.019mg/ml after 120 hours. In the control media,
pentose concentrations increased more rapidly than in the stress media, with pentose
concentrations rising to 0.02mg/ml after 48 hours and peaking at 0.022mg/ml at the 96
hour point. The pentose concentrations after 24 hours clearly were not sufficient to
account for the carbohydrate levels found in the Dubois assay, meaning that there must
be substantial hexose production in the media of both control and stress cultures.
Gas Chromatography
Upon realizing that there must be significant production of hexose sugars in the
media of both xylose cultures, spent fluid samples were sent away to Paprican labs in
Montreal for carbohydrate analysis. Results indicated significant glucose production in
both control and stress samples, as well as the production of some mannose. In the
control fractions, samples contained; 244mg/L xylose, 388mg/L mannose and 894mg/L
of glucose. This is a considerable amount of hexose production given that initial xylose
levels in the media were 3g/L. This equates to a 29% conversion of xylose to glucose
and a 10% conversion to mannose in the control cultures. In the stress cultures, samples
contained, 275mg/L xylose, 427mg/L mannose and 568mg/L of glucose. Therefore,
more conversion of xylose to mannose occurred in the stress cultures and less conversion
to glucose. The conversion of xylose to glucose in the stress fractions was 18% and for
mannose 14 %. Thus, the control media appeared to be much more efficient at producing
glucose, while being slightly less efficient at producing mannose. The actual
concentrations received from the gas chromatography results differed when compared
with those from both the Dubois and Bial’s assay. However, the assay results are still
helpful when trying to understand relative levels of carbohydrates along various isolation
time points. The efficiency of glucose production, especially in control cultures, was
quite significant and could be a sufficient driving force with regards to the economic
viability of this process. Further trials would be beneficial in trying to elucidate even
higher levels of glucose and mannose from the xylose. Given that pentose concentrations
were lowest at the 24 hour point, this would be the optimal point for hexose isolation.
Xylose Dehydrogenase Characterization
Initially a BN PAGE was performed to test if xylose dehydrogenase activity
occurred in control and stress fractions. Activity was noted in membrane fractions of
both control and stress cultures. In order to better understand the manner in which xylose
is metabolized in my cultures I performed another BN PAGE to observe the activity of
XDH between control and stress fraction. High activity of XDH would suggest xylose
was being metabolized through the Weinberg pathway for quick entry into the TCA or
entry into the gluconeogenic pathway for glucose production. If XDH levels were low
this would indicate xylose reduction into xylitol by xylose reductase and subsequent entry
into the Pentose Phosphate Pathway. I also wanted to determine the affinity of XDH for
either nadp+ or nad+. This would give insight as to whether XDH is more important in
generating nadph for antioxidant defense, or nadh to help balance the overall energy
budget of the cell. Cellular membrane fractions were isolated 30 hours after inoculation
when the cultures were already confluent. The activity appeared to be greater in control
fractions in all four reaction mixtures prepared. XDH demonstrated a clear preference
for nadp+ over nad+. This was illustrated by the fact that although there was a
reasonably strong activity band at nad+ levels of 0.5mM, when the cofactor concentration
was reduced to 0.1 mM there was little to no XDH activity. On the other hand, XDH
produced strong band at both concentration of nadp+ with the reaction mixture containing
the 0.5mM NADP+ showing the most substantial activity bands.
The above results which illustrate that nadp+ dependant XDH is active in both
control and stress cultures shows that the Weinberg pathway is being used predominantly
to metabolize xylose. Strong activity bands in the control fractions indicate that the
nadp+ dependant pathway has a significant role in xylose degredation. Upregulation of
nadp+ dependant XDH would suggest that ATP levels in the cell are being achieved
through the quick entry into the TCA cycle provided by the Weinberg pathway. It is clear
however that much of the pyruvate pool created by the degredation of xylose is being
routed though gluconeogenesis in order to build up glucose for storage. XDH bands in
peroxide fractions were not as strong as in control fractions. The time to confluency and
protein concentrations at confluency for the two cultures were quite similar. This
indicates that in the peroxide cultures, xylose was being metabolized through a
combination of the Weinberg pathway as well as through the xylose reductase mediated
pathway. Peroxide has been shown to increase the production of ROS, and consequently
induce ETC disfunction as well as inhibit key TCA enzymes.(Mailloux et al).
Xylose Reductase Pathway Weinberg Pathway
Figure 11 Metabolic Pathways for DXylose in Soil Bacteria
In peroxide stressed cells it is likely that xylose was being metabolized into
mannose as a means to produce energy anearobically to compensate for lowered aerobic
ATP production. This could easily occur through the transformation of ribulose 5
phosphate, a PPP intermediate, into fructose 5 phosphate. Fructose 5 phosphate could
then be isomerized into mannose 6 phosphate and subsequently transformed into
mannose coupled with the production of one mole of ATP by mannose kinase. This
appears to have been occurring as higher production of mannose in stressed cells
illustrates up regulation of mannose kinase activity. Xylose metabolism into fructose 6
Citric Acid Cycle
Xylotal Lactone
Nadp+ Nadph
pyruvate
? Xylose reductase
XDH
Alpha KetoGlutarate
Gluconeogenesis
Citric Acid Cycle
Xylotal Lactone
Nadp+ Nadph
pyruvate
? Xylose reductase
XDH
Alpha KetoGlutarate
Gluconeogenesis
phosphate quickly through xylose reductase and the PPP would account for down
regulation of XDH and the weaker activity bands observed in peroxide fractions.
Optimization and Future Research
In order to try and further increase the amount of xylose being converted into
glucose two more 20mM xylose cultures were started. Low phosphate and 0.02mM Zinc
cultures were started. These cultures were both hypothesized to shut down the TCA
cycle. Zinc was used to try to knock out TCA enzymes, whereas low phosphate cultures
should have limited the proliferation of the bacteria. Both of the above processes should
have limited proliferation of the bacteria and steered the metabolism in the cultures
toward the production of storage polymers, hopefully glucose. The growth profiles of the
zinc and low Phosphate cultures were similar to the control and peroxide cultures in their
biomass at confluency; however confluency did not occur till 48 hours post inoculation.
There was limited time to analyze the metabolites in the cultures of these bacteria and
only a Bial’s test was performed. Pentose concentrations in the media after confluency
were even lower than in the control and peroxide stressed cultures and it is possible that
slightly more glucose was produced. Without performing gas chromatography or HPLC
analysis it is impossible to stipulate as to what metabolites were being produced and how
large their volumes.
Future steps forward for this project would involve studying more closely the
metabolites in the zinc and low phosphate cultures. However, a slower time to
confluency is a slight hindrance in their feasibility as cultures could only be refreshed
half as often as in the control and peroxide media. It appears that the control culture was
the most efficient at producing glucose. By growing these cultures and large media
cyclers one could refresh the media and isolate glucose and mannose every 24 hours.
The glucose and mannose are both valuable products, and large amounts of glucose could
be further fermented into ethanol or butanol for use as biofuels. Further enzyme analysis
should be performed on key metabolic enzymes such as pyruvate dehydrogenase. The
understanding of enzyme regulation at key points would allow for even better growth
media to be designed for maximum production of value added products.
In conclusion it is clear that nutrient engineering can be a useful tool in guiding
the metabolism of microbes into producing value added products. The project uncovered
a novel method for transforming xylose into More research and practice needs to go into
understanding the effects of various media on metabolic enzymes in order to maximize
the efficiency of these processes.
References
Ashley S. Solid progress: hydrogenfuel storage for cars gets a material boost. Sci Am. 2005 Jun;292(6):268.
Atsumi S, Hanai T, Liao J.C.. NonFermentative Pathways for Synthesis of BranchedChain Higher Alcohols as Biofuels. Nature. January 2008. Vol 451
Chandrakant P, Bisaria U. Simultaneous Bioconversion of Cellulose and Hemicellulose to Ethanol. Critical Review of Biotechnology. 2001.
Crowley TJ. Causes of climate change over the past 1000 years. Science. July 14 2000. 2707
Gong Cs, Cao Nj, Du J, Tsao GT. Ethanol Production From Renewable Resources. Adv Biochem Eng Biotechnology. 1999;65; 20741
HahnHagerdal B, Linden T, Senac T, Skoog K. Ethanolic Fermentation of Pentoses in Lignocellulosic hydrolysates. Appl Biochem Biotechnology. 1991 2829: 13144
Horton, Moran, Scrimgeour, Perry and Rawn. Principles of Biochemistry. Prentice Hall. Toronto. 2006
Katzen R, Fowler D. Ethanol from Lignocellulosic Waste with Utilization of Recombinant Bacteria. Appl Biochemical Biotechnology. Spring 1994, 4546 697707
Lee J. Biological Conversion of Lignocellulosic Biomass to Ethanol. Journal of Biotechnology. 1997. July. 56. 124
Lin Y, Tanaka S. Ethanol Fermentation from Biomass Resourses: Current State and Prospects. Applied Journal of Biotechnology. February 2006. 69: 62742
Sedlack M, Ho NW. Production of Ethanol from Cellulosic Hydrolsates Using Genetically Engineered Saccharomyces Yeast Capable of Cofermenting Glucose and Xylose. Appl Biochem Biotechnol. 2004 spring 113116; 40316
Singh R, Mailloux R, PoisseauxDos J, Appanna V. Oxidative Stress Evokes a Metabolic Adaptation that Favors Increased NADP Synthesis and Decreased NADH Production in Pseudomonas fluerescens. Journal of Bacteriology. September 2007. p66656675
Stephens C, Christen B, Fuchs T, Sundaram U, Watanabe K, Jenal U. Genetic Analysis of a Novel Pathway for D Xylose Metabolism in Caulobacter crescentus. Journal of Bacteriology. December 2006. 143806
Sun Y, Cheng J. Hydrolysis of Lignocellulosic Materials for Ethanol Production: A Review. Bioresource Technology May 2002. 83. 111