such as "Introduction", "Conclusion"..etc
Our approach was to develop a microbial system that reports sugar
levels in a reaction that mimics the NREL lignocellulosic ethanol
production process . The presence of microbes in this system overcomes the feedback inhibition problems associated with enzyme-only methods. Escherichia coli strain
CA8404 was selected as the microbe because it carries the crp* mutation
which reduces catabolite repression and thereby allows both the 5- and
6-carbon sugars produced from corn stover hydrolysis reactions to be
metabolized simultaneously . As the E. coli metabolize the sugars from the hydrolysis reaction, cell mass (X) changes at a rate of dX/dt.
This change in culture cell mass can be monitored by light-scattering
measurements of culture turbidity. However, certain reactions,
including lignocellulosic biomass hydrolysis, produce substances that
interfere with light-scattering measurements. In order to more easily
monitor change in cell culture mass, the E. coli strain CA8404 was modified to produce a visual marker, green fluorescent protein (GFP) [10,11].
The sharp emission peak and specific wavelength requirement for
excitation allow for a much greater specificity of detection than does
light scattering alone. The version of GFP used in this study (S65T)
has a maximum excitation wavelength of 490 nm and a maximum emission
wavelength of 510 nm .
Site-directed mutagenesis was conducted according to the Stratagene product QuikChange II® Site-Directed
Mutagenesis kit in order to produce a pPNptGreen plasmid without a
functional GFP fluorophore (Stratagene, La Jolla, CA). Approximately
300 bp into the GFP coding sequence, DNA encoding a glutamate residue
(GAA) was changed to encode a stop codon (TAA). (Primers for
mutagenesis: Forward: GATGACGGGAACTACAAGACACGTGCTTAAGTCAAGTTTGAAGG;
Reverse: CCTTCAAACTTGACTTAAGCACGTGTCTTGTAGTTCCCGTCATC.) The new plasmid
was designated pPNptOchre. The original pPNptGreen plasmid and the
pPNptOchre plasmid without the functional GFP fluorophore were
transformed separately into E. coli strain CA8404 to produce the two strains, crp*-gfp and crp*-gfp-.
At grain maturity, cobs were removed from the corn plants and all
corn stover samples were cut at approximately six inches above the soil
by a forage chopper. Approximately 0.8 kg of sample (wet weight) at a
moisture content of about 35% was collected from each plot and samples
were dried at 55°C for one week. The material from each sample was
ground by a hammermill with a 1 mm screen. We did not evaluate the
distribution of different botanical tissues in the ground material;
however, the ground material appeared uniform. To minimize the effect
of a non-uniform distribution of biological material in the sample we
used multiple subsamples in each experiment.
To characterize the product inhibition of the enzyme preparation Multifect® A-40
(a cellulase/hemicellulase mixture from Genencor Intl.), we carried out
hydrolysis reactions in the presence or absence of 10 mM d-glucose
(CAS# 50-99-7, Sigma-Aldrich Inc., St. Louis, MO). Each treatment was
run with four replicates using 5 mg of a stover sample treated with a
1:20 dilution of enzyme Multifect® A-40 in citrate-phosphate
buffer (21 ml 0.1 M citric acid and 29 ml 0.2 M sodium phosphate, in a
final volume of 100 ml, pH 5.5). Hydrolysis was conducted at 60°C for
90 min. Following hydrolysis, the tubes were centrifuged for 1.5 min.
at 10,000 × g in a microcentrifuge (Spectrafuge, Orem, UT).
An aliquot of the supernatant from the hydrolysis reaction was measured
with a hexokinase glucose assay kit (Sigma-Aldrich Inc., St. Louis,
MO). The absorbance was measured at 340 nm (OD340) using the
MRXII plate reader by DYNEX (Magellan Biosciences Company, Chelmsford,
MA). The absorbance value was converted to glucose yield with a
standard curve constructed by plotting OD340 values versus glucose concentrations following analysis of a series of solutions with known glucose concentrations.
Cultures of E. coli crp*-gfp and crp*-gfp- paired by treatment were grown in modified 1 × M9 minimal media . The M9 media was modified by the addition of Kanamycin (50 μg/ml),
thiamine (0.01% w/v), and ammonium chloride (5 mg/ml). Also, different
carbon source concentrations were provided to the cultures than the
carbon source described by Sambrook and Russell .
d-glucose (CAS# 50-99-7, Sigma-Aldrich Inc., St. Louis, MO), and
d-xylose (CAS# 58-86-6, Sigma-Aldrich Inc., St. Louis, MO) solutions
were made in the appropriate concentrations indicated in each
experimental procedure below. All sugar solutions were
filter-sterilized and frozen. Sugar mixtures were combined from
separate, sterilized glucose and xylose sugar solutions. Cultures were
grown in clear, 96-well cell culture plates (Product # 92096, Techno
Plastic Products, Trasadingen, Switzerland) and sealed with AirPore™
seals (Qiagen, Valencia, CA) in order to ensure that enough oxygen was
available to the cultures. The plates were then securely fastened down
in the Innova 4300 incubator shaker (New Brunswick Scientific, Edison,
New Jersey), and allowed to incubate with shaking at 37°C and 225 rpm.
When it was time to take a measurement, the AirPore™ seal was removed
only from the wells to be measured, and absorbance (OD595)
measurements were taken by the MRXII plate reader (Dynex – a Magellan
Biosciences Company, Chelmsford, MA). The samples from the wells to be
measured were then transferred into a black, 96-well cell culture plate
(Corning Incorporated Life Sciences, Lowell, MA), and fluorescence
measurements (excitation wavelength: 485 nm, emission wavelength: 535
nm) were taken by the SpectraFluor Plus plate reader (Tecan US,
Research Triangle Park, NC). Note that these wavelengths (595 nm for
absorbance, 485 nm for excitation, and 535 nm for emission) were used
consistently throughout the study. The AirPore™ seal was replaced on
the clear 96-well plate and returned to the incubator. To obtain a
value for GFP-specific fluorescence for each culture pair, the
fluorescence reading of the crp*-gfp- strain was subtracted from the fluorescence reading of the crp*-gfp strain.
establish whether it was possible to use GFP to detect differences in
changes in culture cell mass in response to sugars, cultures of E. coli crp*-gfp and crp*-gfp- were
grown in modified 1 × M9 minimal media containing 2, 4, or 8 mg/ml
d-glucose. Absorbance and fluorescence were measured every 2 h for 22
h, and the GFP-specific fluorescence was determined.
To determine the sensitivity and dynamic range of the microbial system, cultures of E. coli crp*-gfp and crp*-gfp- were
grown in modified 1 × M9 minimal media containing d-glucose in
concentrations ranging from 0.025 mg/ml to 6.0 mg/ml. Absorbance and
fluorescence measurements were taken 20 h after inoculation, and the
GFP-specific fluorescence was determined.
To determine the response time of the
microbial system, glucose was added to the reaction when the culture
reached stationary phase. Two sets of three replications of both E. coli strains crp*-gfp and crp*-gfp- were
grown in modified 1 × M9 minimal media containing 2 mg/ml d-glucose for
20 h. After 20 h, half of the cultures (one set) were randomly selected
to receive an addition of 8 mg/ml d-glucose for a total of three
replications each of spiked cultures and unspiked cultures. Absorbance
and fluorescence were measured every 2 h, and the GFP-specific
fluorescence was determined.
Another way we examined the
response time of the microbial system was by stopping protein
production when the culture was in mid-log phase. Six replications of
both E. coli strains crp*-gfp and crp*-gfp- were
grown in modified 1 × M9 minimal media, containing 20 mg/ml d-glucose.
Chloramphenicol was added to a random selection of half of the cultures
after 13 h for a total of three replications each of cultures with
chloramphenicol and without chloramphenicol. Absorbance and
fluorescence were measured every hour, and the GFP-specific
fluorescence was determined.
The microbial system described
here, referred to as simultaneous saccharification and catabolism (SSC)
was used to analyze corn stover samples of five different corn
varieties. For each sample to be analyzed, 25.0 ± 0.2 mg of dried and
ground corn stover was weighed into two separate 14 ml sterile test
tubes (BD Biosciences, San Jose, CA). Two tubes were used to control
for variations in fluorescence of the corn stover samples: an
experimental tube to be inoculated with crp*-gfp and a control tube to
be inoculated with crp*-gfp-. The difference in the fluorescence of these two tubes was used to determine the GFP-specific fluorescence. Then 1150 μl of 0.5% (v/v) sulfuric acid were added to each tube, and the tubes were incubated at 100°C for 1 h . The tubes were allowed to cool for 15 min after incubation, after which 3850 μl
of bacterial media inoculum (2 × M9 media inoculated with the
appropriate bacterial culture) was added to each tube. 1 l of bacterial
media inoculum contained 620 ml sterile water, 330 ml 5 × M9 salts, 6.6
ml 1 M MgSO4, 164.2 μl 1 M CaCl2, 1.7 ml thiamin at 10%, 8.3 ml kanamycin at 10 mg/ml, and 33 ml crp*-gfp or crp*-gfp- liquid culture (grown overnight at 37°C in 1 × M9 media to an OD595 of ~0.6). In addition, 25 μl of 1:1 GC220: Multifect® Xylanase
(Genencor Intl.) were added to each tube. The tubes were allowed to
incubate with shaking at 37°C and 225 rpm. Samples containing 100 μl
of 0, 2, 4, 6, 8, 10, 12, 14, 16, or 18 mg/ml sugar at ratios of 37.5
xylose: 62.5 glucose in place of corn stover were included as positive
controls. Absorbance and fluorescence were measured after 20 h of
incubation by allowing the stover particles to settle in the culture
tube and transferring 100 μl of the culture to a 96-well plate.
The GFP-specific fluorescence values were computed by subtracting the fluorescence from crp*-gfp- cultures
from crp*-gfp cultures. These values were then analyzed by ANOVA in
order to characterize variation in the experiment. When variation was
significant, a student's t-test was performed on each pair to
compare means of the samples. The coefficient of variance (CV) was also
computed in order to determine the variation in measurements for each
The SSC method described
above was used to analyze stover samples from five corn varieties.
Absorbance and fluorescence were measured every 2 h for 24 h and once
at 36 h.
Enter the code exactly as it appears. All letters are case insensitive, there is no zero.