Previous Article | Next Article 
Applied and Environmental Microbiology, January 2001, p. 377-386, Vol. 67, No. 1
Natural Products Research, Eli Lilly and
Company, Indianapolis, Indiana 46285
Received 30 June 2000/Accepted 17 October 2000
A major barrier in the discovery of new secondary metabolites from
microorganisms is the difficulty of distinguishing the minor fraction
of productive cultures from the majority of unproductive cultures and
growth conditions. In this study, a rapid, direct-infusion electrospray
mass spectrometry (ES-MS) technique was used to identify chemical
differences that occurred in the expression of secondary metabolites by
44 actinomycetes cultivated under six different fermentation
conditions. Samples from actinomycete fermentations were prepared by
solid-phase extraction, analyzed by ES-MS, and ranked according to a
chemical productivity index based on the total number and relative
intensity of ions present in each sample. The actinomycete cultures
were tested for chemical productivity following treatments that
included nutritional manipulations, autoregulator additions, and
different agitation speeds and incubation temperatures. Evaluation of
the ES-MS data from submerged and solid-state fermentations by paired
t test analyses showed that solid-state growth
significantly altered the chemical profiles of extracts from 75% of
the actinomycetes evaluated. Parallel analysis of the same extracts by
high-performance liquid chromatography-ES-MS-evaporative light
scattering showed that the chemical differences detected by the ES-MS
method were associated with growth condition-dependent changes in the
yield of secondary metabolites. Our results indicate that the
high-throughput ES-MS method is useful for identification of
fermentation conditions that enhance expression of secondary metabolites from actinomycetes.
Many traits of microbes, both
positive and negative, are attributed to expression of biologically
active secondary metabolites. These metabolites are typically produced
only under specific physiological conditions, and many are not
essential for growth (6, 33). Although the
physiological role of most secondary metabolites remains an area
of speculation (25), these compounds have been ascribed
apparent roles, including cell differentiation, cell signaling or
communication, nutrient sequestering, and defense (6, 7, 12, 21,
26, 33).
The foremost value of secondary metabolites to humanity has been in
providing the basis for new commercial drugs, both directly (e.g.,
penicillin) and indirectly (e.g., synthetic or semisynthetic compounds
derived from secondary metabolites [8]). In recent years, several factors have increased the difficulty of discovering secondary metabolites with the potential to be drug leads. These factors include the difficulty of managing the compatibility of complex
natural extracts with high-throughput and ultra-high-throughput screening methods, the reality that most discoveries are rediscoveries of known compounds, and competition from laboratory-synthesized chemical diversity, the supply of which has been dramatically increased
by combinatorial methods (4, 14). However, human awareness
of microbial biodiversity has expanded tremendously in the past few
years, so we now recognize that less than a few percent of the
biodiversity on earth have been evaluated in drug-screening programs
(37). This awareness has renewed interest in evaluating microbes as a potential source of new secondary metabolites. The number
of microbes and the biological diversity of microbes are so large that
many strategies are being invoked to cope. Examples include cloning to
express secondary metabolite pathways in surrogate hosts
(36) and development of special methods to cultivate and elicit secondary metabolism in new microbial groups (34).
While the traditional methods of cultivation and elicitation, as well as the newer strategies, are intended to give access to previously unknown secondary metabolites, all methods are limited by the large
number of samples that must be evaluated before meaningful inferences
can be made about the effectiveness of one treatment relative to
another with regard to the ultimate objective, production of secondary metabolites.
Feedback on expression of secondary metabolites comes from two classes
of analyses, biological and chemical (40). Because of the
many vagaries of biological assays (e.g., nonlinear responses, chemical
interference, lack of correlation between bioassay data, and domination
of activity outcomes by a few common metabolites), we favor more
general chemical analyses to provide data to guide improved expression
of secondary metabolites. At present, the best chemical analyses depend
on initial separation of natural product extracts (by thin-layer
chromatography or high-performance liquid chromatography [HPLC])
followed by detection of the resolved secondary metabolites by suitable
detectors (15, 23). However, the benefits of superior
resolution resulting from chromatographic separation of natural product
extracts are often offset by lower throughput because of the greater
time, reagents, and labor necessary for chromatography. Ideally, basic
comparisons of groups of cultures or expression conditions could be
done at a lower cost and with higher throughput yet still account for
differences in the concentrations and diversities of secondary
metabolites present in extracts.
In the accompanying paper we describe a direct-infusion electrospray
mass spectrometry (ES-MS) method that has the sensitivity and
throughput required for analysis of large numbers of natural product
extracts (20); we demonstrated the effectiveness of this
method for ranking cultures grown under one set of common conditions.
In this study we extended the application to the more difficult problem
of comparing the effects of different growth conditions on expression
of secondary metabolites by actinomycetes, and we corroborated
conclusions resulting from the rapid method with the results of the
best available HPLC method.
Organisms and cultivation.
Cultures of 44 actinomycetes
(Table 1) were started from cryogenic
stocks in a vegetative medium that contained (per liter) 30 g of
tryptic soy broth (Difco Laboratories, Detroit, Mich.), 3 g of
yeast extract (Sigma Chemical Co., St. Louis, Mo.), 2 g of
MgSO4, 5 g of glucose, and 4 g of maltose. At the
mid-log phase, 60 µl (~0.4% inoculum) of each vegetative culture
was transferred into 12 ml of complex growth medium A, which contained
(per liter) 10 g of glucose, 40 g of potato dextrin (Avedex,
Keokuk, Iowa), 15 g of cane molasses (Cargill, Minneapolis,
Minn.), 10 g of Hy-case amino (Sheffield Products, Norwich, N.Y.),
1 g of MgSO4, and 2 g of CaCO3; the
final pH was 7.0. Other complex fermentation media used for submerged
fermentations included medium B, which contained (per liter) 5 g
of soybean flour Nutrisoy (Archer Daniels Midland, Decatur, Ill.),
10 g of glucose, 10 g of glycerin, 5 g of soluble starch
(Difco), 20 g of potato dextrin, 0.1 g of
FeCl2 · 4H2O, 0.1 g of
ZnCl2, 0.1 g of MnCl2 · 4H2O, 0.5 g of MgSO4 · 7H2O, 5 g of corn steep powder (Sigma), 3 g of
CaCO3, 1 g of phytic acid (Sigma), and 10 g of
cane molasses (final pH, 7.0); medium C, which contained (per liter)
15 g of hexaglycerol dioleate, 5 g of corn steep powder, 3 g
of CaCO3, 5 g of glucose, 50 g of lactose,
10 g of soybean flour Nutrisoy, 5 g of peptone, 2 g of NH4SO4, 0.1 g of FeCl2
· 4H2O, 0.1 g of ZnCl2, 0.1 g of
MnCl2 · 4H2O, and 0.5 g of
MgSO4 · 7H2O (final pH, 7.0); and medium D, which contained (per liter) 10 g of glucose, 40 g of potato dextrin (Avedex), 15 g of cane molasses (Cargill), 10 g of
Hy-case amino (Sheffield Products), 1 g of MgSO4, 2 g
of CaCO3, and 3 g of Na2HPO4 (final
pH, 7.0).
0099-2240/01/$04.00+0 DOI: 10.1128/AEM.67.1.377-386.2001
Copyright © 2001, American Society for Microbiology. All rights reserved.
Use of Direct-Infusion Electrospray Mass
Spectrometry To Guide Empirical Development of Improved Conditions for
Expression of Secondary Metabolites from Actinomycetes
ABSTRACT
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
INTRODUCTION
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
MATERIALS AND METHODS
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
TABLE 1.
Detection of solid-state-dependent changes in the
chemical compositions of extracts from 44 actinomycetes grown in
submerged and solid-state fermentations by using medium A
-irradiated, vented
polypropylene cap lined with a gas-permeable membrane (Performance
Systematix, Inc., Caledonia, Mich.). Submerged fermentations were
incubated at 30°C for 7 days on an orbital shaker (stroke length, 2 in.) at 110 rpm. For solid-state fermentations, Noble agar (Difco) was
added to fermentation medium A to a final concentration of 1.5%.
Solid-state fermentations were incubated under growth conditions that
were identical to those used for submerged fermentations, except that
solid-state fermentations were incubated without shaking. The majority
of growth was restricted to the surface of the agar, except for the
substrate mycelia, which penetrated approximately 1.5 mm into the agar
for most cultures. Contamination was assessed for all fermentations at
the end of the incubation period by removing approximately 20 µl of
broth and spreading it on tryptic soy agar (Difco). Culture purity was assessed after 2 days of incubation at 30°C, and contaminated fermentations (containing more than one microorganism) were discarded. The rate of contamination for fermentations completed in this study was
less than 0.8%.
Liquid fermentations were prepared for chemical analysis by addition of
an equal volume of absolute ethanol to the original fermentation
vessel. Solid-state fermentations were prepared for chemical analysis
by two methods. In the initial experiments the fermentations were
prepared so that they were analogous to the liquid fermentations
described below. Later, we modified the cultivation method by
introducing a nylon membrane between the agar and the colony to allow
separation for quantitation (see below). For the initial solid-state
fermentations, grown directly on agar, samples were prepared by
solublization of secondary metabolites by the protocol described below
for submerged fermentations, except that a lower ratio of ethanol to
water (42:58, vol/vol) was used. The lower ratio of ethanol to water
was based on gas chromatographic analyses of the ethanol concentration
during the 16-h extraction process for five independent solid-state
fermentations of Saccharopolyspora erythraea. The results
showed that 8 to 10% of the water present in the extraction solution
was lost during extraction of solid-state fermentations. This loss of
water was compensated for by using an extraction solution with a lower
ratio of ethanol to water (42:58, vol/vol). Analyses of ethanol were
performed by injecting 1 µl of an extract into a Hewlett-Packard
model 5890 gas chromatograph equipped with a flame ionization detector
and a DB wax column (15 m by 0.53 mm; J. W. Scientific, Folsom,
Calif.). The gas chromatography program parameters were as follows:
isocratic run time, 3.5 min; oven temperature, 80°C; and injector and
detector temperature, 230°C. For both submerged and solid-state
fermentations, vessels containing ethanol were agitated for 2 h on
an orbital shaker (stroke length, 2 in.) at 200 rpm, and then the
contents were allowed to settle for 16 h at 4°C. The aqueous
ethanol extracts were then filtered through a 100-µm-pore-size nylon
screen and transferred to 96-well plates for chemical analysis. Dry
cell weights of submerged fermentations were determined by using the water-washed particulate (10,000 × g, 15 min)
fractions from 6-ml samples of the fermentation broths. The particulate
materials were transferred to preweighed aluminum weighing dishes and
dried under a vacuum (
80 kPa) at 70°C for 16 h.
A-factor
(2-isocapryloyl-3R-hydroxymethyl--butyrolactone) was
isolated from Streptomyces griseus ATCC 10137 by
extraction of biomass with ethyl acetate, followed by reverse-phase
chromatography of the lyophilized extract as previously described
(3, 17). Purified A-factor was solublized in ethanol,
filtered through a 0.2-µm-pore-size filter, and added to sterile
fermentation vessels to a final concentration of 100 ng/ml. This
concentration was based on studies of A-factor production in S. griseus IFO13350, which showed that the concentration of A-factor
reached a maximum value (25 ng/ml) during log-phase growth
(3). The ethanol was removed before addition of medium A
by evaporation of the solvent under a vacuum (80 kPa, 25°C) for
1 h.
SPE of actinomycete extracts.
Polar materials were removed
from 0.7-ml samples of actinomycete extracts by solid-phase extraction
(SPE) on Empore octadecyl SD high-performance 96-well extraction disk
plates (3M Company, St. Paul, Minn.). The SPE stationary phase was
conditioned before use by sequential washing with 2 ml of distilled
H2O, with 2 ml of methanol, and finally with 2 ml of 1 mM
ammonium acetate (pH 5.5) in distilled H2O. The ethanol was
removed from a 0.7-ml sample of actinomycete extract under a vacuum
(80 kPa) at 20°C for 16 h. The residual aqueous material was
resuspended to a total volume of 0.5 ml with 1 mM ammonium acetate (pH
5.5) and loaded on the conditioned SPE stationary phase. The SPE
stationary phase was washed with 5 ml of 1 mM ammonium acetate (pH
5.5), and then the secondary metabolite-containing fraction was eluted
with 0.7 ml of a solution consisting of 70% (vol/vol) acetonitrile,
30% (vol/vol) methanol, and 6.5 mM ammonium acetate (pH 5.5). The
eluate, enriched in secondary metabolites, was transferred into
microwell plates and analyzed by ES-MS and by HPLC-ES-MS.
Quantification of secondary metabolites from solid-state
fermentations.
Although the initial solid-state fermentation
analyses were performed by extraction of mycelia grown directly on agar
(see above), we extracted mycelia grown on nylon membranes for
quantification of cell mass and secondary metabolites. The
actinomycetes were grown on 0.64-cm-diameter, matched-weight (24.0 ± 0.2 mg) filter disks that were placed on the agar surface of the
solid-state fermentation medium. Biomass and secondary metabolite
contents were determined directly by using the growth present on disks that were placed on 2 ml of fermentation medium containing 1.5% Noble
agar (Difco). Fermentations were completed in 24-well tissue culture
plates (Falcon no. 3047; Becton Dickinson Co., Lincoln Park, N.J.) by
transferring 10 µl of a vegetative culture to the center of a disk.
Initially, the following four types of disks were evaluated: Millipore
Immobilon N+ (pore size, 45 µm; catalog no. 202-00), Schleicher & Schuell Nytran (pore size, 0.2 µm; catalog no. 78179), Gelman
Sciences Biotrace polyvinylidene difluoride (PVDF) (pore size, 0.45 µm; catalog no. 66542), and Gelman Sciences Biotrace nitrocellulose
(pore size, 0.45 µm; catalog no. 66489). The Schleicher & Schuell
filters were ultimately selected for use (see below). Dry cell weight
and chemical analyses were completed in parallel, and three disks were
used for each analysis. For dry cell weight determinations, disks were
removed from the surface of the medium, placed in preweighed aluminum
weighing dishes, and dried under a vacuum (80 kPa) at 70°C for
16 h. The mean weight obtained from four uninoculated filter disks
was subtracted from individual dry weight values to determine the dry
weights of the samples. For chemical analyses, disks were removed from the agar surface and each disk was placed in a 2-ml microcentrifuge tube containing 1 ml of an aqueous 50% (vol/vol) ethanol solution. Each filter disk was vortexed vigorously for 1 min and then stored for
24 h at 4°C. Following storage, the extract was vortexed for 1 min and centrifuged at 4,000 × g for 5 min, and the
supernatant was analyzed by ES-MS or HPLC-ES-MS.
80 kPa) at 25°C for 1 h. The dried
disks were transferred to microcentrifuge tubes and extracted as
described above.
Chemical and statistical analyses. Chemical analyses of filtered ethanol extracts or of SPE eluates were performed by HPLC-ES-MS as described by Julian et al. (23) or by ES-MS as described by Higgs et al. (20). A standard mixture consisting of caffeine, m-cresol purple, Spinosad (a mixture of isomeric spinosyn factors A and D; Dow Agrosciences, Indianapolis, Ind.), narasin A, and tylosin (100 µg/ml each) was injected at the beginning, at the end, and after every 10th sample to assess instrument stability and performance during the analyses. Data from the ES-MS studies were evaluated with a UNIX workstation by using the productivity index algorithms described by Higgs and coworkers (20). Background corrections were completed for all treatments by subtracting the chemical productivity score for the uninoculated treatment (20). Data for ES-MS experiments are reported below in chemical productivity units (cp units) per milligram (dry weight of cells, and 1 cp unit was defined as 1/6,171th of the ES-MS signal response obtained with the standard mixture consisting of caffeine, m-cresol purple, Spinosad, narasin A, and tylosin, each at a concentration of 100 µg/ml.
Statistical evaluation of data and the experimental design was performed with JMP statistical discovery software (version 3; SAS Institute, Inc., Cary, N.C.).| |
RESULTS |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
Reproducibility of ES-MS measurements and fermentation output.
In the accompanying paper we describe a high-throughput ES-MS method to
quantify compounds present in natural product extracts (20). In this method, the mass-to-charge ratios and the
intensities of ions present in ES-MS spectra are used to measure the
chemical productivity of natural product extracts. The reproducibility of these ES-MS-based measurements is influenced by changes in instrument sensitivity or performance, the efficiency of sample preparation procedures, and growth parameters, such as temperature, incubation time, and plating methods. Instrument reproducibility was
assessed throughout the experiment by injecting the standard mixture
first, last, and after every 10th experimental sample. Chemical
productivity measurements for 16 control injections of the standard
mixture that were interspersed among 88 experimental samples (total
number of samples, 104) revealed a mean of 6,171 cp units, a standard
deviation of 185 cp units (Fig. 1), and a coefficient of variance of 5.8%. The coefficients of variance for all
bracketing control injections evaluated in this study (n = 56) were less than or equal to 7.1%. Lower
concentrations of the standard mixture yielded mean chemical
productivity values that were linearly related to the known
concentrations between 5 and 100 µg/ml. The standard deviations for
lower concentrations of the standard mixture were also observed to
decrease in a nearly linear fashion between concentrations of 5 and 100 µg/ml.
|
Measurement of growth condition-dependent changes in secondary
metabolites from 44 actinomycetes.
Growth condition-dependent
changes in the chemical compositions of extracts were evaluated for 44 actinomycetes cultivated under six different fermentation conditions by
using the paired t test. The mean differences and standard
errors for paired t test comparisons between the treatments
shown in Table 2 were used to quantify
treatment effects. A comparison of extracts from two independent
fermentations of the 44 actinomycetes, completed in control medium A,
revealed a low mean difference and standard error (Table 2). Similar
results (mean difference values, 
10%) were observed for fermentation
extracts from the 44 actinomycetes cultivated for different growth
periods (5, 7, and 12 days), at different incubation temperatures (25 and 30°C), and at different agitation speeds (110 and 165 rpm) by
using medium A (data not shown). The low mean difference values and low
standard errors for these comparisons indicated that growth conditions
did not differ significantly in the capacity to stimulate secondary
metabolism. The sum of the mean differences for paired t
tests between treatments was used to identify fermentation conditions
that most effectively stimulated secondary metabolism in actinomycetes
(Table 2). The results of this analysis demonstrated that solid-state
fermentation in medium A and submerged fermentation in medium B
provided extracts with the greatest chemical productivity, while all
other fermentation conditions provided lower levels of chemical
productivity.
|
Development of a filter growth method to facilitate measurement of yields of secondary metabolites from solid-state fermentations. The paired t test analysis showed that the solid-state growth conditions provided extracts with the greatest chemical productivity values compared to the values obtained with other fermentation conditions. However, our initial attempts to measure dry cell weight to calculate specific productivity values were hindered by an inability to obtain reproducible biomass samples due to the attachment of substrate mycelia to the agar surface. Therefore, a method for growing organisms on the surface of a filter disk was developed in order to prevent substrate mycelium attachment to the agar surface. Four different filter supports, including two nylon membranes with different pore sizes (0.2 and 0.45 µm), one PVDF membrane, and one nitrocellulose membrane, were tested to evaluate the effect of immobilized growth on the yield of erythromycin from S. erythraea. Similar yields of erythromycin (20.8 ± 4.2 µg of erythromycin/mg [dry weight] of cells) were obtained with nitrocellulose and nylon membranes, while no growth was observed on the PVDF membranes. The nylon membrane with the 0.2-µm pores (Nytran) was chosen for all subsequent solid-state yield determinations since it exhibited greater mechanical strength than the 0.45-µm-pore-size nylon or nitrocellulose membranes and was resistant to changes in shape caused by heat or pressure during sterilization.
The effect of the Nytran nylon membrane on expression of secondary metabolites was assessed by comparing the yields of erythromycin from S. erythraea grown directly on solid-state medium A and grown on nylon membranes supported by solid-state medium A. No apparent differences in the area of growth or the morphology of S. erythraea were observed between solid-state fermentations grown directly on the agar and solid-state fermentations grown on the agar-supported nylon membrane. Furthermore, no qualitative chemical differences in the HPLC-ES-MS profiles of ethanol extracts were observed between S. erythraea grown on the membrane surface and S. erythraea grown directly on the agar surface. Equivalent erythromycin yields based on surface area of growth were obtained when S. erythraea was grown directly on the agar and when it was grown on the membrane (324 ± 43 and 311 ± 34 µg of erythromycin · cm
2, respectively).
These results indicated that the membrane surface did not significantly
influence secondary metabolism in S. erythraea.
An experiment utilizing the ES-MS methods was performed with the group
of 44 actinomyetes to confirm that secondary metabolism was not
influenced by growth on the nylon membrane surface. The mean difference
and standard error (37 ± 31 cp units · mg [dry weight]
of cells1) for the paired t test analysis of
ES-MS data for actinomycetes grown on solid-state medium A or the nylon
membrane supported by solid-state medium A were found to be similar to
the mean difference and standard error for repeat fermentations of the
44 actinomycetes completed in medium A (39 ± 33 cp units
· mg [dry weight] of cells1) (Table 2). These results
provide additional evidence that secondary metabolism in actinomycetes
grown under solid-state conditions is not influenced by growth on a
membrane surface.
The recovery efficiency for the secondary metabolite extraction process
was measured by applying purified tylosin (internal standard) to
S. spadicus biomass present on nylon membranes. Analyses, performed in triplicate by adding 50 and 100 µg of solublized tylosin
to S. spadicus biomass present on nylon membranes, resulted in levels of recovery of tylosin of 93% ± 2.4% and 96% ± 1.9%, respectively. These results were similar to the recovery efficiencies obtained for submerged fermentations and provided justification for
comparison of solid-state and submerged growth conditions.
Characterization of treatment effects associated with solid-state
growth.
Actinomycetes cultivated under solid-state conditions
provided the greatest level of chemical productivity when these
conditions were compared to the other growth conditions evaluated in
this study. A paired t test of the results for extracts from
the 44 actinomycetes cultivated in medium A under submerged or
solid-state conditions showed that significant differences in
expression of metabolites occurred in response to growth conditions
(Fig. 2A). Differences in expression of
metabolites for the two growth conditions were also quantified by
analyzing the distribution of the differences between productivity
values (solid state minus submerged) for individual cultures. The 95%
confidence intervals for this distribution (182 x 467 cp units · mg [dry weight] of cells1) were used
to place treatment responses associated with the ES-MS measurements
into subsets. Treatment effects were assigned to three expression
categories (solid state enhanced, nonresponsive, and solid state
suppressed) based on the position of data points in relation to the
95% confidence intervals. The assignments to these categories are
shown in Table 1. Approximately 41% (18 of 44) of the actinomycetes
tested showed enhanced chemical productivity when they were grown under
solid-state conditions, while a slightly lower number (34%, 15 of 44 actinomycetes) showed reduced chemical productivity when they were
grown under solid-state conditions. The treatment effects due to
solid-state growth did not significantly influence chemical
productivity measurements for 25% (11 of 44) of the actinomycetes
evaluated.
|
Correlation of growth condition-dependent changes detected
by ES-MS with changes in the yields of known secondary
metabolites from actinomycetes.
The 44 actinomycetes evaluated in
this study were chosen, in part, due to their established production of
a number of known, chemically diverse secondary metabolites. Secondary
metabolites present in extracts were separated by reverse-phase
chromatography, and the properties of compounds resolved were compared
to the properties of authentic standards of known secondary metabolites based on UV-visible light absorption, chromatographic retention, and
positive- and negative-ion ES-MS spectra in order to establish the
identities of the compounds. The secondary metabolite concentrations for compounds listed in Table 3 were
determined by comparison of the areas under the curve in HPLC
evaporative light scattering chromatograms to the areas under the curve
for authentic standards. Concentrations were converted to yields by
using cell dry weight measurements for actinomycetes grown under
solid-state and submerged conditions (Table 3). Solid-state
fermentation was found to influence the yields of all of the secondary
metabolites listed in Table 3 except erythromycin. The yield of
erythromycin remained unchanged when the fermentation conditions were
switched from submerged to solid state. The higher concentration of
erythromycin in submerged fermentations was offset by the twofold
increase in biomass that occurred under submerged fermentation
conditions (Table 3; Fig. 3). Solid-state
growth was found to enhance the yield of secondary metabolites for
three of the six actinomycetes shown in Table 3. The greatest change in
the yield of secondary metabolites for cultures grown under solid-state
conditions occurred with hygromycin, whose yield increased more than
4,000-fold compared to the yield achieved under submerged growth
conditions. Similar improvements in the yields of nogalamycin, the
iron-binding siderophore, and rimocidin were observed under solid-state
conditions (Table 3). In addition to the improvements in yield for some
secondary metabolites, solid-state growth was observed to significantly reduce the yields of several secondary metabolites, including tylosin, calcimycin, guanidyl fungin methyl ether, and
N-demethyl streptomycin (Table 3; Fig. 3).
|
|
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
The inability to recognize and directly measure general chemical productivity in natural product extracts by high-throughput, chemically based approaches has been a barrier to progress in discovery of new secondary metabolites for pharmaceutical and commercial applications. Historically, the secondary metabolite discovery process has been guided by activity-based screening approaches that identify active extracts based on their ability to affect a specific biological assay (4, 40). Activity-based screening often results in enrichment of compounds that are already known and is subject to high false-positive rates due to the complex nature of natural product extracts (27). One approach that has been employed to improve the quality of biological screening efforts has been to prescreen natural product extracts by chemical methods in an attempt to ensure a high level of chemical diversity in screening libraries. This general approach, referred to as physicochemical screening (41), is based on coupled separation and detection of secondary metabolites present in natural product extracts by various methods, including thin-layer chromatography-UV-visible light absorbance (15), HPLC-UV-visible light absorbance, HPLC-ES-MS (23), and HPLC-nuclear magnetic resonance spectroscopy (1, 30). In addition to its application in prioritization of natural product extracts entering biological screening analyses, chemical screening has been used as the basis for selection of cultures and growth conditions that support expression of secondary metabolites. Monaghan and coworkers (28) used HPLC coupled with UV-visible light absorbance detection to show that rare fungal metabolites are more often associated with cultures that show high levels of chemical productivity. Culture productivity was assessed by the numbers and intensities of chromatographic peaks present in chromatograms. This study showed that chemical productivity criteria could be used to select new sources of natural products or as a means to improve secondary metabolite expression for existing natural product sources.
While chemical screening approaches have shown considerable promise as ways to expedite discoveries in natural product research, they have not gained widespread use due to limitations associated with low sample throughput and the challenges associated with acquiring meaningful information from many chromatograms. In this report we describe a rapid ES-MS method for measuring chemical productivity of actinomycete extracts and application of the method to evaluate growth conditions and other treatments that influence secondary metabolism in actinomycetes. This compound-centered profiling method differs from prior screening approaches used in natural product discovery in that it provides direct information on the composition of a natural product extract based on the number and intensity of ions present in the sample without the cost of chromatographic separation. This general measure of chemical productivity was shown to positively correlate with the yields of specific secondary metabolites or the total apparent yield of metabolites present in extracts from submerged and solid-state fermentations. The results of this study show that ES-MS is a useful surrogate for HPLC methods for identifying cultures and growth conditions that provide high levels of chemical productivity. Identification of productive extracts is accomplished without prior knowledge of target compounds that are present in the samples and is achieved with high throughput (~50 samples/h) compared to other chemical screening strategies that require separation by high-resolution chromatography methods (24, 28).
Statistical analysis of ES-MS productivity scores by the paired t test provided a means to segregate the actinomycete population based on the magnitude of growth condition-dependent responses. This approach allowed us to rank cultivation conditions for individual cultures based on growth condition-dependent differences in the chemical productivity scores. Through use of this method, it was demonstrated that solid-state fermentation in medium A and submerged fermentation in medium B provided the highest levels of chemical productivity compared to other cultivation conditions. Parallel analyses of the same extracts from solid-state and submerged fermentations by both HPLC-ES-MS and ES-MS confirmed the ES-MS results and showed that the chemical productivity score was sensitive to most changes in the expression of secondary metabolites that occurred in response to growth conditions.
Solid-state fermentation was shown to provide the highest level of chemical productivity and diversity when it was compared to other cultivation conditions evaluated in this study. Solid-state growth conditions have previously been shown to elicit altered physiological states in a few, well-characterized filamentous microorganisms (2, 5, 32, 35). For example, mycotoxin production by the filamentous fungus Aspergillus flavus is enhanced more than twofold on the basis of weight when growth conditions are changed from submerged to solid-state fermentation (18, 19). Similar improvements in the yields of antibiotics and other secondary metabolites have been observed for a small group of actinomycetes and spore-forming nonfilamentous bacteria grown on solid substrates (22, 32, 39). In addition to changes in the expression of small molecules, solid-state fermentation has been shown to influence the expression of extracellular enzymes and the cellular localization of enzymes in actinomycetes and fungi (2, 35).
While this study and other studies have documented the advantages of solid-state fermentation processes for production of secondary metabolites, widespread application of solid-state procedures in fermentation processes has been hindered by difficulties in measuring the key fermentation process variables, including biomass, pH, nutrient concentration, and temperature (5, 31). Direct measurement of filamentous microorganism biomass on solid substrates is often complicated by attachment of substrate mycelia to the surface. Because of the difficulties in directly measuring cell biomass, a number of previous studies have utilized indirect ways to measure biomass, including glucosamine, total sugar, and DNA contents, and the rate of carbon dioxide production to indirectly monitor cell growth (9, 10, 38). Unfortunately, these indirect biomarkers provide only estimates of biomass and often cannot be used in studies of uncharacterized microorganisms without employing assumptions concerning the physiology of the microorganisms. To eliminate these problems, we developed a method for growing actinomycetes on an agar-supported nylon membrane. The yield of secondary metabolites produced in a solid-state fermentation can be directly quantified from the weight of biomass present on the membrane and the concentration of secondary metabolites extracted from the biomass. Growth of actinomycetes on the membrane surface had the additional benefit of providing extracts that were virtually free of contamination from the complex growth medium. Thus, the procedure is a way to produce high-quality actinomycete extracts that can be introduced directly into a mass spectrometer without the expense of the SPE step that was necessary without the membrane method. While growth on a filter membrane reduced problems associated with attachment of biomass to the agar surface or with contaminating medium components, the method may underestimate the concentrations of certain highly polar secondary metabolites that may diffuse from the filter disk during growth.
Treatments that enhanced chemical productivity in actinomycetes often
created nonequivalent sample matrices. The most notable examples of
treatment-dependent matrix interference included interference from
residual medium components or other polar compounds associated with the
growth media. An ability to suppress matrix effects associated with
individual treatments was necessary to measure differences in
expression of secondary metabolites that occurred in response to the
various treatments. One approach to minimize matrix interference was to
select treatments that did not directly contribute to the detector
response of the mass spectrometer. The treatments in this group were
specifically physical parameters (i.e., agitation rate and incubation
temperature) and addition of compounds that did not ionize and
compounds whose masses were below the mass range scanned by the mass
spectrometer (-butyrolactones and Na2HPO4 [medium D], respectively). While this approach provided a higher level of assurance that treatment effects did not contribute to changes
in the chemical productivity measurements, it severely limited the
choice of potential treatments that could be evaluated. A second
strategy, one that has been commonly employed in previous natural
product research efforts, focused on selectively removing matrix
interference while simultaneously enriching for compounds of interest
(secondary metabolites) in the samples by solid-phase extraction. This
strategy is based on the assumption that the majority of valuable drug
leads come from moderately to highly nonpolar compounds that have
masses ranging from 250 to 700 g/mol (4). Methods for
enrichment of secondary metabolites from chemically complex matrices
include extraction with water-immiscible organic solvents, such as
ethyl acetate and chloroform, and solid-phase extraction on nonpolar
stationary phases (13). In this study, solid-phase
extraction on a C18 stationary phase was chosen as a means
to enrich secondary metabolites from the complex organic matrices. The
solid-phase extraction procedure ensured that ES-MS measurements
responded mainly to the changes in the expression of target compounds
having the druglike chemical and physical properties mentioned above.
Many laboratory investigations have shown that secondary metabolism in microorganisms is influenced by environmental parameters that affect cell physiology and biochemistry (13, 25, 33, 40). The results of this study confirmed the importance of selecting the appropriate growth conditions for expression of secondary metabolites from actinomycetes and quantified the relative effects of these conditions on secondary metabolism for a small set of actinomycetes by using a new chemical screening strategy. The results indicate that the ES-MS method is an efficient, high-throughput method for detecting and quantifying changes that occur in the expression of secondary metabolites from actinomycetes. This approach could be used to optimize conditions for expression of secondary metabolites from other microorganisms, to detect sources of variability in a single microorganism, or to optimize secondary metabolite extraction and sample preparation methods.
| |
ACKNOWLEDGMENTS |
|---|
We thank Jeff Gygi, Mike Goodwin, John Scheuring, Dale Duckworth, Matt Clemens, and Mark Strege (Eli Lilly and Company) for insightful discussions and for the analysis of actinomycete extracts by ES-MS and HPLC ES-MS/ELSD.
| |
FOOTNOTES |
|---|
* Corresponding author. Mailing address: Natural Products Research, Eli Lilly and Company, Lilly Corporate Center, DC 1533, Indianapolis, IN 46285. Phone: (317) 276-3584. Fax: (317) 276-5281. E-mail: hilton{at}lilly.com.
Present address: National Swine Research Center, USDA-ARS, Ames, IA 50011.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
Materials and Methods
Results
Discussion
References
|
|---|
| 1. | Abel, C. B. L., J. C. Lindon, D. Noble, B. A. M. Rudd, P. J. Sidebottom, and J. K. Nicholson. 1999. Characterization of metabolites in intact Streptomyces citricolor culture supernatants using high-resolution nuclear magnetic resonance and directly coupled high-pressure liquid chromatography-nuclear magnetic resonance spectroscopy. Anal. Biochem. 270:220-230[CrossRef][Medline]. |
| 2. | Acuna-Arguelles, M. E., M. Gutierrez-Rojas, G. Viniegra-Gonzalez, and E. Favela-Torres. 1995. Production and properties of three pectinolytic activities produced by Aspergillus niger in submerged and solid-state fermentation. Appl. Microbiol. Biotechnol. 43:808-814[Medline]. |
| 3. |
Ando, N.,
K. Ueda, and S. Horinouchi.
1997.
A Streptomyces griseus gene (sga A) suppresses the growth disturbance caused by high osmolarity and high concentration of A-factor during early growth.
Microbiology
143:2715-2723 |
| 4. | Ashton, M. J., M. C. Jaye, and J. S. Mason. 1996. New perspectives in lead generation. II. Evaluating molecular diversity. Drug Discov. Today 1:71-78[CrossRef]. |
| 5. | Barrios-Gonzalez, J., and A. Mejia. 1996. Production of secondary metabolites by solid-state fermentation. Biotechnol. Annu. Rev. 2:85-121[Medline]. |
| 6. |
Bennett, J. W., and R. Bentley.
1989.
What's in a name? microbial secondary metabolism.
Adv. Appl. Microbiol.
34:1-28.
|
| 7. | Beppu, T. 1992. Secondary metabolites as chemical signals for cellular differentiation. Gene 115:159-165[CrossRef][Medline]. |
| 8. | Demain, A. L. 1999. Pharmaceutically active secondary metabolites of microorganisms. Appl. Microbiol. Biotechnol. 52:455-463[CrossRef][Medline]. |
| 9. | Desgranges, C., C. Vergoignan, M. Georges, and A. Durand. 1991. Biomass estimation in solid-state fermentation: manual biochemical methods. Appl. Microbiol. Biotechnol. 35:200-205. |
| 10. | Desgranges, C., C. Vergoignan, M. Georges, and A. Durand. 1991. Biomass estimation in solid-state fermentation: on-line measurements. Appl. Microbiol. Biotechnol. 35:206-209. |
| 11. | Dreux, M., M. Lafosse, and L. Morin-Allory. 1996. The evaporative light scattering detector: a universal instrument for non-volatile solutes in LC and SFC. LC-GC Int. 9:148-153. |
| 12. | Dunny, G. M., and B. A. Leonard. 1997. Cell-cell communication in gram-positive bacteria. Annu. Rev. Microbiol. 51:527-564[CrossRef][Medline]. |
| 13. | Franco, C. M. M., and L. E. L. Coutinho. 1991. Detection of novel secondary metabolites. Crit. Rev. Biotechnol. 11:193-276[Medline]. |
| 14. | Gayo, L. M. 1998. Solution-phase library generation: methods and applications in drug discovery. Biotechnol. Bioeng. 61:95-106[CrossRef][Medline]. |
| 15. | Grabley, S., and R. Thiericke. 1999. Bioactive agents from natural sources: trends in discovery and application. Adv. Biochem. Eng. Biotechnol. 64:101-154[Medline]. |
| 16. | Guiochon, G., A. Moysan, and C. Holley. 1988. Influence of various parameters on the response factors of the evaporative light scattering detector for a number of non-volatile compounds. J. Liq. Chromatogr. 11:2547-2570[CrossRef]. |
| 17. | Hara, O., and T. Beppu. 1982. Mutants blocked in streptomycin production in Streptomyces griseus: the role of A-factor. J. Antibiot. 35:349-358[Medline]. |
| 18. | Hesseltine, C. W. 1977. Solid-state fermentation, part 1. Proc. Biochem. 12:24-27. |
| 19. | Hesseltine, C. W. 1977. Solid-state fermentation, part 2. Proc. Biochem. 12:29-32. |
| 20. |
Higgs, R. E.,
J. A. Zahn,
J. D. Gygi, and M. D. Hilton.
2001.
Rapid method to estimate the presence of secondary metabolites in microbial extracts.
Appl. Environ. Microbiol.
67:371-376 |
| 21. | Horinouchi, S., and T. Beppu. 1992. Autoregulatory factors and communication in actinomycetes. Annu. Rev. Microbiol. 46:377-398[CrossRef][Medline]. |
| 22. | Jermini, M. F. G., and A. L. Demain. 1989. Solid-state fermentation for cephalosporin production by Streptomyces clavuligerus and Cephalosporium acremonium. Experientia (Basel) 45:1061-1065. |
| 23. | Julian, R. K., R. E. Higgs, J. D. Gygi, and M. D. Hilton. 1998. A method for quantitatively differentiating crude natural extracts using high-performance liquid chromatography-electrospray mass spectrometry. Anal. Chem. 70:3249-3254[CrossRef]. |
| 24. | Maier, A., C. Maul, M. Zerlin, I. Sattler, S. Grabley, and R. Thiericke. 1999. Biomolecular-chemical screening: a novel screening approach for the discovery of biologically active secondary metabolites. J. Antibiot. 52:945-951[Medline]. |
| 25. | Marahiel, M. A., T. Stachelhaus, and H. D. Mootz. 1997. Modular peptide synthetases involved in nonribosomal peptide synthesis. Chem. Rev. 97:2651-2673[CrossRef][Medline]. |
| 26. | McCann, P. A., and B. M. Pogell. 1979. Pamamycin: a new antibiotic and stimulator of aerial mycelia formation. J. Antibiot. 32:673-678[Medline]. |
| 27. | Metsa-Ketela, M., V. Salo, L. Halo, A. Hautala, J. Hakala, P. Mantsala, and K. Ylihonko. 1999. An efficient approach for screening minimal PKS genes from Streptomyces. FEMS Microbiol. Lett. 180:1-6[Medline]. |
| 28. |
Monaghan, R. L.,
J. D. Polishook,
V. J. Pecore,
G. F. Bills,
M. Nallin-Omstead, and S. L. Streicher.
1995.
Discovery of novel secondary metabolites from fungi is it really a random walk through a random forest?
Can. J. Bot.
73(Suppl. 1):S925-S931.
|
| 29. | Montgomery, D. C. 1997. Design and analysis of experiments, 4th ed. John Wiley and Sons, New York, N.Y. |
| 30. | Moore, J. 1999. NMR screening in drug discovery. Curr. Opin. Biotechnol. 10:54-58[CrossRef][Medline]. |
| 31. | Murthy, M. V. R., N. G. Karanth, and K. S. M. S. Raghava Rao. 1993. Biochemical engineering aspects of solid-state fermentation. Adv. Appl. Microbiol. 38:99-147[CrossRef]. |
| 32. | Ohno, A., T. Ano, and M. Shoda. 1992. Production of antifungal antibiotic, iturin in a solid-state fermentation by Bacillus subtilis nb22 using wheat bran as a substrate. Biotechnol. Lett. 14:817-822[CrossRef]. |
| 33. | Rose, A. H. 1979. Production and industrial importance of secondary products of metabolism, p. 1-33. In A. H. Rose (ed.), Secondary products of metabolism. Academic Press, New York, N.Y. |
| 34. |
Schimmel, T. G.,
A. D. Coffman, and S. J. Parsons.
1998.
Effect of butyrolactone I on the producing fungus, Aspergillus terreus.
Appl. Environ. Microbiol.
64:3707-3712 |
| 35. | Selvakumar, P., L. Ashakumary, A. Helen, and A. Pandey. 1996. Purification and characterization of glucoamylase produced by Aspergillis niger in solid-state fermentation. Lett. Appl. Microbiol. 23:403-406[Medline]. |
| 36. |
Seow, K. T.,
G. Meurer,
M. Gerlitz,
E. Wendt-Pienkowski,
C. R. Hutchinson, and J. Davies.
1997.
A study of iterative type II polyketide synthases, using bacterial genes clones from soil DNA: a means to access and use genes from uncultured microorganisms.
J. Bacteriol.
179:7360-7368 |
| 37. |
Stein, J. L., and M. I. Simon.
1996.
Archaeal ubiquity.
Proc. Natl. Acad. Sci. USA
93:6228-6230 |
| 38. | Sugama, S., and N. Okazaki. 1979. Growth estimation of Aspergillus oryzae cultured on solid media. J. Ferment. Technol. 57:408-412. |
| 39. | Yang, S. S., and M. Y. Ling. 1989. Tetracycline production with sweet potato residue by solid-state fermentation. Biotechnol. Bioeng. 33:1021-1028[CrossRef]. |
| 40. | Yarbrough, G. G., D. P. Taylor, R. T. Rowlands, M. S. Crawford, and L. L. Lasure. 1993. Screening microbial metabolites for new drugs: theoretical and practical issues. J. Antibiot. 46:535-544[Medline]. |
| 41. | Zähner, H., H. Drautz, and W. Weber. 1982. Novel approaches to metabolite screening, p. 51-70. In J. D. Bu'lock, L. J. Nisbet, and D. J. Winstanley (ed.), Bioactive microbial products: search and discovery. Academic Press, London, United Kingdom. |
Applied and Environmental Microbiology, January 2001, p. 377-386, Vol. 67, No. 1
0099-2240/01/$04.00+0 DOI: 10.1128/AEM.67.1.377-386.2001
Copyright © 2001, American Society for Microbiology. All rights reserved.
This article has been cited by other articles:
-
Kulanthaivel, P., Kreuzman, A. J., Strege, M. A., Belvo, M. D., Smitka, T. A., Clemens, M., Swartling, J. R., Minton, K. L., Zheng, F., Angleton, E. L., Mullen, D., Jungheim, L. N., Klimkowski, V. J., Nicas, T. I., Thompson, R. C., Peng, S.-B.
(2004). Novel Lipoglycopeptides as Inhibitors of Bacterial Signal Peptidase I. J. Biol. Chem.
279: 36250-36258
[Abstract] [Full Text] -
Higgs, R. E., Zahn, J. A., Gygi, J. D., Hilton, M. D.
(2001). Rapid Method To Estimate the Presence of Secondary Metabolites in Microbial Extracts. Appl. Environ. Microbiol.
67: 371-376
[Abstract] [Full Text]
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»