Biology Reference
In-Depth Information
3
SAMPLING BACTERIAL CULTURES FOR METABOLOME
ANALYSIS
Sampling for bacterial metabolome analysis is technically very challenging since all
changes in the physiochemical milieu can instantaneously alter the metabolome.
Generally, there are two options to obtain a realistic snapshot of intracellular metab-
olite concentrations, either the cells are sampled within less than 1 s or the cell's
environment is maintained constant during sampling. Rapid sampling requires spe-
cial devices that are designed to simultaneously withdraw the sample and arrest
metabolism. Some examples are a sampling tube device (
Weuster-Botz, 1997
), an
automated sampling device for pulse experiments (
Schaefer
et al.
, 1999
), the
stopped-flow sampling device (
Buziol
et al.
, 2002
), a mini-plug flow reactor called
BioScope (
Visser
et al.
, 2002
), a single tube heat exchanger (
Schaub
et al.
, 2006
) and
a low pressure
in situ
sampling device (
Hiller
et al.
, 2007
). Although such devices
minimize the time between sampling and the arrest of metabolism, there is the prob-
lem of cell disintegration when using a cold quenching solution to arrest metabolism
(usually aqueous methanol at
50
C). This problem of cell damage and concomitant
metabolite leakage was first observed for amino acids in
Corynebacterium glutami-
cum
and blamed on a general cold shock during quenching (
Wittmann
et al.
, 2004
).
Later, the authors showed that various other species of bacteria leak metabolites
during cold methanol quenching (
Bolten
et al.
, 2007
). Quenching solutions with
protecting agents such as glycerol were used to reduce leakage; however, the loss
of metabolites was still significant (
Villas-Bˆas and Bruheim, 2007; Link
et al.
,
2008
). Since the problem of metabolite leakage remains unsolved, we do not recom-
mend cold methanol quenching. An alternative method is simultaneous quenching
and extraction of the whole culture medium, referred to as whole cell broth extraction
(
Schaub
et al.
, 2006; Bolten
et al.
, 2007; Taymaz-Nikerel
et al.
, 2009
). A major
drawback with whole cell broth extraction is the high level of contamination of
the samples with salts from the medium, thus reducing the sensitivity and selectivity
of LC-MS analysis. Furthermore, this method requires correction for metabolites in
the extracellular milieu. Accurate metabolite quantification using whole cell broth
extraction is possible at high biomass concentrations because of the more favourable
ratio of intracellular to extracellular volume compared with low biomass concentra-
tion samples.
Since no generally accepted and applicable sampling method has emerged yet,
we describe here a fast filtration protocol that produces minimal errors if applied
in combination with the cultivation conditions described in
Section 2
. Sampling
by quick filtration and subsequent washing has been discussed intensively.
Bolten
et al.
(2007)
consider fast filtration a suitable sampling technique if cells are washed
with appropriate washing solutions. However, they restrict applicability of the
method to metabolites with a relatively low turnover.
Rabinowitz (2007)
concludes
that the method is excessively likely to cause artefacts due to alterations to the meta-
bolome during filtering and washing. The success of this method critically depends
