9. Microbial Carbon Cycling in Pelagic Ecosystems: Microbial Methods for Ecosystem Scientists
145
then measures the uptake and/or respiration of the
labeled compound along with a chemical or kinetic
measurement of the total pool size of that molecule.
The fraction utilized per unit time (obtained from
the tracer addition) is multiplied by the pool size
for the total rate of utilization of that substrate.
This approach has revealed that relatively large
fractions of total BP can often be supported by a
few classes of molecules. For example, by measuring both BP and the kinetics of the uptake of dissolved free amino acids (DFAA) Kirchman et al.
(1993) found that about 20% of total BP was supported by DFAA during the spring bloom in the
Eastern North Atlantic (Kirchman et al. 1993). Using a similar approach, Rich et al. (1996) found that
although dissolved neutral monosaccharides
(DNMS) were in low concentration in the equatorial pacific, the uptake of glucose alone supported
15 to 47% of BP. In interpreting this work one has
to keep in mind that what is measured is the proximate substrate and not its ultimate source. It is still
interesting to learn where the DFAA or DNMS
comes from. Further, it is intriguing to consider the
type of molecule supporting the remaining, large
fraction of BP not explained by the known simple
substrates.
In both marine and freshwaters, much of the
DOC is in large, high-molecular weight polymers.
Further, particulate organic carbon (POC) can be
the ultimate source of the small molecules in turbid
systems and in sediments. Bacteria generally cleave
very large molecules into simple compounds using
extracellular enzymes. An approach to examining
the array of complex substrates that support bacterial production has been to examine the suites of
these exoenzymes present in different environments. The idea is to choose broad classes of enzymatic processes of interest and assay for these
using model substrates that release either a colored
or fluorescent endproduct or intermediate (see Sinsabaugh et al. 1997). This approach has also been
used with some success in marine planktonic aggregates (marine snow, Karner and Herndl 1992),
freshwater lakes (Chrost et al. 1989), and estuarine
systems (Hoppe 1993). Recently, Karner and Rassoulzadegan (1995) examined (X- and B-glucosidase
and aminopeptidase activity in the Mediterranean
Sea and found striking variations in the amounts
and proportions of these enzymes at time scales
from the diel to the seasonal. This study suggests
the need to examine enzymatic responses on several
time scales before strong conclusions about patterns can be drawn.
From an ecosystem perspective, one of the most
intriguing applications of the exoenzyme approach
has been done in river and streams using techniques that had been developed largely for working in sediments (Sinsabaugh et al. 1997; Findlay
et al. 1998). Sinsabaugh et al. (1997) recently began looking at enzyme activity in the water column. In this approach, unfiltered water is challenged with a series of model substrates at a range
of concentrations. The model substrates contain 4methylumbelliferyl (MUF) attached to a substrate
of interest. To measure fatty acid esterase, for example, the model substrate is MUF-acetate; to
measure B-N-acetylglucosaminidase the model
substrate is MUF-N-acetyl-B-glucosaminide. The
enzyme cleaves the MUF, which becomes fluorescent. The fluorescence is proportional to MUF concentration. Relative new technology has made this
type of assay rapid and many samples can be processed. The reactions are performed in multi-well
microliter plates and fluorescence is read in a microfluor platereader. This degree of automation allows one to investigate enzyme kinetics over a
range of concentrations and obtain both Km and
V max values for each assay. This approach has recently been applied a range of coastal ecosystems
in which the activities of eight different enzymes
were assayed in five ecosystems (Hopkinson et al.
in press). The multivariate data were then subjected
to principal components analysis with interesting
results (Fig. 9.4). Two of the eight principal components accounted for 62% of the total variability.
Plotted on the space defined by these two, all of the
ecosystems separated, indicating that substrate use
was quite different among them. Samples near the
upper right (e.g., Columbia River) would have
higher standing stock of substrates that can be degraded by peptidases and esterases. Samples near
the origin (e.g., Satilla River) are enriched in xylosidases and glucosaminidase. The implication is
that the bacteria in the Satilla River are more dependent on complex plant polysaccharides and
amino sugars while those in the Columbia are more
dependent on proteins and fatty acids. Coupled with
other descriptions about bacterial production, activity, and substrate composition, these enzyme tech-
145
then measures the uptake and/or respiration of the
labeled compound along with a chemical or kinetic
measurement of the total pool size of that molecule.
The fraction utilized per unit time (obtained from
the tracer addition) is multiplied by the pool size
for the total rate of utilization of that substrate.
This approach has revealed that relatively large
fractions of total BP can often be supported by a
few classes of molecules. For example, by measuring both BP and the kinetics of the uptake of dissolved free amino acids (DFAA) Kirchman et al.
(1993) found that about 20% of total BP was supported by DFAA during the spring bloom in the
Eastern North Atlantic (Kirchman et al. 1993). Using a similar approach, Rich et al. (1996) found that
although dissolved neutral monosaccharides
(DNMS) were in low concentration in the equatorial pacific, the uptake of glucose alone supported
15 to 47% of BP. In interpreting this work one has
to keep in mind that what is measured is the proximate substrate and not its ultimate source. It is still
interesting to learn where the DFAA or DNMS
comes from. Further, it is intriguing to consider the
type of molecule supporting the remaining, large
fraction of BP not explained by the known simple
substrates.
In both marine and freshwaters, much of the
DOC is in large, high-molecular weight polymers.
Further, particulate organic carbon (POC) can be
the ultimate source of the small molecules in turbid
systems and in sediments. Bacteria generally cleave
very large molecules into simple compounds using
extracellular enzymes. An approach to examining
the array of complex substrates that support bacterial production has been to examine the suites of
these exoenzymes present in different environments. The idea is to choose broad classes of enzymatic processes of interest and assay for these
using model substrates that release either a colored
or fluorescent endproduct or intermediate (see Sinsabaugh et al. 1997). This approach has also been
used with some success in marine planktonic aggregates (marine snow, Karner and Herndl 1992),
freshwater lakes (Chrost et al. 1989), and estuarine
systems (Hoppe 1993). Recently, Karner and Rassoulzadegan (1995) examined (X- and B-glucosidase
and aminopeptidase activity in the Mediterranean
Sea and found striking variations in the amounts
and proportions of these enzymes at time scales
from the diel to the seasonal. This study suggests
the need to examine enzymatic responses on several
time scales before strong conclusions about patterns can be drawn.
From an ecosystem perspective, one of the most
intriguing applications of the exoenzyme approach
has been done in river and streams using techniques that had been developed largely for working in sediments (Sinsabaugh et al. 1997; Findlay
et al. 1998). Sinsabaugh et al. (1997) recently began looking at enzyme activity in the water column. In this approach, unfiltered water is challenged with a series of model substrates at a range
of concentrations. The model substrates contain 4methylumbelliferyl (MUF) attached to a substrate
of interest. To measure fatty acid esterase, for example, the model substrate is MUF-acetate; to
measure B-N-acetylglucosaminidase the model
substrate is MUF-N-acetyl-B-glucosaminide. The
enzyme cleaves the MUF, which becomes fluorescent. The fluorescence is proportional to MUF concentration. Relative new technology has made this
type of assay rapid and many samples can be processed. The reactions are performed in multi-well
microliter plates and fluorescence is read in a microfluor platereader. This degree of automation allows one to investigate enzyme kinetics over a
range of concentrations and obtain both Km and
V max values for each assay. This approach has recently been applied a range of coastal ecosystems
in which the activities of eight different enzymes
were assayed in five ecosystems (Hopkinson et al.
in press). The multivariate data were then subjected
to principal components analysis with interesting
results (Fig. 9.4). Two of the eight principal components accounted for 62% of the total variability.
Plotted on the space defined by these two, all of the
ecosystems separated, indicating that substrate use
was quite different among them. Samples near the
upper right (e.g., Columbia River) would have
higher standing stock of substrates that can be degraded by peptidases and esterases. Samples near
the origin (e.g., Satilla River) are enriched in xylosidases and glucosaminidase. The implication is
that the bacteria in the Satilla River are more dependent on complex plant polysaccharides and
amino sugars while those in the Columbia are more
dependent on proteins and fatty acids. Coupled with
other descriptions about bacterial production, activity, and substrate composition, these enzyme tech-