Isolation of Typical Marine Bacteria by Dilution Culture: Growth, Maintenance, and Characteristics of Isolates under Laboratory Conditions - FRITS SCHUT

1. APPLIED AND ENVIRONMENTAL MICROBIOLOGY, July 1993, P. 2150-2160 Vol. 59, No. 7
0099-2240/93/072150-11$02.00/0
Copyright ©) 1993, American Society for Microbiology
Isolation of Typical Marine Bacteria by Dilution Culture:
Growth, Maintenance, and Characteristics of Isolates
under Laboratory Conditions
FRITS SCHUT,l* EGBERT J. DE VRIES,' JAN C. GOTITSCHAL,1 BETSY R. ROBERTSON,2
WIM HARDER,' RUDOLF A. PRINS,' AND DON K. BUTTON2
Department of Microbiology, Biological Centre, University of Groningen, P.O. Box 14,
9751 NNHaren, The Netherlands, 1 and Institute of Marine Science,
University ofAlaska, Fairbanks, Alaska 99775 10802
Received 1 February 1993/Accepted 30 April 1993
Marine bacteria in Resurrection Bay near Seward, Alaska, and in the central North Sea off the Dutch coast
were cultured in filtered autoclaved seawater following dilution to extinction. The populations present before
dilution varied from 0.11 x 109 to 1.07 x 109 cells per liter. The mean cell volume varied between 0.042 and
0.074 ,um3, and the mean apparent DNA content of the cells ranged from 2.5 to 4.7 fg ofDNA per cell. All three
parameters were determined by high-resolution flow cytometry. All 37 strains that were obtained from very
high dilutions of Resurrection Bay and North Sea samples represented facultatively oligotrophic bacteria.
However, 15 of these isolates were eventually obtained from dilution cultures that could initially be cultured
only on very low-nutrient media and that could initially not form visible colonies on any of the agar media
tested, indicating that these cultures contained obligately oligotrophic bacteria. It was concluded that the cells
in these 15 dilution cultures had adapted to growth under laboratory conditions after several months of nutrient
deprivation prior to isolation. From the North Sea experiment, it was concluded that the contribution of
facultative oligotrophs and eutrophs to the total population was less than 1% and that while more than half of
the population behaved as obligately oligotrophic bacteria upon first cultivation in the dilution culture media,
around 50%o could not be cultured at all. During one of the Resurrection Bay experiments, 53% of the dilution
cultures obtained from samples diluted more than 2.5 x 105 times consisted of such obligate oligotrophs. These
cultures invariably harbored a small rod-shaped bacterium with a mean cell volume of 0.05 to 0.06 P,m3 and
an apparent DNA content of 1 to 1.5 fg per cell. This cell type had the dimensions of ultramicrobacteria.
Isolates of these ultramicrobacterial cultures that were eventually obtained on relatively high-nutrient agar
plates were, with respect to cell volume and apparent DNA content, identical to the cells in the initially
obligately oligotrophic bacterial dilution culture. Determination of kinetic parameters from one of these small
rod-shaped strains revealed a high specific affinity for the uptake of mixed amino acids (a'A, 1,860 liters/g of
cells per h), but not for glucose or alanine as the sole source of carbon and energy (a'A, 200 liters/g of cells
per h). The ultramicrobial strains obtained are potentially a very important part of picoplankton biomass in
the areas investigated.
Nondifferentiating heterotrophic marine bacteria are often
thought to survive the oligotrophic conditions that prevail in
oceans in a state of dormancy (78) or starvation (65, 74), i.e.,
in a state of nongrowth. A possible rationale for this hypoth-esis
is that although generally less than 1% of the cells
present in natural seawater can be cultured and isolated on
standard laboratory media (13, 26, 48, 54), most of the
bacterial population is metabolically active. The determina-tion
of in situ rates of oxygen consumption (19, 21), amino
acid uptake (10, 20, 28, 49, 58), glucose uptake (6, 24),
protein synthesis (53, 76), DNA synthesis (23, 27, 67), RNA
synthesis (50), and cell division (22, 36) and the determina-tion
of electron transport system activity by using tetra-zolium
salts (10, 72, 82) or fluorescein diacetate (60) and of
the cell's energy charge (51), as well as the development of
activity counts by microautoradiography (23, 43, 64, 79) or
direct viable counts by using the DNA gyrase inhibitor
nalidixic acid (54), have led to the conclusion that more than
60% of the cells exhibit some sort of metabolic activity.
However, the organisms responsible for these activities have
* Corresponding author.
yet to be identified and characterized, and as a result there is
little insight into the ecophysiological roles of individual
subpopulations. In this respect, community structure analy-sis
of bacterioplankton populations using rRNA sequences
as a tool to identify different genotypes appears to be very
promising. It is now evident that numerous cells present in
oceanic waters have 16S rRNA sequences not related to any
known sequences from cultivated organisms (11, 30, 31, 75).
The main problems in performing investigations into the
autecology and physiological properties of marine bacteria
are currently the isolation and cultivation of dominant ma-rine
bacterial species. Whether the isolates obtained by
cultural methods represent important indigenous bacterial
species will remain unknown unless the culturability of the
cells present in natural samples can be increased consider-ably
or until it is proven that genetic and physiological
characteristics for cultured isolates and indigenous cells are
similar.
As determined for natural assemblages of bacterioplank-ton,
the typical characteristics of marine bacteria include
small cell size (9, 57, 76), even down to ultramicrobacterial
sizes (cells with diameters less than 0.3 ,um or cell volumes
less than 0.1 ,um3) (5, 61, 80), a low apparent DNA content
2150

2. ISOLATION OF OLIGOTROPHIC ULTRAMICROBAC7TERIA 2151
compared with that of laboratory-cultured cells (71), and a
high uptake affinity for substrates (6, 15). Furthermore, there
are strong indications that marine bacteria require low
substrate concentrations for growth, i.e., that they are
obligately oligotrophic. The definitions used throughout this
study are essentially the generally accepted ones (42), with
oligotrophs defined as bacteria that have the ability to grow
at low substrate concentrations either obligately or faculta-tively
and eutrophs defined as organisms restricted to growth
at high substrate concentrations. Recently, the subject of
oligotrophy and ultramicrobacteria in relation to their sub-strate-
sequestering abilities was discussed (15, 34). Gener-ally,
platable cell counts (viable counts) from pelagic sam-ples
increase significantly when a low-nutrient medium is
used (2, 12, 17, 24, 45, 81). For this reason, the noncultura-bility
of marine bacteria has been attributed to their obli-gately
oligotrophic nature (24, 45), which would then imply
that these bacteria are not dormant (78) but capable of
growth at ambient substrate concentrations (24, 46).
So far, very few data have been collected on truly obli-gately
oligotrophic marine bacteria. In fact, it is not even
clear whether a demarcation line between oligotrophic and
eutrophic bacteria exists at all. Eutrophic organisms can be
retrained to become obligate oligotrophs (42), and obligate
oligotrophs can be converted into facultative oligotrophs (61,
63, 81).
Neither physiological traits nor morphological character-istics
can be readily used to identify dominant marine
bacteria. Normal isolates on plates have been shown to form
viable but nonculturable cells (69, 73, 74 [and references
therein]) as well as ultramicrocells (66, 77 [and references
therein]) upon starvation or exposure to other stress condi-tions.
In a study by Moyer and Morita (66), both the DNA
content and cell size of a marine vibrio were reduced
considerably upon prolonged nutrient deprivation, resulting
in the morphology characteristic of indigenous pelagic bac-teria.
However, since physiological properties of an isolate
under laboratory conditions will not necessarily reflect the
organism's natural state, information on the quantitative
importance of the organism in its natural habitat and isola-tion
in pure culture are needed before laboratory data can be
taken back to the field.
The aim of this study was to isolate dominant marine
bacteria by cultural methods. We used unamended filtered
(0.2-,um-pore-size filter) and subsequently autoclaved sea-water
as dilution medium to study heterotrophic marine
bacteria by applying dilutions to the extent of extinction. At
the same time, we established a protocol for determining the
viability of cells in the seawater sample. The ability to
generate cultures from very high (106-fold) seawater dilu-tions
reflected the high level of viability of the cells present
in the inoculum. Viability values between 30 and 40% were
frequently observed. The theory, procedures, and initial
results of this promising technique have been described
elsewhere (16). In order to investigate the trophic character
of the cultures and isolates, we have determined the range of
carbon source concentrations that can support the growth of
isolates by replica plating and by cultivation in synthetic
seawater medium. Results indicate the presence of obli-gately
and facultatively oligotrophic bacteria at the sampling
time and upon first cultivation. However, an unspecified
adaptation leading to the ability to form colonies on solid
medium containing higher nutrient concentrations was ob-served
for all obligately oligotrophic cultures tested. Iso-lates,
once obtained, invariably represented facultatively
oligotrophic bacteria, i.e., bacteria able to grow at relatively
high and very low substrate concentrations.
Furthermore, we determined whether a representative
dilution culture and an isolate obtained from such a culture
exhibited specific affinity values (15) comparable to those
generally observed for natural bacterioplankton (14, 34). It is
concluded that dilution cultures and isolates exhibit uptake
affinities for mixed amino acids comparable to those of
natural bacterioplankton.
MATERIALS AND METHODS
Sampling sites. Seawater samples were collected off the
Seward Marine Center pier at Resurrection Bay, Seward,
Alaska, on 10 December 1989, 24 March 1990, and 28 August
1990 (60°03'N, 149°25'W). The sampling depth was 10 m.
Experiments with the samples collected on these three dates
are referred to as Resurrection Bay I, Resurrection Bay II,
and Resurrection Bay III, respectively. North Sea water
samples were obtained at the oyster grounds approximately
250 km north of the Dutch Island of Texel on 21 February
1991, during a cruise of the MS Aurelia of The Netherlands
Institute for Sea Research (54°75'N, 4°75'E). The sampling
depth was 20 m.
Preparation of dilution series and monitoring of growth.
Dilution series were prepared and growth of the resulting
dilution cultures was monitored as described previously (16).
Unless stated otherwise, 20 tubes were inoculated for every
dilution. For any tube, the observation frequency was lim-ited
to a maximum of once every 3 to 4 days. This observa-tion
frequency, together with the relatively high detection
level for populations in the dilution cultures (5 x 104cells per
ml), generally resulted in detection of cultures in the late log
phase of growth at the earliest.
Subculturing and isolation. Subcultures of the dilution
cultures were made in filtered-autoclaved seawater (FAS)
and in synthetic seawater medium (MPM), containing the
following (in grams per liter of Milli Q-purified water [Milli-pore
Corp., Bedford, Mass.]): NaCl, 30.0; MgCl2- 6H20),
1.0; Na2SO4, 4.0; KCl, 0.70; CaCl2. 2H2O, 0.15; NH4Cl,
0.50; NaHCO3, 0.20; KBr, 0.10; SrCl2. 6H20, 0.04; H3B03,
0.025; KF (Sigma Chemical Company, St. Louis, Mo.),
0.001; morpholinepropanesulfonic acid (MOPS) buffer (Sig-ma)
(pH 7.8), 2.09; KH2PO4, 0.27; trace element solution,
1.0 ml/liter; vitamin solution, 2.0 ml/liter; and a carbon
source as mentioned in the text.
The phosphates, trace elements, vitamins, and carbon
sources were added separately to the autoclaved basal salts.
The trace element solution was developed from the data by
Goldberg (32) and consisted of the following (in milligrams
per liter of distilled water): Na2EDTA, 1,000; FeCl3 6H20,
2,000; LiCl, 1,000; AlC13, 50; NaVO3. 14H20, 5; K2Cr207,
0.15; MnCl2 4H20, 80; CoC12. 6H20, 5; NiCl2 6H20, 20;
CuCl2, 20; ZnC12, 60; Na2SeO3 - 5H20, 15; RbCl, 150;
Na2MoO4. 2H20, 75; SnC12 2H20, 1.5; KI, 80;
BaC12. 2H20, 50; and Na2WO4 2H20, 15. The vitamin
solution was as previously described (39) and consisted of
the following (in milligrams per liter of distilled water):
para-aminobenzoic acid, 100; folic acid, 50; thioctic acid, 50;
riboflavin, 100; thiamine, 200; nicotinic acid amide, 200;
pyridoxamine, 500; panthothenic acid, 100; cobalamin, 100;
and biotin, 20.
Dilution cultures were spread plated onto full-strength
marine nutrient agar consisting of the following (in grams per
liter of distilled water): nutrient broth (BBL Microbiology
Systems, Cockeysville, Md.), 8; NaCl, 27; and agar (BBL
VOL. 59, 1993

3. 2152 SCHUT ET AL.
Microbiology Systems), 15. When growth occurred, isolates
were obtained from these plates. Dilution cultures that could
not be cultured on marine nutrient agar or broth were
subcultured and maintained in FAS or MPM containing 0.5
to 2.0 mg of Casamino Acids (Difco Laboratories, Detroit,
Mich.) per liter. Spread plating of dilution cultures that could
not be plated onto marine nutrient agar was also done on
ZoBell 2216E agar (MPM with 5 g of Bacto Peptone [Difco]
per liter and 1 g of yeast extract [BBL] per liter) and on MPM
agar without vitamins but with various concentrations of a
complex substrate mixture. This mixture was prepared as a
concentrated stock solution containing the following (per
liter): D-glucose, 6.9 g; D-fructose, 6.9 g; D-galactose, 6.9 g;
Na-acetate, 9.5 g; L-lactic acid (90%), 6.4 ml (Fluka Chemie,
Buchs, Switzerland); Na-glycolate, 11.3 g; Na-succinate, 8.1
g (BDH Chemicals Ltd., Poole, England); mannitol, 7.0 g;
ethanol (96%), 7.0 ml; glycerol, 6.3 ml; Na-benzoate, 4.8 g;
salicylic acid, 4.6 g (BDH Chemicals Ltd.); Casamino Acids,
76.5 g; and peptone, 29.7 g. The pH was set at 7.5 with
NaOH. The carbon content of this stock was approximately
100 g of carbon per liter. MPM agar plates were prepared
with various concentrations of the mixed-substrate stock
solution to yield organic carbon concentrations from 1 to
10,000 mg of C per liter, called MPM1 to MPM10,000. When
these low-nutrient agar plates were used, the agar was
washed subsequently with ethanol (96%), acetone, and eth-anol
(96%) and then washed three times with Milli Q-purified
water, when substantial amounts of sugars (1 mM reducing
sugars in first H20 washing) and amino acids could be
removed from the agar.
Unless stated otherwise, all chemicals used were obtained
from Merck (Darmstadt, Germany) and were analytical
grade. All vitamins were obtained from Sigma Chemical
Company.
Growth temperature of dilution cultures was 10°C in the
dark. Stationary-phase cultures were stored at 5°C in the
dark. Spread plate incubations occurred in 10°C dark incu-bator
rooms and at room temperature (20°C) in the light. The
isolates were then grown at 30°C in the dark.
The population density and morphology of the cultures
were determined by flow cytometry and epifluorescence
microscopy as described below.
Epifluorescence microscopy. For epifluorescence micros-copy,
a modification of Porter and Feig's (70) nuclear
staining method was used. Subsamples (1 ml each) were
withdrawn aseptically from a culture, fixed in 0.2% (wt/vol)
gluteraldehyde, and made permeable by treatment with 0.1%
(vol/vol) Triton X-100 to improve stain penetration. Cells
were then stained for 15 min with 0.5 ,ug of 4',6-diamidino-
2-phenylindole (DAPI) (Sigma Chemical Co.) per ml, col-lected
onto a 0.1-,um-pore-size black, polycarbonate mem-brane
filter (Poretics Corp., Livermore, Calif.), and rinsed
with 4 ml of a prefiltered (0.2-,um-pore-size Nuclepore) 3%
sodium chloride solution. The filter was fitted between a
microscope slide and a cover glass with immersion oil
(Leitz). The cells were observed by first focusing on the filter
surface and slowly raising the focal point until the cells were
in focus. Cells were counted at a magnification of x 1,000 by
using a Leitz dialux 20 EB microscope equipped with a
100-W Hg arc lamp and a 365-nm-wavelength optical band-pass
filter. At least 10 fields were counted, or observation
was continued until 300 cells were counted.
Flow cytometry. Flow cytometry was performed on a
Ortho Cytofluorograf IIS as described previously (71), ex-cept
that samples were preserved with 0.5% formaldehyde,
stored cold, made permeable by treatment with 0.1% Triton
0
so
su
q
0
w
c3)
M"
0s
35
30
25
20
15
10
5
APPL. ENVIRON. MICROBIOL.
Energ.
Not energ.
+ cCCCP
0
0 20 40 60 80 100
Time (sec)
FIG. 1. Time course of D-['4C]glucose uptake by strain RB2256
with (A) or without (-) 2-min preincubation with L-alanine (10 F.M)
in order to energize the cells or with the protonophore CCCP (5 ,uM)
(*). The glucose concentration was 20 nM. prot., protein.
X-100, and stained with DAPI (0.5 ,ug/ml) at 10°C for 1 h
before analysis. Data were analyzed as described previously
(16) and are plotted in bivariate histograms of apparent DNA
per cell versus cell volume.
Uptake experiments and determination of specific affinity.
Dilution cultures were subcultured in MPM on 2 mg of
Casamino Acids per liter. It was found that the low cell
densities obtained by using these low substrate concentra-tions
required rigorous pelleting procedures in order to
obtain sufficient cell material. Therefore, cells were spun
down at 250,000 x g for 20 min, washed twice in MOPS-buffered
MPM basal salts (pH 7.8) (MPMBS), resuspended
to a cell population of approximately 3.5 x 105 cells per ml
(22 ,ug of cells per liter), and incubated with nanomolar
concentrations of 3H-labelled mixed amino acids (NET-250
L-amino acid mixture, 28 Ci/mmol; NEN) at 10°C. Within 10
min, 1-ml subsamples were withdrawn from the incubation
culture and filtered through a 0.1-,um-pore-size Nuclepore
filter. The filter was rinsed with 2 ml of MPM, dried, and
counted in a liquid scintillation counter, using a toluene-based
scintillation cocktail.
Isolated strains were grown at 30°C in MPM with 2 mM
D-glucose, 4 mM L-alanine, or 250 mg of Casamino Acids per
liter. Since cells tended to form little clumps at these high
substrate concentrations and even formed strands on MPM
plus glucose, sedimentation efficiencies were usually near
95% at 10,000 x g. However, by using electron microscopy,
no significant differences were observed between the sizes of
individual cells within such strands or clumps and those of
cells grown at low substrate concentrations (unpublished
results). Cells were then harvested by centrifugation at
10,000 x g for 20 min at 4°C, washed in MPMBS (pH 7.8),
and resuspended to a density of 3 mg of protein per ml. Cells
were stored on ice until measurement. For glucose uptake,
10 pL1 of this cell suspension was resuspended in 100 p.l of
MPMBS (pH 7.8) and preincubated for 2 min at 30°C in the
presence of 10 p.M L-alanine in order to energize the cells.
This procedure was necessary to facilitate immediate uptake
of the substrate tested (Fig. 1). Glucose was used to energize
the cells during alanine uptake. The uptake reaction was

4. ISOLATION OF OLIGOTROPHIC ULTRAMICROBACTERIA 2153
TABLE 1. Characteristics of bacterioplankton collected from two sites determined by flow cytometryz
Location Depth Date Population density MCV6 DNAC Carbon biomassd
(m) (cells/ml, 106) (pm3) (fg/cell) (pg of C/liter)
Resurrection Bay 0 3 Apr. 89 0.71 0.042 4.7 10.4
Resurrection Bay I 10 10 Dec. 89 0.83 0.048 2.5 13.9
Resurrection Bay II 10 24 Mar. 90 0.20 0.052 2.7 3.6
Resurrection Bay III 10 28 Aug. 90 1.07 0.074 3.7 27.7
North Sea 10 16 Feb. 91 0.11 0.061 3.3 2.4
a The two sites were Resurrection Bay near Seward, Alaska, and near the oyster grounds, North Sea, The Netherlands.
b MCV, mean cell volume.
Apparent DNA content (population mean).
d The conversion factor 0.35 x 10-12 g of carbon per p.m3 (8) was used.
started with various concentrations of D-[U-'4C]glucose or
L-[U-'4C]alanine (Amersham Corp., Amersham, United
Kingdom). The reaction was stopped by quickly adding 3 ml
of ice-cold MPMBS (pH 7.8). Cells were filtered through a
0.2-,um-pore-size cellulose nitrate filter (BA 83; Schleicher
and Schuell GmbH, Dassel, Germany) and rinsed with 3 ml
of ice-cold MPMBS. Filters were immediately immersed in
Packard Scintillator 299 scintillation cocktail and counted in
a Packard 2000CA liquid scintillation counter.
Protein levels were determined by the method of Lowry et
al. (59), using bovine serum albumin as the standard.
Specific affinity was calculated from the relationship a'A =
Vma,IK,, where Vma. is the maximum specific uptake rate
and K, is the half-saturation concentration for uptake after
converting maximum uptake rates to units of substrate
accumulated per gram of cells (dry weight) per hour (15).
RESULTS
Enumeration of bacteria in raw seawater samples. Bacteri-oplankton
populations in samples taken from Resurrection
Bay ranged from a usual 0.20 x 106 to 1.07 x 10' cells per ml
for the Resurrection Bay experiments, as determined by flow
cytometric analysis. The sampled North Sea population was
0.11 x 10' cells per ml. On the basis of these data, bacterial
biomass was 7.0 to 82.3 p.g/liter (Table 1).
Growth of cells in dilution culture. By epifluorescence
microscopy, the limit of detection of cells present in dilution
culture was determined to be approximately 5 x 104 cells per
ml. With final populations seldom exceeding 106/ml, growth
was generally not observed until the late log growth phase or
after 30 days (Fig. 2). Therefore, any morphological change
in the appearance of the cells or change in population
composition during growth of the dilution culture could not
be detected. The following observations were all made
during the Resurrection Bay experiments.
Several dominant cell types were observed in the different
tubes in which growth occurred. In various dilution cultures,
a mixture of cell types was present. Besides very small and
quite large rod-shaped organisms, spirilla, vibrios, and cocci
could also be recognized. The cultures generated from the
highest dilution series consisted mostly of one single cell
type, whereas those from the lower dilution series were
generally mixed cultures with large cells (Table 2). Micro-scopic
observations are supported by flow cytometric data
showing that cultures from lower dilutions also contain more
contour regions in the bivariate histogram (Table 2) (also see
reference 16).
The most frequently observed organism, in 9 of 17 positive
tubes of the highest dilution series (dilutions of 1 x 106 and
5 x 105) during the Resurrection Bay II (March 1990)
experiment, was a small, straight rod, with a cell volume of
approximately 0.05 to 0.06 ,um3 (length, 0.9 p.m; diameter,
0.3 p.m), and an apparent DNA content of 1.0 to 1.5 fg of
DNA per cell (Fig. 3). It is likely that this organism was
overgrown in the lower dilution series by organisms that are
able to attain higher maximum populations densities, since
maximum population densities differed considerably be-tween
cultures in which small rods were dominant and
cultures in which vibrios, large rods, or a variety of different
cell types were present (Fig. 2). Furthermore, in cultures in
which the larger rods were predominant, the presence of
subpopulations consisting of smaller cells could often be
recognized by flow cytometry (see Fig. 4 in reference 16).
Some of these cells were small rods according to micro-scopic
observations. Small rod-shaped organisms were the
dominant cell type in over 50% of the cultures that were
generated from the highest dilution series. Since all of these
cells were shown to have a uniform cell and colony morphol-ogy
upon isolation (discussed below) and taking into account
the fact that the overall viability of the sampled population
1-
P-ci
ot
0
8
6
4
2
0
o 20 40 60 80
Time (days)
FIG. 2. Growth curves of representative examples of each of the
various cell types in dilution cultures obtained during the experi-ments
with Resurrection Bay I (December 1989) and Resurrection
Bay II (March 1990) samples. Cultures are identified according to
the dominant cell type present after 2 months at 10°C (mixed culture
[0], vibrio culture [A], large-rod culture [E], and small-rod culture
[0]). Small-rod cultures attain maximum population densities of
around 0.25 x 106 cells per ml within the course of 2 months,
whereas cultures with other dominant cell types attain maximum
values of around 1.0 x 106 cells per ml. The horizontal line at log 4.5
cells per ml represents the detection limit.
VOL. 59, 1993
i
/I
o/
, I
I I
I I
/ I ,
,
1, I ,
,I
/ / I ,
/ I ,,, I
I , ,
I ,
,
I

5. 2154 SCHUT ET AL.
APPL. ENvIRON. MICROBIOL.
TABLE 2. Cell types in dilution cultures generated from the Resurrection Bay II (March 1990) experiment'
Dilution Dominant cell No. of cultures MCVb DNAC No. of subpopulationsd type generated (xm3) (fg/cell) (histogram regions)
1 x 106 Small rods 2 0.051 1.0 1
Large rods 1 0.231 2.2 3
Cocci 1 0.054 2.2 1
S x 105 Small rods 7 0.058 1.4 1
Large rods 3 0.153 2.5 3
Cocci 1 NDe ND ND
Vibrios 2 0.096 3.4 2
2.5 x 105 Small rods 4 0.055 0.3 3
Large rods 4 0.097 1.8 3
Vibrios 3 0.148 2.7 2
Coccoid (phage infectedf 2 ND ND ND
2.5 x 104 Vibrios 3 0.114 1.5 3
Mixed 6 >0.2 >4 >4
Coccoid (phage infectedf 1 ND ND ND
2.5 x 103 Mixed 10 >0.2 >4 >4
a Determined by flow cytometry and epifluorescence microscopy after 42 days of incubation at 10'C. For each dilution, a total of 20 tubes was inoculated. When
there was more than one culture of a cell type, MCV and DNA content data for only one culture are given.
b MCV, mean cell volume.
c Apparent DNA content (population mean).
d Number of subpopulations recognized by flow cytometry, specifically, the number of cell types present in one sample (see text; also see reference 16).
e ND, not determined.
f Possible phage infections leading to loss of the culture. The cell type was invariably coccoid.
was calculated to be between 29 and 69%, it was inferred culture and of isolated strains. (i) Resurrection Bay studies.
that this small rod might form a very important subpopula- Extensive subculturing studies were performed on the dilu-tion
within the natural bacterioplank-lton at the time of tion cultures of the Resurrection Bay II (March 1990) exper-sampling.
iment. All dilution cultures venerated could be subcultured
Trophic ranges of bacteria in raw seawater and in dilution in FAS, using inocula of 10 to 107 cells per tube. Subcul-
Apparent DNA content (fg/cell)
1000 -
c 800
.a
0 600
'0
e 400
0-
0
#°200
O L
0
0.1 10 0.1 10
.I ...I
0.1
af) s
0.01 IO
0.001
I I
. . . . II . 1 . , . 11
... I . . - I . . TV I -
A B
0
a -.11 <::Z;) .6 Al10
. 4
200 400 600 800 1000 0 200 400 600 800 1000
Log DAPI-DNA fluorescence intensity
FIG. 3. Bivariate apparent DNA content per cell versus cell volume histograms of a representative Resurrection Bay seawater sample
(December 1989) (A) and of a typical small-rod culture obtained by extinction dilution with an inoculum size of one cell per tube (B). Contour
lines reflect population density. DNA content was measured as DAPI-stained DNA fluorescence and is therefore apparent DNA content (16).

6. ISOLATION OF OLIGOTROPHIC ULTRAMICROBACTERIA 2155
TABLE 3. Cell populations present in dilution tubes of Resurrection Bay II (March 1990) and colony counts on various solid mediaa
Dilution culture Total Colony counts (CFU/ml of culture) on the following plate type:
population MPM1 MPM10 MPM100 MPM1,000 MPM10,000 ZoBell 2216E
RB2256 6.2 x 105 7.9 x 103 2.2 x 10 1.2 x 104 0 0 6.7 x 105
RB2259 4.8 x 105 1.0 x 105 1.6 x 105 9.3 x 104 1.1 x 105 0 5.0 x 105
RB2510 1.0 x 106 1.6 x 104 4.8 x 105 4.7 x 105 2.1 x 105 0 4.5 x 105
RB2512 1.1 x 106 2.9 x 105 5.4 x 105 4.2 x 105 3.5 x 105 0 1.0 X 106
RB2518 3.6 x 105 6.2 x 103 1.3 x 105 9.2 x 103 0 0 8.4 x 105
RB2519 8.0 x 105 3.2 x 104 4.1 x 105 2.8 x 105 3.2 x 105 0 4.7 x 105
RB2109 1.1 x 106 7.2 x 104 5.0 x 105 1.5 x 105 5.0 x 103 0 9.6 x 105
a Dilution cultures stored at 5°C for 1 year were spread onto six agar plate types. Counts were obtained in duplicate, using six plate types as described in
Materials and Methods. Final CFU counts were obtained after 32 days of incubation at 20°C.
I Total counts (in cells per milliliter) present in dilution tubes immediately prior to plating. Counts were obtained by using epifluorescence microscopy as
described in Materials and Methods.
tures could even be obtained from the two highest dilution
series after 2 years of storage at 5°C. All dilution cultures
except those that contained small rod-shaped bacteria could
be subcultured on full-strength marine nutrient agar or broth
(eutrophic medium), and all dilution cultures could be cul-tured
in synthetic seawater medium (MPM) containing 2 mg
of Casamino Acids per liter (oligotrophic medium). There-fore,
according to definitions used, only the small-rod-containing
cultures harbored obligately oligotrophic bacteria
as the single cell type and all other cultures represented
facultative oligotrophs.
Although these small-rod dilution cultures could not be
subcultured in high-nutrient medium, low-level (2-mg/liter)
Casamino Acid additions allowed further growth of the
population. Progressive subculturing of these dilution cul-tures
in MPM with 2 mg of Casamino Acids per liter yielded
strands of organisms that often exceeded 100 pm in total
length. Transfer to FAS again yielded single cells.
On repeated occasions dtiring and after growth of the
dilution culture, aliquots of the small-rod dilution cultures
were spread plated. Six months after the cultures had
reached the stationary growth phase, it was possible to
obtain colonies from these cultures on ZoBell 2216E agar
and on MPM agar. These colonies were barely visible to the
naked eye and could be properly observed and counted only
by using a binocular at a magnification of x25. It took
approximately 6 weeks at room temperature for these colo-nies
to appear. This isolation procedure could be repeated
with cultures that had been stored for 1 year at 5°C (Table 3).
Colony counts differed with substrate concentration and
type of medium used. The highest numbers were obtained on
ZoBell 2216E agar, representing between 60 and 100% of the
population present in the dilution culture. Counts on the
defined complex substrate mixture in MPM agar were high-est
with 10 mg of C per liter (MPM10). Colony counts
represented around 50% of the population present in the
dilution culture. No colony counts were obtained on MPM
agar with 10,000 mg of C per liter (MPM10,000). The single
colony type obtained from small-rod cultures of the Resur-rection
Bay II experiment was a smooth, translucent, and
yellow-pigmented colony. Flow cytometric analysis of pure
cultures generated from these colonies revealed that the cell
volume and apparent DNA content of these cells were
similar to those of the cells in the original dilution culture
(Table 4). Incomplete segregation and strand formation were
also observed for these isolates during growth on 2 mM
D-glucose, but not on 4 mM L-alanine.
(ii) North Sea studies. During the experiments with the
North Sea samples, the trophic range of the organisms
present in the original seawater sample was investigated with
various carbon concentrations in the dilution medium. For
this purpose, FAS in the dilution tubes was supplemented
with the mixed-carbon-source stock solution also used for
MPM agar to final dissolved organic carbon concentrations
of 5, 50, 500 and 5,000 mg of C per liter. After 2 months of
incubation at 10°C, of the 160 tubes containing FAS supple-mented
with the four substrate concentrations and inocu-lated
with 1 ml of seawater diluted 2 x 106 and 1 x 106 times,
none produced cell populations above the detection level.
This result indicated that the growth of dilution cultures was
completely inhibited by the addition of 5 mg of C per liter.
Viable counts from the seawater sample on ZoBell 2216E
agar were 0.5 to 1% of DAPI-stained total counts (data not
shown). Isolates obtained from direct spread plating of the
North Sea sample on ZoBell 2216E agar were, however, able
to produce cultures in the carbon-supplemented dilution
tubes at all organic carbon concentrations tested, indicating
that these organisms were not present in the inocula at the
very high dilutions used. These spread-plated isolates resem-bled
the organisms generally obtained by standard isolation
procedures with a cell volume and an apparent DNA content
of 0.20 to 0.68 ,um3 and 3.8 to 4.6 fg per cell, respectively.
From these observations, it was concluded that the contri-bution
of cells present in the original North Sea sample and
able to grow at high substrate concentrations (i.e., faculta-tive
oligotrophs or eutrophs) to the total population was less
than 1% and that the majority of the cells behaved as
obligately oligotrophic bacteria upon first cultivation in the
TABLE 4. Comparison of flow cytometry data of a raw seawater
sample, a typical dilution culture, and an isolate of
Resurrection Bay II experiment*
Medium Pop(uclelaltsi/omnl,d1e0n6s)ity M(ACmV3b) (DfgN/cAelCl)
Raw seawater sample 0.20 0.052 2.7
FASd Dilution culture 0.71 0.072 1.1
Subculture 0.24 0.051 1.0
MPMC
1 mg of C/liter, Ala 7.2 0.06 1.7
100 mg of C/liter, Ala 65.5 0.09 1.7
a Identification code of the dilution tube and isolate was RB2256. This
organism was one of the two small rod isolates from a dilution of 1 x 106
(Table 2).
b Mean cell volume.
c Apparent DNA content per cell (population mean).
d Dilution culture in filtered autoclaved seawater.
Isolate in synthetic seawater medium supplemented with alanine.
VOL. 59, 1993

7. APPL. ENVIRON. MICROBIOL.
dilution culture medium. The total culturable fraction of this
seawater sample was between 42 and 63% (16).
All 16 dilution cultures in unamended FAS generated from
the North Sea sample could be subcultured in FAS, but none
of the dilution cultures could be subcultured in ZoBell 2216E
broth. However, more than 1 year after reaching the station-ary
growth phase, seven of these dilution cultures could be
isolated on low-nutrient agars and ZoBell 2216E agar,
whereas none of these cultures had produced colonies on
any of the agars tested in the preceding period. The numbers
of colonies appearing on spread plates accounted for approx-imately
80% of the cells present in the dilution cultures,
according to direct microscopic counts (data not shown).
This result showed that the remarkable adaptation period
eventually leading to the isolation of formerly obligately
oligotrophic bacteria on plates that was observed during the
Resurrection Bay II experiment was reproducible. One
strain (strain NS 1619) could be isolated 4 months after
reaching the stationary growth phase. This strain showed
close resemblance to the nine small, rod-shaped strains from
Resurrection Bay II, both in cell and colony morphology,
with a cell volume of 0.07 ,um3, an apparent DNA content of
0.9 fg per cell, and yellow pigmentation. The strains that
were isolated after 1 year were also rod shaped, but colonies
were not pigmented. Flow cytometric data of these cells are
not yet available.
All isolates eventually obtained from the North Sea exper-iment
could be cultured on the total range of substrate
concentrations tested (FAS through ZoBell 2216E broth),
and it was concluded that all isolates represented faculta-tively
oligotrophic bacteria after isolation.
Specific affinity of dilution cultures and isolates. Uptake
experiments were performed on a representative member of
the small rod isolates of the Resurrection Bay II experiment,
strain RB2256, and on a representative small-rod dilution
culture from this same experiment, culture RB2109.
Time course uptake experiments with dilution cultures
revealed that the uptake of mixed amino acids was linear for
up to 10 min. High cell density uptake experiments with
strain RB2256 were linear during the first 30 s (Fig. 1). The
presence of 5 ,uM CCCP (carbonyl cyanide m-chlorophenyl-hydrazone)
completely inhibited the uptake of glucose,
indicating that uptake may be directly driven by proton
motive force. For chemostat-grown cells, preincubation with
the metabolizable substrate alanine was essential to allow
immediate uptake of glucose. This result may indicate that
cells became deenergized during the washing procedure that
preceded the uptake experiment. Immediate uptake at
equally high rates could also be accomplished by using
phenazine metasulfate and ascorbate as the energizing sys-tem.
The increase in slope in the uptake curve by whole cells
that were not preincubated with alanine implies gradual cell
energization by glucose metabolism (Fig. 1). It may thus be
inferred that cell energization is necessary only in the course
of such short-term (1-min) uptake experiments but not in the
longer-term (10-min) uptake experiments with dilution cul-tures.
For the isolated strain RB2256, half-saturation concentra-tions
for uptake (K,) and maximum specific uptake rates
(Vma,) varied with the growth conditions. The lowest K,
values for glucose were obtained at low dilution rates in the
chemostat in the presence of a second substrate (alanine) (a
detailed description of these experiments is being prepared).
Eadie-Hofstee plots for glucose and alanine uptake during
nutrient-limited growth in continuous culture are presented
in Fig. 4.
50
,40
~30
E 20 o
10
0 1 2 3 4
V/S (mL/min/mg prot.)
FIG. 4. Eadie-Hofstee plots of substrate uptake data for glucose
(0) and alanine (-). Data were obtained with strain RB2256 grown
on medium supplemented with glucose (2 mM) and alanine (4 mM)
in continuous culture at a dilution rate of 0.025/h. prot., protein.
Calculated specific affinities for the dilution culture
RB2109 and for strain RB2256, together with data from the
literature for other marine bacteria are presented in Table 5.
When bi- or multiphasic kinetics are described for pure
cultures, both high- and low-affinity systems are presented.
DISCUSSION
The occurrence of bacterial growth in unamended filtrates
of seawater has been documented extensively. In order to
obtain conditions as natural as possible, filtered seawater (4,
35, 41, 47, 55, 57, 58, 76), autoclaved seawater (17), UV-irradiated
seawater (37), and filtered-autoclaved seawater (7,
38) have been used as the culture media for marine bacteria.
In many studies, the mean cell size of the cultured bacteria
is greater than that of cells normally observed in seawater (4,
26, 52). This phenomenon may be the result of overgrowth of
the dominant population by fast-growing atypical cells or the
result of a morphological alteration by the dominant cells.
Conclusive evidence for either of these possibilities can be
obtained only from genomic analysis of the cells.
The small cell size and low apparent DNA content of the
small rod-shaped isolates obtained in the course of this study
(even when grown in richer media) compared with the large
cell size and high apparent DNA content of cells in cultures
from lower dilution series suggest overgrowth by atypical
bacteria in low dilution series. Significant release of utiliz-able
organic substrates may occur during filtration and
sterilization of seawater (26, 33, 58). That this phenomenon
occurs is reflected in the fact that inocula of a few cells can
reach populations of 106 cells per ml in unamended seawater
from which all cells have been removed by filtration and that
has subsequently been autoclaved (FAS medium). An eutro-phic
organism, waiting for conditions to improve, may in this
system overgrow an autochthonous oligotrophic bacterium
that is not able to respond quickly (80) and has a low
maximum specific growth rate. Extinction dilution is then
the only remedy to obtain liquid cultures of dominant
autochthonous organisms from seawater samples. Whether
subsequently obtained isolates truly represent dominant
marine bacteria remains to be determined by in situ oligo-nucleotide
probe hybridization.
2156 SCHUT ET AL.

8. ISOLATION OF OLIGOTROPHIC ULTRAMICROBACTERIA 2157
VOL. 59, 1993
TABLE 5. Specific affinities for substrate uptakea by marine samples and known marine isolates calculated from
various references and of a dilution culture and isolate of the Resurrection Bay II experiment
V.. a' (liter/g of Rernc
Sample or organism Region Substrate K, (nM) (nmol/min/ A InstoitOcueatfnoegraphy, ~~~mg Reference of cells) cells/h)
Seawater'-C Coastal seawater (Scripps Glucose 2.3d 0.23 6,000 6
Institute of Oceanography,
La Jolla, Calif.)
Kumano-nada Sea (Japan)
Aarhus Harbour (Denmark)
Seawaterlh,e
GL-7f
RB2109 (dilution culture)
RB2256 (isolate)
Vibric sp. strain S14
(starved)
Serratia maninorubra
Ant 300g
Vibric sp. strain S14 (log
phase)
Vibrio sp. strain S14
(starved)
Corynebacterium sp.
strain 198
RB2256
Pseudomonas sp. strain
486
RB2256
Vibrio sp. strain S14 (log
phase)
Pseudomonas sp. strain
RP-303
Alteromonas haloplanktis
(Pseudomonas sp.
strain B16)
LN-155h
Ajstrup Beach (Denmark)
Sargasso Sea
Coastal seawater (Scripps
Institute of Oceanography,
La Jolla, Calif.)
Resurrection Bay (Seward,
Alaska)
Resurrection Bay
Botany Bay (New South
Wales, Australia)
Pacific Ocean (off California)
Antarctic convergence
Botany Bay
Botany Bay
Cook Inlet (Alaska)
Resurrection Bay
Nansei Shoto area (Japan)
Resurrection Bay
Botany Bay
Nansei Shoto area (Japan)
Isolated from Pacific clams
Coastal seawater (Scripps
Institute of Oceanogrphy,
La Jolla, Calif.)
Glucose
Glucose
Glucose
Glucose
Glucose
Glucose
Glutamate
Glycine
Glutamate
Alanine
Glutamate
Alanine
Amino acids
Glucose
Amino acids
Amino acids
Glucose
Glucose
Arginine
Arginine
Glucose
Leucine
Leucine
Glucose
Glucose
Glucose
Proline
Alanine
Leucine
Leucine
Glucose
Proline
Alanine
Alanine
Glucose
Glucose
Glucose
32d
150d
2,400d
58,000d
590,000d
85_194d
46-911d
83-262d
67
157
18
31
75
200
550
6,400
17
4,500
4,600
760
20,000
2,667
100-20,000
13,600
200
2,000-5,000
760
20,000
3,600
1,800
46,000
190,000
180
6,600
180,000
0.68 1,275
1.3 520
4.2 105
24 25
110 11
0.7-2.9 494-906
1.5-124 1,000-39,600
2.2-34 1,600-7,700
0.7 647
1.3 489
1.6 5,300
2.1 4,100
633
6.7 5,300
4,320
6
5.2
60
0.16
0.66
37
1,860
567
563
550
9
483
24
49
58
38
This report
Preliminary data
3
40
29
3
5.6 442 62
6.7 20
388 14
4-20
6
0.7
5-17
1.7
10.7
0.15
1.3
6.9
20.8
0.01
0.05
0.39
92-220
26
210
133-180
134
32
2.5
43
9
6.6
3
0.43
0.13
This report
1
This report
62
1
25
68
a Losses of label due to respiratory formation of CO2 are not taken into account.
b Bacterial biomass is calculated by assuming commonly observed cell volumes of 0.065 ,um3, a volume-to-carbon (C) biomass conversion factor of 0.35 x 1012
g of C per pLm3 and assuming cell carbon to be 50% of cells (dry weight) (8).
c Population not known, assumed density of 1 x 106 cells per ml.
d Values are not corrected for the presence of ambient substrate and therefore reflect K, + S,.
e Calculated from uptake rate of known population at added substrate concentration of 4.35 nM.
f Biomass for GL-7 estimated at 0.9 x 10-14 g (dry weight) per cell (cell dimensions, 0.2 by 0.5 pLm. Formula for volume: rr2 (4/3r + L - W), where L is cell
length and Wis cell width.
g Biomass for Ant 300 estimated at 1.5 x 10-12 g (dry weight) per cell (cell volume, 2.2 p.m3 [29]).
h Biomass for LN-155 estimated at 14 x 10-14 g (dry weight) per cell (cell volume, 0.20 p.m3 [18]).
An important part of the dilution cultures obtained con-tained
bacteria with an obligately oligotrophic character
upon first subcultivation. After a period of up to 1 year in the
stationary growth phase, during which the cultures were left
in the dark at 5°C, the ability to grow on medium with a
higher carbon content was displayed by 13 small-rod cul-tures
from the Resurrection Bay II experiment and by 8
small-rod cultures from the North Sea experiment. It is not
known whether specific mutations or other genetic changes
initiated this alteration in trophic character or whether
(de)repression or induction mechanisms of metabolic or
energy-transducing pathways are at work here. Whatever
the case, the ability to adapt to eutrophic growth conditions
shows that the trophic character of seawater bacteria may be
subject to alteration once they are separated from their
natural environment, suggesting that obligate oligotrophy
Seawaterb
Seawatere
(

9. APPL. ENvIRON. MICROBIOL.
reflects a life history, rather than a fixed physiological
characteristic.
Since the extinction dilution method favors the growth of
predominant species, the cultures generated from small
inocula (<10 cells) can be addressed as cultures of typical
marine bacteria. The isolates obtained from these dilutions
are indeed similar in size and apparent DNA content to those
of the bacteria normally present in seawater (Fig. 2 and
Tables 1 and 4). The 0.05 to 0.06 ,um3 cell volume is only 20
to 25% that of a starved typical marine organism Pseudo-monas
sp. strain T2 isolated on marine agar (71) and is equal
to the remarkably low cell volume exhibited by Vibio sp.
strain ANT 300 (volume of normal log-phase cell, 2.2 um3)
during its starvation-survival response, imposed by a period
of nutrient-limited growth followed by a period of more than
2 months of total nutrient deprivation (66). The apparent
DNA content of 1.5 fg per cell is only 38% of a single-copy
Eschenichia coli genome (44). This fact, together with the
mean value of 2.7 fg per cell for the bacteria in the original
seawater sample (Table 1) shows the remarkably low appar-ent
DNA contents of both indigenous marine bacteria and
the isolates described here.
The absence of correlation between values for mean cell
volume and DNA content in raw seawater samples (Table 1)
and in one of the isolated strains (Table 4) may be explained
by changes in population composition during the year or
differences in the physiological state of the cells (also see
reference 66). The low apparent DNA content of the small
rod isolates may, on the other hand, also reflect resistance to
DAPI penetration by these cells or a systematic change in
DNA fluorescence with cell size (16). Triton X-100 perme-abilization
of cells improves DAPI fluorescence but is not
always consistent. Therefore, DNA content is reported as
apparent DNA content per cell. Hoechst 33258-based fluo-rescence
measurements of cell extracts are planned. If DNA
measurements are indeed correct, these results indicate that
bacteria with the reported low cell volume and DNA content
are not necessarily exhibiting a starvation-survival response
(66) but may be actively growing bacteria.
It is noteworthy that the small rod-shaped cells isolated in
the course of this study exhibit a high specific affinity for
mixed amino acids, while their affinity for glucose or alanine,
supplied as the sole carbon and energy source does not set
them apart from other marine isolates. In fact, the affinities
obtained for these two substrates would result in doubling
times of up to half a year at concentrations in the environ-ment
of around 10 nM (formula 23 in reference 14). At this
moment, specific affinity values for mixed-amino-acid up-take
by other described marine isolates are unknown, but the
possible presence of high values is of great interest and
should be investigated in the future. Multiple-substrate uti-lization
would be the organism's only possibility to create
fluxes of organic carbon that can support realistic growth
rates. The reported specific affinity for mixed amino acids by
RB2256 would enable this organism to multiply every 4 days
at ambient concentrations of 100 nM mixed amino acids. In
this context, Law and Button (56) have shown the decrease
in growth threshold concentrations for glucose in the chemo-stat
when additional amino acids were supplied in the
medium. They already suggested that specific affinities of
around 400 liters/g of cells per hour for each component of
their balanced mixture of 21 substrates would enable cells to
grow with doubling times of a few days at oceanic substrate
concentrations.
The remarkably high specific affinity value for glucose of
strain GL-7, as presented in Table 5, is somewhat doubtful.
The specific uptake rate by this organism was calculated
from reference 38, using the reported cell size of 0.2 by 0.5
,um. However, cell size-to-biomass conversions are subject
to considerable error. A cell with essentially the same
dimensions of 0.24 by 0.54 ,um or 0.16 by 0.46 ,um has a 53%
larger or 40% lower cell volume, respectively. Such errors
may result in a significant deviation in the calculated specific
uptake rate. The same may be noted for Ant-300 and LN-155
in Table 5.
Although the values for specific affinities obtained from
pure cultures are generally somewhat lower than those
determined for natural samples, it is premature to conclude
on the basis of these results that cells isolated thus far are
essentially different from cells present in the environment.
Physiological conditioning such as starvation of cells has
been shown to increase the substrate-sequestering abilities
of Vibrio sp. strain S14 (Table 5). Growth conditions pre-vailing
in the ocean will be very difficult to reproduce under
laboratory conditions, and data can never be realistically
extrapolated to the oceanic situation when the mechanism of
obligate oligotrophy is not clearly understood. To achieve
understanding, it is essential that representative organisms
be investigated, even if those organisms have lost their
obligately oligotrophic character. We are currently setting
up for 16S rRNA sequence analysis, together with oligonu-cleotide
probing of natural waters, to prove whether the
isolates presented in this study are indeed an important part
of the bacterioplankton population in the areas investigated.
ACKNOWLEDGMENTS
This work was supported in part by grants from the Foundation
for Sea Research (SOZ), the Netherlands, the Department of
Biology and the Bureau Buitenland of the University of Groningen
in The Netherlands, the Royal Dutch Academy of Sciences, The
Netherlands Institute for Sea Research, the Biological Oceanogra-phy
section of the U.S. National Science Foundation, and personal
funds of R. A. Prins.
REFERENCES
1. Akagi, Y., and N. Taga. 1980. Uptake of D-glucose and L-pro-line
by oligotrophic and heterotrophic marine bacteria. Can. J.
Microbiol. 26:454-459.
2. Akagi, Y., N. Taga, and U. Simidu. 1977. Isolation and distri-bution
of oligotrophic marine bacteria. Can. J. Microbiol. 23:
981-987.
3. Albertson, H., T. Nystrom, and S. Kjelleberg. 1990. Starvation-induced
modulations in binding protein-dependent glucose
transport by the marine Vibrio sp. S14. FEMS Microbiol. Lett.
70:205-210.
4. Ammerman, J. W., J. A. Fuhrman, A. Hagstrom, and F. Azam.
1984. Bacterioplankton growth in seawater. I. Growth kinetics
and cellular characteristics in seawater cultures. Mar. Ecol.
Prog. Ser. 18:31-39.
5. Andersen, J. I. W., and W. P. Heffernan. 1965. Isolation and
characterization of filterable marine bacteria. J. Bacteriol. 90:
1713-1718.
6. Azam, F., and R. E. Hodson. 1981. Multiphasic kinetics for
D-glucose uptake by assemblages of natural marine bacteria.
Mar. Ecol. Prog. Ser. 6:213-222.
7. Baxter, M., and J. M. Sieburth. 1984. Metabolic and ultrastruc-tural
response to glucose of two eurytrophic bacteria isolated
from seawater at different enriching concentrations. Appl. En-viron.
Microbiol. 47:31-38.
8. Bj0rnsen, P. K. 1986. Automatic determination of bacterio-plankton
biomass by image analysis. Appl. Environ. Microbiol.
51:1199-1204.
9. B0rsheim, K. Y., G. Bratbak, and M. Heldal. 1990. Enumeration
2158 SCHUT ET AL.

10. ISOLATION OF OLIGOTROPHIC ULTRAMICROBACTERIA 2159
and biomass estimation of planktonic bacteria and viruses by
transmission electron microscopy. Appl. Environ. Microbiol.
56:352-356.
10. Bright, J. J., and M. Fletcher. 1983. Amino acid assimilation and
electron transport system activity in attached and free-living
marine bacteria. Appl. Environ. Microbiol. 45:818-825.
11. Britschgi, T. B., and S. J. Giovannoni. 1991. Phylogenetic
analysis of a natural bacterioplankton population by rRNA gene
cloning and sequencing. Appl. Environ. Microbiol. 57:1707-
1713.
12. Buck, J. D. 1974. Effects of medium composition on the
recovery of bacteria from sea water. J. Exp. Mar. Biol. Ecol.
15:25-34.
13. Buck, J. D. 1979. The plate count in aquatic microbiology, p.
19-28. In J. W. Costerton and R. R. Colwell (ed.), Native
aquatic bacteria: enumeration activity and ecology. American
Society for Testing Materials, Philadelphia.
14. Button, D. K. 1985. Kinetics of nutrient-limited transport and
microbial growth. Microbiol. Rev. 49:270-297.
15. Button, D. K. 1991. Biochemical basis for whole-cell uptake
kinetics: specific affinity, oligotrophic capacity, and the meaning
of the Michaelis constant. Appl. Environ. Microbiol. 57:2033-
2038.
16. Button, D. K., F. Schut, P. Quang, R. Martin, and B. R.
Robertson. 1993. Viability and isolation of typical marine oligo-bacteria
by dilution culture: theory, procedures and initial
results. Appl. Environ. Microbiol. 59:881-891.
17. Carlucci, A. F., S. L. Shimp, and D. B. Craven. 1986. Growth
characteristics of low-nutrient bacteria from the northeast and
central Pacific Ocean. FEMS Microbiol. Ecol. 38:1-10.
18. Carlucci, A. F., S. L. Shimp, and D. B. Craven. 1987. Bacterial
response to labile dissolved organic matter increases associated
with marine discontinuities. FEMS Microbiol. Ecol. 45:211-
220.
19. Christensen, J. P., T. G. Owens, A. H. Devol, and T. T. Packard.
1980. Respiration and physiological state in marine bacteria.
Mar. Biol. (New York) 55:267-276.
20. Coffin, R. B. 1989. Bacterial uptake of dissolved free and
combined amino acids in estuarine waters. Limnol. Oceanogr.
34:531-542.
21. Craig, H. 1971. The deep metabolism: oxygen consumption in
abyssal ocean water. J. Geophys. Res. 76:5078-5086.
22. Davis, P. G., and J. M. Sieburth. 1984. Estuarine and oceanic
microflagellate predation of actively growing bacteria: estima-tion
by frequency of dividing-divided bacteria. Mar. Ecol. Prog.
Ser. 19:237-246.
23. Douglas, D. J., J. A. Novitsky, and R. 0. Fournier. 1987.
Microautoradiography-based enumeration of bacteria with esti-mates
of thymidine-specific growth and production rates. Mar.
Ecol. Prog. Ser. 36:91-99.
24. Eguchi, M., and Y. Ishida. 1990. Oligotrophic properties of
heterotrophic bacteria and in situ heterotrophic activity in
pelagic seawaters. FEMS Microbiol. Ecol. 73:23-30.
25. Fein, J. E., and R. A. MacLeod. 1975. Characterization of
neutral amino acid transport in a marine pseudomonad. J.
Bacteriol. 124:1177-1190.
26. Ferguson, R. L., E. N. Buckley, and A. V. Palumbo. 1984.
Response of marine bacterioplankton to differential filtration
and confinement. Appl. Environ. Microbiol. 47:49-55.
27. Fuhrman, J. A., and F. Azam. 1982. Thymidine incorporation as
a measure of heterotrophic bacterioplankton production in
marine surface waters: evaluation and field results. Mar. Biol.
66:109-120.
28. Fuhrman, J. A., and R. L. Ferguson. 1986. Nanomolar concen-trations
and rapid turnover of dissolved free amino acids in
seawater: agreement between chemical and microbiological
measurements. Mar. Ecol. Prog. Ser. 33:237-242.
29. Geesey, G. G., and R. Y. Morita. 1979. Capture of arginine at
low concentrations by a marine psychrophilic bacterium. Appl.
Environ. Microbiol. 38:1092-1097.
30. Giovannoni, S. J., T. B. Britschgi, C. L. Moyer, and K. G. Field.
1990. Genetic diversity in Sargasso Sea bacterioplankton. Na-ture
(London) 345:60-63.
31. Giovannoni, S. J., E. F. DeLong, T. M. Schmidt, and N. R. Pace.
1990. Tangential flow filtration and preliminary phylogenetic
analysis of marine picoplankton. Appl. Environ. Microbiol.
56:2572-2575.
32. Goldberg, D. E. 1965. Minor elements in sea water, p. 163-196.
In J. P. Riley and G. Skirrow (ed.), Chemical oceanography,
vol. 1. Academic Press, London.
33. Goldman, J. C., and M. R. Dennett. 1985. Susceptibility of some
marine phytoplankton species to cell breakage during filtration
and post-filtration rinsing. J. Exp. Mar. Biol. Ecol. 86:47-58.
34. Gottschal, J. C. 1992. Substrate capturing and growth in various
ecosystems. J. Appl. Bacteriol. Symp. Suppl. 73:39S-48S.
35. Hagstrom, A., J. W. Ammerman, S. Henrichs, and F. Azam.
1984. Bacterioplankton growth in seawater. II. Organic matter
utilization during steady-state growth in seawater cultures. Mar.
Ecol. Prog. Ser. 18:41-48.
36. Hagstrom, A., K. Larsson, P. Horstedt, and S. Normark. 1979.
Frequency of dividing cells, a new approach to the determina-tion
of bacterial growth rates in aquatic environments. Appl.
Environ. Microbiol. 37:805-812.
37. Hamilton, R. D., and A. F. Carlucci. 1966. Use of ultraviolet-irradiated
seawater in the preparation of culture media. Nature
(London) 211:483-484.
38. Hamilton, R. D., K. M. Morgan, and J. D. H. Strickland. 1966.
The glucose uptake kinetics of some marine bacteria. Can. J.
Microbiol. 12:995-1003.
39. HejthuUsen, J. H. F. G., and T. A. Hansen. 1986. Interspecies
hydrogen transfer in co-cultures of methanol-utilizing acidogens
and sulphate-reducing or methanogenic bacteria. FEMS Micro-biol.
Ecol. 38:57-64.
40. Hodson, R. E., and F. Azam. 1979. Occurrence and character-ization
of a phosphoenolpyruvate:glucose phosphotransferase
system in a marine bacterium, Serratia marinorubra. Appl.
Environ. Microbiol. 38:1086-1091.
41. Hofle, M. G. 1984. Degradation of putrescine and cadaverine in
seawater cultures by marine bacteria. Appl. Environ. Micro-biol.
47:843-849.
42. Hood, M. A., and M. T. MacDonell. 1987. Distribution of
ultramicrobacteria in a Gulf Coast estuary and induction of
ultramicrobacteria. Microb. Ecol. 14:113-127.
43. Hoppe, H.-G. 1976. Determination of properties of actively
metabolizing bacteria in the sea, investigated by means of
microautoradiography. Mar. Biol. 36:291-302.
44. Ingraham, J. L., 0. Maal0e, and F. C. Neidhardt. 1983. Growth
of the bacterial cell. Sinauer Associates, Inc., Sunderland,
Mass.
45. Ishida, Y., M. Eguchi, and H. Kadota. 1986. Existence of
obligately oligotrophic bacteria as a dominant population in the
South China Sea and the West Pacific Ocean. Mar. Ecol. Prog.
Ser. 30:197-203.
46. Ishida, Y., K. Fukami, M. Eguchi, and I. Yoshinaga. 1989.
Strategies for growth of oligotrophic bacteria in the pelagic
environment, p. 89-93. In T. Hattori, et al. (ed.), Recent
advances in microbial ecology. Japanese Scientific Society
Press, Tokyo.
47. Jannasch, H. W. 1967. Growth of marine bacteria in limiting
concentrations of organic carbon in seawater. Limnol. Ocean-ogr.
12:264-271.
48. Jannasch, H. W., and G. E. Jones. 1959. Bacterial populations in
sea water as determined by different methods of enumeration.
Limnol. Oceanogr. 4:128-140.
49. J0rgensen, N. 0. G., and M. S0ndergaard. 1984. Are dissolved
amino acids free? Microb. Ecol. 10:301-316.
50. Karl, D. M. 1979. Measurement of microbial activity and growth
in the ocean by rates of stable ribonucleic acid synthesis. Appl.
Environ. Microbiol. 38:850-860.
51. Karl, D. M., and 0. Holm-Hansen. 1978. Methodology and
measurement of adenylate energy charge ratios in environmen-tal
samples. Mar. Biol. 48:185-197.
52. Kirchman, D., H. Ducklow, and R. Mitchell. 1982. Estimates of
bacterial growth from changes in uptake rates and biomass.
Appl. Environ. Microbiol. 44:1296-1307.
53. Kirchman, D. L., S. Y. Newell, and R. E. Hodson. 1986.
VOL. 59, 1993

11. APPL. ENVIRON. MICROBIOL.
Incorporation versus biosynthesis of leucine: implications for
measuring rates of protein synthesis and biomass production by
bacteria in marine systems. Mar. Ecol. Prog. Ser. 32:47-59.
54. Kogure, K., U. Simidu, and N. Taga. 1980. Distribution of viable
marine bacteria in neritic seawater around Japan. Can. J.
Microbiol. 26:318-323.
55. Landry, M. R., L. W. Haas, and V. L. Fagerness. 1984.
Dynamics of microbial plankton communities: experiments in
Kaneohe Bay, Hawaii. Mar. Ecol. Prog. Ser. 16:127-133.
56. Law, A. T., and D. K. Button. 1977. Multiple-carbon-source-limited
growth kinetics of a marine coryneform bacterium. J.
Bacteriol. 129:115-123.
57. Lee, S., and J. A. Fuhrman. 1987. Relationships between
biovolume and biomass of naturally derived marine bacterio-plankton.
Appl. Environ. Microbiol. 53:1298-1303.
58. Li, W. K. W., and P. M. Dickie. 1985. Growth of bacteria in
seawater filtered through 0.2 pm Nuclepore membranes: impli-cations
for dilution experiments. Mar. Ecol. Prog. Ser. 26:245-
252.
59. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall.
1951. Protein measurement with the Folin phenol reagent. J.
Biol. Chem. 193:265-275.
60. Lundgren, B. 1981. Fluorescein diacetate as a stain for meta-bolically
active bacteria. Oikos 36:17-22.
61. MacDonell, M. T., and M. A. Hood. 1982. Isolation and charac-terization
of ultramicrobacteria from a Gulf Coast estuaxy.
Appl. Environ. Microbiol. 43:566-571.
62. Marden, P., T. Nystrom, and S. Kjelleberg. 1987. Uptake of
leucine by a marine gram-negative heterotrophic bacterium
during exposure to starvation conditions. FEMS Microbiol.
Ecol. 45:233-241.
63. Martin, P., and R. A. MacLeod. 1984. Observations on the
distinction between oligotrophic and eutrophic marine bacteria.
Appi. Environ. Microbiol. 47:1017-1022.
64. Meyer-Reil, L.-A. 1978. Autoradiography and epifluorescence
microscopy combined for the determination of number and
spectrum of actively metabolizing bacteria. Appl. Environ.
Microbiol. 36:506-512.
65. Morita, R. Y. 1988. Bioavailability of energy and its relationship
to growth and starvation survival in nature. Can. J. Microbiol.
34:436-441.
66. Moyer, T., and R. Y. Morita. 1989. Effect of growth rate and
starvation-survival on the viability and stability of a psychro-philic
marine bacterium. Appl. Environ. Microbiol. 55:1122-
1127.
67. Newell, S. Y., and R. D. Fallon. 1982. Bacterial productivity in
the water column and sediments of the Georgia (USA) coastal
zone: estimates via direct counting and parallel measurements
of thymidine incorporation. Microb. Ecol. 8:33-46.
68. Nissen, H., P. Nissen, and F. Azam. 1984. Multiphasic uptake of
D-glucose by an oligotrophic marine bacterium. Mar. Ecol.
Prog. Ser. 16:155-160.
69. Oliver, J. D., L. Nilsson, and S. Kjelleberg. 1991. Formation of
nonculturable Vibno vulnificus cells and its relationship to the
starved state. Appl. Environ. Microbiol. 57:2640-2644.
70. Porter, K. G., and Y. S. Feig. 1980. The use of DAPI for
identifying and counting aquatic microflora. Limnol. Oceanogr.
25:943-948.
71. Robertson, B. R., and D. K. Button. 1989. Characterizing
aquatic bacteria according to population, cell size, and apparent
DNA content by flow cytometry. Cytometry 10:70-76.
72. Rodriguez, G. G., D. Phipps, K. Ishiguro, and H. F. Ridgway.
1992. Use of a fluorescent redox probe for direct visualization of
actively respiring bacteria. Appl. Environ. Microbiol. 58:1801-
1808.
73. Roszak, D. B., and R. R. Colwell. 1987. Metabolic activity of
bacterial cells enumerated by direct viable count. Appl. Envi-ron.
Microbiol. 53:2889-2893.
74. Roszak, D. B., and R. R. Colwell. 1987. Survival strategies of
bacteria in the natural environment. Microbiol. Rev. 51:365-
379.
75. Schmidt, T. M., E. F. DeLong, and N. R. Pace. 1991. Analysis of
a marine picoplankton community by 16S rRNA gene cloning
and sequencing. J. Bacteriol. 173:4371-4378.
76. Simon, M., and F. Azam. 1989. Protein content and protein
synthesis rates of planktonic marine bacteria. Mar. Ecol. Prog.
Ser. 51:201-213.
77. Smigielski, A. J., B. J. Wallace, and K. C. Marshall. 1990. Genes
responsible for size reduction of marine vibrios during starva-tion
are located on the chromosome. Appl. Environ. Microbiol.
56:1645-1648.
78. Stevenson, L. H. 1978. A case for bacterial dormancy in aquatic
systems. Microb. Ecol. 4:127-133.
79. Tabor, P. S., and R. A. Neihof. 1984. Direct determination of
activities for microorganisms of Chesapeake Bay populations.
Appl. Environ. Microbiol. 48:1012-1019.
80. Torrella, F., and R. Y. Morita. 1981. Microcultural study of
bacterial size changes and microcolony and ultramicrocolony
formation by heterotrophic bacteria in seawater. Appl. Environ.
Microbiol. 41:518-527.
81. Yanagita, T., T. Ichikawa, T. Tsuji, Y. Kamata, K. Ito, and M.
Sasaki. 1978. Two groups of bacteria, oligotrophs and eutrophs:
their distributions in fresh and sea water areas in the central
northern Japan. J. Gen. Appl. Microbiol. 24:59-88.
82. Zimmerman, R., R. Iturriaga, and J. Becker-Birck. 1978. Simul-taneous
determination of the total number of aquatic bacteria
and the number thereof involved in respiration. Appl. Environ.
Microbiol. 36:926-935.
2160 SCHUT ET AL.