Limnol.
Oceanogr., 33(5), 1988, 1153-I 165
0 1988, by the American
Society of Limnology
and Oceanography,
Inc.
Methylamine uptake by zooxanthellae-invertebrate
symbioses:
Insights into host ammonium environment and nutrition
Christopher F. D ‘Elia
Chesapeake Biological Laboratory, University of Maryland,
Center for Environmental and Estuarine Studies, Solomons 20688-0038
Clayton B. Cook
Bermuda Biological Station for Research, 17 Biological Station Lane, Ferry Reach GE-O 1
Abstract
Cnidarians with endosymbiotic algae (=zooxanthellae) take up dissolved inorganic nutrients
from seawater, but neither the physiological mechanisms nor the effect ofhost nutrition on transport
kinetics is known. We used the NH,+ analogue [14C]methylamine ([14C]MA) to examine these
aspects of NH,+ uptake by a sea anemone (Aiptasia pallida) and a coral (Madracis decactis).
Both intact symbioses and isolated zooxanthellae took up [14C]MA. In anemones, uptake rates
per algal cell increased with time after feeding. Uptake rates for isolates from hosts unfed for 710 d were linear for at least 200 min, slightly light-dependent, and conformed to Michaelis-Menten
kinetics (K, = 68 PM; I’, = 3.8 mol lo-l8 cell-’ s-l). Isolates from well-fed hosts took up [14C]MA
much less rapidly at all concentrations tested and did not exhibit saturable uptake kinetics. NH,+
competitively inhibited [14C]MA uptake by isolated algae (inhibition constant = 4.0 PM) and
reduced [14C]MA uptake by intact symbiotic anemones. We hypothesize that [14C]MA (and by
analogy NH,+) uptake occurs by a “depletion-diffusion” mechanism in intact symbiotic anemones
with zooxanthellae maintaining very low intracellular [14C]MA and NH,+ concentrations in host
tissue and that [14C]MA uptake kinetics will be useful in evaluating the nutritional status of corals
and similar symbiotic associations under field conditions.
A key feature of symbioses between marine invertebrates and endozoic algae (e.g.
corals and zooxanthellae) is their ability to
remove net quantities of dissolved inorganic nutrients from seawater at low environmental concentrations (e.g. Muscatine 1980;
Wilkerson and Trench 1986). Although the
mechanism is unknown, three lines of evidence indicate that algal endosymbionts play
an important role: uptake is often light-dependent (Muscatine and D’Elia 1978; Wilkerson and Trench 1986); symbiont-free
animals cannot effect uptake (Kawaguti
1953; Muscatine and D’Elia 1978; D’Elia
1977); and, isolated zooxanthellae exhibit
Acknowledgments
This research was performed using the facilities of
the Bermuda Biological Station for Research, Inc., and
was supported by U.S. National Science Foundation
grants OCE 85-16599 (to C.F.D.) and OCE 86-02190
and BSR 84-07946 (to C.B.C.).
Contribution 1174 from the Bermuda Biological Station.
We thank J. Hayes, C. Reidy, and S. Ferguson for
technical assistance, D. Yellowlees for providing us
with a copy of an unpublished manuscript, and K. L.
Webb and G. Muller-Parker for making helpful comments on the manuscript.
saturation kinetics comparable to free-living algae and rates of nutrient transport that
compare closely to those for the intact association (D’Elia et al. 1983).
D’Elia (1977) suggested the mechanism
for dissolved nutrient uptake by invertebrate-algal symbioses: Zooxanthellae as sites
of active transport for nutrients deplete the
animal tissue of POd3- (or, by analogy, NH,+
or N03-) creating a concentration gradient
through which additional nutrient diffuses
passively into the animal tissue from seawater (Fig. 1). This putative mechanism,
which has not been the subject of much investigation, was more formally defined by
D’Elia et al. (1983) as the “depletion-diffusion” hypothesis. Alternative hypotheses
also exist. For example, Rees (1987), who
tested the effects of metabolic inhibitors
3-(3,4 dichlorophenyl)1,l -dimethylurea
(DCMU)
and methionine
sulfoximine
(MSX) on the NH,+ flux of symbiotic and
aposymbiotic
animals of the freshwater
green hydra symbiosis, hypothesized that
the host and not the algal endosymbiont was
responsible for NH,+ retention and uptake
in that symbiotic association.
1153
1154
D’Elia and Cook
In the present paper, we examine the uptake of the NH,+ analogue [ 14C]methylamine
([ 14C]MA) by the anemone Aiptasia pallida
and its zooxanthellae.
This radiolabeled
NH,+ analogue has proven to be a convenient and rapid tool for assessing uptake and
transport kinetics (see Balch 1986 for a discussion of uptake and transport terminology) of NH,+ for taxonomically
diverse organisms, including fungi (e.g. Hackette et al.
1970), free-living and symbiotic bacteria
(Kleiner
and Fitzke 198 1; Wiegel and
Kleiner 1982), cyanobacteria (Turpin et al.
1984), microalgae (Wheeler 1980), macroalgae (Wheeler 1979; MacFarlane and Smith
1982), lichens with algal endosymbionts
(Tapper 1983), and angiosperms (Smith
1982). With few exceptions, [14C]MA has
proved useful in understanding NH,+ transport, and generally the “affinity”
of the
transport system has been shown to be
greater (i.e. lower KS)for NH,+ than methylamine (MA), rendering MA a poor competitor with NH,+. For ecological studies in
environments with variable or high NH4+
concentrations, the fact that MA is a poor
competitor kinetically with NH,+ may yield
results that are difficult to interpret (Wheeler and McCarthy 1982). As we show here,
however, such a difference in kinetic behavior can be exploited for understanding
NH,+ metabolism and flux in symbiotic organisms where uptake and regenerative processes seem compartmentalized.
Although
a recent study has examined the transport
of MA by isolated zooxanthellae (Gunnersen et al. 1988), to our knowledge no previously published reports have used the
NH,+ analogue as a means to understanding
the nutrient fluxes and nutritional
state of
the intact symbiosis.
Our objectives were fivefold: to develop
a sensitive and rapid [14C]MA-based assay
for NH,+ uptake capacity by isolated zooxanthellae; to evaluate the effect of host
holozoic feeding history on the [ 14C]MA uptake kinetics of freshly isolated endosymbionts; to estimate host intracellular NH4+
concentrations indirectly from the inhibition kinetics of NH4+ on [14C]MA uptake;
to compare [14C]MA uptake kinetics obtained for isolated zooxanthellae with those
obtained with intact symbioses; and, to de-
Host
Tissue
Zooxanthella
Nutrlent
Fig. 1. Schematic representation of the depletiondiffusion hypothesis.
velop a convenient assay for the nutritional
sufficiency of invertebrates with algal endosymbionts. Our results are consistent with
the diffusion-depletion
hypothesis and suggest that a field assay for nutrient sufficiency
of symbiotic organisms can be developed
by evaluating [ 14C]MA uptake kinetics.
Materials and methods
Culture of anemones-The
details of the
culture techniques have been described in
detail by Cook et al. (1988). In brief, a clone
of A. pallida was maintained in an incubator
(12 : 12 L/D, 80 PEinst m-2 s-l, 25°C) and
fed Artemia daily for at least 1 week before
use. Some anemones (oral disc 4-7 mm)
were selected from these cultures and maintained individually
without
feeding in
capped culture tubes containing 20 ml of
glass-fiber filtered (Whatman GF/AE) Sargasso Sea water (GFSSW) obtained from the
surface at Hydrostation
“S,” 22 km SE of
Bermuda. Zooxanthellae
from anemones
maintained in these culture tubes showed
signs of nutrient limitation within 20-30 d
after feeding had ceased (Cook et al. 1988).
Aposymbiotic anemones- Anemones with
greatly reduced algal populations were produced by the method of Steen and Muscatine (1987). After a 4-h exposure to low temperature
(4”C), these anemones
were
maintained in the dark without feeding at
25°C with daily changes of filtered GFSSW
for 2 weeks. Thereafter, the anemones were
kept in the dark and fed weekly. Microscopic examination of tentacle squashes revealed few zooxanthellae in these anemo-
Methylamine uptake by symbioses
nes, which for convenience
we term
aposymbiotic.
Collection of coral -The scleractinian
coral Madracis decactis was collected at a
depth of 3 m from the West End ledge reefs
in Bermuda and kept in unfiltered seawater
with indirect natural sunlight in the seawater system at the Bermuda Biological Station without additional feeding for 3 d before use.
[14C]methylamine uptake by isolated zooxanthellae-Except for dark incubations, all
[14C]MA experiments were conducted under the incubation
conditions
described
above. Care was taken to avoid NH,+ contamination that would affect MA uptake.
All glassware was rinsed first in 1.O N HCl
and next with GFSSW, known to contain
virtually
undetectable
NH,+ concentrations.
Zooxanthellae were isolated from anemones as follows: 4-5 anemones (4-7-mm oral
disc diam) were homogenized for 1 min in
a glass tube and pestle tissue grinder with
- 2 ml of GFSSW; the homogenate was centrifuged at high speed (- 7 50 x g) on an
IEC clinical centrifuge for l-2 min and the
supernatant removed by Pasteur pipet; the
algal pellet was resuspended in -2 ml of
GFSSW and rehomogenized in a clean tissue grinder. This procedure was repeated at
least three times to produce a pellet consisting of zooxanthellae and a few isolated
nematocysts but very little animal tissue.
Cell suspensions were diluted with GFSSW
to produce densities between 5 x lo5 and
1.5 x 1O6 cells ml-l, and cell counts were
obtained from four counts with a Neubauer
hemacytometer. Chlorophyll
a was determined by extracting an aliquot of the stock
cell suspension with 90% acetone, grinding
with a tissue grinder, centrifuging the acetone extract, and reading absorbances at 663
and 630 nm. The equations of Jeffrey and
Haxo ( 1968) were used to calculate Chl a
from absorbances. Some determinations
were made with a Turner model 100 fluorometer, calibrated spectrophotometrically.
Before [14C]MA was added, zooxanthellae were resuspended in GFSSW and allowed to sit in indirect light for 30 min to
ensure that any residual NH,+ was removed
from the medium. [ 14C]methylamine (ICN)
1155
was obtained as methylamine-HCl
in 1.5
ml of ethanol (sp act, 40 mCi mmol-l). The
initial solution was diluted with 3.75 ml of
distilled/deionized
water and kept refrigerated until use. [ 14C]MA was added to 0.51.O-ml zooxanthellae
suspensions
in
GFSSW in polycarbonate clinical centrifuge
tubes to yield desired MA concentrations
(generally 1O-20 PM) without additional
carrier. Initial experiments were conducted
to ensure that the time-course of uptake was
linear at all substrate concentrations and to
verify that substrate limitation
was not a
problem during the 30-min (or less) incubation. For experiments involving competi tive inhibition with NH4+, the desired NH,+
concentration was achieved by adding 5100 ~1 of stock solutions of NH,Cl in distilled water. Centrifuge tubes were placed in
a culture incubator held at 25°C and zooxanthellae were suspended every 5-10 min.
At sampling times, algal suspensions were
filtered through 25-mm (0.45~pm pore size)
Millipore filters, and the medium was collected with an Amicon model VFMl filter
manifold. Filters were rinsed 2 x with -2
ml of GFSSW, sucked dry at vacuum pressures of < 125 mm of Hg, and placed in 5 .O
ml of Opti-Fluor
(United Technologies/
Packard). Aliquots of filtrates were also collected for counting. 14C was counted with a
Packard 45 30 liquid scintillation
counter,
calibrated with the manufacturer’s quenched
14Cstandards to calculate disintegrations per
minute (dpm). All samples were counted to
a precision of 1.0% at the 99% C.L. MA
uptake was calculated from dpm [14C]MA
taken up using the supplier’s stated specific
activity.
Incorporation of [14C]methylamine by
isolated zooxanthellae-To
determine if
zooxanthellae metabolize MA, we incubated zooxanthellae isolated from anemones
unfed for 13 d with [14C]MA (2.5 &i ml-‘,
14 PM MA; total volume, 25 ml). At intervals, 1 ml of cell suspension was removed
and filtered through 0.45-pm filters. Aliquots (50 ~1) of the filtrate were sampled for
[14C]MA counts in the medium. A second
l-ml sample was extracted with an equal
volume of 10% trichloroacetic
acid for 30
min at room temperature and then filtered.
These filters were rinsed with -2 ml of 5%
D’Elia and Cook
1156
TCA and then GFSSW. The dry filters were
counted as described above. Total 14C activity in cells was calculated from depletion
of [14C]MA in the medium, and substrate
incorporation
was expressed as the percentage of this activity in cells after TCA
extraction.
[14C]methylamine uptake by intact symbioses- Because of difficulties in obtaining
organisms of exactly equal size and morphology to obtain uptake rates at given concentrations, we decided instead to monitor
the time-course of [14C]MA depletion for
individual
specimens during short experiments in which the biomass used was large
relative to the amount of substrate added.
This approach minimizes effects due to accumulation of substrate by the organism.
All experimental
incubations
were conducted in 25-ml Nunc plastic vials in an
incubator held at 25°C. Single anemones (47-mm oral disc diam) were carefully isolated and allowed to reattach at the bottom
of a sample vial in 2.5 ml of GFSSW. Once
all anemones had re-expanded, an appropriate aliquot (5-10 ~1) of [14C]MA stock
was added to produce the desired MA concentration. Incubation solutions were bubbled very gently with intramedic tubing attached to an air pump to facilitate mixing.
At intervals, 50-~1 samples of medium were
removed for 14C counting as described
above. Uptake rates between sampling intervals were calculated from dpm, specific
activity, and medium volume.
A l-cm finger of M. decactis was placed
in a Nunc vial with 5 ml of GFSSW. When
the polyps expanded, [14C]MA was added
to yield a final concentration of 0.4 PC1 ml-l,
and the coral incubated as above. Samples
of medium were taken and analyzed as for
the anemones.
Kinetics calculations-The
well-known
means used to characterize nutrient uptake
is the so-called Michaelis-Menten
equation:
Jf=-
v,
x s
KS + s
(1)
where V is uptake rate, V, the constant for
maximum uptake rate, K, the half-saturation constant, and S the substrate concentration. For isolated zooxanthellae
when
Michaelis-Menten
kinetics were observed,
Michaelis-Menten
kinetic coefficients and
their standard errors were calculated directly using the uptake rates of MA determined at different substrate concentrations
as described above and the weighted, nonlinear regression estimates developed by
Wilkinson (196 1). For intact associations
where size and morphological
differences
between anemones made it impossible to
obtain exactly comparable specimens for
replicate samples at given concentrations,
we used the “perturbation”
method (Harrison and Davis 1977) and followed the depletion of [14C]MA from the medium. This
approach requires obtaining the uptake rates
between sampling intervals and correcting
for the effect of sample removal on incubation volume (D’Elia 1977). To reduce the
effect of sample variability
on rate determinations, which are very sensitive to small
errors, we first fitted a third-order polynomial regression to the uptake curve
S = A + Bt + Ct2 i- Dt3.
(2)
We next took the first derivative with respect to time of the fitted depletion curve
at each sampling time
dS/dt = B + 2Ct + 3Dt2
(3)
where A, B, C, and D are constants, S is
substrate concentration,
and t is time, to
obtain the uptake rate at a given time. This
value was then corrected for incubation volume and biomass values, as appropriate, to
obtain uptake rates. An S/ V vs. S (“Woolf “)
linear transformation (cf. Neame and Richards 1972) was used to determine if Michaelis-Menten kinetics applied.
To calculate the inhibition
constant, K,,
for NH,+ on [14C]MA uptake, we used the
“algebraic” method described by Neame and
Richards (1972), for which the applicable
equation is
K,=v’x
v-
y
-KS x
I
S x KS
(4)
where V, is the rate of uptake in the presence
of inhibitor,
V the rate of uptake in the absence of inhibitor, I the inhibitor concentration, S the substrate concentration, and
K, the half-saturation constant for substrate
uptake, which was as determined for hosts
unfed for 7-l 0 d.
Methylamine uptake by symbioses
0.6,
1157
I
0.0
0
60
120
160
240
TIME (min)
Fig. 3. Time-courses of [ 14C]MA uptake in the light
and dark expressed as dpm per cell by zooxanthellae
isolated from Aiptasia pallida unfed for 11 d. The dark
treatment included preincubation in the dark for 5 min.
Curves are empirical fits by second-order polynomial
regressions.
0.01 . . . . ’ . . . . ’ . . . . ’ . . . . 1
0
50
100
150
200
TIME (min)
Fig. 2. A. Time-course of [14C]MA uptake expressed as 14Cdisintegrations per minute (dpm) per
cell by zooxanthellae freshly isolated from symbiotic
Aiptasia unfed for 13 d. Curve was fitted by leastsquares linear regression. B. TCA-insoluble fraction in
zooxanthellae during experiment.
Results
f4C]methylamine uptake by isolated zooxanthellae - A time-course of [ 14C]MA uptake was biphasic for zooxanthellae freshly
isolated from anemones (A. pallida) unfed
for 13 d. Under the conditions of this experiment (MA concn, 18 PM; 5.8 x 1O5cells
ml-l), an initial nonlinear phase existed only
during the first 5-10 min of uptake (Fig.
2A). For the rest of the experiment (200
min), [14C]MA uptake was linear with time
and did not show evidence of feedback inhibition of uptake by product accumulation.
To determine whether [14C]MA taken up
was assimilated into protein, we extracted
samples obtained during the above experiment with 5% TCA. During the first 3 min
-6% of the radioactivity
remained in the
TCA-insoluble
fraction (Fig. 2B). After 20
min, < 3% of the 14Cwas found in the TCAinsoluble fraction (Fig. 2B), indicating that
negligible quantities of MA were assimilat-
ed into protein. In a separate experiment,
only 5% of the 14Ctaken up by isolated zooxanthellae was TCA soluble after a 4-h incubation. The high TCA-insoluble
fraction
on this longer time scale may indicate the
association of MA with the NH,+ porter or
metabolic assimilation of MA into protein.
To determine if light affects the uptake of
[14C]MA by isolated zooxanthellae, we isolated algae from anemones starved for 11
d. Half of this pooled cell suspension was
used to measure uptake in the light; the remainder was preincubated in darkness for
5 min before adding [‘“C]MA, with subsequent maintenance in darkness. There was
little difference in uptake over the first 30
min, but uptake decreased in the dark subsequently (Fig. 3). Thus the depletion of an
energy substrate produced by photosynthesis may be a factor in [14C]MA uptake.
Zooxanthellae isolated from anemones of
different particulate feeding history showed
vastly different kinetic patterns of nutrient
uptake (Fig. 4). Zooxanthellae obtained from
a host unfed for 2 months exhibited saturable uptake kinetics and the highest rate of
MA uptake per cell: using the Wilkinson
(196 1) weighted, nonlinear regression estimates for Michaelis-Menten
coefficients, we
obtained a maximal uptake rate (V,) and
standard error of the mean of 12.4 * 0.43 x
lo-l8 mol cell-’ s-l and a half-saturation
constant
(K,) and standard
error of
119.6 + 56.8 PM. Zooxanthellae
isolated
from hosts unfed for 7-10 d (using pooled
D’Elia and Cook
1158
A
‘-
2
UNFED 2 MONTHS
4
5
5
0
10
20
30
40
50
60
70
a
I ot-..-..--.-...-.--.-.....’
60
0
120
240
100
TIME (min)
METHYLAMINE CONCENTRATION (/AA)
Fig. 4. [14C]methylamine uptake rates vs. methylamine concentration (calculated from dpm and sp act)
for zooxanthellae isolated from anemones not fed brine
shrimp for 1 d, 7-10 d, and 2 months. Curves plotted
for 2-month and 7-10-d unfed treatments were fitted
from Michaelis-Menten coefficients with the method
of Wilkinson (196 1) as described in text. Data from
7-10-d unfed treatments are pooled from three separate experiments. Data from 1-d unfed treatment were
fitted with a least-squares linear regression.
A
--If
-
-
-
.
’
-
-
.
.
’
.
-
10
data from three separate experiments) exhibited saturable uptake kinetics with a lower cell-specific maximum uptake rate (V, =
3.83k0.67 x lo-l8 mol cell-’ s-l) and a KS
of 68.4+ 19.0 PM. Zooxanthellae
isolated
from anemones fed to repletion daily with
brine shrimp exhibited the lowest cell-specific uptake rates; uptake appeared to be
nonsaturable and was better explained by a
diffusional model than by a Michaelis-Menten model.
[14C]methylamine uptake by intact symbioses-symbiotic
A. pallida unfed for 27 d
rapidly depleted [ 14C]MA from the medium
z0
: \
:
APOSYMBIOTIC
O\
F
20 .A-A~-~-A-A-A-~-A-A~A-A
5z
15-
8
lo-
\
O\O
‘0
.
.
’
c
3’.
0
5
MWINE
10
15
CONCENTRATION (/A)
Fig. 6. A. Depletion curves of [14C]MA (i.e. substrate concentration vs. time, S vs. T) typical for symbiotic Aiptasia pallida unfed for at least 7 d (0, l ) and
fed to repletion (A). Fitted curves are third-order polynomials. B. Rate of uptake vs. concentration (i.e. V vs.
S) curves from panel A-uptake rates shown are in
scaled dimensionless units for comparison of the shapes
of the curves. C. The data in panel B for Aiptasia unfed
for at least 7 d expressed as S/V vs. S, which for Michaelis-Menten kinetics should yield a straight line.
S/V vs. S data for anemones fed to repletion are off
scale and not shown.
SYMBIOTIC
\
E
z
5-
0,
.‘~“~‘.~..~“~’
0
‘o-o-o
0
0
30
60
90
120
150
TIME (min)
Fig. 5. Depletion of methylamine (MA) from medium by symbiotic and aposymbiotic Aiptasia. MA
concentration was calculated knowing dpm and the
[14C]MA sp act.
containing Sargasso Sea water and no added
NH4+, but unfed aposymbiotic A. pallida
did not affect the MA content of the incubation seawater (Fig. 5). This finding implies that zooxanthellae play an important
role in uptake. Other depletion experiments
for symbiotic anemones yielded similar
curves (Fig. 6A) to that shown in Fig. 5. V
Methylamine uptake by symbioses
3
0
1
2
3
METHYLAMINE
4
5
6
7
CONCENTRATION
6
9
0
10
1159
60
120
180
240
TIME (min)
(,uM)
8
Fig. 7. Methylamine uptake rates vs. methylamine
concentration on a per zooxanthella basis obtained from
[ 14C]MA depletion curves with intact symbiotic anemones (four separate experiments-different symbols) as
compared with the kinetic curve obtained for isolated
zooxanthellae from hosts unfed for 7-10 d (solid line;
data of Fig. 4). Anemones used for uptake by intact
symbioses were unfed for - 1 week.
-0
/O
,0-O
B
o.O’ORo
-6
vs. S (Fig. 6B) and S/ Vvs. S (Fig. 6C) curves
derived from the depletion data do not yield
what would be expected if simple MichaelisMenten kinetics applied: the transport rate
was zero at a positive substrate or “threshold” concentration and the S/ V vs. S transformation, which should yield a straight line
intercepting the abscissa at -I&, was curvilinear.
Another indication of the role of the zooxanthellae in the uptake of MA in the intact
symbiosis comes by comparing cell-specific
uptake rates of [ 14C]MA in intact symbioses
and isolated zooxanthellae (Fig. 7). Clearly,
cell-specific uptake rates were somewhat
lower for the intact symbioses over the
ranges of concentrations
examined.
At
higher MA concentrations, cell-specific uptake rates in intact symbioses were about
70% of those for isolated zooxanthellae, although at lower concentrations (5 2 PM) uptake by the intact symbiosis, but not isolated algae, ceased.
The symbiotic coral A4. decactis also removed [ 14C]MA from the medium (Fig. 8A).
As in the case of [14C]MA uptake by symbiotic A. pallida, the kinetic curves obtained
did not conform to Michaelis-Menten
kinetics: over the concentration range examined, there was no evidence of saturation,
and depletion of [ 14C]MA from the medium
ceased at a substrate concentration of - 1.5
PM (Fig. 8B, C).
0’
LLI 205
2‘A
15.
Li
<
z
8
lo-
-
.
-
.
’
.
-
.
.
’
.
1
.O
,oOO
c
,0°0
m
O0
5-
/O
w
0
M~HYLAMNE &NCENTRATIO~~~M)
Fig. 8. A. Depletion of [14C]MA from medium by
symbiotic coral Madracis decactis. Curve fitted from
third-order polynomial regression. B. Uptake rate vs.
methylamine concentration (i.e. I/ vs. S), curve obtained from the fitted depletion curve in panel Auptake rates shown are in scaled dimensionless units
for comparison of the shape of the curve. C. The same
data in panel B with S/V vs. S transformation.
Competitive inhibition of [14C]methylamine uptake by NH,+ - [ 14C]MA uptake by
zooxanthellae
isolated from an anemone
host unfed for 8 d was readily inhibited by
the presence of NH4+ in the medium at low
concentrations (Fig. 9). Based on past studies of MA uptake by a wide variety of algae,
we assumed that the inhibition
was competitive and calculated an inhibition
constant (IQ for NH,+ inhibition of MA uptake
of 4.05 PM for zooxanthellae isolated from
D’Elia and Cook
1160
A-A
O-O
0-O
L
--lo’
a4
’
’
8
8
8
10
0
AiMONI”hi
CONCEtTRATIO;
(PM)
Fig. 9. Percent inhibition of [14C]MA uptake vs.
ammonium concentration for zooxanthellae isolated
from Aiptasia unfed for 7-l 0 d. The initial MA concentration was 5 FM and did not change significantly
during the experiment.
hosts unfed for 7-10 d; this value compares
to half-saturation
constants for NH,+ uptake ranging from 5 to 22 PM calculated by
D’Elia et al. (1983) for zooxanthellae isolated from several tropical marine invertebrate hosts.
When NH,+ was supplied to the medium
during uptake of [‘“C]MA
by an intact
anemone-algal
symbiosis, the uptake of
[14C]MA was inhibited in accordance with
the NH,+ concentrations supplied (Fig. 10).
For A. pallida which had not been fed for
15 d, NH,+ at 10 PM substantially reduced
[14C]MA uptake, although with time, depletion continued at a rate comparable to
that before NH,+ was introduced. A concentration of 50 PM NH,+ virtually halted
[ 14C]MA depletion (not shown).
Discussion
Depletion-d@usion hypothesis-The
data
we present here are consistent with the depletion-diffusion
hypothesis (D’Elia et al.
1983) for nutrient uptake by invertebrates
with algal endosymbionts. According to this
hypothesis, net nutrient flux into the host
does not depend on active transport at the
host cell membrane, but rather on the concentration gradient between the host cytoplasm and external seawater: when the nutrient concentration in the seawater exceeds
the nutrient concentration in the host cytoplasm, the nutrient diffuses into the host,
and vice versa. We hypothesize that the nutrient concentration in the host cytoplasm
-0
60
120
CONTROL
1pcM SPIKE
1OpM SPIKE
160
240
TIME (min)
Fig. 10. Inhibition effects of NH,’ spikes on
[14C]MA depletion by intact Aiptasia pallida unfed for
8 d. NH,’ spikes were added to experimental treatments at 60 min to achieve final concentrations of 1
and 10 FM.
is regulated, in turn, by two interrelated factors: the nutritional
status of the host as it
affects host nutrient regeneration rates and
the nutritional status of the alga as it affects
the alga’s nutrient uptake rate. Thus, when
uptake of the nutrient by the zooxanthellae
exceeds the rate of regeneration of the nutrient by the host tissue, the nutrient concentration in the host cytoplasm must decrease. Conversely,
when host nutrient
regeneration exceeds nutrient uptake by
zooxanthellae, the concentration in the cytoplasm must increase. Host cytoplasm nutrient levels, in turn, affect the diffusion of
nutrients across the host membrane.
[14C]MA is a useful tracer to test the depletion-diffusion
hypothesis for NH,+ because its kinetic behavior enables indirect
estimation of the internal concentration of
NH,+ in the host cytoplasm. MA is a poor
competitor kinetically with NH4+, and as
such, is a poor surrogate for quantifying
NH,+ uptake directly in ecological studies.
However, its competitive kinetic disadvantage is actually an advantage for studying
NH,+ uptake by algal-invertebrate
associations: the fact that [14C]MA is taken up by
symbiotic associations implies that NH,+
concentrations in the host cytoplasm are not
high enough for substantial competitive inhibition of uptake to occur.
It is implicit in the symbiosis literature
that the host intracellular environment is a
nutrient-rich
one that provides endosymbionts with a surfeit of nutrients (e.g. Cook
197 1; Wilkerson and Trench 1986). Evi-
1161
Methylamine uptake by symbioses
A.
Seawater
Host Tissue
Zooxanthella
Protein Catabolism
of Well-Fed Host
Produces Much
Ammonium
B.
Seawater
Host Tissue
Zooxanthella
Ammonium
Protein Catabolism
of Unfed Host
Produces Little
Ammonium
Fig. 11. Schematic representation of the hypothesis
to explain the effect of NH,+ on [14C]MA uptake by
corals and anemones with endosymbiotic zooxanthellae. A. High rates of NH,’ production by catabolic
processesin a well-fed host increase cytoplasmic NH,+
concentration and competitively inhibit MA uptake by
depletion-diffusion mechanism. B. Low rates of NH,+
production in unfed host result in low cytoplasmic concentrations of NH,+ that do not competitively inhibit
MA uptake by the depletion-diffusion mechanism.
dence in the present paper suggests that although the endosymbiont does receive NH,+
from host sources, the intracellular
environment is very low in concentration and
dependent on the feeding history of the host.
NH,+ concentrations in host tissue have
been measured in several studies, although
as Anderson and Burris (1987, p. 456)
pointed out “it is difficult . . . to determine
what the local nitrogen environment is for
intracellular symbionts such as zooxanthellae . . . .” Measurements of NH,+ concentrations in tissue homogenates have been
reported from corals (5-50 PM: Crossland
and Barnes 1977) and the anemone Aiptasia
pulchella (40 PM: Wilkerson and Muscatine
1984) but the disruption of animal tissue
undoubtedly generates NH,+ from deamination reactions when cellular structure is
destroyed. Moreover, homogenization
will
evenly distribute tissue metabolites such as
NH,+. Whether small-scale local depletion
of NH,+ exists near endozoic zooxanthellae
is unknown and cannot be determined on
the basis of NH,+ determinations
on crude
homogenates, but small-scale variations in
NH4+ concentrations
in host tissue seem
plausible. Wilkerson and Trench (1986)
suggested that zooxanthellae in the hemocytes of giant clams may be isolated from
host NH,+ sources, as the hemolymph fluid
has low NH,+ content.
The results shown in Fig. 10 suggest that
cytoplasmic NH,+ concentrations
are indeed low in A. pallida unfed for 8 d. [ 14C]MA
uptake by the intact symbiosis occurred at
comparable rates for three anemones until
NH,+ was added to the medium. Due to its
competitive
advantage, NH,+ interfered
with [ 14C]MA uptake; inhibition
increased
with NH,+ concentration (Fig. 10). [ 14C]MA
uptake apparently
resumed only after
enough time had passed for depletion of
added NH,+ from both the medium and the
anemone tissue.
Figure 11 illustrates our hypothesis to explain why [14C]MA uptake can occur when
NH,+ uptake by zooxanthellae should be
kinetically favored. Figure 11A shows that
when host cell regenerative processes produce NH,+ in concentrations that exceed the
uptake capacity of zooxanthellae, NH,+ will
interfere competitively
with [ 14C]MA uptake. We attribute the results of NH,+ pulses
(Fig. 10) and the reduced uptake of MA by
zooxanthellae from well-fed anemones (Figs.
4 and 6) to this effect. However, hosts that
do not feed are likely to have lower catabolic
rates, and thus uptake capacities of the zooxanthellae will be higher. Under these conditions of depleted cytoplasmic levels, NH,+
concentrations are not high enough to interfere with [ 14C]MA uptake (Fig. 11B).
On the basis of our proposed mechanism
of [‘“C]MA uptake, the data in Fig. 7 can
be used to estimate NH,+ concentrations of
host tissue. Rearranging Eq. 4, we obtain
S + KS
z=-x-v- K
(5)
x K
v;
K
where variables are as in Eq. 4. With our
calculated value of K, for NH,+ on MA of
1162
D’Elia and Cook
- 4 PM, and the assumptions that K, for MA
is -70 PM and that the uptake of MA by
intact symbioses at 4 PM occurs at -50%
of the rate expected for isolated zooxanthellae (Fig. 7) the cytoplasmic inhibitor
(NH,+) concentration would be 5 PM. We
believe that this approach is likely to overestimate host NH4+ concentration
I because the calculation of K, involves the
quantity (V - V,) in the denominator (Eq.
4) which will decrease if analytically “undetectable” NH,+ is present, and because
host cell membranes and cytoplasm present
additional barriers to substrates reaching
symbionts from ambient seawater- barriers that undoubtedly reduce uptake compared to isolated zooxanthellae and thus affect the kinetic coefficients obtained. Other
factors such as the host organism’s surface
morphology (cf. Gavis 1976) would also affect the rate of uptake in intact symbioses
relative to isolated zooxanthellae.
We emphasize that the kinetics of uptake
by intact associations are complicated by
the spatial and morphological relationships
of host and symbiont and that the hypothesized depletion-diffusion
mechanism is an
oversimplification.
Zooxanthellae are separated from the seawater medium by at least
two host cell membranes (the outer plasmalemma and a vacuolar membrane), plus
a variable amount of host tissue. The role
of host cell membranes is considered to be
passive in the diffusion-depletion
hypothesis, but it seems improbable that these
membranes are readily permeable to both
anions and cations.
The term “diffusion” sensu Muscatine and
D’Elia (1978) vs. “diffusion”
in the depletion-diffusion
hypothesis as used here is a
potential source of confusion that requires
clarification. The former use of the term was
invoked to explain NH,+ uptake kinetics by
corals. Such kinetic behavior can be resolved into two simultaneous and additive
terms: a saturable Michaelis-Menten
component with high substrate affinity, and a
concentration-dependent
component that
can be modeled either as Fickian diffusion
or by a very low substrate affinity (i.e. high
IQ Michaelis-Menten
uptake process. As
is the first in
used in this paper, “diffusion”
a sequence of events that we believe occurs
during nutrient uptake: specifically, it is a
concentration-dependent
movement of a
nutrient or nutrient analogue from seawater
into host cytoplasm and may or may not be
facilitated by an enzyme on the plasmalemma. Active mediation by animal membranes is not necessary for this transport to
occur, if the nutrient concentration in seawater exceeds the concentration in host tissue. The next event in nutrient uptake creates
the concentration gradient that enables the
first event to occur: we postulate that active
transport, conforming to Michaelis-Menten
kinetics, is responsible for nutrient uptake
by the zooxanthellae. As sequential events,
diffusion and active transport components
are mathematically
multiplicative,
not additive terms, sensu “diffusion
limitation”
(Pasciak and Gavis 1975).
Nutritional status of host and zooxanthellae-It is well established that invertebrates harboring zooxanthellae excrete less
NH,+ than do nonsymbiotic
counterparts
(Kawaguti 1953; Szmant-Froelich
and Pilson 1977; Muscatine and D’Elia 1978).
When they are fed to repletion, symbiotic
hosts also capable of capturing zooplankton
prey excrete NH,+ at greater rates than do
hosts deprived of particulate food (SzmantFroelich and Pilson 1977, 1984). Well-fed
hosts probably regenerate NH,+ at greater
rates both because of higher rates of protein
catabolism and because NH4+ is liberated
during the digestive process (Szmant-Froelich and Pilson 1977, 1984). Thus, factors
such as feeding, that increase host catabolic
and excretory processes should increase the
cytoplasmic nutrient concentration and thus
retard or reverse nutrient diffusion into the
host from seawater-effectively
reducing
uptake of the nutrient from seawater. A
compartmentalized
model involving
simultaneous uptake and regeneration of nutrient in accordance with nutritional history
could resolve why the direction and magnitude of net nutrient fluxes of different algal-invertebrate associations varies. Such a
model was developed for phosphorus fluxes
of corals (D’Elia 1977).
The effects of feeding history on MA uptake confirm our previous findings that the
zooxanthellae of A. pallida become nutritionally altered when the host is deprived
Methylamine uptake by symbioses
of prey food for a week or more. We have
shown elsewhere that starvation of A. palZida produces changes in growth rate, C : N
ratio, and other parameters of its zooxanthellae that are typical of nutrient-limited
algae (Cook et al. 1988). These changes presumably result from the depletion of host
tissue nutrient sources. Consistent with such
an interpretation
is Wilkerson and Muscatine’s (1984) observation that nitrate uptake
by A. pulchella could be induced only after
a month without feeding. They concluded
that host tissue levels of NH,+ had become
sufficiently depleted so that the induction of
nitrate reductase (Wilkerson and Trench
1985) could occur in the zooxanthellae. In
our study, we found higher uptake of MA
by zooxanthellae freshly isolated from unfed
anemones than by those isolated from wellfed anemones (Fig. 4; cf. Fig. 6B). This finding conforms to the predictions of the diffusion-depletion
hypothesis and to changes
in MA transport and nitrogen limitation that
have been reported in other organisms. For
example, Hackette et al. (1970) found similar changes in kinetic patterns for a filamentous fungus, Penicillium chrysogenum,
under varying conditions of nitrogen sufficiency, Wright and Syrett (1983) found that
N deprivation increased the rate of MA uptake in Phaeodactylum 50-fold, and Pelley
and Bannister (1979) found that uptake for
N-starved Chlorella was several orders of
magnitude higher than for N-replete Chlo-
rella.
[14C’MA transport and metabolism by isolated zooxanthellae- [ 14C]MA has been used
to investigate NH,+ fluxes in aquatic systems (e.g. Wheeler and McCarthy 1982), but
more often to investigate NH,+ transport
mechanisms of algae, fungi, and bacteria (e.g.
Hackette et al. 1970). Several studies have
shown that time-course of the uptake of MA
by bacteria, cyanobacteria, and microalgae
is biphasic, with an initial rapid uptake phase
followed by a slower linear phase (Gordon
and Moore 1981; Balch 1986; Wright and
Syrett 1983). The initial rapid phase may
relate to the transport mechanism itself,
while the second slower phase may relate to
the metabolic transformation
of MA to
g-methylglutamine
by glutamine synthetase
(GS). In studies with cyanobacteria (Bous-
1163
siba et al. 1984; Turpin et al. 1984; Kerby
et al. 1986), the second phase can be eliminated with the GS inhibitor
methionine
sulfoximine
(MSX). In other studies that
have demonstrated the biphasic nature of
MA uptake in the dinoflagellate Gonyaulux
(Balch 1986) and the diatom Phaeodactylum (Wright and Syrett 1983), the second
phase was not related to the metabolism of
MA.
In the present study, we observed biphasic MA uptake by isolated zooxanthellae
(Fig. 2A). The initial rapid phase of uptake
lasted < 10 min; during this period, we observed the highest proportion of TCA-insoluble MA (Fig. 2B). Incorporation of MA
into protein would be unlikely during this
short time frame since MA is not an amino
acid. Accordingly, the likely interpretation
of these data is that the TCA-insoluble
material represents MA bound either to the
porter molecule or to some other component of the algal cell membrane. Such an
interpretation would be consistent with the
notion that the rapid uptake phase results
from the operation of the transport system.
We observed the slower, linear phase to
continue for at least 3 h, suggesting that uptake rate during this time was not affected
by product accumulation. During this phase
we obtained a KS: K, ratio of 17 for zooxanthellae that were isolated from hosts
starved for 2 weeks. This ratio is close to
the mean value of 25 and well within the
range of values summarized by Koike et al.
(1983), who suggested that this value is
characteristic of NH,+ transport systems.
In contrast, our I’,,, value was about 30
pmol (g cell N)- l min-‘, roughly an order
of magnitude less than values Koike et al.
( 198 3) reported for several species of marine phytoplankton. We believe that this difference is real, but stress that the different
experimental
approaches taken to determine coefficients render it difficult to compare the results of different investigators.
Since we collected algal samples on membrane filters, we were unable to resolve uptake events on time scales < 1 min. Using
a rapid centrifugal separation technique,
Gunnersen et al. (1988) have investigated
the transport of MA during the first minute
by zooxanthellae isolated from Australian
1164
D’Elia and Cook
corals. They reported a rapid phase lasting
< 1 min. Although their approach is preferred for kinetic measurements, the longer
time scale approach we used will be more
appropriate for comparison with uptake
studies for the intact association, particularly with application to field studies. Despite the methodological
differences between our approach and theirs, both studies
obtained comparable KSvalues for MA uptake (68 vs. 34 PM) and K, for NH,+ (4.05
vs. 6.6 PM).
[14C]methylamine uptake as a field assay
of N suficiency of zooxanthellae- [14C]MA
has been used as a field assay of NH,+ uptake by phytoplankton,
although its usefulness is limited by the strong competitive
effect of NH,+ (Wheeler and McCarthy
1982). In contrast,
the advantages
of
[14C]MA as a tool to investigate the NH,+
environment of zooxanthellae in A. pallida
suggest its utility in experimentally
examining nitrogen sufficiency in field populations of corals and other zooxanthellar associations. We have hypothesized elsewhere
that nutrient limitation might occur in these
zooxanthellae in the field (Cook and D’Elia
1987). Our single observation of [14C]MA
uptake by the coral M. decactis indicates
that NH,+ levels may be low in natural coral
populations. The kinetics of [14C]MA uptake by zooxanthellae of A. pallida reflect
the feeding history and “nutrient environment” of the host. Such information in conjunction with other parameters of nutrient
sufficiency of zooxanthellae should provide
valuable insights into the nutrient dynamics
of corals under field conditions.
References
ANDERSON, S. L., AND J. E. BURRIS. 1987. Role of
glutamine synthetase in ammonia assimilation by
symbiotic marine dinoflagellates (zooxanthellae).
Mar. Biol. 94: 451-458.
BALCH, W. M. 1986. Exploring the mechanism of
ammonium uptake in phytoplankton with an ammonium analogue, methylamine. Mar. Biol. 92:
163-171.
BOUSSIBA, S., W. DILLING, AND J. GIBSON. 1984.
Methylammonium transport in Anacyctis nidulans R-2. J. Bact. 160: 204-210.
COOK, C. B. 1971. Transfer of ‘S-labeled material
from food ingested by Aiptasia sp. to its endosymbiotic zooxanthellae, p. 218-224. In H. M. Len-
hoff et al. [eds.], Experimental coelenterate biology. Univ. Hawaii.
AND C. F. D’ELIA. 1987. Are natural populaiions of zooxanthellae ever nutrient-limited?
Symbiosis 4: 199-2 12.
-,
AND G. MULLER-PARKER. 1988. Host
felding and nutrient sufficiency for zooxanthellae
in the sea anemone Aiptasia pallida. Mar. Biol.
98: 253-262.
CROSSLAND, C. J., AND D. J. BARNES. 1977. Nitrate
assimilation enzymes from two hard corals Acropora acuminata and Goniastrea australensis.
Comp. Biochem. Physiol. 57B: 15l-l 57.
D’ELIA, C. F. 1977. The uptake and release of dissolved phosphorus by reef corals. Limnol. Oceanogr. 22: 301-3 15.
p,
S. L. DOMOTOR,AND K. L. WEBB. 1983. Nutrient uptake kinetics of freshly isolated zooxanthellae. Mar. Biol. 75: 157-167.
GAVIS, J. 1976. Munk and Riley revisited: Nutrient
diffusion transport and rates of phytoplankton
growth. J. Mar. Res. 34: 161-179.
GORDON, J. K., AND R. A. MOORE. 198 1. Ammonium
and methylammonium transport by the nitrogenfixing bacterium Azotobacter vinelandii. J. Bact.
148: 435-442.
GUNNERSEN, J., D. YELLOWLEES, AND D. J. MILLER.
1988. The ammonium/methylammonium
uptake system of Symbiodinium
microadriaticum.
Mar. Biol. 97: 593-596.
HACKETTE, S. L., G. E. SKYE, C. BURTON, AND I. H.
SEGEL. 1970. Characterization of an ammonium
transport system in filamentous fungi with methylammonium-14C as the substrate. J. Biol. Chem.
245: 424 l-4250.
HARRISON, P. J., AND C. 0. DAVIS. 1977. Use of the
perturbation technique to measure nutrient uptake
rates of natural phytoplankton populations. DeepSea Res. 24: 247-255.
JEFFREY, S. W., AND F. T. HAXO. 1968. Photosynthetic pigments of symbiotic dinoflagellates (zooxanthellae) from corals and clams. Biol. Bull. 135:
149-165.
KAWAGUTI, S. 1953. Ammonium metabolism of the
reef corals. Biol. J. Okayama Univ. 1: 17 l-l 76.
KERBY, N. W., P. ROWELL, AND W. D. P. STEWART.
1986. The uptake and metabolism of methylamine by N,-fixing cyanobacteria. Arch. Microbiol. 143: 353-358.
KLEINER, D., AND E. FITZKE. 198 1. Some properties
of a new electrogenic transport system: The ammonium (methylammonium) carrier from Clostridium pasteurianum.
Biochim. Biophys. Acta
641: 138-147.
KOIKE, I., D. G. REDALJE, J. W. AMMERMAN, AND 0.
HOLM-HANSEN. 1983. High-affinity uptake of an
ammonium analogue by two marine microflagellates from the oligotrophic Pacific. Mar. Biol, 74:
161-168.
MACFARLANE, J. J., AND F. A. SMITH. 1982. Uptake
of methylamine by Ulva rigida: Transport of cations and diffusion of free base. J. Exp. Bot. 33:
195-207.
MUSCATINE, L. 1980. Uptake, retention, and release
Methylamine uptake by symbioses
of dissolved inorganic nutrients by marine algainvertebrate associations, p. 229-244. In C. B.
Cook et al. [eds.], Cellular interactions in symbiosis and parasitism. Ohio State.
AND C. F. D’ELIA. 1978. The uptake, retention, and release of ammonium by reef corals.
Limnol. Oceanogr. 23: 725-734.
NEAME, K. D., AND T. G. RICHARDS. 1972. Elementary kinetics of membrane carrier transport. Wiley.
PASCIAK, W. J., AND J. GAVIS. 1975. Transport hmited nutrient uptake rates in Ditylum brightwellii.
Limnol. Oceanogr. 20: 604-6 17.
PELLEY, J. L., AND T. T. BANNISTER. 1979. Methylamine uptake in the green alga Chlorella pyrenoidosa. J. Phycol. 15: 110-112.
REES, T. A. V. 1987. The green hydra symbiosis and
ammonium. 1. The role of the host in ammonium
assimilation and its possible regulatory significance. Proc. R. Sot. Lond. Ser. B 229: 299-314.
SMITH, F. A. 1982. Transport of methylammonium
and ammonium ions by Elodea densa. J. Exp. Bot.
33: 22 l-232.
STEEN, R. G., AND L. MUSCATINE. 1987. Low temperature evokes rapid exocytosis of symbiotic algae by a sea anemone. Biol. Bull. 172: 246-263.
SZMANT-FROELICH, A., AND M. E. Q. PILSON. 1977.
Nitrogen excretion by colonies of the temperate
coral Astrangia danae with and without zooxanthellae. Proc. 3rd Int. Coral Reef Symp. 1: 417423.
-,
AND -.
1984. Effects of feeding frequency and symbiosis on nitrogen metabolism and
respiration of the coral Astrangia danae. Mar. Biol.
81: 153-162.
TAPPER, R. C. 1983. Uptake of methylamine by symbionts of the lichen, Cladonia convoluta (algal
symbiont: Trebouxia). New Phytol. 95: 61-67.
TURPIN, D. H., S. A. EDIE, AND D. T. CANVIN. 1984.
In vivo nitrogenase regulation by ammonium and
methylamine and the effect of MSX on ammo-
1165
nium transport in Anabaena flos-aquae. Plant
Physiol. 74: 70 l-704.
WHEELER, P. A. 1979. Uptake of methylamine (an
ammonium analogue) by Macrocystis pyrifera
(Phaeophyta). J. Phycol. 15: 12-17.
-.
1980. Use of methylammonium as an ammonium analogue in nitrogen transport and assimilation studies with Cyclotella cryptica (Bacillariophyceae). J. Phycol. 16: 328-334.
AND J. J. MCCARTHY. 1982. Methylammonium uptake by Chesapeake Bay phytoplankton:
Evaluation of the use of the ammonium analogue
for field uptake measurements. Limnol. Oceanogr.
27: 1129-l 140.
WIEGEL, J., AND D. KLEINER. 1982. Survey of ammonium (methylammonium) transport by aerobic
N,-fixing bacteria- the special case of Rhizobium.
FEMS (Fed. Eur. Microb. Sot.) Microbial. Lett.
15: 61-63.
WILKERSON, F. P., AND L. MUSCATINE. 1984. Uptake
and assimilation of dissolved inorganic nitrogen
by a symbiotic sea anemone. Proc. R. Sot. Lond.
Ser. B. 221: 71-86.
-,
AND R. K. TRENCH. 1985. Nitrate assimilation by zooxanthellae maintained in laboratory
culture. Mar. Chem. 16: 385-393.
-,
AND -.
1986. Uptake of dissolved inorganic nitrogen by the symbiotic clam Tridacna
gigas and the coral Acropora sp. Mar. Biol. 93:
237-246.
WILKINSON, G. N. 196 1. Statistical estimations in
enzyme kinetics. Biochem. J. 80: 324-332.
WRIGHT, S. A., AND P. J. SYRETT. 1983. The uptake
of methylammonium and dimethylammonium by
the diatom, Phaeodactylum
tricornutum.
New
Phytol. 95: 189-202.
Submitted: 16 November 1987
Accepted: 15 February 1988
Revised: 23 May 1988