© 2000 Wiley-Liss, Inc.

Cytometry 39:173–178 (2000)

Original Articles

Cytometric Quantification of Nitrate Reductase by

Immunolabeling in the Marine Diatom
Skeletonema costatum
Frank J. Jochem,1* G. Jason Smith,2 Yu Gao,3 Richard C. Zimmerman,2 Alejandro Cabello-Pasini,4
Donald G. Kohrs,2 and Randall S. Alberte5
1
The University of Texas at Austin, Marine Science Institute, Port Aransas, Texas
2
Moss Landing Marine Laboratories, Moss Landing, California
3
Rutgers, The State University of New Jersey, New Brunswick, New Jersey
4
Universidad Autonoma de Baja California, Ensenada, Mexico
5
PhycoGen, Inc., Portland, Maine
Received 23 August 1999; Revision Received 22 November 1999; Accepted 24 November 1999

Background: The uptake of nitrate by phytoplankton is a Blot densiometry in NH4-grown cells. The protocol al-
central issue in biological oceanography due to its impor- lowed observation of early stages of NR induction during
tance to primary production and vertical flux of biogenic an upwelling simulation. NR abundance increased after
carbon. Nitrate reductase catalyzes the first step of nitrate the nutrient shift to reach a new physiological “steady-
assimilation, the reduction of NO3 to NO2. A cytometric state” 96 hrs later. NR activity exhibited diel variation with
protocol to detect and quantify relative changes in nitrate maxima at mid-day. NR abundance as estimated by both
reductase (NR) protein content of the marine centric flow cytometry and Western analysis exhibited a hyper-
diatom Skeletonema costatum is presented. bolic relationship to NR activity. This pattern suggests
Methods: Immunolabeling of NR protein was achieved post-translational activation of NR protein.
with polyclonal antibodies raised against S. costatum NR. Conclusions: The presented protocol allows the differ-
Antisera specific to a NR protein subunit and to a NR entiation of NH4- versus NO3-grown algae as well as the
polypeptide sequence were compared, and cytometric monitoring of early stages in the induction of nitrate
results of NR protein abundance were related to Western assimilatory capacities. Cytometry 39:173–178, 2000.
analyses. Changes in cellular NR abundance and activity © 2000 Wiley-Liss, Inc.
were followed during an upwelling simulation experi-
ment in which S. costatum was exposed to a shift from
ammonia to nitrate as major nitrogen source. Key terms: Skeletonema costatum; flow cytometry; ni-
Results: NR protein could be detected in NO3-grown cells trate reductase; immunolabeling; upwelling; phytoplank-
and at extremely low levels hardly discernible by Western ton bloom

The uptake of nitrate by phytoplankton is a central issue ing as much as 25% of photosynthesis-generated energy
in biological oceanography due to its importance to pri- (5). A comprehensive understanding of intrinsic fea-
mary production and vertical flux of biogenic carbon. For tures of NO3 metabolism in phytoplankton can yield
an oceanic ecosystem, the rate of “new” production (re- substantial insights into the biological mechanisms un-
lying on external inputs of nitrate in contrast to “regener- derlying the dynamics of new production (6). Despite
ated” production thriving on ammonia; 1) directly relates
to the sinking flux of biogenic material (2). Therefore,
nitrate utilization rates should set the upper limit for Grant sponsor: German Research Council; Grant number: DFG Jo
carbon supply rates to benthic communities (3). 192/5-5,6 to FJJ, Grant sponsor: Office of Naval Research to GJS, Grant
sponsor: National Science Foundation to RCZ.
Assimilatory nitrate reductases (NR; EC 1.6.6.1) cata-
*Correspondence to: Frank J. Jochem, Marine Science Institute, 750
lyze the first step of nitrate assimilation, the reduction Channel View Drive, Port Aransas, TX 78373.
of NO3 to NO2, in plants and fungi (4), thereby expend- E-mail: frank@emcs.de
174 JOCHEM ET AL.

growing knowledge of NR activity in diverse phyto- specific antibodies were produced by injecting New Zea-
plankton (7) and attempts to correlate NR activity to land White Rabbits with a synthetic peptide coupled to
algal growth rates and the environment’s nitrogen con- the KLH antigen and comprised of the sequence APAAKS-
dition (8, 9, 10), important aspects of NR regulation and DRITA derived from the sequence of a 45 kDa V8 protease
control at the cellular and molecular level have not fragment of S. costatum NR. Both sets of crude antisera
been addressed adequately (6, 11). were enriched for NR-specific titer by affinity purification
Attempts to provide mechanistic information on intrin- against the NR subunit polypeptide as described previ-
sic processes driving NO3 metabolism have been impeded ously (11).
due to technical difficulties in isolating and characterizing NR protein abundance was determined by standard
the key biochemical components (6). The present work Western analysis (16). Proteins in crude cell extracts were
describes an immunofluorescence-based cytometric pro- separated by SDS-PAGE and transferred electrophoreti-
tocol to detect and quantify changes in NR protein con- cally onto nitrocellulose membranes by a BioRad (Her-
tent in the marine centric diatom Skeletonema costatum cules, CA) Mini Trans-Blot威. Non-specific protein binding
and the application of this protocol to investigate the sites on the membranes were blocked prior to antibody
dynamics of NR protein and activity induction during incubation by 2% Carnation non-fat dry milk in phosphate
changes in nitrogen source and supply rate. Additionally, buffer saline (PBS, pH 7.4). NR subunit polypeptides were
the utility of flow cytometric analysis of NR protein abun- visualized by immunodecoration with an affinity-purified
dance, as a proxy for nitrate assimilation capacity, is as- NR polyclonal Ig from S. costatum (11). Duplicate blots
sessed. S. costatum was chosen due to its ecological were digitized at 300 dots per inch (dpi) and 256 gray
importance; being a prominent bloom-forming species in levels on an HPScanJet 4P. Densities of immunoreactive
diverse marine systems upon nitrate entrainment into the bands were quantified from the digitized images using the
upper water column by upwelling, wind mixing or land UTHSCSA ImageTool (V 1.25) software package (Univer-
runoff (12, 13). sity of Texas Health Science Center, San Antonio, TX;
freeware) by using non-immunoreactive regions of the
MATERIALS AND METHODS membrane as the baseline intensity.
Skeletonema costatum (CCMP 1332) was cultured in For whole cell immunofluorescence assays, cells were
0.2 ␮m filtered and autoclaved Monterey Bay seawater concentrated by centrifugation at 2500 g for 10 min at
with f/2 nutrients (14, 15) containing 62 ␮M NH4 (60 ␮M 15°C. Media was aspirated away from the cell pellet on
added, 2 ␮M in natural seawater) and 22 ␮M NO3 (from ice, and the cells were flash fixed by resuspension in 96%
natural seawater). Light (200 ␮mol quanta m-2 s-1) was ethanol at ⫺20°C. After 24 h at ⫺20°C, cells were con-
provided under a 16:8 light/dark cycle. centrated by centrifugation, resuspended in fresh 96%
For the simulated upwelling nitrogen-shift chemostat ethanol at ⫺20°C, and stored until processed for immu-
experiment, a 20 L polycarbonate carboy was inoculated nofluorescence. Prior to antibody labeling, cells were cen-
with a one-liter culture (5⫻105 cells ml-1) of S. costatum trifuged and rinsed twice with a solution of 3% BSA in PBS.
and medium was supplied at a dilution rate of 0.12 d-1. Non-specific binding sites were blocked in 3% BSA in PBS
This low f-ratio (i.e. the ratio of ammonia over total inor- for 1 h on ice, and subsequent incubation with 2% DMSO
ganic nitrogen concentration; 0.26) condition was main- (dimethylsulfoxide) in the blocking buffer for 30 min
tained for 14 d prior to sampling at the mid-photocycle for increased cell permeabilization. Cells were washed as
3 d prior to the N-source manipulation. At the onset of above, resuspended in the BSA-PBS buffer containing the
light (t0), the medium supply was changed to high f-ratio primary NR antibody at a 1:100 dilution (ca 1 ␮g ␮l-1) and
(0.92; 2 ␮M NH4, 100 ␮M NO3) and dilution rate (0.86 incubated for 1 h at room temperature. After washing 3
d-1). The culture was sampled at 4 h intervals during the times with 0.05% Tween-20 in PBS, cells were resus-
light and 2 h intervals during the dark periods for the pended in BSA-PBS buffer containing a 1:200 dilution of
following 5 d. goat-anti-rabbit IgG-FITC conjugate (Sigma Chemicals) for
In vitro extractable NR activity was estimated from 30 min at room temperature. Finally, all cells were washed
filtered cells that were frozen in liquid N2, resuspended in with the PBS-Tween-20 buffer to remove unbound anti-
0.1 M imidazole buffer (pH 7.5) containing 1 mM EDTA, 1 body complexes.
␮M FAD, 5% (v/v) glycerol and protease inhibitors (120 A parallel series of cells was treated only with the
␮M Pefabloc, 20 ␮M leupeptin, 10 ␮g ml-1 chymostatin; all secondary antibody to control for background fluores-
Sigma Chemicals, St. Louis, MO), and sonicated. The crude cence and non-specific binding. Affinity purified antibody
extracts were incubated in 0.1 M potassium phosphate stocks for immunofluorescence assays were pre-absorbed
buffer (pH 7.5) containing 1 mM EDTA, 100 ␮M NADH, 10 with fixed S. costatum cells that had been grown with
mM KNO3 and 1 ␮M FAD for 30 min at 15°C. NO2 formed NH4 as the sole nitrogen source in order to reduce the
during incubation was measured colorimetrically using 2% titer of non-NR-specific binding to cellular structures.
sulfanilamide in 4 N HCl and 0.2% N-1-naphtyl-ethylenedi- Cytometric analysis of 10,000 cells per sample was
amine (NED; Sigma Chemicals) at 540 nm in a Hewlett- performed on a Becton Dickinson FACSort flow cytom-
Packard 8452A diode array spectrophotometer. eter equipped with a 488 nm argon laser. S. costatum
Polyclonal antibodies against a S. costatum NR subunit cells were identified and discriminated against (low
were obtained as described in Gao et al. (11). Epitope- amount of) debris in 2-parameter histograms of Forward
 CYTOMETRIC QUANTIFICATION OF NITRATE REDUCTASE 175

FIG. 2. NR specific fluorescence (mean values, relative units) of NH4-

and NO3-grown Skeletonema costatum labeled by different protocols and
antisera. C ⫽ crude affinity purified antiserum, A ⫽ affinity purified
antiserum pre-sorbed to NH4-grown S. costatum, S ⫽ NR subunit probe,
P ⫽ NR polypeptide sequence probe.

tentially greater number of epitopes accessible to the Ig

pool generated against the entire NR subunit and suggests
FIG. 1. Frequency histograms of green fluorescence (FL1, 530/30 nm)
of Skeletonema costatum cells: after fixation (no label), with pre-immune
that, for low abundance targets like NR protein, multiva-
rabbit antibodies control (sec.ab cont.), with anti-NR grown in NH4- lent Ig pools are better than single epitope targeted Ig.
medium (NR-ab; ⫹NH4) and NO3-medium (NR-ab; ⫹NO3). Pre-absorption of anti-NR protein Ig stocks with NH4-
grown S. costatum improved differentiation of NO3-utiliz-
ing from NH4-utilizing cells and is recommended for stud-
Angle Light Scatter (FSC) versus Side Angle Light Scatter ies monitoring N-source dependent processes. The titer of
(SSC). Green fluorescence of the FITC-conjugated sec- the NR peptide epitope Ig was insufficient to overcome
ondary antibody was recorded through a 530/30 nm loss of Ig during the pre-absorption phase.
bandpass filter on a 1024 channels, 4 decades log scale, During the simulated upwelling experiment, immuno-
and the data were processed using WinList威 software labeling by NR subunit antibodies pre-absorbed to NH4-
(Verity, Topsham, ME). grown S. costatum resolved low levels of NR protein
under low f-ratio conditions and the initial and instanta-
RESULTS AND DISCUSSION neous increase in cellular NR abundance upon the shift to
Polyclonal NR protein subunit antiserum provided suf- high f-ratio conditions (Fig. 3). Such resolution was not
ficient immunolabeling to detect NR protein in Skel- obtained by Western Blot densiometry (6). The low-level
etonema costatum grown under both nitrate and ammo- NR abundance revealed by immunocytometry was linked
nia conditions. Although the protocol yielded some to extremely low but measurable NR activity (NRA).
increase in fluorescence upon incubation with secondary Again, these results may result from minor amounts of
Ig alone compared to untreated cells, treatment of NH4- NO3 in the Monterey Bay seawater used for culture prep-
and NO3-grown S. costatum with affinity purified anti-NR aration or reflect low constitutive NR levels that have been
Ig followed by the secondary Ig resulted in sufficiently demonstrated earlier for this species (17). Non-specific
high and differential increases in FITC fluorescence (Fig. binding can account for only a part of this low fluores-
1). The differences in fluorescence reflected differences in cence signal since all presented results were corrected for
nitrate assimilation between these two cultures. The low non-specific binding of the secondary antibody, FITC-la-
NR contents under ammonia growth might be related to beled goat-anti-rabbit IgG, and non-specific fluorescence
minor amounts of nitrate in the Monterey Bay water de- signals of the applied anti-NR probe were only about 10
rived culture medium (see Materials and Methods), but relative units.
low constitutive NR levels were reported in S. costatum The fast increase in NR abundance upon NO3 addition
grown on synthetic ammonia medium free of nitrate as has been demonstrated previously in S. costatum and
well (17). assigned to the initiation of NR transcripts (17), and sim-
Comparison of different anti-NR immunoprobe stocks ilar results are reported for the marine diatom Thalassio-
revealed a consistently higher fluorescence signal with the sira weissfloggii (18). After 48 hrs, NR protein leveled
anti-NR subunit (S) as compared to the NR peptide (P, into a new physiological balance. Although the scatter in
epitope-specific antibody), even when applied at the same the data is quite high and temporal resolution too low to
Ig concentration (Fig. 2). This difference reflects the po- resolve diel patterns, it appears that NR abundance was
176 JOCHEM ET AL.

FIG. 3. Temporal changes of cellular NR protein

abundance (NR ab., mean relative fluorescence units
of anti-NR) and NR activity (NR act.; fmol N cell-1 h-1)
during the upwelling simulation chemostat experi-
ment. Nitrogen supply was changed from NH4 to
NO3 at t⫽0, light/dark cycle is indicated during the
first 114 hrs after nutrient shift when more than one
daily sampling was performed.

FIG. 5. Protein specific NR activity (NRA/NR ab. ⫻ 1000; relative units)

during the upwelling simulation chemostat experiment, mid-day samples
only.

edra was also shown to exhibit a diel maximum NR

abundance at mid-day, as revealed by immunogold label-
ing (19).
Detection of NR protein abundance by flow cytometry
showed good agreement with densiometric analysis of west-
ern blots (Fig. 4b), and both methods exhibit a hyperbolic
relationship to enzymatic assay of NR activity (NRA) in the
light (Fig. 4a). Maximum NRA occurred during mid-light
cycle (Fig. 3) as shown previously for natural diatom popu-
lations, S. costatum and Thalassiosira weissfloggii, and the
rhodophyte Gracilaria tenuistipitata (17, 18, 20, 21). This
FIG. 4. a: Relation of NR protein abundance derived from densiometry is consistent with the observed induction of NRA by light
of Western Blots and flow cytometry (both relative units) to NR activity
(fmol N cell-1 h-1). Filled circles ⫽ cytometry, light-phase samples; hollow and repression in the dark (6, 17, 22). Therefore, no relation
circles ⫽ cytometry, dark-phase samples; squares ⫽ Western analysis of exists between NR protein abundance and NRA in dark-
mid-day samples; b: relation between NR abundance (relative units)
derived from densiometry of Western Blots and flow cytometry for mid-
phase samples (Fig. 4a).
day samples (r2⫽0.74, P ⬍ 0.001). The hyperbolic relation of NR protein to NRA in light-
phase samples is related to the less pronounced initial in-
crease in NRA as compared to NR abundance (Fig. 3). The
higher in the light than during the dark periods. This data suggest that the shift in f-ratio and nitrogen supply rate
result corresponds to an earlier short-term nitrate-pulse initiated NR synthesis, but the weak increase in NRA imme-
experiment with S. costatum in which maximum NR diately after the shift might be related to activation of low NR
abundance seemed to occur towards the end of the light protein contents already present during NH4 growth. A
cycle (17). The marine dinoflagellate Gonyaulax poly- strong increase in NRA occurred only after 2 days when NR
 CYTOMETRIC QUANTIFICATION OF NITRATE REDUCTASE 177

FIG. 6. Forward Angle Light Scatter versus anti-NR fluorescence at different times during the upwelling simulation experiment.

abundance had reached a new and higher “steady state”. homologous amino acid domains corresponding to redox
This pattern points towards non-biosynthetic effects on NRA centers the overall similarity in amino acid sequences
regulation consistent with the high degree of post-transla- from different plant and fungal NR is less than 50% (23).
tional control of this enzyme known for S. costatum (17). As There is evidence that NR from different algal classes are
a consequence, protein-specific NRA increased 2 d after the diverse and distinct in terms of biochemical characteris-
nitrogen shift (Fig. 5). tics and regulatory features (6, 7, 8, 11). These diversities
The present NR antiserum does not differentiate active might be related to the fact that diatoms present an evo-
from inactive forms of NR protein, and activity of NR lutionary lineage distinct from that leading to land plants
protein varies along physiological adaptation processes. as documented by rDNA sequence comparison (24).
Differentiation of active and inactive NR by the applied Polyclonal antibodies raised against squash NR showed
antiserum was not expected since specific Ig to target the fair cross-reactivity with several dinoflagellates, although
activation sites (phosphorylation, co-factor binding, Fe- this study may have been biased by non-specific binding
heme) would have to be designed for this task. Single cell, (3). NR antibodies specific to dinoflagellates (25) may
flow cytometric analysis of NR immunolabeling did, how- extend applications. Further antisera for species of inter-
ever, provide improved sensitivity for NR detection under est in specific study areas or specific to other algal classes
conditions of low NRA, indicating that this technique is are to be developed and more taxonomically universe
particularly valuable for monitoring early events during probes established for routine field application with di-
the induction of nitrate assimilatory capacity. Additionally, verse and complex phytoplankton communities. The gen-
this assay provides strong signal discrimination between eration of immunoprobes against widely conserved pep-
phytoplankton acclimated to NH4-supported versus NO3- tide sequence domains offers one approach for generic
supported growth. Flow cytometry provides the direct immunoassays (26). However, the observation that Ig
demonstration of population wide response to higher pools targeting single epitopes may not yield sufficient
f-ratio and nitrogen supply rate that cannot be observed by signal for detection of low abundance targets, such as NR,
test tube or filter paper measured induction of NR. Two- suggests that this approach may have limited application.
parameter-plots of Forward Angle Light Scatter (FSC) as a Still the present study demonstrates that the coupling of
relative measure for cell size versus NR immunofluores- flow cytometry with immunological techniques can be
cence suggest that all cells accumulated NR protein during used to assess nitrogen nutrition in phytoplankton and
the 24 hrs induction phase, followed by volumetric offers new insights into the study of nitrogen pathways.
growth of the induced cells: At all sampling intervals, no
high/low NR subpopulations were discernible as would LITERATURE CITED
have resulted from higher NR abundance in some but not 1. Dugdale RC, Goering JJ. Uptake of new and regenerated forms of
all cells, but the culture always provided a picture of one nitrogen in primary productivity. Limnol Oceanogr 1967;12:196-206.
homogeneous population (Fig. 6). 2. Eppley RW, Peterson BJ. Particulate organic matter flux and plank-
tonic new production in the deep ocean. Nature 1979;28:677-680.
Field application is limited, however, by the restricted 3. Balch WM, Yentsch CM, Reguera B, Campbell W. Examining nitrate
cross-reactivity of the NR antibody at hand. The polyclonal reductase by phytoplankton with an immunoassay. In: Yentsch CM,
S. costatum NR antiserum showed positive immunolabel- Mague FC, Horan PK, eds. Immunochemical approaches to coastal,
estuarine and oceanographic questions. Lecture notes on coastal and
ing in other marine diatoms (Thalassiosira pseudonana, estuarine studies. New York: Springer; 1988. p 263-276.
Phaeodactylum tricornutum) but not in non-diatom spe- 4. Beevers L, Hagman RH. Nitrate and nitrite reduction. In: Stumph PK,
Conn EE, Miflin BJ, eds. The biochemistry in plants, a comprehensive
cies such as Amphidinium carterae, Emiliania huxleyi, treatise, vol. 5: amino acids and derivates. New York: Academic
Isochrysis galbana, Dunaliella tertiolecta, and Synecho- Press;1980. p 115-168.
coccus sp. (6); it, thus, appears to be specific to diatoms. 5. Solomonson LP, Barber MJ. Assimilatory nitrate reductase: functional
properties and regulation. Annu Rev Plant Physiol Plant Mol Biol
Antigenecity of nitrate reductases from different phyto- 1990;41:225-253.
plankton species seems divergent (11, 18) and despite 6. Gao Y. Nitrate assimilation in the marine diatom Skeletonema costa-
178 JOCHEM ET AL.

tum (Bacillariophyceae): Biochemical characterization and environ- responses of diatoms to variable levels of irradiance and nitrogen
mental regulation. Ph.D. thesis, University of Southern California. Los availability: growth of Skeletonema costatum in simulated upwelling
Angeles; 1997;211 p. condition. Limnol Oceanogr 1992;37:989-1007.
7. Berges JA. Algal nitrate reductases. Eur J Phycol 1997;32:3-8. 18. Vergara JJ, Berges JA, Falkowski PG. Diel periodicity in nitrate reduc-
8. Berges JA, Harrison PJ. Nitrate reductase activity quantitatively pre- tase activity and protein levels in the marine diatom Thalassiosira
dicts the rate of nitrate incorporation under steady state light limita- weissfloggii (Bacillariophyceae). J Phycol 1998;34:952-961.
tion: a revised assay and characterization of the enzyme in three 19. Fritz L, Stringher CG, Colepicolo P. Imaging oscillations in Gon-
species of marine phytoplankton. Limnol Oceanogr 1995;40:82-93. yaulax: a chloroplast rhythm of nitrate reductase visualized by im-
9. Berges JA, Harrison PJ. Relationships between nitrate reductase ac- munocytochemistry. Braz J Med Biol Res 1996;29:111-117.
tivity and rates of growth and nitrate incorporation under steady-state 20. Berges JA, Cochlan WP, Harrison PJ. Laboratory and field responses of
light or nitrate limitation in the marine diatom Thalassiosira pseud- algal nitrate reductase to diel periodicity in irradiance, nitrate exhaus-
onana (Bacillariophyceae). J Phycol 1995;31:85-95. tion, and the presence of ammonium. Mar Ecol Prog Ser 1995;124:
10. Joseph L, Villareal TA. Nitrate reductase activity as a measure of 259-269.
nitrogen incorporation in Rhizosolenia formosa (H. Peragallo): in- 21. Lopes PF, Oliveira MC, Colepicolo P. Diurnal fluctuation of nitrate
ternal nitrate and diel effects. J Exp Mar Biol Ecol 1998;229:159-176.
reductase activity in the marine red alga Gracilaria tenuistipitata
11. Gao Y, Smith J, Alberte RS. Nitrate reductase from the marine diatom
(Rhodophyta). J Phycol 1997;33:225-231.
Skeletonema costatum. Plant Physiol 1993;103:1437-1445.
12. Guillard RRL, Kilham P. The ecology of marine planktonic diatoms. 22. Gao Y, Smith GJ, Alberte RS. Light regulation of nitrate reductase in
In: Werner D, ed. The biology of diatoms. Berkeley: University of Ulva fenestrata (Chlorophyceae). I. Influence of light regimes on
California Press; 1977. p. 372-469. nitrate reductase activity. Mar Biol 1992;112:691-696.
13. Round FE, Crawford RM, Mann DG. The diatoms: biology and mor- 23. Campbell WH, Kinghorn JR. Functional domains of assimilatory ni-
phology of the genera. Cambridge: University Press; 1990. trate reductases and nitrite reductases. Trends Biochem Sci 1990;15:
14. Guillard RRL, Ryther JH. Studies of marine planktonic diatoms. I. 315-319.
Cyclotella nana Hustedt and Detonula confervacea Cleve. Can J 24. Gunderson JH, Elwood H, Ingold A, Kindle K, Sogin ML. Phylogenetic
Microbiol 1962;8:229-239. relationships between chlorophytes, chrysophytes, and oomycetes.
15. Guillard RRL. Culture of phytoplankton for feeding marine inverte- Proc Natl Acad Sci USA 1987;84:5823-5827.
brates. In: Smith WL, Chanley MH, eds. Culture of marine invertebrate 25. Fritz L, Stringher CG, Colepicolo P. Immunolocalization of nitrate
animals. New York: Plenum Press; 1975. p. 26-60. reductase in the marine dinoflagellate Gonyaulax polyedra (Pyrro-
16. Timmons TM, Dunbar BS. Protein blotting and immunodetection. In: phyta). J Phycol 1996;32:632-637.
Deutscher MP, ed. Guide to protein purification. San Diego: Aca- 26. Smith GJ, Gao Y, Alberte RS. Immunological probes of new produc-
demic Press; 1990. p. 679-688. tion: universal and taxon-specific markers for nitrate reductase in
17. Smith GJ, Zimmerman RC, Alberte RS. Molecular and physiological marine algae. EOS 1996;76:62.