sign in sign up
<<
Home , Cyanidioschyzon, Diatom, Fragilariophyceae

J. Phycol. 41, 95–104 (2005) r 2005 Phycological Society of America DOI: 10.1111/j.1529-8817.2005.04090.x

CHARACTERIZATION OF (BACILLARIOPHYCEAE) GENES AND THEIR DETECTION IN MARINE COMMUNITIES1

Andrew E. Allen,2 Bess B. Ward Department of Geosciences, Princeton University, Princeton, New Jersey 08544, USA and Bongkeun Song Department of Biological Sciences, University of North Carolina at Wilmington, Wilmington, North Carolina 28401-1003, USA

The complete assimilatory nitrate reductase (NR) New production, in the that gene from the pennate diatom Phaeodactylum trico- results from inputs of from outside the nutum Bohlin was sequenced from cDNA and com- euphotic zone, determines the amount of carbon avail- pared with NR sequences from fungi, green , able for utilization by higher trophic levels. It is largely vascular , and the recently sequenced regulated by the rate and magnitude of nitrate (NO3 ) of the centric diatom pseudonana Has- uptake and assimilation by pelagic microbes (Dugdale le and Heimdal CCMP1335. In all the major and Goering 1967). Within marine phytoplankton, di- eukaryotic nitrate reductase (Euk-NR) functional atoms are among the best competitors for high levels of domains, diatom NR gene sequences are generally NO3 (42 mM) (Wilkerson et al. 2000). Therefore, in 50%–60% identical to and alga sequences at regions where high rates of nutrient supply are sus- the amino acid level. In the less conserved N-ter- tained, such as environments, and on con- minal, hinge 1, and hinge 2 regions, homology to tinental margins, often constitute a large other NR sequences is weak, generallyo30%. Two fraction of the photosynthetic (Kudela and PCR primer sets capable of amplifying Euk-NR Dugdale 2000). The ability of diatoms to thrive under from plants, algae, and diatoms were designed. upwelling-induced nutrient-rich conditions makes One primer set was used to amplify a 750-base them the basis for the world’s shortest and most effi- pair (bp) NR fragment from the cDNA of five addi- cient food webs (Steele 1974). Some of the worlds larg- tional diatom strains. The PCR amplicon spans part est fisheries are driven and maintained primarily by of the well-conserved dimer interface region, the diatom based new production (Jickells 1998). Diatom more variable hinge 1 region, and part of the con- produces approximately 25% of the served cytochrome b heme binding region. The sec- 45–50 billion tons of organic carbon fixed annually in ond primer set, targeted to the dimer region, was the sea (Nelson et al. 1995). used to amplify an approximately 400-bp fragment Nitrogen uptake and assimilation differ in that up- of the NR gene from DNA samples collected in take is defined as NO3 transport across the membrane Monterey Bay, California and in central New Jersey and NO3 assimilation is the reduction and subsequent inner (LEO-15 site) . Only incorporation of reduced N (nitrogen) into cellular diatom-like NR sequences were recovered from biomass. Diatoms, like other photosynthetic eukar- Monterey Bay samples, whereas LEO-15 samples yotes, reduce NO3 to NO2 using an assimilatory ni- yielded NR sequences from a range of photosyn- trate reductase (NR) (E.C.1.6.6.1-3) before thetic . The prospect of using DNA- and incorporating N into biomass. Nitrate reduction ap- RNA-based methods to target the NR genes of dia- pears to be the rate-limiting process in nitrogen acqui- toms specifically is a promising approach for future sition (Campbell 1999). Regulation of the eukaryotic physiological and ecological experiments. NR by different environmental conditions has Key index words: diatoms; DNA sequence; nitrate; been studied in vascular plants, , and dia- nitrate reductase; nitrogen; upwelling toms. For example, in the marine green alga Dunaliela tertiolecta, NR mRNA transcripts have been shown to be Abbreviations: Euk-NR, eukaryotic nitrate red- nitrate inducible and ammonium repressible, and uctase; NR, nitrate reductase; RACE, rapid ampli- transcription is not induced under nitrogen depletion fication of cDNA ends (Song and Ward 2004). In the freshwater green algae Chlorella vulgaris, Chlamydomonas reinhardtii, and Volvox carteri, NR transcription was induced, more rapidly than in D. tertiolecta, by the removal of repressors such as ammonia and metabolites of ammonia (Cannons 1Received 4 June 2004. Accepted 26 October 2004. and Pendleton 1994, Quesada and Ferna´ndez 1994). 2Author for correspondence: e-mail [email protected]. In diatoms, NR enzyme assays indicate that NR activity

95 96 ANDREW E. ALLEN ET AL. is nitrate inducible and ammonium repressible as well soli-Guillard National Center for Culture of Marine as activated and temperature dependent (Lomas Phytoplankton (CCMP, West Boothbay Harbor, ME, USA) and Glibert 1999, 2000, Gao et al. 2000). Nitrate red- and the Plymouth Culture Collection of Algae (PCC, Ply- uctase activity in diatoms is quantitatively related to mouth, UK). The following strains were used: Phaeodactylum triconutum Bohlin (Bacillariophyceae, clone CCMP 630), NO3 incorporation under steady-state growth condi- Chaetocerous mulleri Lemmermann (Coscinidiscophyceae, tions (Berges and Harrison 1995). clone CCMP 1316), Asterionellopsis glacialis (Castracane) Information currently available concerning the reg- Round (, clone CCMP 139), Thalassiosira ulation of NR in diatoms was derived from in vitro ex- weissflogii (Grun.) Fryxell et Hasle (Coscinidiscophyceae, periments using protein extracts or antibody probing clone CCMP 1336), Coscinodiscus granii Gough (Cos- for protein abundance (Smith et al. 1992, Berges and cinidiscophyceae, clone CCMP 1817), Entomoneis cf_alata Eh- renberg (Bacillariophyceae, clone CCMP 1522), Amorpha sp. Harrison 1995, Vergara et al. 1998, Lomas 2004). Al- (Bacillariophyceae, clone CCMP 1405), Thalassiosira oceanica though these studies have undoubtedly increased our (Hustedt) Hasle et Heimdal (Bacillariophyceae, clone CCMP understanding of factors that regulate N uptake and 1005), and costatum (Bacillariophyceae, clone in diatoms, there are limitations to these PCC 582). methods. Protein based enzyme activity assays depend Batch cultures (1 L) were grown under continuous fluores- cent light at 181 C in 0.2 mm filtered and autoclaved New Jersey on the extraction of intact protein in the presence of coastal amended with f/2 nutrients with NO3 as the saturating substrates and cofactors, and it is difficult to sole nitrogen source (Guillard 1975). Cells were harvested in relate in vivo measurements of enzyme activity quanti- late log phase by filtration onto a 47-mm diameter 3.0-mm pore tatively to in situ processes (Berges and Harrison 1995). size Durapore filter (Millipore, Bedford, MA, USA) and trans- Likewise, 15N tracer techniques, which historically ferred by scraping with a sterile razor blade into a sterile have been used to estimate rates of NO3 incorporation 2.0-mL microcentrifuge tube. and to investigate the effect of environmental factors on RNA isolation: Total RNA was extracted from cultures us- assimilation and uptake rates, have well-documented lim- ing an Ambion RNAqueous-4PCR kit (Ambion, Austin, TX, USA) following the manufacturer’s instructions. The extract- itations. Additions of labeled nitrogen may alter the dy- ed RNA was treated with DNAse (2 U) (1 h, 371 C) to elim- namics of nutrient-limited systems (Dugdale and inate genomic DNA contamination. The RNA integrity was Wilkerson 1986). The 15N based methods can also suffer examined by visualizing the RNA on a denaturing agarose from containment artifacts such as enhanced and gel. Quantity and purity of the RNA was evaluated by meas- altered light (Collos et al. 1993). Finally, and most impor- uring the A260/A280 ratio with a spectrophotometer. Ex- 1 tantly, 15N techniques and enzyme based assays do not tracted total RNA was stored at 80 until used for experiments. provide a high degree of phylogenetic resolution when Degenerate rapid amplification of cDNA ends (RACE) PCR applied to samples. It is very difficult to estimate primer design and overall strategy for NR genes in cultured al- nitrogen incorporation or protein activity for particular gae: Degenerate oligonucleotide primers for the assimilatory taxa, , or even broadly defined functional groups NR gene were designed based on a DNA alignment of three (i.e. photosynthetic organisms vs. heterotrophic ). plant and three algal genes that encode the NR protein. In- One means to overcome some of these problems in- itially, three primers, NRPt547F, NRPt1000F, and NRPt 1389R, were designed to be used in nested RACE PCR re- volves conducting assays to collect information at the actions (Table 1, Fig. 1). Briefly, the forward primer NRPt level of DNA and RNA. To address ecological questions 547F and the 30 RACE universal primer mix generated a concerning variability in the diversity, abundance, and PCR product that was used as the template in a nested PCR expression of eukaryotic NR genes, it is first necessary to reaction with the primers NRPt1000F and NRPt1389R. This construct a sequence database so that appropriate reaction produced robust 390-bp PCR products with each of probes can be designed to detect groups and clades of the diatom samples. The PCR products were cloned and se- interest. We present here the first published, complete quenced, and gene specific PCR primers were designed to amplify the complete NR coding sequence with 30 and 50 and partial, NR sequences for diatoms. These data will RACE PCR. The 30 and 50 RACE PCR products from P. tri- enable the investigation of transcriptional variability cornutum were cloned and sequenced because this culture within different groups of phytoplankton in response amplified most robustly. þ 0 0 to availability of NO3 ,NH4 , and dissolved organic ni- Complete 3 and 5 RACE PCR products were obtained by trogen and will pave the way for in situ ecological inves- using the NRPt30 RACE and NRPt50 RACE primers, which tigations designed to evaluate the response of marine were designed specifically for the P.tricornutum NR gene (Table 1). Primers were designed from sequence information ob- diatom populations to differing oceanic N regimes. A tained by cloning and the 390-bp PCR product preliminary analysis of eukaryotic nitrate reductase obtained with degenerate primers (above). To obtain the 50 (Euk-NR) genes retrieved from clone libraries indicates region of the P.tricornutum NR gene, the reverse transcription a of diatoms in Monterey Bay samples and a reaction to synthesize 50 RACE cDNA was modified. Total RNA higher proportion of diatoms relative to other types of was used as a template with a SMARToligo primer (BD Science phytoplankton at 1 m compared with 15 m in inner con- Clontech, Palo Alto, CA, USA) and an NR gene specific reverse primer (NRPt50 RACE), which replaced the oligo(dT)-primer tinental shelf waters off the central coast of New Jersey. in the manufacturer’s instruction. The 50 RACE PCR amplifi- cation was performed with the P. tricornutum NR gene specific reverse primer (NRPt50 RACE) and the 50 RACE universal MATERIALS AND METHODS primer mix (Table 2). NR from diatom cultures. Strains and growth conditions: Dia- RACE PCR amplification: Total RNA was used as a tem- tom strains used in this study were obtained from the Prova- plate to synthesize oligo(dT)-primed cDNA using a RACE DIATOM NITRATE REDUCTASE GENES 97

TABLE 1. Oligonucleotide primers used in this study.

Primer Sequence (50 to 30) Amino acid sequence NRPt547F CGYAARGARAAYATG RKEQNM NRPt907F GGYGGNMGNATGRTBAAGTGGCT GGRMIKWL NTPt1000F GRDGGHTGGTGGTACAAGCC GGWWYKP NRPt1389R GTTGTTYMYCATBCCCAT MGMGNN NRPt1727R ATVGAVTCDGCKCCACCNGGRTG HPGGIDSI NRPt2325R GGGVGTGATRCCHGTVCCDCCVGC AGGTGITTP NRPt3’RACE AAGGAGCAGCCAACGGACTACGGAATGT NRPt5’RACE ACATTCCGTAGTCCGTTGGCTGCTCCTT

The primers are labeled according to position numbers that refer to the position in the P.tricornutum NR DNA sequence. kit (BD Science Clontech) following the manufacturer’s in the first round of PCR and NRPt1000F and NRPt1727R instructions. The 30 RACE PCR was performed with the for- primers in the second round. A total of 2.0 mL of the RT re- ward NR primer NRPt547F (1.0 mM) and the 30 RACE uni- action was used as a template in first round PCR reactions and versal primer mix (0.2 mM) in a total volume of 25 mL 1.5 mL first round PCR product was used as the template in containing 2.5 mLof10 Advantage II PCR bufferTM (BD second round reactions (Table 2). The second round PCR re- Biosciences/Clonetech, Palo Alto, CA, USA), 0.2 mM of each action with primers NRPt1000F and NRPt1727R produced an deoxyribonucleoside triphosphate, 1 U Advantage Taq po- approximately 730-bp product on each of the diatom tem- lymerase, and 1.5 mLof30 RACE cDNA. The second round plates. The PCR products from S. coastatum, A. glacialis, T. oce- of nested PCR was performed with the forward and reverse anica, Amorpha sp., and C. mulleri were selected for cloning and primers NRPt1000F and NRPt1389R (1.0 mM each), and the sequencing because they generated the strongest PCR prod- first round PCR product (1.5 mL)wasusedasthetemplatein ucts. Cloned PCR products were sequenced M13F and M13R 25-mL second round reactions. All thermal cycling conditions primers (see below). and PCR reactions are given in Table 2. The PCR products NR sequences from seawater samples. Collection of environmen- were purified, cloned, and sequenced (see below). Clones tal samples: Water samples from Monterey Bay were collected containing 30 and 50 RACE PCR products were sequenced from 10 m at a mid-bay station (361 450 N, 1221 010 W) in with several internal primers designed by gene walking. April and October 1998. Seawater (4 L) was concentrated Degenerate PCR primer design and reverse transcriptase (RT)- using tangential flow filter cassettes (Filtron Ultrasette 300 PCR amplification: The complete genome sequence for the KD open channel, Filtron Technology Corp., Northborough, centric diatom CCMP1335 (http:// MA, USA). The retentate was collected on 47-mm diameter genome.jgi-psf.org/thaps1/) became available after we had 0.2 mm pore size Supor filters (Gelman, Covina, CA, USA) completed sequencing the P.tricornutum NR gene. The DNA and frozen immediately. The filters were archived at 801 C sequence alignments were therefore constructed with the until DNA was extracted. complete P. tricornutum and T. p s e u d o n a n a NR sequences, Other seawater samples were collected from 1 and 15 m at complete mRNA sequences from vascular plant and alga rep- the Node A site of the Rutgers University Marine Field Station resentatives, and the 390-bp fragment obtained with the de- Long-Term Ecosystem Observatory (LEO-15) (http://ma- generate primers NRPt1000F and NRPt1389R from eight rine.rutgers.edu/mrs/LEO/LEO15.html), which is located in other diatoms (as described above). inner continental shelf waters off the central coast of New Jer- Degenerate primers capable of amplifying NR gene frag- sey (391 270 N, 741 150 W). Approximately 1 L of water from ments from vascular plants, green algae, and diatoms were se- each depth was collected on a 0.2-mm sterivex filter capsule lected. The NRPt1000F and NRPt1389R primers did not (Millipore) using a peristaltic pump. Samples were collected at require modification, and three additional primers were de- 12:00 on 22 July 2002, flash frozen in liquid nitrogen, trans- signed: RPt907F, NRPt1727R, and NRPt2325R (Table 1, Fig. ported on dry ice, and stored until analysis in a 801 C freezer 1). First round RT cDNA synthesis reactions were primed with at the lab. the reverse primer NRPt2325R. Reactions consisted of 2.0 mL DNA isolation: DNA was purified from the sterivex filters of 10 RT buffer, 2.0 mL of dNTP mix (5mM each), 1.0 mLof using a DNA extraction kit (Gentra, Minneapolis, MN, USA) Omniscript Reverse Transcriptase (Qiagen, Inc., Valencia, CA, following the directions of the manufacturer, except that the USA), 1 mL Rnasin (10 U/mLin1 RT buffer) (Promega, volume of the lysis buffer and of all other reagents was dou- Madison, WI, USA), approximately 100 ng of total RNA, bled. DNA samples purified from seawater were 1.0 mM of the reverse primer NRPt2325R, and RNase free wa- used as the template for nested PCR reactions with ter to 20 mL. Reverse transcriptase reactions were followed by NRPt907F and NRPt2325R primers in the first round and nested PCR reactions with NRPt907F and NRPt2325R primers NRPt1000F and NRPt1389R primers in the second round.

FIG. 1. Deduced model of Phaeodactylum tricornutum CCMP 630 nitrate reductase based on sequence homology to Arabidopsis thaliana NIA2 (GenBank accession no. J03240) (Campbell 1999). All primer locations are shown. The numbers, based on the P. tricornutum sequence, signify the boundaries of the five structural domains and the hinge regions. identities are explained in the text. 98 ANDREW E. ALLEN ET AL.

TABLE 2. PCR reactions and thermal cycling conditions

PCR reaction F primer R primer Template bpa Thermal cycling conditions 30 RACE (Rnd1) NRPt547F 30 Univ. mix Oligo (dT)-primed cDNA NAb 30 (941 C, 30 s; 621 C, 30 s; 721 C, 3 min) 30 RACE (Rnd2) NRPt1000F NRPt1389R Partial 30 RACE PCR 389 30 (941 C, 30 s; 551 C, 30 s; 721 C, 30 s) Complete 30 NRPt30 RACE 30 Univ. mix Oligo (dT)-primed cDNA 1600 30 (941 C, 30 s; 681 C, 30 s; 721 C, 3 min) Complete 50 NRPt50 RACE 50 Univ. mix Oligo (dT)-primed cDNA 1300 30 (941 C, 30 s; 681 C, 30 s; 721 C, 3 min) NR RT-PCR (Rnd1) NRPt907F NRPt2325R NRPt2325R-primed cDNA 1419 30 (941 C, 30 s; 581 C, 30 s; 721 C, 2 min) NR RT-PCR (Rnd2) NRPt1000F NRPt1727R NR RT-PCR (Rnd1 730 30 (941 C, 30 s; 551 C, 30 s; 721 C, 1 min) NR nested (Rnd1) NRPt907F NRPt2325R Genomic DNA NA 30 (941 C, 30 s; 581 C, 30 s; 721 C, 3 min) NR nested (Rnd2) NRPt1000F NRPt1389R NR nested (Rnd1) 389 30 (941 C, 30 s; 551 C, 30 s; 721 C, 30 s)

abp, refers to the length of the PCR amplicon. bNA, not applicable. Indicates that these first round nested PCR reactions were not visualized by gel electrophoresis. All PCR reactions included a hot start at 941 C for 5 min and a final extension step at 721 C for 7 min.

Approximately 10 ng of DNA was added to first round (Campbell 1999) (Fig. 1). Three sequence regions reactions, and 1.5 mL of first round reaction was used as the without similarity to other and varying in template in second round reactions (Table 2). sequence among different NRs are the N-terminal Cloning and sequencing PCR products: The amplified prod- region, the hinge 1 region, and the hinge 2 region. ucts were examined on 1.0% agarose gels by electrophoresis and then were purified using the Qiaquick gel extraction kit Phylogenetically, diatoms cluster together with high (Qiagen), cloned, and sequenced (Allen et al. 2001). The bootstrap support, separate from green and red al- cleaned PCR products were used for direct cloning with the gae, vascular plants, fungi, and the TOPO-TA cloning system (Invitrogen, Carlsbad, CA, USA) Phytophthora (Fig. 2). following the manufacturer’s instructions (as described by By aligning the complete diatom NR proteins with Allen et al. 2001). The ligated were transformed in those from plants, we were able to estimate domain high transforming efficiency Escherichia coli TOP10F’ (In- vitrogen) following the manufacturer’s instructions. The delineations for the diatom NR subunits (Figs. 1 and transformed cells were plated on Luria-Bertrani agar plates 3). Based on the sequence homology and organization containing 50 mg mL 1 kanamycin. Four clone libraries of the major functional domains, diatom NR genes ap- were constructed with the nested pair of primers NRPt907F pear to be structurally very similar to plant and algal and NRPt2325R and NRPt1000F and NRPt1389R: one each NR genes. Twenty-one invariant residues have been from the Monterey Bay April and October samples and the 1- identified in plant Euk-NR genes that are important m and 15-m LEO-15 samples. Ten clones each from the LEO- 15 1 m and Monterey Bay April libraries and five clones each for functions such as binding of nitrate, heme, and from the LEO-15 15 m and Monterey Bay October libraries FAD. Diatom sequences are conserved in all these po- were screened for inserts. The inserts were sequenced using sitions except in the case of a serine residue, in the Big-Dye terminator chemistry (Applied Biosystems, Foster hinge 1 region that is thought to be important for light City, CA, USA) and an ABI model 310 automated DNA se- dependent regulatory phosphorylation (Tischner quencer (Applied Biosystems). Cloned PCR products from 2000). One unique feature that is well conserved be- sweater samples were sequenced with M13F and M13R primers. tween the P.tricornutum and T. pseudonana NR gene se- Phylogenetic analysis of the NR gene: The amino acid se- quences of complete and partial NR genes from phylogenet- quences is an insertion that is identical at 10 of 11 ically representative major taxa in GenBank were aligned amino acid sites in the Mo- molybdopterin domain be- with the diatom sequences generated in this study. The NR tween the nitrate and molybdopterin binding residues gene sequences were aligned using the ClustalX multiple se- (in the 275–300 residue region, Fig. 3). quence alignment algorithm. Phylogenetic trees were in- The N-terminal region of diatom NR genes, which ferred and drawn using the TREECON software package has received attention recently because of its potential (version 1.3b) (Van de Peer and De Wachter 1997) and the Kimura two-parameter model for inferring evolutionary dis- role in temperature sensitivity (Berges et al. 2002), has tances. Bootstrap estimates (1000 replicates) of confidence a length of 82 and 89 amino acids for P.tricornutum and intervals were also made using the algorithms available in the T. pseudonana, respectively. This is more similar in TREECON package. length to the approximately 90 amino acid residue N-terminal region in vascular plants than to the green algal N-terminal regions of approximately 50 amino RESULTS AND DISCUSSION acids. Aspartic and glutamic acid-rich regions, which Full-length diatom NR sequences. The full-length di- are important for light regulation via phosphorylation atom NR amino acid sequences from P. tricornutum in plants (Pigaglio et al. 1999), are missing in the CCMP 630 and T. p s e u d o n a n a CCMP1335 3 H are diatom N-terminal regions. Overall, the diatom NR 65% identical to one another and are each, on aver- N-terminal regions have very little homology to one age, 40% identical to NR sequences from green algae another or to those of other NR genes, although there and vascular plants. In plants, the NR monomer is are two short segments of three to five residues that are composed of five structural domains, the boundaries identical between the diatoms and one segment that of which have been defined: Mo-molybdopterin, shares similarity with the green alga N-terminal re- dimer interface, cytochrome b,FAD,andNADH gions (Fig. 3). The diatom N-terminal regions contain DIATOM NITRATE REDUCTASE GENES 99

FIG. 2. Inferred phylogenetic relationships of complete eukaryotic assimilatory nitrate reductase amino acid sequences. Bootstrap values greater than 50 (of 100) are shown at the nodes. The sulfite oxidase gene, a putative evolutionary ancestor of eukaryotic as- similatory nitrate reductase from Arabidopsis thaliana, was used to root the tree. GenBank accession numbers are given next to their designated sequence. Sequences contain approximately 900 amino acids. a conserved cysteine residue that is absent in the other from plants, green, and . The aligned re- lineages. The function of the N-terminal-region re- gion is part of the well-conserved dimer interface re- mains to be discovered. Berges et al. (2002) and Gao gion, the more variable hinge 1 region, and part of et al. (2000) demonstrated the relatively low tempera- the conserved cytochrome b heme binding region ture optima of T. pseudonana and other diatoms and (Fig. 3). As a group, the 7 diatom amino acid NR se- suggested that diatoms and other chromophytic algal quences are an average of 77% and 75% identical to NR have a common feature that enables high one another and 53% and 50% identical to plant and activity at relatively low temperatures. Interestingly, as alga NR sequences in the dimer interface and cyto- noted by Berges et al. (2002), a deletion of the NR N- chrome b heme binding region of the NRPt1000F terminal region in the plant Nicotina caused a drop in and NRPt1727R amplicon, respectively. In the less the thermal optimum of the NR enzyme from 301 Cto conserved hinge 1 region, however, diatom sequenc- 151 C (Nussaume et al. 1995). es are an average 58% identical to one another and Partial diatom NR sequences from conserved and vari- only 25% identical to plant and algal sequences. able domains. The approximately 730-bp NR gene The approximately 390-bp gene fragments that fragments that were amplified from additional dia- were amplified with the NRPt1000F and NRPt1389R tom strains with the primers NRPt1000F and primers from additional diatom strains and DNA ex- NRPt1727R were aligned with full-length genes tracted from filtered seawater were translated into 100 ANDREW E. ALLEN ET AL.

Fig.3. continued on next page DIATOM NITRATE REDUCTASE GENES 101

FIG. 3. ClustalX alignment of the complete nitrate reductase protein from the diatoms Thalassiosira pseudonana and Phaeodactylum tricornutum, the green algae Chlamydomonas reinhardii and Dunaliella teriolecta, the red alga merola, and the vascular plant Arabidopsis thaliana. Additional diatom gene fragments amplified with the primers NRPt1000F and NRPt1727R are included. Numbering of the alignment is based on the A. thaliana NIA2 sequence (GenBank accession no. J03240). Asterisks denote amino acids that have been identified as key invariant residues in plant nitrate reductase sequenced (Campbell 1999). Closed circles denote residues that are nearly specific to diatoms within the alignment. amino acids and aligned. Phylogenetic analysis of the NR sequences. Seven of the 25 NR sequences fall alignment, which includes 10 partial NR sequences outside of the diatom clade and are associated with from cultured diatoms, indicated that the diatom NR eukaryotic organisms without particular affinity to sequences again form a distinct clade apart from algal any sequenced representatives. and plant sequences, with significant bootstrap support The four LEO-15 15-m sequences that form a tight (Fig. 4). Therefore, although the NRPt1000F and group with bootstrap support of 100 are approximate- NRPt1389R primers target a short fragment of the ly 95% identical to one another and only 45% identical highly conserved dimer interface region where se- to diatom and other known NR sequences. Three oth- quences from diverse organisms share relatively high er sequences retrieved from the LEO-15 samples fall levels of homology (generally 450%), diatoms retain outside the diatom cluster; the Leo212(15m)2 and their group properties with even higher levels of iden- Leo212(1m)1 sequences form a distinct clade that is tity (70%–80%). 5% more similar to vascular plants than to any other Detection of NR sequences from diatoms and other group, and the Leo212(1m)2 sequence is deeply eukaryotes in seawater. Eightof10clonesscreened branching and does not cluster with any of the exist- from the LEO-15 1 m and Monterey Bay April librar- ing NR sequences. Each of these unknown marine en- ies each contained Euk-NR gene fragments. Most of vironmental representatives share conserved amino the clones screened from the LEO-15 15 m (5 of 5) acid motifs with the known sequences, but each of and Monterey Bay October (4 of 5) libraries con- them also contains unique and, in the case of the well- tained Euk-NR gene fragments. The cloned frag- supported cluster of four LEO-15 sequences, well-con- ments that were not Euk-NR gene fragments were served motifs. The challenge still remains to uncover not similar to one another and are probably the result the biochemical and phenotypic significance of such of PCR artifacts associated with highly degenerate unique motifs in otherwise very conserved regions of primers. Many (18 of 25) of the Euk-NR sequences the NR gene. retrieved from Monterey Bay and LEO-15 cluster Phylogenetic comparisons and ecological applica- with a high level of confidence with known diatom tions. In the past decade, research on Euk-NR has 102 ANDREW E. ALLEN ET AL.

FIG. 4. Inferred phylogenetic relationships of the approximately 130 amino acid eukaryotic assimilatory nitrate reductase gene fragments amplified with the primers NRPt1000F and NRPt1389R. Bootstrap values greater than 500 (of 1000) are shown at the nodes. The tree is rooted through the homologous region of the Arabidopsis thaliana sulfite oxidase gene. received substantial attention from plant scientists, N incorporation (Berges and Harrison 1995, Vergara biochemists, and ecologists working across many taxa et al. 1998, Lomas and Glibert 1999, Lomas 2004). (Berges 1997, Campbell 2001). Exciting new findings Vergara et al. (1998) developed a model to account and techniques include the observation in plants that for their measurements of NR protein activity and NRtranscriptioniscontrolledinpartbyphotosyn- abundance and concluded that diel variations in NR thetic electron flow and mechanistically linked to activity are controlled primarily by transcriptional CO2 reduction (Sherameti et al. 2002) and the de- regulation. velopment of an immunolabeling cytometric meth- Despite these significant increases in understanding od to quantify NR protein abundance in diatoms NR activity dynamics, basic studies at the transcrip- (Jochem et al. 2000). Also, in diatoms, fundamental tional level for biogeochemically important marine experiments on the effect of different environmental phytoplankton such as diatoms have not been possible regulators such as nitrogen source, light, and tem- because of the absence of sequence data. Although the perature on protein activity and abundance have T. pseudonana genome has been completely sequenced, shown that NR activity is highly influenced by it is important to examine a range of sequences from N source, has a diel cycle, is correlated with internal many diatoms to conduct ecologically meaningful carbon pools, and is likely the rate-limiting step of experiments. New techniques based on microarray DIATOM NITRATE REDUCTASE GENES 103 technology and real-time PCR rely on sequence data community composition at the NR gene level. Assays from a diversity of environmentally important taxa. designed to detect Euk-NR mRNA in various phyto- Eukaryotic NR is present in vascular plants; red, plankton lineages, therefore, will be useful for sam- green, and ; all the lineages of eukaryotic pling the nutrient status and activity of specific groups phytoplankton; and in some heterotrophic organisms within phytoplankton communities. Delineating par- such fungi and nonphotosynthetic . ticular conserved and variable regions of the Euk-NR Eukaryotic NR does not have a prokaryotic homolog, gene in important marine phytoplankton such as dia- and have a different form of NR. toms is the first step toward introducing such powerful Eukaryotic assimilatory NR therefore is thought to methods to aquatic microbial ecology. have originated from a single evolutionary event ear- The alignment in Figure 3 illustrates the different ly in the history of eukaryotes, before the symbiotic levels of conservation that occur between regions of the origin of (Stolz and Basu 2002). Diatoms NR gene that are thought to have evolved independ- were derived by secondary endosymbiosis whereby a ently and to be under differing levels of selection nonphotosynthetic acquired a (Zhou and Kleinhofs 1996, Campbell 1999). In the by engulfing a photosynthetic eukaryote, probably a poorly conserved hinge 1 region, which does not con- red algal (Armbrust et al. 2004). Phylo- tain detectable similarity between diatoms and other genetically, diatom NR does not have any particular taxa, a diatom signature is retained. This region could affinity to the red algae Cyanoidioschyzon merolae, rela- be useful for designing diatom-specific PCR primers or tive to green algae and vascular plants (Figs. 2–4). It probes targeted to particular groups within the dia- appears therefore that in secondary endosymbiotic lin- toms. eages, such as the diatom, NR is also derived from the Based on data collected in this study, it is clear that heterotrophic host. Eukaryotic NR sequences tend to diatom NR genes can be targeted as a coherent group be relatively well conserved and consistent with rRNA- at the DNA and RNA level. It should be noted that the defined taxonomic lines. As a group, for example, dia- primers designed in this study for the amplification of toms form a deep coherent branch distinct from algae diatom NR genes do not amplify heterotrophic bacte- and plants (Fig. 2). At this point it is unclear whether or rial or eukaryotic dissimilatory NR genes. It is recog- not this clade includes chromophytes other than diatoms. nized that some groups of heterotrophic marine The four nested PCR primers that generated these bacteria do assimilate nitrate aerobically using an NR 389-bp PCR products were intentionally designed to enzyme with molybdopterin and heme cofactors (Allen amplify a diverse range of phytoplankton Euk-NR et al. 2001), but such bacterial NR genes (nasA) are not gene fragments. At the time of primer design, the homologous to eukaryotic NR genes. Eukaryotic dis- only photosynthetic taxa for which Euk-NR sequences similatory genes (Zvyagil’skaya et al. 1996) are also not were available were green algae, plants, and diatoms. homologous to eukaryotic assimilatory NR genes. This Therefore, it is likely that our PCR primers are not is fortunate because it facilitates the investigation of the capable of amplifying NR gene fragments from some response of these ecologically distinct groups at the groups of eukaryotic phytoplankton. Nevertheless, transcriptional level to environmental variables and phylogenetic analysis of the sequences obtained from will provide insight into community successional pat- the seawater samples indicates that we retrieved some terns in changing nutrient regimes, such as upwelling sequences of unknown identity that are in fact NR and events. An important priority for the future is devel- most likely do not belong to any of the major taxa used oping NR sequence databases that include other for primer selection (Fig. 4). Degenerate-nested PCR groups of ecologically important marine phytoplank- facilitated the amplification of sequences from highly ton, such as dinoflagellates, haptophytes, and pi- divergent taxa, which would not have been possible coeukaryotes. with other methods. Although we did not sequence enough clones from We thank Alan Milligan for assistance in maintaining the any one library to draw any robust conclusions regard- diatom cultures and Lee Kerkhoff for help in collecting the LEO-15 samples. Comments by John Berges helped to ing the distribution of various type of NR genes, it is improve the quality of the manuscript. This research was interesting to note that all the Monterey Bay sequences supported by grants to B. B. W. from the U.S. Department retrieved belong to diatoms. This is consistent with the of Energy and the National Science Foundation. ecology of Monterey Bay, where diatoms dominate production and NO3 assimilation (Kudela and Dug- dale 2000). At the LEO-15 site, six of eight sequences Allen, A. E., Booth, M. G., Frischer, M. E., Verity, P.G., Zehr, J. P.& retrieved from 1 m belonged to the diatom group, Zani, S. 2001. Diversity and detection of nitrate assimilation genes in marine bacteria. Appl. Environ. Microbiol. 67:5343–8. whereas all five of the sequences retrieved from 15 m Armbrust, E. V., Berges, J. A., Bowler, C., Green, B. R., Martinez, grouped outside of the diatoms, suggesting the pres- D., Putnam, N. H., Zhou, S. G., Allen, A. E., Apt, K. E., ence of distinct communities at different depths. This Bechner, M., Brzezinski, M. A., Chaal, B. K., Chiovitti, A., pattern probably reflects differences in diversity and Davis, A. K., Demarest, M. S., Detter, J. C., Glavina, T., Goodstein, D., Hadi, M. Z., Hellsten, U., Hildebrand, M., community composition between the depths sampled Jenkins, B. D., Jurka, J., Kapitonov, V. V., Kroger, N., Lau, W. at the LEO site and Monterey Bay site, which could be W. Y., Lane, T. W., Larimer, F. W., Lippmeier, J. C., Lucas, described more quantitatively with a larger study of S., Medina, M., Montsant, A., Obornik, M., Parker, M. S., 104 ANDREW E. ALLEN ET AL.

þ Palenik, B., Pazour, G. J., Richardson, P. M., Rynearson, T. A., Lomas, M. W. & Glibert, P. M. 1999. Interactions between NH4 Saito, M. A., Schwartz, D. C., Thamatrakoln, K., Valentin, K., and NO3 uptake and assimilation: comparison of diatoms and Vardi, A., Wilkerson, F. P. & Rokhsar, D. S. 2004. The genome dinoflagellates at several growth temperatures. Mar. Biol. of the diatom Thalassiosira pseudonana: Ecology, , and 133:541–51. metabolism. Science 306:79–86. Lomas, M. W. & Glibert, P. M. 2000. Comparisons of nitrate up- Berges, J., Varela, D. & Harrison, P. 2002. Effects of temperature take, storage, and reduction in marine diatoms and flagellates. on growth rate, cell composition and nitrogen metabolism in J. Phycol. 36:903–13. the marine diatom Thalassiosira pseudonana (Bacillar- Nelson, D. M., Treguer, P., Brezezinski, M. A., Leynaert, A. & iophyceae). Mar. Ecol. Prog. Ser. 225:139–46. Queguiner, B. 1995. Production and dissolution of biogenic Berges, J. A. 1997. Minireview: algal nitrate reductases. Eur. J. silica in the —revised global estimates, comparison with Phycol. 32:3–8. regional data and relationship to biogenic sedimentation. Berges, J. A. & Harrison, P. J. 1995. Nitrate reductase activity Global Biogeochem. Cycles 9:359–72. quantitatively predicts the rate if nitrate incorporation under Nussaume, L., Vincentz, M., Meyer, C., Boutin, J. & Caboche, M. steady state light limitation: a revised assay and characteriza- 1995. Post-translational regulation of nitrate reductase by light tion of the enzyme in three species of marine phytoplankton. is abolished by an N-terminal deletion. Plant Cell 7:611–21. Limnol. Oceanogr. 40:82–93. Pigaglio, E., Durand, N. & Meyer, C. 1999. A conserved acidic motif Cannons, A. C. & Pendleton, L. C. 1994. Possible role for mRNA in the N-terminal domain of the nitrate reductase is necessary stability in the ammonium-controlled regulation of nitrate for the inactivation of the enzyme in the dark by phosphor- reductase expression. Biochem. J. 297:561–5. ylation and 14-3-3 binding. Plant Physiol. 119:219–29. Campbell, W. H. 1999. Nitrate reductase structure, function and Quesada, A. & Ferna´ndez, E. 1994. Expression of nitrate assimila- regulation: bridging the gap between and phys- tion related genes in Chlamydomonas reinhardtii. Plant Mol. Biol. iology. Annu. Rev. Plant Physiol. Plant Mol. 50:277–303. 24:185–94. Campbell, W. H. 2001. Structure and function of eukaryo- Sherameti, I., Soporyl, S. K., Trebicka, A., Pfannschmidt, T. & tic NAD(P)H:nitrate reductase. Cell. Mol. Life Sci. 58: Oelmuller, R. 2002. Photosynthetic electron transport deter- 194–204. mines nitrate reductase gene expression and activity in higher Collos, Y., Descolas-Gros, C. & Fontugne, M. 1993. Chemical, iso- plants. J. Biol. Chem. 277:46594–600. topic and enzymatic monitoring of free and enclosed seawa- Smith, G. J., Zimmerman, R. C. & Alberte, R. S. 1992. Molecular ter—implications for primary production estimates in and physiological responses of diatoms to variable levels of ir- incubation bottles. Maine Ecol. Progr. Series 93:49–54. radiance and nitrogen availability: growth of Skeletonema cos- Dugdale, R. C. & Goering, J. J. 1967. Uptake of new and regen- atum in simulated upwelling conditions. Limnol. Oceanogr. erated forms of nitrogen in primary . Limnol. 37:989–1007. Oceanogr. 12:196–206. Song, B. K. & Ward, B. B. 2004. Molecular characterization of the Dugdale, R. C. & Wilkerson, F. P. 1986. The use of 15N to measure assimilatory nitrate reductase gene and its expression in the nitrogen uptake in eutrophic : experimental consider- marine green alga Dunaliella tertiolecta (Chlorophyceae). Jour- ations. Limnol. Oceanogr. 31:673–89. nal of 40:721–731. Gao, Y., Smith, G. J. & Alberte, R. S. 2000. Temperature depend- Steele, J. H. 1974. The Structure of Marine Ecosystems. Harvard Uni- ence of nitrate reductase activity in marine phytoplankton: versity Press, Cambridge, MA, 128pp. biochemical analysis and ecological implications. J. Phycol. Stolz, J. F. & Basu, P. 2002. Evolution of nitrate reductase: molec- 36:304–313. ular and structural variations on a common theme. Chem. Guillard, R. R. L. 1975. Culture of phytoplankton for feeding ma- Biochem. 3:198–206. rine invertebrates. In Smith, W. & Chanley, M. [Eds.] Culture Tischner, R. 2000. Nitrate uptake and reduction in higher and of Marine Invertebrate . Plenum Press, New York, pp. lower plants. Plant Cell Environ. 23:1005–24. 29–60. Van de Peer, Y. & DeWachter, R. 1997. Construction of evolution- Jickells, T. 1998. Nutrient biogeochemistry of the coastal zone. Sci- ary distance trees with TREECON for Windows: accounting ence 281:217–22. for variation in nucleotide substitution rate among sites. Com- Jochem, F. J., Smith, G. J., Gao, Y., Zimmerman, R. C., Cabello- put. Appl. Biosci. 13:227–30. Pasini, A., Kohrs, D. G. & Alberte, R. S. 2000. Cyto- Vergara, J. J., Berges, J. A. & Falkowski, P.G. 1998. Diel periodicity metric quantification of nitrate reductase by immunolabeling of nitrate reductase activity and protein levels in the marine in the marine diatom Skeletonema coastatum. Cytometry 39: diatom Thalassiosira weissflogii (Bacillariophyceae). J. Phycol. 173–8. 34:952–61. Kudela, R. M. & Dugdale, R. C. 2000. Nutrient regulation of Wilkerson, F. P., Dugdale, R. C., Kudela, R. M. & Chavez, F. P.2000. phytoplankton productivity in Monterey Bay, California. Deep- Biomass and productivity in Monterey Bay, California: contri- Sea Res. II 47:1023–53. bution of the large phytoplankton. Deep-Sea Res. II 47:1003–22. Lomas, M. W. 2004. Nitrate reductase and urease enzyme activity Zhou, J. & Kleinhofs, A. 1996. Molecular evolution of nitrate red- in the marine diatom Thalassiosira weissflogii (Bacillar- uctase genes. J. Mol. Evol. 42:432–42. iophyceae): interactions among nitrogen substrates. Mar. Zvyagil’skaya, R. A., Vartapetyan, B. B. & Lvov, N. P. 1996. Nitrate Biol. 144:37–44. dissimilation in eukaryotes. Appl. Biochem. Microbiol. 32:165–9.