Báo cáo khoa học: A new molecular tool for transgenic diatoms Control of mRNA and protein biosynthesis by an inducible promoter–terminator cassette docx

A new molecular tool for transgenic diatoms
Control of mRNA and protein biosynthesis by an inducible
promoter–terminator cassette
Nicole Poulsen
1,2
and Nils Kro
¨
ger
1,2,3
1 Biochemie I, Universita
¨
t Regensburg, Germany
2 School of Chemistry & Biochemistry, 3 School of Materials Science & Engineering, Georgia Institute of Technology, Atlanta, GA, USA
Diatoms (Bacillariophyceae) are a group of unicellular
algae that are of enormous ecological importance,
being responsible for about 40% of the primary bio-
logical production in the oceans [1,2]. In addition to
their role in global carbon cycling, diatoms have
recently attracted interdisciplinary interest because of
their extraordinary ability to produce intricately
shaped, nanostructured silica as their cell wall material
[3–5].
In contrast to the wide interest in diatom biology,
genetic manipulation of diatoms is still in its infancy.
With the recent completion of the Thalassiosira pseu-
donana genome project [6] and establishment of an
expressed sequence tag (EST) databank for Phaeod-
actylum tricornutum [7], there is now an urgent
demand for genetic tools to analyze the function of
diatom genes in vivo. Genetic transformation meth-
ods are at present available for the diatom species

Keywords
Cylindrotheca fusiformis; diatom
transformation; green fluorescent protein
(GFP); inducible gene expression; nitrate
reductase
Correspondence
N. Kro
¨
ger, School of Chemistry &
Biochemistry, Georgia Institute of
Technology, 770 State St, Atlanta,
GA 30332-0400, USA
Fax: +1 404 894 7452
Tel: +1 404 894 4228
E-mail:
Website: />faculty/kroger/
Notes
Nucleotide sequence data for cffcpA-1A and
CfNR are available in the GenBank database
under accession numbers DQ060240 and
DQ060241
(Received 25 March 2005, revised 7 May
2005, accepted 11 May 2005)
doi:10.1111/j.1742-4658.2005.04760.x
Research in diatom biology has entered the postgenomic era since the
recent completion of the Thalassiosira pseudonana genome project. How-
ever, the molecular tools available for genetic manipulation of diatoms are
still sparse, impeding the functional analysis of diatom genes in vivo. Here
we describe the first method for inducible gene expression in transgenic
diatoms. This method uses a DNA cassette containing both promoter (Pnr)

and terminator (Tnr) elements derived from the nitrate reductase gene of
the diatom Cylindrotheca fusiformis. By using green fluorescent protein
(gfp) cDNA as a reporter gene, it is demonstrated that gene expression
under the control of the Pnr ⁄ Tnr cassette is switched off when cells are
grown in the presence of ammonium ions and becomes switched on within
4 h when cells are transferred to medium containing nitrate. Incubating
cells in nitrogen-free medium switches on transcription of the gfp gene, yet
gfp mRNA does not become translated into protein. This block on trans-
lation is released by the addition of nitrate, resulting in rapid onset of GFP
production with a drastically reduced delay time of only 1 h. Altogether we
have demonstrated that the Pnr ⁄ Tnr cassette enables inducible gene expres-
sion and control of both the level and timing of mRNA and protein
expression in transgenic diatoms.
Abbreviations
BLE (ble), bleomycin binding protein (gene); fcp, fucoxanthin chlorophyll a ⁄ c binding protein gene; egfp (egfp), enhanced green fluorescent
protein (gene); NR, nitrate reductase; Pd, promoter of frua3 gene; Pfcp (Tfcp), promoter (terminator) of fucoxanthin chlorophyll a ⁄ c binding
protein gene; Pnr (Tnr), promoter (terminator) of C. fusiformis nitrate reductase gene; SOEing, splicing by overlap extension.
FEBS Journal 272 (2005) 3413–3423 ª 2005 FEBS 3413
P. tricornutum [8], Cylindrotheca fusiformis [9], Cyclo-
tella cryptica and Navicula saprophila [10]. So far, in
all but one case, only constitutive expression of intro-
duced genes has been achieved. The one exception
involved expression of a green fluorescent protein
(GFP) fusion protein in P. tricornutum, under control
of the promoter fcpA, derived from a gene encoding
a fucoxanthin chlorophyll a ⁄ c binding protein [8]. In
this instance, expression of the GFP fusion protein
was repressed after a 7 day incubation in the dark
and induced by 24 h exposure of the cells to light
[11]. This method is not generally applicable for stud-

ies on diatom cell biology as diatom growth is totally
inhibited in the dark. For example, analyzing the
effect of introduced proteins or RNA-mediated gene
interference only at certain developmental stages (e.g.
cell division) requires regulated expression of intro-
duced genes within a much shorter time scale. There-
fore, to study diatom biology using molecular genetic
techniques, promoters need to be identified that
enable rapid and tightly controlled expression of
genes in transgenic diatoms.
Previous physiological studies in diatoms have
shown that the activity of nitrate reductase (NR), the
rate-limiting enzyme in nitrogen assimilation, is regula-
ted by the nitrogen source present in the medium. NR
activity is suppressed by ammonium and induced when
ammonium is replaced by nitrate [12,13]. In the green
algae Chlamydomonas reinhardtii [14,15], Chlorella
vulgaris [16] and Dunaliella tertiolecta [17], ammo-
nium-dependent suppression of NR activity is due to
down-regulation of NR gene expression as well as
post-transcriptional regulation. These studies have
prompted us to speculate that the promoter of the
diatom NR gene may be a suitable molecular genetic
tool for regulating transgenic protein expression in
diatoms. However, the unexpected discovery by gen-
ome sequence analysis of a complete urea cycle in the
diatom T. pseudonana appeared to imply that a more
complex regulatory network may control nitrogen
metabolism in diatoms, possibly also involving the NR
step [6]. Therefore, a thorough analysis was required

of the applicability of diatom NR promoters to drive
inducible gene expression in transgenic diatoms.
Only recently the first two sequences of diatom NR
genes have become available from T. pseudonana [6]
and P. tricornutum [18], yet in neither organism has
NR gene expression been analyzed as a function of the
nitrogen source in the medium. In this study we have
isolated the NR gene (CfNR) from a genomic DNA
library of the diatom C. fusiformis and analyzed CfNR
levels in response to different nitrogen sources. Using
GFP as reporter protein, we have demonstrated that
the 5¢-UTR and 3¢-UTR of CfNR allow control of
both the timing and level of expression of introduced
genes in transgenic C. fusiformis.
Results
Increasing the efficiency of C. fusiformis
transformation
Previously, DNA fragment Pd, from the 5¢-UTR of
the frua3 gene, was the only established promoter in
C. fusiformis to drive expression of the selection mar-
ker protein bleomycin-binding protein (BLE), as well
as other introduced proteins [9]. However, only mod-
erate numbers of transformants and relatively low lev-
els of heterologous protein expression were obtained
[9]. In contrast, promoter fcpA from a gene encoding
a fucoxanthin chlorophyll a ⁄ c binding protein (fcp)
has been successfully used to obtain high expression
levels of foreign proteins in transgenic P. tricornutum
[19]. This promoter is constitutively active in light
but not functional in C. fusiformis (N. Kro

¨
ger, unpub-
lished observation). Therefore, the promoter of a
C. fusiformis fcp gene has been isolated and tested for
its applicability in C. fusiformis transformations. Prim-
ers were designed based on a C. fusiformis fcp cDNA
sequence (cffcpA-3) available from the NCBI database
(see Experimental procedures) and used to amplify
a 441-bp fragment from C. fusiformis genomic DNA.
This DNA fragment was used as a probe for screening
a C. fusiformis genomic DNA library. From a phage
reacting positive in this screen, the sequence of a com-
plete fcp gene including 5¢-UTR and 3¢-UTR was
determined. This gene contained no introns and sur-
prisingly was not identical with the cffcpA-3 cDNA
sequence, but perfectly matched the cffcpA-1A cDNA
sequence (GenBank accession number AY125580).
The two sequences share 93% sequence identity at the
nucleotide level.
To generate an fcp promoter-based expression vector
for C. fusiformis termed pCfcp, 1624 bp of the 5¢-UTR
(termed Pfcp) and 504 bp of the 3¢-UTR (termed Tfcp)
from the cffcpA-1A genomic DNA were cloned into
pBluescript flanking a short region containing three
unique restriction sites (EcoRV, XbaI, NotI), allowing
easy insertion of genes. The ble gene, which confers
resistance to the antibiotic zeocin, was ligated with
pCfcp, generating pCfcp-ble. This plasmid was used for
C. fusiformis transformation by microparticle bombard-
ment, yielding typically 36 ± 4 zeocin-resistant trans-

formants per 10
7
cells (using 1 lg plasmid), whereas an
average of only 11 ± 1 zeocin-resistant transformants
per 10
7
cells were obtained using 1 lg plasmid pPd-ble
Inducible promoter for transgenic diatoms N. Poulsen and N. Kro
¨
ger
3414 FEBS Journal 272 (2005) 3413–3423 ª 2005 FEBS
(the transformant numbers represent averages from
three transformation experiments). Thus, C. fusiformis
transformation using the new plasmid pCfcp-ble was
3–4 times more efficient than the previous method.
Cloning of the NR gene from C. fusiformis and
mRNA expression studies
Degenerate primers corresponding to the highly
conserved NR motifs W-W-Y-K-P-E ⁄ D-Y ⁄ F and
W-N-L-M-G-M were used in RT-PCR yielding a 375-
bp product which exhibited 69% and 77% sequence
identity with the corresponding NR sequence regions
from T. pseudonana and P. tricornutum, respectively.
Screening of the C. fusiformis genomic DNA library,
using the 375-bp PCR product as a probe, led to the
identification of a phage clone that contained the
entire C. fusiformis NR gene (CfNR) on a single 5.4-kb
BamHI DNA fragment. RACE PCR was used to
determine the 5¢ end of the CfNR cDNA, allowing
unequivocal identification of the gene’s start codon.

On the basis of these data, the CfNR gene is made up
of 2619 bp of intron-less sequence encoding a 873-
amino acid polypeptide which exhibits 69% and 72%
sequence identity with the predicted polypeptide
sequences of the NR genes from T. pseudonana and
P. tricornutum, respectively (Fig. 1).
To investigate the effect of the nitrogen source on
expression of the CfNR gene, C. fusiformis cells were
preconditioned for 2 weeks in medium containing
ammonium chloride as the sole nitrogen source (ammo-
nium medium). After being washed with nitrogen-free
medium, the cells were transferred to medium contain-
ing ammonium (NH
4
+
), nitrate (NO
3
–
), a 1 : 1 mixture
of ammonium and nitrate (NH
4
+
⁄ NO
3
–
) or kept in
nitrogen-free medium (–N). After an incubation period
of 24 h, NR expression was monitored by RT-PCR
analysis (Fig. 2). The CfNR gene was expressed both in
the presence of nitrate and under conditions of nitrogen

starvation, but not in ammonium-containing medium.
Ammonium proved to be an inhibitor of CfNR expres-
sion, as shown by the lack of CfNR mRNA in the pres-
ence of equal molar amounts of nitrate and ammonium
(Fig. 2). These results demonstrate that the CfNR gene
Fig. 1. Alignment of NR polypeptide sequences from diatoms. The sequence alignment was performed using CLUSTALW [40]. C.f., Cylindroth-
eca fusiformis NR (this study); P.t., Phaeodactylum tricornutum NR (GenBank accession number AY579336); T.p., Thalassiosira pseudonana
NR [6]. Amino acids identical with the CfNR polypeptide sequence are indicated by asterisks. The CfNR polypeptide exhibits the typical NR
domain structure containing the molybdopterin domain (aa 54–295), dimerization domain (aa 321–447), heme domain (aa 519–592), FAD
domain (aa 623–728) and NADH domain (residues 744–858) [27]. A unique 17-amino-acid insertion in the molybdopterin-binding domain iden-
tified in the two other diatom NR genes [18] is also conserved in CfNR (aa 211–227).
N. Poulsen and N. Kro
¨
ger Inducible promoter for transgenic diatoms
FEBS Journal 272 (2005) 3413–3423 ª 2005 FEBS 3415
expression can be easily switched on and off by varying
the nitrogen source in the medium. To evaluate the
applicability of the CfNR gene’s regulatory elements to
drive inducible expression of foreign proteins in trans-
genic diatoms, a GFP-based reporter gene system was
established.
Construction of a vector for inducible gene
expression
A chimeric gene was constructed consisting of the egfp
coding sequence [20] flanked by 775 bp of the 5¢-UTR
(termed Pnr) immediately upstream of the start ATG
of the CfNR gene and 571 bp of the 3¢-UTR (termed
Tnr) immediately after the stop codon. The chimeric
gene was cloned into pCfcp-ble yielding plasmid
pNICgfp (Fig. 3A). The plasmid DNA (3 lg) was

introduced into C. fusiformis by microparticle bom-
bardment, and transformants were selected on zeocin-
containing plates. From the over 100 zeocin-resistant
clones obtained, 33 clones were analyzed for GFP
expression by fluorescence microscopy, of which 22
clones were positive. The fluorescence intensities of
eight of these clones were quantified by fluorimetry,
using excitation at 485 nm and monitoring emission at
510 nm. Owing to the chlorophyll content, C. fusifor-
mis wild-type cells exhibited noticeable fluorescence in
these measurements, yet fluorescence intensities of dif-
ferent GFP-expressing transformants were 7- to 50-
fold higher than in wild-type cells (Fig. 3B). Variation
in GFP fluorescence intensities between different
transformant clones has previously been observed in
P. tricornutum expressing GFP or GFP fusion proteins
under the control of the constitutive fcp promoter
[8,19]. As the introduced genes become randomly
integrated into the diatom’s genome [8], the variation
in GFP expression levels may result from differences
in copy numbers or location of the introduced genes
within the genome. Clone 31 exhibited the highest
intensity of GFP fluorescence and therefore was cho-
sen for further analysis. In growth medium containing
ammonium as the sole nitrogen source, the fluores-
A
B
C
Fig. 3. Structure of transformation plasmid pNICgfp and analysis of
GFP expressing C. fusiformis transformants. (A) Restriction map of

the part of plasmid pNICgfp containing the egfp gene flanked by
the Pnr ⁄ Tnr cassette and ble gene flanked by cffcpA-1A promoter
(Pfcp) and terminator (Tfcp) sequences. (K, KpnI; H, HindIII;
E, EcoRI; N, NotI; E105, Eco105I; E5, EcoRV; S, SacI). (B) Fluores-
cence intensity (excitation 485 nm, emission 510 nm) of C. fusifor-
mis wild-type and transformant clones (C#) expressing GFP. Cell
concentration was 1· 10
7
ÆmL
)1
for each clone. (C) GFP fluores-
cence intensity (excitation 485 nm, emission 510 nm) of transform-
ant C31 in different growth media, containing nitrate (1.5 m
M
KNO
3
), ammonium (1.5 mM NH
4
Cl) or mixtures of nitrate and
ammonium as nitrogen source (nitrate concentration was 1.5 m
M,
ammonium concentrations were: Q10, 0.15 m
M; Q25, 0.06 mM;
Q50, 0.03 m
M; Q100, 0.0015 mM).
Fig. 2. Influence of nitrogen source on the expression of CfNR
mRNA. C. fusiformis wild-type cells were grown in ammonium
medium and then transferred to different medium containing nitrate
(NO
3

–
), ammonium (NH
4
+
), a 1 : 1 mixture of nitrate and ammo-
nium (NO
3
–
⁄ NH
4
+
) or lacking any nitrogen (– N). After a 24-h incu-
bation period, RNA was isolated from each sample and RT-PCR
was performed to analyze CfNR mRNA expression (NR). As a posit-
ive control RT-PCR analyses for the constitutively expressed
cffcpA-1A mRNA (fcp) were performed using the same RNA prepa-
rations.
Inducible promoter for transgenic diatoms N. Poulsen and N. Kro
¨
ger
3416 FEBS Journal 272 (2005) 3413–3423 ª 2005 FEBS
cence intensity was slightly above wild-type levels (not
shown), yet dramatically increased when ammonium
was replaced by nitrate (Fig. 3C). These results show
that the Pnr ⁄ Tnr cassette in plasmid pNICgfp retains
the inducible property of the regulatory sequences that
drive expression of the CfNR gene.
As ammonium acted as an inhibitor of CfNR expres-
sion (Fig. 2), we investigated the possibility of control-
ling the level of GFP expression by growing Clone

31 in medium containing mixtures of ammonium and
nitrate. At relative molar concentrations of nitrate vs.
ammonium of 50 (Q50) and 100 (Q100), GFP fluores-
cence levels were virtually indistinguishable from fluor-
escence levels of cells incubated in medium containing
only nitrate (Fig. 3C). However, at molar ratios of
nitrate to ammonium of 10 (Q10) and 25 (Q25), the lev-
els of GFP fluorescence were 44% and 83%, respect-
ively, of the fluorescence levels of nitrate-grown cells.
Thus, by adjusting appropriate relative concentrations
of ammonium and nitrate, it is possible to down-regu-
late, rather than completely shut off, the expression of
genes that are under control of the Pnr ⁄ Tnr cassette.
Decoupling of transcription and translation
of gfp mRNA
To further evaluate the properties of the Pnr ⁄ Tnr
expression cassette, Clone 31 cells preconditioned in
ammonium medium were subjected to nitrogen starva-
tion, and gene expression was monitored by RT-PCR.
In agreement with the result obtained with wild-type
cells (Fig. 2, lane –N), expression of both CfNR and
gfp genes was found to be switched on in the trans-
formant (Fig. 4B). Surprisingly, when fluorimetry (not
shown) and fluorescence microscopy were used, no
GFP fluorescence was detected in Clone 31 cells
(Fig. 4A), indicating the absence of functional GFP.
Western blot analysis confirmed that GFP was indeed
absent from nitrogen-starved Clone 31 cells (Fig. 4B),
ruling out the possibility that GFP was present in a
nonfluorescent form. After the addition of nitrate,

GFP fluorescence developed in Clone 31 cells
(Fig. 4A), demonstrating that inhibition of GFP pro-
duction in nitrogen-starved cells was reversible. Alto-
gether these results indicate that gfp mRNA did not
become translated until nitrate was present, implying
that, in nitrogen-starved cells, protein production from
genes flanked by the NR promoter (Pnr) and termina-
tor (Tnr) is controlled at the post-transcriptional level.
On the basis of these results, we assumed that
decoupling of mRNA and protein expression in nitro-
gen-starved cells may provide a useful tool to obtain
control over the timing of gene expression, as the built
up pool of transgenic mRNA may very rapidly become
translated into protein after the addition of nitrate. To
investigate this, we analyzed by fluorimetry the kinetics
of GFP production in Clone 31 cells in nitrate medium
after preconditioning in ammonium medium and nitro-
gen-free medium, respectively. After transfer of ammo-
nium-preconditioned cells to nitrate medium, a lag
phase of 5 h was observed before GFP expression
became noticeable. Beyond this time, fluorescence lev-
els increased with a doubling time of 1.5 h (Fig. 5A).
RT-PCR analysis demonstrated that gfp mRNA
expression started 4 h after the transfer of the cells to
nitrate medium, thus preceding the onset of GFP
fluorescence by about 2 h. Cells preconditioned in
nitrogen-free medium exhibited a comparable rate of
increase of GFP fluorescence (2 h doubling time) after
A
B

Fig. 4. Influence of nitrogen source on mRNA expression and the
formation of GFP protein. (A) Fluorescence images of C. fusiformis
Clone 31 cells in nitrogen-free medium (–N) and in nitrate medium
(NO
3
–
). Each micrograph represents an overlay of a transmission
light microscopy image and two different fluorescence images. The
green color shows GFP fluorescence and the red color depicts
chloroplast autofluorescence (bar, 10 lm). (B) Comparison of gfp
mRNA expression and GFP protein expression in nitrate medium
(NO
3
–
), ammonium medium (NH
4
+
) and nitrogen-free medium (–N).
The bottom row shows a GFP-specific western blot from total
extracts of Clone 31 cells after 24 h of incubation in the indicated
media. The rows above show the results from RT-PCR analysis for
gfp mRNA (gfp) expression in Clone 31 cells from the same three
cultures. As controls, expression of CfNR mRNA (NR) and the con-
stitutive cffcpA-1 A mRNA (fcp) were monitored by RT-PCR using
the same RNA preparations.
N. Poulsen and N. Kro
¨
ger Inducible promoter for transgenic diatoms
FEBS Journal 272 (2005) 3413–3423 ª 2005 FEBS 3417
transfer to nitrate medium. However, the lag phase

for the onset of GFP fluorescence was drastically
reduced to only 1 h, which probably corresponds to
the time required for the extremely slow process of
GFP chromophore formation [21]. Therefore, addition
of nitrate to cells preconditioned on nitrogen-free
medium appears to enable virtually instantaneous
induction of protein expression from genes that are
controlled by the Pnr ⁄ Tnr cassette.
Discussion
In this study, we have isolated from a C. fusiformis
genomic library an fcp gene (cffcpA-1A) and a nitrate
reductase gene (CfNR). From these genes 5¢-UTRs (pro-
moters) and 3¢-UTRs (terminators) were used to
construct a transformation vector for inducible gene
expression in C. fusiformis. Flanking the zeocin resist-
ance gene ble by promoter (Pfcp) and terminator (Tfcp)
regions from cffcpA-1A improved the transformation
efficiency for C. fusiformis about fourfold over previ-
ously used transformation vectors. Presumably, the
increased transformation rate is due to the exceptional
strength of the fcp promoters [22]. Therefore, more BLE
protein may be produced in C. fusiformis transformants
by using a Pfcp-based vector compared with the previ-
ously used Pd-containing vectors, which allows more
transformants to grow using extremely high zeocin con-
centrations (1 mgÆmL
)1
) required for suppression of
C. fusiformis wild-type growth. In future the promoter
Pfcp may be a useful tool for generating higher levels

of expression of other transgenic proteins.
With the CfNR gene sequence in hand, we were able
to demonstrate by RT-PCR that CfNR expression can
be simply regulated at the transcriptional level by vary-
ing the nitrogen source in the medium (Fig. 4). Regu-
lation of the NR gene transcript in C. fusiformis is
similar to the green alga C. reinhardtii, as, in both
organisms, NR mRNA production is switched off in
the presence of ammonium and induced by nitrate or
nitrogen starvation [23–25]. This regulatory pattern is
preserved in the Pnr ⁄ Tnr cassette driving gfp expres-
sion in C. fusiformis cells that have been transformed
using the pNICgfp plasmid. Remarkably, induction of
Pnr ⁄ Tnr-driven gfp expression has different outcomes
depending on whether nitrate or nitrogen starvation is
used as the inducer. Biosynthesis of GFP is inhibited
in nitrogen-free medium, and protein is only produced
in the presence of nitrate (Fig. 4B). As the gfp mRNA
coding region is a highly unlikely target for post-tran-
scriptional regulation in a diatom, we assume that this
effect is mediated by regions in the CfNR-derived 5¢-
UTR or 3¢-UTR of the gfp mRNA. Nutrient-depend-
ent, post-transcriptional regulation of eukaryotic gene
expression mediated by the UTRs of mRNA molecules
is well characterized for iron metabolism in mammals
[26], and recently evidence has been presented that the
stability of NR mRNA in the green alga Chlorella vul-
garis is mediated via the 5¢-UTR [27]. Therefore, we
speculate that the UTRs in the CfNR mRNA may
contain target sites for nitrate-dependent regulators of

translation or mRNA stability. Interestingly, in C. fusi-
formis, mRNA expression of AMT (encoding ammo-
nium transporter proteins) and NAT (encoding nitrate
transporter proteins) genes becomes strongly up-regu-
lated when cells are transferred from ammonium
medium to –N medium [28,29], suggesting that the
expression of different proteins involved in nitrogen
metabolism may be controlled by the same mechanism.
However, it is at present unknown if nitrogen-starved
A
B
Fig. 5. Kinetics of gfp mRNA and GFP protein expression in C. fusi-
formis. (A) Development of fluorescence intensity in Clone 31 (C31)
and wild-type (wt) cells. Cells were preconditioned in ammonium
medium and then transferred either directly to nitrate medium
(NO
3
–
) or incubated for 24 h in nitrogen-free medium before nitrate
was added (–N ⁄ NO
3
–
). The x-axis indicates the time after addition
of nitrate. (B) RT-PCR analysis of gfp mRNA expression in Clone 31
cells. Cells were preconditioned in ammonium medium and then
directly transferred to nitrate medium. Hours indicate the time after
addition of nitrate. At each time point RT-PCRs were performed
using primers specific for gfp mRNA (gfp) and the constitutively
expressed cffcpA-1A mRNA (fcp), respectively.
Inducible promoter for transgenic diatoms N. Poulsen and N. Kro

¨
ger
3418 FEBS Journal 272 (2005) 3413–3423 ª 2005 FEBS
cells synthesize AMT and NAT protein or if transla-
tion of AMT and NAT mRNAs is inhibited, as we
have demonstrated for gfp mRNA.
Exploiting the ability of the Pnr ⁄ Tnr cassette to
decouple the transcription of a chosen gene from its
translation into protein represents a valuable experi-
mental tool. The subsequent addition of nitrate enables
rapid protein production from introduced genes.
Furthermore, the addition of both nitrate and appro-
priate concentrations of ammonium allows, within lim-
its, the control of the amount of induced protein. The
observed lag phase of 1 h for Pnr ⁄ Tnr-controlled pro-
tein expression most likely represents an underestimate
of the speed of induction, because de novo formation
of fluorescent GFP from the unfolded polypeptide is a
very slow process, exhibiting a half time of 84.3 min
[21]. The following observation is consistent with this
assumption. In a Saccharomyces cerevisiae transform-
ant carrying the gfp gene under control of the GAL1
promoter, GFP fluorescence starts 2.5 h after galactose
addition [30], yet GAL1-controlled expression of other
S. cerevisiae proteins already occurs < 10 min after
induction with galactose [31]. Therefore, we expect that
the experimental methods developed in the present
work should allow analysis of the role of diatom pro-
teins in short-lived cellular processes such as cyto-
kinesis and valve and girdle band formation which are

completed in less than 1 h. In future, analysis of cell
division or valve formation in C. fusiformis transform-
ants carrying a gene of interest under control of the
Pnr ⁄ Tnr cassette may be performed as outlined in the
following. A transformant grown in ammonium med-
ium will be starved of silicic acid to arrest the cells at
the G1 ⁄ S boundary [32]. After subsequent incubation
in nitrogen-free medium to induce mRNA expression,
silicic acid will be added to initiate cell division and sil-
ica formation. Concomitantly with or at appropriate
times after silicic acid replenishment, nitrate will be
added to induce instantaneous expression of the pro-
tein of interest, allowing observation of the protein’s
influence on the progression of the cell cycle and the
silica biogenesis. As the regulation of NR expression
appears to be very similar throughout the diatom
realm [12,13], the Pnr ⁄ Tnr cassette of C. fusiformis rep-
resents a paradigm for establishing inducible gene
expression systems also in other diatom species.
Experimental procedures
Culture conditions
C. fusiformis was grown as described previously [33] under
constant light and in artificial seawater medium containing
1.5 mm KNO
3
as sole nitrogen source (nitrate medium).
Where indicated, nitrate was not included in the medium
(nitrogen-free medium) or replaced by 1.5 mm NH
4
Cl

(ammonium medium) or a mixture of 0.75 mm NH
4
Cl +
0.75 mm KNO
3
(ammonium + nitrate medium).
Cloning of the cffcpA-1A gene
To generate a selection marker for use in C. fusiformis,we
first cloned the fcp gene and used its promoter and termina-
tor sequences to drive expression of the zeocin resistance
gene ble. To this end, C. fusiformis genomic DNA was
extracted [33] and gene-specific oligonucleotides (sense: fcp1
5¢-AGAGCGAACTTGGTGCCCAG-3¢; antisense: fcp2
5¢-GCACGTCCGTTGTTCAATTC-3¢) were designed based
on a C. fusiformis fcp precursor cDNA sequence available
from the NCBI database (cffcpA-3; GenBank accession
number AY125583). Thirty cycles of PCR produced a 441-
bp DNA fragment, which was cloned into the pGEMT
vector (Promega, Madison, WI, USA) and sequenced. The
sequence obtained matched perfectly the database sequence.
To screen the C. fusiformis genomic DNA library (in
kEMBL3) [33] the 441-bp fcp DNA fragment was used as a
probe after labeling with digoxygenin (Roche, Mannheim,
Germany) according to the manufacturer’s instructions.
Phage DNA of one positive clone was analyzed by diges-
tion with different restriction enzymes and subsequent
Southern blotting using the same probe as above. Two
BamHI-digested DNA fragments (1.9 kb and 1.46 kb) that
hybridized to the probe were cloned into the BamHI site of
pUC18 and sequenced, resulting in pUC18 ⁄ fcp1.9kb (cover-

ing 278 bp of fcp coding sequence preceded by 5¢-UTR)
and pUC18 ⁄ fcp1.46kb (covering 357 bp of the fcp coding
sequence followed by 3¢-UTR).
Construction of vector pCfcp for constitutive
gene expression
The 1.9-kb insert of the pUC18 ⁄ fcp1.9kb plasmid was sub-
cloned into the KpnI–PstI sites of pBluescriptII SK+, gen-
erating pBluescript ⁄ fcp1.9kb. To introduce a cloning site
between the 5¢-UTR and 3¢-UTR of the fcp gene, a short
165-bp fragment of the fcp 5¢-UTR was amplified by PCR
from pUC18 ⁄ fcp1.9kb using the sense primer 5¢-GAT
CTTTGC
TACGTACGAACG-3¢ and the antisense primer
5¢-GCTCTAGAGATATCTAGTCTTTGTGATAAAGAAA
ATTATG-3¢. The resulting 165-bp PCR product contained
an Eco105I restriction site (underlined) and an EcoRV
(bold) and XbaI (italic) restriction site, which were both
introduced by the antisense primer. The PCR product,
which covered part of the 5¢-UTR starting 12 bp upstream
of the start ATG, was then cloned into the Eco105I–XbaI
sites of pBluescript ⁄ fcp1.9kb, generating pBluescript ⁄
fcp1.6kb, which covers bp )12 to ) 1613 upstream of the
fcp gene’s start ATG. The fcp terminator was ampli-
N. Poulsen and N. Kro
¨
ger Inducible promoter for transgenic diatoms
FEBS Journal 272 (2005) 3413–3423 ª 2005 FEBS 3419
fied by PCR from pUC18 ⁄ 1.46kb using the sense pri-
mer 5¢-G AAT
GCGGCCGCATTGCTTGTTGAGAAATA

GG-3¢, which introduced a NotI restriction site (under-
lined) and the antisense primer 5¢-CG
GAGCTCTGG
AAGCATGAAGTACTGCCA-3¢, which introduced a SacI
restriction site (underlined). The 524-bp PCR product was
digested with NotI and SacI and cloned into the NotI ⁄ SacI
sites of pBluescript ⁄ fcp1.6kb, generating the C. fusiformis
expression vector pCfcp. Genes to be inserted into the
pCfcp vector require the sequence 5¢-ATCAAAACAACC
AAA-3¢ immediately upstream of the start codon because
vector pCfcp lacks bp ) 1to)12 of the promoter.
Construction of zeocin resistance plasmid
pCfcp-ble
The ble gene [34] (GenBank accession number X52869) was
amplified from pZEOSV (Invitrogen, Carlsbad, CA, USA)
by PCR using sense primer 5¢-ATCAAAACAACCAAAA
TGGCCAAGTTGACCAGTGC-3¢ and antisense primer
5¢-GAAT
GCGGCCGCTCAGTCCTGCTC CTCGGCCAC-3 ¢,
which introduced a NotI restriction site (underlined). The
resulting 386-bp PCR product was digested with NotI and
cloned into the EcoRV ⁄ NotI site of pCfcp to generate
pCfcp-ble.
Cloning of the CfNR gene
Extraction of poly(A)-rich RNA from C. fusiformis and
synthesis of cDNA coupled to oligo(dT)
25
magnetic beads
(Dynal Biotech, Hamburg, Germany) was as described [9].
Degenerate oligonucleotides NR1 (5¢-TGGTGGTAYAAR

CCNGANT-3¢) and NR2 (5¢-CATNCCCATNARRTTC
CA-3¢) were designed based on the alignment of the
deduced amino-acid sequences of nine NR genes from algae
and higher plants. The cDNA was used as template for 35
cycles of PCR amplification (15 s 94 °C, 15 s 52 °C, 30 s
72 °C) resulting in a 380-bp DNA product which was
cloned into pGEMT vector (Promega) and sequenced. This
380-bp DNA fragment of the CfNR gene was labeled with
digoxigenin and used to screen a C. fusiformis genomic
DNA library as described above. From a positive phage
identified in this screen, a 5.4-kb BamHI DNA fragment
was identified by Southern blot analysis, which hybridized
with the CfNR-specific probe. This DNA fragment was
cloned into the BamHI site of pUC18, generating plasmid
Pnr ⁄ BamHI5.4kb. Sequencing of the 5.4-kb insert revealed
that it covered the complete CfNR gene including the
5¢-UTR and 3¢-UTR. To determine the 5¢ sequence of the
CfNR mRNA, the cDNA was C-tailed as described in [35]
and used in a 5 ¢-RACE PCR using antisense primer
5¢-CAGCTAAACCCAATAGTCTG-3¢ and sense primer
5¢-GGCCACGCGTCGACTAGTACGGGIIGGGIIGGGI
IG-3¢. Thirteen cycles of PCR amplification (12 s 94 °C,
15 s 55 °C, 1.5 min 72 °C) followed by 33 cycles (12 s
94 °C, 15 s 55 °C, 1.5 min 72 °C) with an increment of 5 s
at 72 °C per cycle were performed. A  550-bp DNA frag-
ment was gel purified and re-amplified in a second PCR
using the same antisense primer and the nested sense pri-
mer 5 ¢-GGCCACGCGTCGACTAGTAC-3¢. The resulting
product was cloned into pGEMT vector and sequenced.
Construction of the inducible expression plasmid

pNICgfp
A 456-bp HindIII–EcoRI DNA fragment from plasmid
Pnr ⁄ BamHI5.4kb covering part of the 5¢-UTR of the CfNR
gene was cloned into the HindIII–EcoRI sites of pBlue-
scriptII SK+, generating pNRp. To create the Pnr–gfp
hybrid DNA fragment, the Gene SOEing technique (splicing
by overlap extension) [36] was used. The first PCR for Gene
SOEing was performed using Pnr ⁄ BamHI5.4kb as the tem-
plate with the sense primer SOE-1 (5¢-CCTCTTCTAGC
GAGTCTGG-3¢) and antisense primer SOE-2 (5¢-CTC
GTTGCTCACCATTGTTCAGCGTTGATTTTT-3¢). The
second PCR was performed on an egfp-containing plasmid
(a gift from Dr K. Apt, Martek Biosciences, Columbia,
MD, USA) with sense primer SOE-3 (5¢-AAAAATCA
ACGCTGAACAATGGTGAGCAAAGGGCGAG-3¢) and
antisense primer SOE-4 (5¢-GAAT
GCGGCCGCTTACT
TGTAACAGCTCGTCCATG-3¢), which introduced a
NotI site (underlined). The third PCR was carried out
with both the first two PCR products and primers SOE-1
and SOE-4. The resulting PCR product was digested with
EcoRI and NotI and cloned into the EcoRI–NotI sites of
pNRp. This resulted in a chimeric gene with Pnr (775 bp)
fused to egfp (pPnr-gfp). To introduce Tnr from the 3¢-UTR,
a PCR was performed using Pnr ⁄ BamHI5.4kb as the
template with sense primer 5¢-GAAT
GCGGCCGCGA
ATGTGTGCAAATTGAAGAAC-3¢ and antisense primer
5¢-TTCGAGCTCCGGGGAAACGGTGCCAACTT-3¢,which
introduced a NotI site (underlined) and a SacI site (bold).

The resulting 592-bp DNA fragment was digested with NotI
and SacI and cloned into the NotI–SacI sites of pPnr-gfp
yielding pPnr-gfp-Tnr. The final step of cloning involved the
digestion of pPnr-gfp-Tnr with SacI (blunted) and KpnI and
cloning into the BamHI (blunted)–KpnI sites of the zeocin
resistance plasmid pCfcp-ble (see above) yielding pNICgfp.
All PCR products were checked for the correct sequence by
DNA sequencing.
Diatom transformation
C. fusiformis was transformed by microparticle bombard-
ment using the Biolistic PDS-1000 ⁄ He Particle Delivery
system (Bio-Rad, Hercules, CA, USA) as described [9]. For
selection of transformants, bombarded cells were plated on
artificial seawater medium containing 1.5% agar and
1mgÆ mL
)1
zeocin. After 8 days of incubation of the plates
under C. fusiformis standard growth conditions (see above),
Inducible promoter for transgenic diatoms N. Poulsen and N. Kro
¨
ger
3420 FEBS Journal 272 (2005) 3413–3423 ª 2005 FEBS
individual clones were picked from the plates and inocula-
ted into liquid artificial seawater medium containing
1mgÆmL
)1
zeocin.
RT-PCR
Total RNA was isolated from 5 · 10
7

cells using 1 mL
TRI reagent (Sigma, St Louis, MO, USA) according to
the manufacturer’s instructions. Contaminating DNA was
removed from the RNA preparation by DNase treatment,
followed by an additional RNA purification step using
TRI reagent. RNA concentration was determined photo-
metrically at 260 nm and estimated by agarose gel electro-
phoresis. For first strand cDNA synthesis, 5 lg total
RNA, 25 pmol oligonucleotide 5¢-GCCGCCGAATTCC
CAG(T)
18
-3¢, 500 lm dNTPs and 100 U Superscript III
reverse transcriptase (Invitrogen) were incubated in a
20 lL reaction mix (1· RT buffer; Invitrogen) at 50 °C
for 1 h. After heat inactivation at 70 °C for 15 min, 1 lL
of the reverse transcription reaction mix was used in a
50 lL PCR using 30 cycles for amplification. For the
cffcpA-1A PCR, the first strand cDNA was diluted 1 : 50
before amplification, and for gfp and CfNR PCR the first
strand cDNA was used undiluted.
Studies on CfNR gene expression
To investigate the effect of the type of nitrogen source
on the production of CfNR mRNA, cells were grown in
ammonium medium under constant illumination and agi-
tation. After reaching a density of  1.5 · 10
6
cellsÆmL
)1
,
cells were harvested by centrifugation (2800 g, 5 min) and

washed three times with nitrogen-free medium. The
washed cells were then resuspended in nitrogen-free med-
ium or medium containing either nitrate, ammonium, or
a 1 : 1 mixture of nitrate and ammonium. After 24 h of
incubation, the cells were harvested for RNA isolation,
and RT-PCR was performed. To study the kinetics of
CfNR expression, cells were grown in ammonium med-
ium as described above, and then transferred to fresh
medium containing nitrate. Equal aliquots of cells were
harvested every other hour for RT-PCR analysis and
hourly for fluorescence measurements. The same type of
kinetic analysis was performed in a second experiment, in
which cells were incubated for 24 h in nitrogen-free med-
ium before the addition of nitrate. To exert control on
the amount of expressed GFP protein, cells were grown
to a cell density of  1.5 · 10
6
cellsÆmL
)1
in ammonium
medium, washed three times with nitrogen-free medium
and resuspended in nitrogen-free medium to a final cell
density of 0.5 · 10
6
cellsÆmL
)1
. After a 24 h incubation
period, protein expression was induced by the addition of
various concentrations of NH
4

Cl (final concentrations:
0.15 mm, 0.06 mm, 0.03 mm, 0.015 mm) immediately fol-
lowed by the addition, to each sample, of KNO
3
to a
final concentration of 1.5 mm. The cells were grown for
a further 24 h before fluorescence measurements were
taken.
Fluorescence measurements of GFP
A Shimadzu RF-5301PC spectrofluorophotometer was used
for fluorescence measurements of diatom cells at ambient
temperature. The excitation wavelength was 485 nm, the
emission maximum was at 510 nm, and the slit width at
both wavelengths was 5 nm. Cell suspensions were either
directly loaded into the quartz cuvette (kinetic measure-
ments) or concentrated 10-fold before measurements (from
cells grown in ammonium ⁄ nitrate medium).
Western blot analysis
Separation of proteins by SDS ⁄ PAGE [37], Coomassie
staining of SDS gels [38], and western blot analysis [39] were
performed according to standard protocols. For western
blot analysis, 1.25 · 10
6
cells were harvested from the
respective medium, the cells were rapidly lysed by incuba-
tion for 5 min in SDS sample buffer at 95 °C, and equal
aliquots of the extracts were subjected to SDS ⁄ PAGE. For
detection of GFP, a specific antibody (developed in rabbit;
Clontech, Mountain View, CA, USA) was used and an anti-
rabbit IgG–alkaline phosphatase conjugate (Sigma) as the

secondary antibody.
Microscopy analysis
Confocal imaging was performed using an inverted Zeiss
LSM 510 laser scanning microscope and a 63 · oil immer-
sion objective (Carl Zeiss AG, Jena, Germany). For imaging
the expression of GFP and the chloroplast autofluores-
cence, excitation lines of an argon ion laser of 488 nm were
used with a 505 ⁄ 550-nm bandpass filter for GFP and exci-
tation lines of an HeNe laser of 543 nm with a 585 long
pass filter for chloroplast autofluorescence in the multitrack
facility of the microscope.
Acknowledgements
We are grateful to the following people from the Uni-
versita
¨
t Regensburg: Michael Leiss for experimental
assistance in the initial phase of the project, Peter Heg-
emann for help with spectrofluorimetry, Guido Gross-
mann for assistance with confocal microscopy, and
Gerhard Lehmann for technical assistance. We are
indebted to Ju
¨
rgen Stolz for critically reading the
manuscript. We thank Kirk Apt (Martek Biosciences,
Columbia, MD, USA) for providing an egfp-containing
plasmid. This work was supported by the DFG (SFB-
521-A2) and the Fonds der Chemischen Industrie.
N. Poulsen and N. Kro
¨
ger Inducible promoter for transgenic diatoms

FEBS Journal 272 (2005) 3413–3423 ª 2005 FEBS 3421
References
1 Field C, Behrenfeld M, Randerson J & Falkowski P
(1998) Primary production of the biosphere: integrat-
ing terrestrial and oceanic components. Science 281,
237–240.
2 Nelson D, Treguer P, Brzezinski M, Leynaert A & Que-
guiner B (1995) Production and dissolution of biogenic
silica in the ocean: revised global estimates, comparisons
with regional data and relationship to biogeneic sedi-
mentation. Global Biogeochem Cycles 9, 359–372.
3 Sumper M & Kro
¨
ger N (2004) Silica formation in dia-
toms: the function of long-chain polyamines and silaf-
fins. J Mater Chem 14, 1–8.
4 Coradin T & Lopez P (2003) Biogenic silica patterning:
simple chemistry or subtle biology? Chembiochem 4,
251–259.
5 Hildebrand M & Wetherbee R (2003) Components and
control of silicification in diatoms. Prog Mol Subcell
Biol 33, 11–57.
6 Armbrust EV, Berges JA, Bowler C, Green BR, Marti-
nez D, Putnam NH, Zhou S, Allen AE, Apt KE,
Bechner M, et al. (2004) The genome of the diatom
Thalassiosira pseudonana: ecology, evolution, and meta-
bolism. Science 306, 79–86.
7 Montsant A, Jabbari K, Maheswari U & Bowler C
(2005) Comparative genomics of the pennate diatom
Phaeodactylum tricornutum. Plant Physiol 137, 500–513.

8 Apt K, Kroth Pancic P & Grossman A (1996) Stable
nuclear transformation of the diatom Phaeodactylum tri-
cornutum. Mol Gen Genet 252, 572–579.
9 Fischer H, Robl I, Sumper M & Kro
¨
ger N (1999) Tar-
geting and covalent modification of cell wall and mem-
brane proteins heterologously expressed in the diatom
Cylindrotheca fusiformis (Bacillariophyceae). J Phycol
35, 113–120.
10 Dunahay T, Jarvis E & Roessler P (1995) Genetic trans-
formation of the diatoms Cyclotella cryptica and Navi-
cula saprophila. J Phycol 31, 1004–1012.
11 Kilian O & Kroth PG (2005) Identification and charac-
terization of a new conserved motif within the prese-
quence of proteins targeted into complex diatom
plastids. Plant J 41, 175–183.
12 Lomas M (2004) Nitrate reductase and urease enzyme
activity in the marine diatom Thalassiosira weissflogii
(Bacillariophyceae): interactions among nitrogen sub-
strates. Mar Biol 144, 37–44.
13 Gao Y, Smith GJ & Alberte RS (1993) Nitrate reduc-
tase from the marine diatom Skeletonema costatum (bio-
chemical and immunological characterization). Plant
Physiol 103, 1437–1445.
14 Quesada A & Fernandez E (1994) Expression of nitrate
assimilation related genes in Chlamydomonas reinhardtii.
Plant Mol Biol 24, 185–194.
15 Fernandez E, Schnell R, Ranum LP, Hussey SC, Silflow
CD & Lefebvre PA (1989) Isolation and characteriza-

tion of the nitrate reductase structural gene of Chlamy-
domonas reinhardtii. Proc Natl Acad Sci USA 86,
6449–6453.
16 Cannons AC & Shiflett SD (2001) Transcriptional regu-
lation of the nitrate reductase gene in Chlorella vulgaris:
identification of regulatory elements controlling expres-
sion. Curr Genet 40, 128–135.
17 Song B & Ward B (2004) Molecular characterization of
the assimilatory nitrate reductase gene and its expres-
sion in the marine green alga Dunaliella tertiolecta
(Chlorophyceae). J Phycol 40, 721–731.
18 Allen AE, Ward BB & Song BK (2005) Characteriza-
tion of diatom (Bacillariophyceae) nitrate reductase
genes and their detection in marine phytoplankton com-
munities. J Phycol 41, 95–104.
19 Zaslavskaia L, Lippmeier J, Shih C, Ehrhardt D, Gross-
man A & Apt K (2001) Trophic obligate conversion of
an photoautotrophic organism through metabolic engi-
neering. Science 292, 2073–2075.
20 Zhang G, Gurtu V & Kain SR (1996) An enhanced
green fluorescent protein allows sensitive detection of
gene transfer in mammalian cells. Biochem Biophys Res
Commun 227, 707–711.
21 Reid BG & Flynn GC (1997) Chromophore formation
in green fluorescent protein. Biochemistry 36, 6786–
6791.
22 Kro
¨
ger N (2001) The sweetness of diatom molecular
engineering. J Phycol 37, 657–658.

23 Loppes R, Radoux M, Ohresser MC & Matagne RF
(1999) Transcriptional regulation of the Nia1 gene
encoding nitrate reductase in Chlamydomonas reinhard-
tii: effects of various environmental factors on the
expression of a reporter gene under the control of the
Nia1 promoter. Plant Mol Biol 41, 701–711.
24 Fernandez E & Cardenas J (1982) Regulation of the
nitrate-reducing system enzymes in wild-type and
mutant strains of Chlamydomonas reinhardtii. Mol Gen
Genet 186, 164–169.
25 Fernandez E & Cardenas J (1989) Genetics and regula-
tory aspects of nitrate assimilation in algae. Molecular
and Genetic Aspects of Nitrate Assimilation (Wray JL &
Kinghorn JR, eds), pp. 101–124. Oxford Science Publi-
cations, Oxford.
26 Wilkie GS, Dickson KS & Gray NK (2003) Regulation
of mRNA translation by 5¢- and 3¢-UTR-binding fac-
tors. Trends Biochem Sci 28, 182–188.
27 Campbell WH (1999) Nitrate reductase structure, func-
tion and regulation: bridging the gap between biochem-
istry and physiology. Annu Rev Plant Physiol Plant Mol
Biol 50, 277–303.
28 Hildebrand M & Dahlin K (2000) Nitrate transporter
genes from the diatom Cylindrotheca fusiformis (Bacil-
Inducible promoter for transgenic diatoms N. Poulsen and N. Kro
¨
ger
3422 FEBS Journal 272 (2005) 3413–3423 ª 2005 FEBS
lariophycea): mRNA levels controlled by nitrogen
source and by the cell cycle. J Phycol 36, 702–713.

29 Hildebrand M (2005) Cloning and functional characteri-
zation of ammonium transporters from the marine
diatom Cylindrotheca fusiformis (Bacillariophceae).
J Phycol 41, 105–113.
30 Li J, Wang S, VanDusen WJ, Schultz LD, George HA,
Herber WK, Chae HJ, Bentley WE & Rao G (2000)
Green fluorescent protein in Saccharomyces cerevisiae:
real-time studies of the GAL1 promoter. Biotechnol
Bioeng 70, 187–196.
31 Yarger JG, Halverson HO & Hopper JE (1984) Regula-
tion of galactokinase (GAL1) enzyme accumulation in
Saccharomyces cerevisiae. Mol Cell Biochem 61 , 173–
182.
32 Martin-Jezequel V, Hildebrand M & Brzezinski MA
(2000) Silicon metabolism in diatoms: implications for
growth. J Phycol 36, 821–840.
33 Kro
¨
ger N, Bergsdorf C & Sumper M (1996) Frustulins:
domain conservation in a protein family associated with
diatom cell walls. Eur J Biochem 239 , 259–264.
34 Gatignol A, Durand H & Tiraby G (1988) Bleomycin
resistance conferred by a drug-binding protein. FEBS
Lett 230, 171–175.
35 Poulsen N & Kro
¨
ger N (2004) Silica morphogenesis by
alternative processing of silaffins in the diatom Thalas-
siosira pseudonana. J Biol Chem 279, 42993–42999.
36 Horton RM, Cai ZL, Ho SN & Pease LR (1990)

Gene splicing by overlap extension: tailor-made genes
using the polymerase chain reaction. Biotechniques 8,
528–535.
37 Laemmli UK (1970) Cleavage of structural proteins
during the assembly of the head of bacteriophage T4.
Nature 227, 680–685.
38 Wilson CM (1983) Staining of proteins on gels: compar-
isons of dyes and procedures. Methods Enzymol 91,
236–247.
39 Kro
¨
ger N & Wetherbee R (2000) Pleuralins are involved
in theca differentiation in the diatom Cylindrotheca fusi-
formis. Protist 151, 263–273.
40 Thompson JD, Higgins DG & Gibson TJ (1994) Clus-
tal-W: improving the sensitivity of progressive multiple
sequence alignment through sequence weighting, posi-
tion-specific gap penalties and weight matrix choice.
Nucleic Acids Res 22, 4673–4680.
N. Poulsen and N. Kro
¨
ger Inducible promoter for transgenic diatoms
FEBS Journal 272 (2005) 3413–3423 ª 2005 FEBS 3423