HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Dunstan, S. L. Articles by Campbell, A. K. Search for Related Content PubMed PubMed Citation Articles by Dunstan, S. L. Articles by Campbell, A. K. Social Bookmarking                      What's this?

J Biol Chem, Vol. 275, Issue 13, 9403-9409, March 31, 2000

Cloning and Expression of the Bioluminescent Photoprotein Pholasin from the Bivalve Mollusc Pholas dactylus^*

Sarah L. Dunstan, Graciela B. Sala-Newby^§, Alexandra Bermúdez Fajardo, Kathryn M. Taylor, and Anthony K. Campbell^¶

From the Department of Medical Biochemistry, University of Wales College of Medicine, Heath Park, Cardiff CF14 4XN, United Kingdom

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Pholasin is the photoprotein responsible for luminescence in the bivalve Pholas dactylus and consists of a luciferin tightly bound to a glycosylated protein. It is a sensitive indicator of reactive oxygen species. A full-length clone encoding apopholasin was isolated from a P. dactylus light organ cDNA library. The unprocessed apoprotein contained 225 amino acids, starting with a signal peptide of 20 amino acids, 3 predicted N-linked glycosylation sites, 1 O-linked site, no histidines, and 7 cysteines. The recombinant apoprotein was expressed in cell extracts and insect cells. The size of the apoprotein expressed in cell extracts and the cytosol of insect cells was 26 kDa but that of the fully processed protein was 34 kDa, as was native pholasin. Both the processed and unprocessed recombinant apoproteins were recognized by a polyclonal antibody raised against native pholasin. Acid methanol extracts from Pholas added to recombinant apoprotein resulted in chemiluminescence triggered by sodium hypochlorite but not photoprotein formation. These results have important implications in understanding the molecular evolution of bioluminescence and will allow the development of recombinant pholasin as an intracellular indicator of reactive oxygen species.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Pholas dactylus is a marine, rock-boring, bivalve mollusc, which squirts a blue, luminous secretion (_max 490 nm) from its siphon when disturbed. Its luminescence has been known since ancient times (1) and has an historic place in the biochemistry of bioluminescence, since it was the first organism used to describe the oxygen-requiring luciferin-luciferase reaction that is universal in bioluminescence (2). In fact, the "luciferin" from Pholas that Dubois named is a photoprotein containing the luciferin, a small organic molecule that is the true light emitter. However, the precise chemistry of the luminescence and its evolutionary origin are unknown.

Pholas is essentially a European species, being common in soft, sedimentary rocks along the south coast of United Kingdom from Folkestone to Plymouth and is also found in the Severn estuary. It also occurs in substantial numbers along the French coast, particularly Brittany, and in the Mediterranean. The adult mollusc lives its entire life buried in soft rock and can only be collected at a good spring tide. At night the majority of organisms are seen as a starlight display in rocks never exposed at low water. Light emission has been observed in the wild to change during the lunar cycle, being difficult to provoke at full moon.¹ It is not known whether Pholas luminescence follows a circadian rhythm as do many other luminous organisms (1, 3, 4).

The two main components of the luminescent reaction, pholasin and Pholas luciferase, are found in triangular light organs on the surface of the muscles and in light cords along the edge of the siphon (5, 6). The luciferase is a 150-kDa copper-containing enzyme with peroxidase-like properties (7). Pholasin consists of a luciferin (a small organic moiety) bound tightly to an apoprotein and binds to the luciferase to generate light in the presence of oxygen (Equation 1) (7-9).

(Eq. 1)

However, pholasin alone generates a continuously weak glow (k = 8.12 ± 0.87 × 10⁶ s¹). Pholasin chemiluminescence can be enhanced up to 1 million times by addition of its luciferase or enzymes that can generate reactive oxygen species (8). Pholasin, like aequorin, is a photoprotein (1, 10); the total light emitted is directly proportional to the amount of luminescent protein present. Thus, unlike a normal enzyme, pholasin turns over only once (Equation 2).

(Eq. 2)

Pholasin chemiluminescence provides an indicator for ROS² in biological systems, such as the superoxide released by neutrophils as a result of the oxidative burst. Pholasin has been proved to be a very sensitive indicator, capable of measuring the oxidative burst of one phagocyte (8), or the intracellular oxidative stress of a single myocyte (11). Measurement of ROS in vivo should be a valuable tool to study the involvement of ROS in many pathological conditions including cancer (12) and Alzheimer's disease (13) and also in apoptosis (14, 15), the regulation of phosphorylation events (16) and in aging (17-19).

We have cloned, sequenced, and expressed the apoprotein of pholasin. We reconstituted the luminescent protein using putative luciferin isolated from Pholas. The lack of overall sequence similarity with all the other known luminescent proteins suggests that Pholas luminescence is another example of the independent evolution of bioluminescence.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Live specimens of P. dactylus were collected at Charmouth Beach, Dorset, UK, during periods of low tide and kept in filtered seawater at 8-10 °C. Oligonucleotide primers were prepared using an Applied Biosystems 392 DNA synthesizer and purified as "trityl off." All degenerate oligonucleotides (the degenerate oligonucleotides used are: I, inosine; N, any base; H, A/C/T; R, A/G; and Y, C/T) were obtained from Perkin-Elmer Applied Biosystems. The degenerate oligonucleotides 1, ACI ATH TTY TTY CAR GT; 2, CAR GAR GAR GGN ACI GA; 2A, TCI GTN CCY TCY TCY TG; and N-term, TTY AAY GTI GAY TGG ATG hybridized to the portions of cDNA encoding the peptides sequenced (Fig. 1). The oligonucleotides T3 and T7 hybridize to the Bluescript phagemid 5' and 3', respectively, to the inserted cDNA sequence. Reagents used to construct the cDNA library were from Stratagene (La Jolla, CA). [-³²P]dATP, [-³²P]ATP, and [³⁵S]methionine and Hybond N⁺ nylon membrane were purchased from Amersham Pharmacia Biotech. The coupled transcription/translation (TNT) kit was from Promega (Uppsala, Sweden). The pVL1393 and BaculoGold plasmid and the transfection kit were obtained from PharMingen (San Diego, CA). High Five^TM cells (BTI-TN-5B1-4) were purchased from Invitrogen (Leek, The Netherlands). ExCell 401 insect cell medium was from JRH Biosciences (Lenexa), and TNM-FH insect cell medium was from Sigma.

Restriction enzymes and DNA modification enzymes were obtained from Promega (Southampton, UK). DNA purification kits were from Qiagen (Crawley, UK). All other reagents were Analar quality and obtained from Fisher and Sigma.

Characterization of Native Pholasin-- Native pholasin was purified using the method of Roberts et al. (8). Pholasin was subjected to 10% SDS-PAGE and excised, and internal peptides were generated by in situ trypsin digestion using the method of Rosenfeld et al. (20). Peptides that appeared pure following separation by reverse phase high pressure liquid chromatography were subjected to N-terminal sequencing by Edman degradation on an Applied Biosystems model 471A Protein Sequenator. The sequence of the N terminus of the protein was generated by solid state amino acid sequencing on an Applied Biosystems model 476A Protein Sequencer (21).

Construction of a cDNA Library-- Pholas light organs were dissected out of the animals soon after collection and immediately snap-frozen in liquid nitrogen. Total RNA was purified from this tissue using the method of Groppe and Morse (22) that was developed for RNA extraction from molluscan material high in ribonucleases and mucus. Attempts to purify RNA from light organ tissue using Ultraspec^TM reagent (Biotecx) or the method of Chirgwin et al. (23) resulted in the degradation of the RNA (data not shown). Messenger RNA (mRNA) was selectively purified from total RNA using a column of Oligotex beads (Qiagen) according to the manufacturer's instructions. The cDNA complementary to this mRNA was generated by oligo(dT)-primed reverse transcription using the ZAP cDNA synthesis kit (Strategene). A unidirectional library was constructed in the  ZAP phage.

Generation of Probes and Library Screening-- Degenerate oligonucleotides were designed from portions of amino acid sequence. cDNA was produced by reverse transcription of mRNA and amplified by step down polymerase chain reaction (24) with the degenerate primers N-term and 2A. This produced a DNA fragment of approximately 300 base pairs, which was purified by extraction from an agarose gel. This fragment was subcloned into pGEMT. A positive clone was selected and sequenced. Non-degenerate oligonucleotide probes were designed from this cDNA fragment. The cDNA library was screened with the non-degenerate oligonucleotides according to the method of Sambrook et al. (25). Positive clones were excised into the Bluescript phagemid in vivo by simultaneous infection of the SOLR strain of Escherichia coli with the  vector and the f1 bacteriophage and were then sequenced in both directions.

Identification of the Gene Encoding Pholasin-- Total genomic DNA was isolated from whole Pholas organisms following the methods of Winnepenninckx et al. (26) and Bowtell (27). The resulting preparations of gDNA were amplified with primers designed from the cDNA sequence using proofreading polymerases (rTth DNA polymerase XL and BioXAct polymerase, respectively). The polymerase chain reaction products were purified by gel extraction (Qiagen) and sequenced in both directions.

Expression and Characterization of the Recombinant Protein-- A polyclonal antibody was raised in rabbits by immunization with native pholasin bound to nitrocellulose. Briefly, native purified protein was fractionated by SDS-PAGE and transferred onto nitrocellulose. The nitrocellulose-containing pholasin was excised and dissolved in a minimum volume of dimethyl sulfoxide. An equal volume of sterile phosphate-buffered saline was added, and the resulting suspension, which constituted the immunogen, was sonicated, ready for immunization of the rabbits.

Expression in Cell Extracts-- Apoprotein was produced in cell extracts by coupled in vitro transcription and translation (TNT^TM) of Bluescript clone 40 using T3 polymerase (T3 promoter is included in the plasmid) and rabbit reticulocyte lysate according to the manufacturer's instructions. Microsomal membranes were added to the translation mix to perform post-translational modifications. [³⁵S]Methionine (>1000Ci/mmol) was included to aid quantification and characterization of the protein by SDS-PAGE and autoradiography.

Expression in Insect Cells-- Insect cells were maintained and infected in Excell 401 protein-free medium, pH 6.2, at 27 °C. BamHI sites were engineered onto the cDNA encoding clone 40 with and without the native signal peptide via polymerase chain reaction with a proofreading polymerase (BioXAct, Bioline). This cDNA was subcloned into the BglII site of the pVL1393 vector. Co-transfection of the construct and the BaculoGold plasmid into HighFive insect cells was performed as described previously (28) allowing the formation of active virus in vivo. Supernatant from the virally infected insect cells was used to infect further flasks of cells for protein production.

The medium was harvested at 4 or 6 days post-infection. The apoprotein in the medium was precipitated with saturating ammonium sulfate at 4 °C for 1 h, with stirring. This precipitate was pelleted by centrifugation at 47,800 × g for 30 min at 4 °C. The pellet was resuspended in 2.5 ml of 50 mM sodium phosphate, pH 6, and the ammonium sulfate was removed with a PD10 Sephadex column (Amersham Pharmacia Biotech). Cells expressing protein lacking a signal peptide were pelleted by centrifugation and lysed in Laemmli buffer. These apoproteins were characterized by SDS-PAGE and Western blot, probed with the anti-pholasin antibody, detected with alkaline phosphatase, and quantified by densitometry in comparison with native pholasin.

Extraction of the Luciferin-- As the luciferin may be unstable, the tissue destined for extraction was snap-frozen immediately and stored at 70 °C. Five whole P. dactylus organisms were homogenized in 22.5 ml of 50 mM HEPES, pH 7.5, in a Polytron homogenizer at 4 °C. An equal volume of methanol (high pressure liquid chromatography grade) and 5 ml of 1 N HCl were added to the homogenate followed by incubation on ice for 2 h. The resulting particulate material was pelleted by centrifugation at 47,800 × g for 30 min at 4 °C, and the supernatant was collected. This supernatant was dried in a rotatory evaporator, and the residue was resuspended in 5 ml of 50 mM HEPES, pH 7.5. The sample was loaded onto a SEP-PAC cartridge (Waters) and washed with 5 ml of ethyl acetate. Fractions eluted with methanol were pooled and dried in a vacuum concentrator and stored at 70 °C. Prior to a reactivation experiment, the dried extract was resuspended in 1 ml of HEPES buffer, pH 7.5.

Reactivation of the Recombinant Apoprotein with the Putative Luciferin-- The secreted apoprotein (10.89 µg in 10 µl) was incubated for up to 24 h at 4 °C in the presence of 80 µl of 10 mM Tris-HCl, pH 7.5, 500 mM NaCl, 1 mM EDTA, 0.1% gelatin, and 10 mM ascorbate plus 10 µl of the above extracted luciferin. The final pH of the reactivation mixture was 3.5. Controls consisted of buffer, apoprotein, or luciferin extract only. Samples were assayed for chemiluminescence after 0, 1, 2, 6, and 24 h (Fig. 4). Apoprotein reactivated for 18 h as above was loaded onto 0.025-µm MCE filters (Millipore, Bedford, MA) floating on 10 mM Tris-HCl, pH 7.5, 500 mM NaCl, 1 mM EDTA, 0.1% gelatin, and 10 mM ascorbate and incubated at 4 °C (drop dialysis). Chemiluminescent activity was determined after 18 h. Controls were native pholasin, apopholasin dialyzed without luciferin, and non-dialyzed apopholasin stored at 4 °C. Re-addition of luciferin extract was also investigated to assess the stability of apopholasin during dialysis. Reactivation at a range of pH from 3 to 6 was also investigated using 200 mM phosphate/citrate mixtures.

Quantification of Reconstituted Pholasin-- The chemiluminescence emitted was detected and quantified in a purpose-built luminometer (1). A 50-µl aliquot of reactivation mix or controls (sample without apoprotein, without luciferin, or with neither) was added to 450 µl of 0.1 M sodium barbitone, pH 9, and immediately placed in the luminometer housing. Background light was measured (10 s), and pholasin chemiluminescence was triggered by the injection of 500 µl of 2% NaOCl. Light emission (counts per 10 s) was monitored for a further 2 min.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Isolation and Characterization of the Pholasin cDNA-- The primary Pholas cDNA library contained 10⁶ clones. The non-degenerate oligonucleotide probes hybridized to 2.8 × 10³% of the plaques of the amplified library. The positive clones were excised into the Bluescript phagemid (pBluescript® SK (+/)), and three clones were completely sequenced (termed 40, 3, and 5) (Fig. 1). The open reading frame of these three clones encoded an identical protein of 225 residues from the first methionine to a stop codon. The first methionine was flanked by a Kozak consensus sequence for the initiation of translation (29). The protein encoded by these clones contained all internal peptide sequences determined by sequencing of the native protein and had the same amino acid composition as the native protein (30) (Table I). A signal peptide of 20 amino acids was identified between the first methionine and the N terminus of the mature protein determined by amino acid sequencing. A neural network trained to identify eukaryotic signal peptides suggested that the most likely cleavage site is between positions 20 and 21 (GSG-EE), in agreement with the experimental data (31). Three potential N-linked glycosylation sites with the consensus triplet sequence Asn-Xaa-(Ser/Thr) (32) have been identified in the protein. Thr-216 was identified as a potential site of O-linked glycosylation (33).

View larger version (46K): [in this window] [in a new window]   Fig. 1.   The complete nucleotide and derived amino acid sequence of clone 40. Partial amino acid sequences corresponding to tryptic peptides are underlined. The N-terminal sequence of the purified protein is double underlined. The numbering of the amino acids is based on the position of the first predicted methionine. The stop codon is in bold; an asterisk marks the stop. Amino acids that are putative N-linked glycosylation sites are shown in square brackets. A potential O-linked glycosylation site is shown in parentheses. Bases that are absent or different in clones 3 or 5 are shown in italics. Clone 3 had the first two bases missing at the 5' end of the cDNA and finished at adenosine 804. Clone 5 did not have TATGA at positions 829-833. Both clone 3 and clone 5 have an A-G change at 186, a G-T change at 534, and a T-C change at position 705. Genomic DNA sequence was obtained from bases 17 to 832.

Sequence similarity between the complete recombinant protein and any known protein in the Swiss-Prot data base was very small (<5%). Small segments of the protein sequence were also compared against the sequences in the Swiss-Prot data base using the BLASTP program which identified several proteins with regions of high sequence similarity to regions of the cloned protein. These included several proteins that interact with nucleotides (Table II).

The amino acid sequence of the cloned protein was then searched for known motif sequences by comparison with the PROSITE data base by the MOTIF algorithm. The sequence of the clones did not contain any of these signatures present in the PROSITE data base (34). The C terminus, however, was highly acidic and may bind calcium (35, 36).

The bioluminescent proteins of several bioluminescent organisms have been cloned. The amino acid sequence of pholasin was compared with the sequences of these cloned bioluminescent proteins. A small region of similarity was found between the recombinant protein and the putative luciferin-binding sites of Vargula luciferase (37) and Renilla LBP (38) (Table II) but not to any other bioluminescent protein.

The amplification of the pholasin gene from genomic DNA resulted in a product of the same size as the cDNA. The gDNA sequence was found be identical to that of clone 40 from the cDNA library (Fig. 1), indicating the absence of introns.

Expression and Reactivation of Recombinant Pholasin-- The expression of the protein encoded by clone 40 successfully generated recombinant apopholasin in cell extracts (Fig. 2), prokaryotic cells (data not shown), and eukaryotic cells (Fig. 3). The molecular mass of recombinant pholasin calculated from the amino acid sequence was 23.5 kDa. The average molecular mass of the unprocessed protein generated by in vitro transcription and translation in cell extracts was approximately 26 kDa (Fig. 2). In the presence of canine microsomal membranes, the molecular mass of the recombinant protein increased to approximately 34 kDa, the same as that of native pholasin. Apopholasin lacking a signal peptide remained in the cytosol of insect cells and had a molecular mass of approximately 26 kDa. The addition of a signal peptide resulted in the secretion of the protein into the insect cell medium and an increase of the molecular mass to approximately 33 kDa. Gels run in the absence of -mercaptoethanol revealed the presence of high molecular weight aggregates on Western blots. Less than 0.1% of the recombinant protein was monomeric on non-reducing gels. Recombinant apopholasin expressed in E. coli formed inclusion bodies, which were only disrupted in the presence of 6 M guanidinium hydrochloride, and the removal of this denaturant caused the reoccurrence of protein aggregation (data not shown).

View larger version (51K): [in this window] [in a new window]   Fig. 2.   In vitro transcription and translation of DNA encoding for pholasin. The protein encoded by clone 40 was generated by in vitro TNT and labeled by the incorporation of [35S]methionine. The figure shows the protein products separated by SDS-PAGE and autoradiographed. Transcription and translation of clone 40 resulted in the generation of a protein of approximately 26 kDa, which was not present in the negative controls lacking a template (arrow 1). The addition of microsomal membranes to the TNT reaction resulted in a processed protein of approximately 34 kDa (arrow 2).

View larger version (30K): [in this window] [in a new window]   Fig. 3.   Immunoblot of insect cell medium harvested 4 days post-infection. Native pholasin 5 (lane 1) and 1 µg (lane 2), 18 µg of recombinant pholasin secreted into the medium and partially purified (lane 3), or from an insect cell extract (lane 4) and negative controls, and medium from insect cells infected with a virus containing C9 cDNA (lane 5), or uninfected insect cells (lane 6), were subjected to SDS-PAGE in reducing (blots a and c) or non-reducing conditions (blot b). Western blots were probed with the anti-pholasin polyclonal antiserum. Arrows indicate a dimer of native pholasin and aggregation of recombinant pholasin in a non-reducing gel.

The addition of an acid/methanol extract of whole P. dactylus to the recombinant apopholasin resulted in an increase in chemiluminescence (Fig. 4), indicating the production of an active protein. Controls using acid/methanol extract or apoprotein alone, always in the presence of carrier protein, showed chemiluminescent activity some 4-fold lower than with apopholasin and luciferin. Furthermore, addition of bovine serum albumin (0.1%) had no effect on the apopholasin chemiluminescence or the controls, confirming that the extract containing the putative luciferin reacted specifically with the apoprotein. However, the kinetics of the light emitted from the recombinant apopholasin with luciferin were very different from that of native pholasin. Native pholasin chemiluminescence triggered by hypochlorite resulted in a flash, >95% complete within 40 s. In contrast, addition of hypochlorite to the apopholasin with luciferin resulted in a much slower chemiluminescence, decaying to half-maximum within 120 s. The decay was approximately first order (k = 0.006 s¹). Furthermore, there was no detectable difference in activity at any of the time points (0-24 h). This suggested that, under these conditions, true photoprotein was not formed. This was confirmed by drop dialysis, which resulted in 98 ± 1% (mean ± S.E., n = 3) loss in chemiluminescent activity. This showed that no photoprotein had been formed, since no loss of activity was detected on drop dialysis of native pholasin. Furthermore, reactivation of the dialyzed apoprotein with fresh luciferin resulted in only 15-17% (n = 2) of the apoprotein chemiluminescence being recovered from the filter.

View larger version (25K): [in this window] [in a new window]   Fig. 4.   Time course of hypochlorite-triggered chemiluminescence. A time course of apopholasin reactivation was performed by incubating partially purified recombinant pholasin secreted by insect cells with acid/methanol extract of P. dactylus () for 0 (solid line), 1 (- - -), 2 (- - -), 6 (- -), or 24 h (- -), or incubated without extract (). Controls of buffers only with no apopholasin or acid/methanol extract () and extract alone () were treated in identical conditions. Gelatin (0.1%) was present as protein carrier under all conditions (see "Experimental Procedures"). Chemiluminescence was triggered by the addition of 2% sodium hypochlorite at 10 s and is shown as chemiluminescent counts minus background light. A typical curve obtained by hypochlorite triggering of native pholasin is also shown (×). A representative experiment carried out in duplicate is shown. E, einstein.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

This paper reports the cloning of the cDNA encoding the apoprotein responsible for light emission in the luminous bivalve mollusc P. dactylus. Several criteria were used to confirm that this cDNA did code for apopholasin as follows: the amino acid composition of the recombinant protein and the size of the processed protein matched that of native pholasin, the recombinant protein and native pholasin are both secreted proteins, and the reactivated recombinant protein is chemiluminescent. The predicted amino acid composition was that previously reported for purified pholasin (Table I). The fragments of sequence obtained by protein sequencing were present in the sequence of the recombinant protein. The unprocessed protein had a molecular mass of 26 kDa. However, expression in cell extracts with microsomal membranes resulted in a protein with a molecular mass of 34 kDa, the same as that of native pholasin. Furthermore, both the processed and unprocessed protein cross-reacted with antibodies raised to native pholasin. The protein contained glycosylation sites and a putative signal peptide sequence consistent with its identity as a secretory protein. In vitro transcription-translation in the presence of canine microsomal membranes, or the targeting of protein to the insect cell secretory pathway, resulted in an increase in the size of the recombinant protein likely to be due to glycosylation. Recombinant apopholasin showed a tendency to aggregate in non-reducing gels. This aggregation may be caused by the lack of the luciferin, as aequorin is very unstable without its luciferin (39), or by the high mannose glycosylation, as the recombinant pholasin in the cytosol demonstrated the least aggregation. It is interesting to note that other workers have shown that the hemagglutinin protein for an influenza strain aggregated when half the N-linked sugars were incomplete (40), whereas the unglycosylated protein can fold normally (41). A further explanation may be that the production of recombinant protein overwhelmed the chaperones of the cell, as the expression of enzymes such as protein-disulfide isomerase has increased the expression of active protein in insect cells (42).

A further finding reported here was the isolation of a small moiety by acid methanol extraction that can generate light when added to the recombinant apopholasin (Fig. 4). In native pholasin the luciferin is tightly bound, probably covalently, to the apoprotein. Muller and Campbell (43) were successful in extracting a small fluorescent molecule from native pholasin using butanol.

The kinetics of the light emitted from the recombinant apopholasin with luciferin were very different from that of native pholasin. Native pholasin chemiluminescence resulted in a flash, whereas apopholasin with luciferin resulted in a much slower light emission (Fig. 4). This, together with the lack of any detectable difference in activity at any of the time points of reactivation (0-24 h), and complete loss of chemiluminescence after dialysis, indicated that, under these conditions, true photoprotein was not formed. The loss of chemiluminescent activity upon dialysis was the result of both loss of luciferin (that was not covalently bound) and denaturation of the apopholasin on the filter. We described a similar loss of activity of another photoprotein, apo-aequorin, expressed in the cytosol of HeLa cells which was partially prevented by growing the cells in the presence of coelenterazine (39).

The apopholasin was therefore acting as a luciferase. Thus there may be further, as yet unknown, cofactors and chemical events required to form the full photoprotein. The pH used in the reactivation experiments was acid. The use of a range of pH showed that an increase in pH from 3 to 6 resulted in a decrease of apopholasin chemiluminescence of 40% (data not shown). This was not entirely surprising since pholasin is secreted into the siphon of Pholas, and secretory granules have an acid pH (44). The fact that the clones had some sequence similarity with proteins that interact with nucleotides (Table II) may suggest that pholasin requires a cofactor for reactivation. However, there was no P-loop binding motif ((A,G)X₄GK(S,T) or (A)X₄GK(T)) in the amino acid sequence of these clones, although not all ATP-binding proteins contain this motif. Neither did the cloned apoprotein contain the GXGXXG phosphate binding consensus sequence for the binding NAD (45). However, it should now be possible to use reactivation of recombinant apopholasin as an assay to isolate sufficient luciferin for chemical identification. Apopholasin contained seven cysteine residues, so there must be at least one free cysteine in the completely folded active protein, which might contribute to aggregation via intermolecular disulfide bonds in non-reducing conditions. In addition, cysteines have been implicated in several other bioluminescent reactions. Substitution of the cysteine residues of the photoprotein aequorin with serine resulted in a reduction of chemiluminescence, possibly as a result of reduced luciferin (coelenterazine) binding (46). In addition, there was close sequence homology between the cysteine domains of the luciferases in seven species of firefly. However, firefly luciferase is active after cysteine modification (47). It is also interesting to note that pholasin contains no histidine residues. The C-terminal tryptophan is unusual in proteins. This, together with a proline eight amino acids upstream, is consistent with the need to have a solvent cage that allows oxygen access to the luciferin but protects the excited state generated by the chemiluminescent reaction from quenching by molecules such as oxygen and water in the bulk solvent. GFP exemplifies this solvent cage principle, requiring a rigid barrel to maintain the correct physicochemical environment for fluorescence. Based on mutagenesis experiments (48, 49), we have proposed that in beetle luciferases and coelenterazine photoproteins the C terminus is a crucial part of this solvent cage. Like most bioluminescent photoproteins or luciferases, pholasin has a slow maximum turnover number (~0.8 s¹) compared with most enzymes.

No introns were detected in the genomic DNA coding for apopholasin. This is consistent with findings in the genes of other bioluminescent proteins where the introns are either very short, as in the luminous beetles (50), or absent, as in aequorin (51, 52) and both the luciferase and the LBP of the dinoflagellate Gonyaulax polyhedra (53, 54).

Luminescence is found in molluscs mainly in 20 squid genera, either as symbiotic bacteria or generated endogenously (1, 55), but only in two bivalves, Pholas and Rotocellaria. Five luciferin families have so far been identified in bioluminescence (1, 56) as follows: aldehydes; imidazolopyrazines, the most common chemistry responsible for bioluminescence in the sea; benzothiazole, unique to luminous beetles; linear tetrapyroles; and flavins. Coelenterazine has been identified as the luciferin in some squid (57, 58). Although the small homology of apopholasin with Vargula luciferase and Renilla luciferin-binding protein might indicate that pholasin had an imidazolopyrazine luciferin, Muller and Campbell (43) were unable to reactivate oxypholasin with imidazolopyrazine luciferins, and no coelenterazine nor Vargula luciferin were detected in methanol extracts of P. dactylus. Imidazolopyrazines are proposed to form by cyclization of a tripeptide (FYY or RIW), similar to fluor formation in GFP (SYG), but with both a six- and five-membered ring instead of simply the single five-membered imidazo ring in GFP. The possibility, however, existed that some of the unknown luciferins, including that of pholasin, might be synthesized by cyclization of other tripeptides. This seemed not to be the case in Pholas, since no light emission over background was obtained from apopholasin incubated in the absence of acid methanol extract. Spectral absorbance suggested a flavin or catechol as the most promising candidate for pholasin luciferin (43). The evolutionary origin of bioluminescence presents a considerable challenge for the Darwinian principle of evolution by natural selection (55, 59). The current data support the hypothesis that bioluminescence has originated several times during evolution, since pholasin has no significant homology with other known bioluminescent proteins. It will be interesting to examine other non-bioluminescent bivalves to see if they contain proteins that might show sequence similarity to apopholasin. In order to generate chemiluminescence that is susceptible to natural selection, a solvent cage with a high quantum yield is necessary, which contains a substrate already present in the organism, perhaps as an antioxidant (60).

The results here provide a basis for discovering the chemical structure of Pholas luciferin and the evolutionary origin of its bioluminescence. Furthermore, the ability to express recombinant pholasin in defined compartments of live cells will enable reactive oxygen species to be imaged within individual cells and intact organs for the first time.

	ACKNOWLEDGEMENTS

We thank Dr. M. Wilkinson (University of Liverpool) for sequencing key amino acids for cloning and Dr. C. Van den Berg (University of Wales College of Medicine) for initial help with sequencing.

	FOOTNOTES

^* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in this paper has been submitted to the GenBank^TM/EMBL Data Bank with accession number(s) AJ131051, AJ131052, AJ131053, and AJ131054.

Supported by postgraduate studentships provided by Professor G. H. Elder, C.B.E., and the University of Wales College of Medicine.

^§ Supported by the BHF. Present address: Bristol Heart Institute, University of Bristol, Bristol Royal Infirmary, BS2 8HW, UK.

^¶ To whom correspondence should be addressed. Tel.: 44(0)1222 742951; Fax: 44(0)1222 745440; E-mail: campbellak@cardiff.ac.uk.

¹ A. K. Campbell, unpublished observations.

	ABBREVIATIONS

The abbreviations used are: ROS, reactive oxygen species; GFP, green fluorescent protein; PAGE, polyacrylamide gel electrophoresis; TNT, coupled in vitro transcription and translation.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				E. J. Swindle, J. A. Hunt, and J. W. Coleman A Comparison of Reactive Oxygen Species Generation by Rat Peritoneal Macrophages and Mast Cells Using the Highly Sensitive Real-Time Chemiluminescent Probe Pholasin: Inhibition of Antigen-Induced Mast Cell Degranulation by Macrophage-Derived Hydrogen Peroxide J. Immunol., November 15, 2002; 169(10): 5866 - 5873. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Dunstan, S. L. Articles by Campbell, A. K. Search for Related Content PubMed PubMed Citation Articles by Dunstan, S. L. Articles by Campbell, A. K. Social Bookmarking