DOI: 10.1002/cbic.201402669
Full Papers
Brevetoxin, the Dinoflagellate Neurotoxin, Localizes to Thylakoid Membranes and Interacts with the LightHarvesting Complex II (LHCII) of Photosystem II Ryan T. Cassell,[a] Wei Chen,[a] Serge Thomas,[b] Li Liu,[a] and Kathleen S. Rein*[a] vesting complex II (LHCII) and thioredoxin. The LHCII is essential to non-photochemical quenching (NPQ), whereas thioredoxins are critical to the maintenance of redox homeostasis within the chloroplast and contribute to the scavenging of reactive oxygen. A culture of K. brevis producing low levels of toxin was shown to be deficient in NPQ and produced reactive oxygen species at twice the rate of the toxic culture, implicating a role in NPQ for the brevetoxins.
The brevetoxins are neurotoxins that are produced by the “Florida red tide” dinoflagellate Karenia brevis. They bind to and activate the voltage-gated sodium channels in higher organisms, specifically the Nav1.4 and Nav1.5 channel subtypes. However, the native physiological function that the brevetoxins perform for K. brevis is unknown. By using fluorescent and photoactivatable derivatives, brevetoxin was shown to localize to the chloroplast of K. brevis where it binds to the light-har-
Introduction The brevetoxins, produced by the “Florida red tide” dinoflagellate Karenia brevis, are a suite of neurotoxins that are responsible for massive fish and marine mammal mortalities in the Gulf of Mexico.[1] The brevetoxins are characteristic of a larger class of molecules known as polyether (PE) ladders (Scheme 1), which are produced principally by marine dinoflagellates. These molecules share common structural features, which include a series of trans-fused polyether rings with oxygen atoms alternating across the “top” and “bottom” of the molecules. The brevetoxins have two skeletal backbones: A-type and B-type, and a variety of side chain substituents on the rings distal to the lactone, resulting in about a dozen unique structures.[2] The most abundant of the brevetoxins, PbTx-2 (1), has the B-type skeleton with an a,b-unsaturated aldehyde side chain on the K-ring. PE ladder toxins can accumulate in fish or filter-feeding shellfish, resulting in both human and marine animal poisonings. In humans, the brevetoxins are the causative agents of a syndrome known as neurotoxic shellfish poisoning (NSP), as well as respiratory distress in beach visitors through exposure to aerosolized toxins.[3] Ciguatoxin, maitotoxin (not shown), gambieric acid, and gambierol, produced by the dinoflagellate Gambierdiscus toxicus, are responsible for ciguatera fish poisoning (CSP),[4] whereas yessotoxins from Proto-
ceratium reticulatum have been associated with diarrheic shellfish poisoning (DSP).[5] Since the discovery and structure elucidation of the brevetoxins more than 30 years ago, the intrinsic function of these and other PE ladders has remained elusive. Significant cellular resources are devoted to the production of these large and complex molecules, suggesting that they play an important role for the dinoflagellates. A recent report describes an increase in brevetoxin production in response to sudden changes in salinity, which implicates a role in osmoregulation or osmotic sensing.[6] However, other laboratories have disputed this finding.[7] More recently, an increase in cellular brevetoxin content was observed in response to phosphate limitation, leading to the suggestion that the brevetoxins play an allelopathic role for K. brevis.[8] Other studies report that copepods avoid grazing on K. brevis cells.[9] However, grazing studies that used purified brevetoxins failed to demonstrate a reduction in grazing by rotifers, even in those species that avoid grazing on K. brevis whole cells,[10] nor do purified brevetoxins exhibit allelopathic activity towards the diatom Asterionellopsis glacialis,[11] suggesting that feeding avoidance and allelopathy associated with K. brevis might be due to compounds other than brevetoxins. We reasoned that the localization of brevetoxin to an organelle and the identification of a brevetoxin receptor in K. brevis could implicate an endogenous role for the brevetoxins and other PE ladders. In an effort to better define this role in dinoflagellates, we prepared a fluorescent brevetoxin derivative (2, Figure 1 A) in order to localize brevetoxin to a specific organelle in K. brevis and a biotin-tagged, brevetoxin photoaffinity probe (3, Figure 2 A) to aide in the isolation and identification of a native brevetoxin receptor. We demonstrated that externally applied brevetoxin localizes to the chloroplast, where it binds to the light-harvesting complex II (LHCII) of photosys-
[a] R. T. Cassell,+ W. Chen,+ Dr. L. Liu, Prof. Dr. K. S. Rein Department of Chemistry and Biochemistry, Florida International University 11200 SW 8th St., Miami, FL 33199 (USA) E-mail: [email protected] [b] Prof. Dr. S. Thomas Department of Marine and Ecological Sciences Florida Gulf Coast University 10501 FGCU Blvd. S., Fort Myers, FL 33965 (USA) [+] These authors contributed equally to this work. Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cbic.201402669.
ChemBioChem 0000, 00, 0 – 0
These are not the final page numbers! ÞÞ
1
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers
Scheme 1. Structures of representative polyether ladder toxins produced by dinoflagellates.
Figure 1. A) Synthesis of fluorescent brevetoxin probe 2. For brevity, only the K-ring of PbTx-2 is shown. a) MeOH,*100 %; b) Alexafluor-488 alkyne, tBuOH/ H2O/DMF, CuSO4·-5 H2O, sodium ascorbate, *60 %. B) K. brevis cells with Alexa Fluor 488 alkyne. C) K. brevis cells with fluorescent probe 2. Both (B) and (C) were visualized with a 450–490 nm band pass/515 nm long pass EX/EM filter set. *Yields are based on HPLC analyses.
Results and Discussion
tem II and thioredoxin (Trx). Furthermore, a comparison of photosynthetic parameters of toxic and less toxic cultures suggests that the low toxin culture is deficient in non-photochemical quenching (NPQ), or the dissipation of excess light energy as heat, a central function of LHCII.
&
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
Brevetoxin localization The fluorescent brevetoxin derivative 2 was prepared in two steps from PbTx-2 (1): a thiol-Michael reaction between PbTx-2 and azidothiol 4 provided thiazepine 5, which was subjected to a 1, 3-dipolar cycloaddition reaction with Alexa Fluor 488 2
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Papers
Figure 2. A) Synthesis of brevetoxin photoaffinity probe 3. For brevity, only the K-ring of PbTx-2 is shown. a) K2CO3, KI, DMF, 90 %; b) biotin-PEG4-alkyne, tBuOH/H2O/DMF, CuSO4·5 H2O, sodium ascorbate, 60 %; c) (NH2CH2CH2S)2, NaBH4, EtOH, 75 %; d) DTT, MeOH, *100 %; e) MeOH, *100 %. B) SDS-PAGE of protein fractions from photoaffinity experiments. K. brevis membrane fractions incubated with photoaffinity probe 3 or 8. Lane 1: molecular weight markers. Lane 2: K. brevis membrane fraction incubated with 3 (7 mg protein). Lane 3: pre-incubation membrane fraction (7 mg protein). Lane 4: K. brevis homogenate incubated with 8 (18 mg protein). Lane 5: pre-incubation membrane fraction (6 mg protein). Lane 6: spinach homogenate incubated with 3 (16 mg protein). C) Spinach membrane fractions incubated with probe 3 at different pH. *Yields are based on HPLC-MS analysis of reaction mixture.
alkyne (Figure 1 A). The Alexa Fluor dye was chosen specifically because it is cell-impermeable. The thiol-Michael reaction has been described as a “click” reaction because it is fast, typically quantitative, and can be performed under mild conditions.[12] We found that the Michael adduct spontaneously cyclized to thiazepine, 5. This probe was used to localize brevetoxin within K. brevis. Live K. brevis cells were incubated with either the fluorescent brevetoxin derivative 2 or Alexa Fluor 488 alkyne (4 mm) for 30 min in the dark. After they had been washed with culture medium, the cells were visualized by using a fluorescent microscope equipped with a 450–490 nm band pass/515 nm long pass EX/EM filter set (Figure 1 B and C). Chlorophyll autofluorescence (red) could be clearly seen in both treatments. However, treatment with the fluorescent brevetoxin derivative 2 also demonstrated localization of the fluorescent brevetoxin conjugate (yellow) to the chloroplasts. The control experiment, with Alexa Fluor 488 alkyne alone, failed to label K. brevis cells.
tion. Brevetoxin photoprobe 3 incorporated a photolabile diazirine and a biotin tag. The synthesis of the probe is illustrated in Figure 2 A. Aldehyde 6 was alkylated with 1-azido-6-bromohexane to provide azide 7. A click reaction between 7 and biotin-PEG4-alkyne provided the photolabeled 8. Reductive alkylation of 8 by using cystamine, followed by reduction of the intermediate disulfide, provided thiol 9. Thiol 9 then underwent a thiol-Michael reaction with PbTx-2 (1), followed by cyclization to thiazepine 3. This probe was used to identify a brevetoxin receptor. An homogenate of K. brevis cells was separated into soluble and membrane fractions. These fractions were incubated either with the photolabeled 8 or the brevetoxin photoprobe 3, followed by photolysis and protein isolation using streptavidin immobilized on magnetic beads. No proteins were found when the soluble protein fraction was used. However, incubation and photolysis when using the membrane fraction with 3 consistently revealed a protein with an apparent mass of 27 kD when analyzed by SDS-PAGE (Figure 2 B). A control experiment with 8 did not yield any bands by SDS-PAGE. Analysis of the excised band by in-gel protease digestion liquid chromatography electrospray ionization-tandem mass spectrometry (LC ESIMS/MS) identified two chloroplast-localized proteins: LHCII of the photosynthetic apparatus and thioredoxin. LHCII is highly conserved across species, and we found in subsequent experi-
Brevetoxin binding protein Photoaffinity labeling has become an increasingly popular technique for identification of the targets of biologically active molecules.[13] A typical photoaffinity probe incorporates three groups, a ligand, a photoreactive group, and a tag for purificaChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
These are not the final page numbers! ÞÞ
3
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers ments that our photoaffinity probe also binds to the LHCII of spinach. The light-harvesting complex LHCII is located in the thylakoid membranes of photosynthetic organisms and is composed of three 25 kD subunits with 77 % or higher sequence homology.[14] Each subunit contains three membrane-spanning a-helices. These subunits form homo- or heterotrimers. LHCII functions in the harvesting and transfer of light energy to photosystem II. It is also central to the dissipation of light energy under conditions of excess light through NPQ. NPQ is initiated by a drop in pH on the lumenal side of the thylakoid membranes in response to light. We examined the binding of brevetoxin photoaffinity probe 3 to LHCII in spinach membranes (Figure 2 C) by using a pH range from pH 2 (simulating highlight conditions) to pH 8 (simulating low-light conditions). The 27 kD band increased in intensity with a decrease in pH (or conditions of high light). However, at high pH (or low light conditions), a smaller band appeared around 12 kD. LC ESI-MS/ MS analysis of this smaller band identified five different photosystem I and six different photosystem II membrane proteins (Table S2 in the Supporting Information). Also identified were three proteins which are located in the stroma: thioredoxin, a protein kinase, and Rubisco. It could be that at high pH/low light, brevetoxin is simply released from LHCII into the thylakoid membrane, where it might interact nonspecifically with many proteins. Many of the proteins identified were considerably larger than the 12–15 kD indicated by the markers. Thus it was surprising to find them in this low MW band. We believe that because protease inhibitors were not present in our homogenization buffer, high MW proteins might have been cleaved during the incubation period.
Having determined that brevetoxin binds to LHCII under high light conditions, and knowing that LHCII is integral to light harvesting and NPQ, we sought to compare photosynthetic parameters, including NPQ, in two cultures of K. brevis (Wilson strain) whose brevetoxin content varies by tenfold. Brevetoxin content can vary significantly among K. brevis isolates.[17] The reason the low toxin culture produces so little toxin is not well understood, and we are unaware of any studies which have uncovered any metabolic differences between these two cultures. The NPQ of both cultures were determined in their log phase of growth by pulsed amplitude modulation (PAM) fluorimetry at four excitation wavelengths. Under the same culture conditions, the toxic culture reached an algal biomass of about 1590 mg Chl L 1, versus 240 mg Chl L 1 for the less toxic culture, after 20 days of growth. Both algal cultures exhibited the same fluorescence yield subsequent to the four different excitation wavelengths (470, 520, 645, and 665 nm). The fluorescence yield was always above 0.4, with a lower value at 645 nm (0.45). It peaked at 520 nm (1.0), whereas it was 0.76 and 0.86 at 470 and 665 nm, respectively. The NPQ was calculated for each excitation wavelength by measuring the saturated fluorescence at 3600 mmol photons m 2 s 1 of the 25 min dark and actinic light adapted (363 mmol photons m 2 s 1) algal cultures. NPQ of the toxic culture was 2.1–3.8 times higher than the less toxic culture (Figure 3 A). NPQ for the four excitation wavelengths were also determined at increasing levels of 1 min actinic light adaptation, ranging from 164 to 2064 mmol photons m 2 s 1 (Figure 3 B). NPQ increased with increasing actinic light intensities, but this increase was much lower for the less toxic culture and overall much less than for the toxic culture. To our surprise, a slightly negative NPQ ( 0.05 to 0.01) was observed for the less toxic culture at actinic light intensities 64–764 mmol photons m 2 s 1, whereas all NPQ values were positive for the toxic culture. To the best of our knowledge, a negative NPQ has not been reported for any photosynthetic organism. This finding was reproducible and will be the subject of further study.
Photosynthetic parameters Oxygenic photosynthesis results in the production of reactive oxygen species (ROS) when the input of light energy exceeds the saturation limit of the photosynthetic apparatus. Chloroplasts have evolved a number of mechanisms to mitigate the damaging effects of ROS, which include both the scavenging of ROS and the suppression of ROS production through NPQ. Each LHCII subunit binds 14 chlorophyll molecules and four carotenoids, including two xanthophylls: zeaxanthin (under high light conditions) or violaxanthin, the di-epoxide of zeaxanthin (under low light conditions). The xanthophyll cycle modulates light energy dissipation through the interconversion of zeaxanthin/violaxanthin.[15] Under low light conditions, LHCII selectively binds violaxanthin. However, under high light conditions, violaxanthin is converted to zeaxanthin in a process catalyzed by violaxanthin de-epoxidase (VDE). Zeaxanthin binds to LHCII, which undergoes a conformational change, facilitating the quenching of chlorophyll fluorescence.[16] Under low-light conditions, zeaxanthin is converted back to violaxanthin by the enzyme zeaxanthin epoxidase (ZE), which then replaces zeaxanthin in the LHCII xanthyophyll binding site. In dinoflagellates, zeaxanthin/violaxanthin are replaced by dinoxanthin and diadinoxanthin, respectively.
&
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
Production of ROS by K. brevis Thioredoxins (Trxs) are small oxidoreductases that function to maintain redox homeostasis within the chloroplast by thiol–disulfide exchange reactions. Numerous Trx-dependent regulatory networks have been identified within chloroplasts,[18] including VDE activity.[19] Trxs are also critical to the scavenging of excess ROS by reduction of 2-cysteine peroxiredoxins (2-Cys Prxs), enzymes whose sole function is the detoxification of peroxides.[20] Therefore, the rate of ROS production in the toxic and less toxic cultures was monitored by using the indicator dye 5-(and-6)-carboxy-2’,7’-dichlorodihydrofluorescein diacetate (carboxy-H2DCFDA). The dye is cell-permeable and, after hydrolysis of the acetate groups by esterases, might be oxidized to a fluorescent product. When normalized to chlorophyll fluorescence, the low-toxin-producing culture of K. brevis produces ROS at twice the rate of the high-toxin-producing culture (Figure 3 C). 4
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Papers as no surprise. PE ladder molecules are known to migrate to the cell membranes where they interact with membrane proteins. Brevetoxins have also been demonstrated to aggregate into artificial lipid bilayers.[29] Our findings contradict earlier localization studies: one which failed to identify brevetoxin in isolated chloroplasts by LC-MS/MS,[30] and a published abstract that reported that brevetoxins are located in secretory vesicles by using fluorescent anti-brevetoxin antibodies.[31] A follow-up paper to the latter report has not yet appeared. We also observed that the association with LHCII was reduced at neutral pH Figure 3. A) NPQ for the toxic and less toxic cultures, calculated for a sample adapted at or low light, conditions which were used for chloro363 mmol photons m 2 s 1 actinic light. The toxic culture exhibited two to four times more NPQ relative to the less toxic culture. Error bars represent standard deviation plast isolation. Our findings are consistent with the (n = 6). B) NPQs for the less toxic (^: 470, &: 520, ~: 645, *: 665 nm) and toxic (^: 470, &: localization of another dinoflagellate toxin, okadaic 520, ~: 645, *: 665 nm) algal cultures, calculated for each sample after 1 min of adaptaacid,[32] to the chloroplast and of the cyanobacterial tion to increasing levels of actinic lights. Both cultures exhibited an increase in NPQ with toxin, microcystin, to thylakoid membranes of cyanoactinic light, but a sharper increase in magnitude of the NPQs were found for the toxic culture (n = 4). C) ROS production normalized to chlorophyll fluorescence for the less bacteria.[33] toxic (^) and toxic (&) cultures of K. brevis. We also observed distinct differences in NPQ and ROS production between the toxic and low-toxin strains of K. brevis. This, coupled with the localization Conclusions of brevetoxin to the thylakoid membrane, suggests a role for In higher organisms, PE ladders exhibit extremely potent and brevetoxin in facilitating NPQ. The mechanism by which this varied biological activities. The molecular target of these toxins occurs is undetermined. However, we offer three possible sceis typically a transmembrane protein or ion channel containing narios: 1) brevetoxin interacts directly with LHCII, inducing the one or more a-helices, to which PE ladders bind with extremeconformational change associated with NPQ; 2) brevetoxin acly high affinities (nanomolar to picomolar Kd values). The bretivates an ion channel in the thylakoid membrane; 3) brevetoxvetoxins and ciguatoxin bind to and activate the voltage-gated in self-assembles into transmembrane pores, promoting cation sodium channel (VGSC).[4, 21] To date, at least ten different movement across the thylakoid membrane. LHCII is integral to light harvesting and energy transfer to sodium channel subtypes have been identified, with the brevephotosystem II as well as the dissipation of light energy as toxins and ciguatoxin specifically targeting the skeletal muscle heat, under conditions of high light, through the quenching of and cardiac channel subtypes (Nav1.4 and Nav1.5, respectively). chlorophyll fluorescence through the xanthophyll cycle. AnothThe type B brevetoxins show a lower affinity for the Nav1.5 carer transmembrane protein identified in higher plants, PsbS diac channel subtype.[22] Gambierol and gambieric acid A inhib(photosystem II subunit S), has been hypothesized to act as it binding of the brevetoxins to the VGSC.[23] Brevenal, a nona catalyst for the conformational change that occurs in LHCII toxic molecule containing only five polyether rings binds to during NPQ.[34] Although the gene for PsbS has been found in the VGSC at a site that is distinct from the brevetoxin site.[24] [25] green algae, it has yet to be identified in a dinoflagellate. It is Gambierol is also a potent potassium channel blocker, and absent from the Symbiodinium genome.[35] Furthermore, maitotoxin is a calcium channel activator.[26] A specific molecular target has yet to be identified for yessotoxin. However, tBLASTn analysis of a K. brevis EST library[36] containing 21 000 a recent study on yessotoxin, brevetoxin, and synthetic PE ladunique contigs against PsbS from Arabidopsis thaliana failed to ders has demonstrated that these molecules have strong affinireveal a transcript encoding for this protein. When brevetoxin ties for transmembrane a-helices.[27] Furthermore, a maitotoxin binds to the VGSC, it induces a conformational change from the closed to the open state of the ion channel, stabilizing the photoaffinity probe was used to label membrane proteins in open state and resulting in a depolarization of excitable memred blood cells. An unidentified 23 kD protein was labeled, and branes. Because K. brevis likely does not produce PsbS, we prothe labeling was inhibited in the presence of PbTx-2.[28] pose that brevetoxin binds to the quenched conformation of We have demonstrated that a fluorescent brevetoxin derivaLHCII and helps to stabilize it in the same way that it stabilizes tive localizes to the chloroplast of K. brevis, and photoaffinity the open conformation of the VGSC. labeling with a brevetoxin photoprobe has indicated that breLHCII is highly concentrated in the thylakoid membranes. In vetoxin interacts with LHCII. It could be argued that the labels fact, it has been described as the most abundant membrane influence the physiochemical properties of brevetoxin. The two protein on the planet. It is possible that our photoaffinity labels have distinct properties, yet each of these derivatives probe linked to LHCII simply as a result of its high density in migrate to the chloroplast. The Alexa Fluor dye, in particular, is the thylakoids and that the true brevetoxin target in chloronot membrane-permeable, and the Alexa Fluor brevetoxin conplasts is a less abundant membrane protein, such as an ion jugate (2) would seem even less likely to enter the thylakoid channel. Ion channels selective for chloride, potassium, or divamembrane than native brevetoxin. The finding that brevetoxlent cations such as calcium and magnesium have been found ins localize to the lipophilic thylakoid membranes should come ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
These are not the final page numbers! ÞÞ
5
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers gation for 5 min at 466 g, and the supernatant was discarded. The cells were resuspended in phosphate-buffered saline (PBS) buffer (400 mL) and vortexed for 1 min. This suspension was centrifuged for 10 min at 14 000 g, and the supernatant and the pellet were separated. The pellet was resuspended in PBS (500 mL). An aliquot (50 mL) of each fraction was reserved for a protein assay. Samples were quantitated against BSA standards (0, 1, 0.5, 0.25, 0.125, and 0.0625 mg mL 1) prepared in PBS by using the Coomassie Protein Assay Reagent (BioRad) according to the manufacturer’s instructions. Final protein concentrations ranged from 0.5–0.9 mg mL 1.
in outer- and inner-envelope membranes of chloroplasts, as well as from thylakoid membranes, by using the patch clamp technique and have been hypothesized to regulate photosynthesis by modulating the proton motive force.[37] Finally, brevetoxin has been demonstrated to promote selective cation transport across phosphatidylcholine vesicles containing 0.3 % PbTx-2 (1). Furthermore, the rate of transport is highly sensitive to the addition of cholesterol, which increases lateral membrane pressure and reduces lipid dynamics, thus suggesting the formation of membrane-spanning channels as opposed to carrier mechanisms. Subsequent studies on fluorescent brevetoxin derivatives revealed that brevetoxin aggregates into hexamers in lipid bilayers, forming a pore which mediates ion transport through membranes.[29b] Photosynthetic electron transport in thylakoid membranes results in the generation of transmembrane voltage Dy as well as DpH. Partial dissipation of Dy by counter ion movement could allow for an enhanced DpH and the concomitant activation of NPQ. The ecological consequences of the high variability of brevetoxin content among K. brevis isolates, and presumably a corresponding variation in capacities for NPQ and ROS production, have yet to be determined. PsbB mutants of Arabidopsis that are NPQ-deficient also produce ROS at an elevated rate.[38] These mutants demonstrate a reduced fitness in the field,[39] yet they appear to have the capacity to respond more rapidly to stress[40] and experience decreased herbivory when compared to the wild type.[38] In summary, we submit that PE ladders specifically target a transmembrane protein located in the thylakoid membrane and coincidentally associate with other transmembrane a-helices in higher organisms. However, the finding that brevetoxins might have an intrinsic biochemical role related to photosynthesis does not preclude an extrinsic function as a feeding deterrent. Indeed, secondary metabolites such as salicylic acid have been demonstrated to fill dual roles as feeding deterrents as well as hormone and growth regulators in plants.[41]
Spinach (5 g) was crushed in PBS (20 mL) and concentrated by centrifugation (5 min at 14 000 g). The supernatant was discarded. The pellet was resuspended in PBS buffer (5 mL, and an aliquot (50 mL) was reserved for a protein assay. The final protein concentration was 1.0–2.5 mg mL 1. Incubation of protein homogenates with photoaffinity probe (3) or photoaffinity label (8): Synthetic details for the preparation of 3 and 6–9 and compound characterization data can be found in the Supporting Information. The brevetoxin photoaffinity probe 3 or label 8 were incubated at room temperature with the soluble protein or membrane preparations at a probe/protein ratio of 0.2 mg probe per mg protein for 1 h. After incubation, the mixture was photolyzed (Vilber Lourmat transilluminator, Model FLX-20M) for 20 min at 312 nm. The soluble fraction was applied directly to a suspension of streptavidin beads (200 mL, Dynabeads, 2.8 mm or 1.0 mm, Life Technologies). Triton X-100 (1 %, 200 mL) was added to the membrane fractions (200 mL). The mixture was vortexed and centrifuged for 5 min at 14 000 g and applied to a suspension of streptavidin beads (200 mL). After incubation for 30 min at room temperature, the beads were separated from the solution by using a strong magnet and washed with PBS buffer (4 200 mL). Each wash was analyzed for protein content, and the final wash contained no detectable protein. Protein was eluted from the beads by using 0.1 % SDS solution (3 250 mL) at 100 8C for 5 min. An aliquot of each elution was analyzed by Bradford protein assay, concentrated to dryness by lyophilization (FreeZone, Model 2.5, 86 L min 1), and resuspended in water (10 mL) and loading buffer (2 mL, 0.25 m Tris-HCl, pH 6.8, 15 % SDS, 20 % glycerol, 25 % mercaptoethanol, 0.01 % Bromophenol Blue) for analysis by SDS-PAGE (12 % acryalamide with Tris·glycine SDS, pH 8.3) running buffer. Acrylamide gels were stained with PAGE Blue (Fermentas), according to the manufacturer’s instructions. The 25 kD band was excised from the SDS-PAGE gel. No protein was detected in the soluble fraction from the Bradford assay, nor were bands observed in the gel.
Experimental Section Incubation of K. brevis with the fluorescent brevetoxin derivative 2 or the Alexa Fluor 488 alkyne: Synthetic details for the preparation of 2 and compound characterization data can be found in the Supporting Information. Live K. brevis culture (3 mL) was incubated for 1 h in the dark with either the fluorescent brevetoxin derivative 2 or Alexa Fluor 488 alkyne (Molecular Probes). The final concentration of the Alexa Fluor 488 alkyne was 4 mm. Because the exact concentration of the brevetoxin derivative 2 was not known, the fluorescence of the solution of 2 was adjusted to be equal to that of the Alexa Fluor 488 alkyne. After incubation, the cells were centrifuged for 8 min at 130 g. The supernatant was discarded, and the cells were resuspended in RE medium (1 mL) and centrifuged again for 8 min at 130 g. The supernatant was discarded, and the cells were resuspended in RE medium (1 mL). The live cells were visualized by using a fluorescent microscope (Leica Leitz, Model DMRB) with a 450–490 nm band pass/515 nm long pass EX/EM filter.
In-gel digestion: In-gel digestion and LC-ESI MS/MS analysis was performed by the Proteomics and Mass Spectrometry Facility at the UMass Medical School. Gel slices were cut into 1 1 mm pieces and placed in 1.5 mL Eppendorf tubes with water (1 mL) for 30 min. The water was removed, and ammonium bicarbonate (50 mL, 250 mm) was added. For reduction, a solution of 1,4-dithiothreitol (DTT; 5 mL, 45 mm) was added, and the samples were incubated at 50 8C for 30 min. The samples were cooled to room temperature and, for alkylation, iodoacetamide solution (5 mL, 100 mm) was added and allowed to react for 30 min. The gel slices were washed with water (20 1 mL). The water was removed, ammonium bicarbonate/acetonitrile (50:50, 1 mL, 50 mm) was placed in each tube, and the samples were incubated at room temperature for 1 h. The solution was removed, and acetonitrile (200 mL) was added to each tube, at which point the gels slices turned opaque white. The acetonitrile was removed, and the gel slices were further dried in a Speed Vac. Gel slices were rehydrated in trypsin
Preparation of protein homogenates and membrane suspensions: K. brevis culture (200–400 mL) was concentrated by centrifu-
&
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
6
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Papers (50 mL, 2 ng mL 1, Sigma) in 0.01 % ProteaseMAX Surfactant (Promega)/50 mm ammonium bicarbonate. Additional bicarbonate buffer was added to ensure complete submersion of the gel slices. Samples were incubated at 37 8C for 21 h. The supernatant of each sample was removed and placed in a separate 1.5 mL Eppendorf tube. Gel slices were further dehydrated with acetonitrile/1 % formic acid (80:20, 100 mL). The extract was combined with the supernatants of each sample. The samples were dried in a Speed Vac. Samples were dissolved in 25 mL of 5 % acetonitrile in 0.1 % trifluoroacetic acid prior to injection on LC-MS/MS.
spectral signature for each algal culture was determined at the beginning of the experiment. The maximum quantum efficiency of the photosystem II (PSII) photochemistry (referred to as “yield”) was measured by using a 25 min dark-adapted algal sample, thus allowing all reaction centers to be closed. Minimal fluorescence “F0” was recorded at 16 mmol photons m 2 s 1, and the maximal fluorescence “Fm” was determined after a 0.5 s saturating actinic red light pulse. The yield was calculated as: yield = Fv/Fm = (Fm F0)/Fm. The NPQ that estimated the rate of constant heat loss from PSII was calculated as follows: NPQ = (Fm/Fm’)/Fm’, where Fm’ is the maximum fluorescence measured after a 0.5 s saturating actinic red pulse of the lightadapted sample.
LC-MS/MS on Q Exactive: For peptide sequencing, a 3.0 mL aliquot was directly injected onto a custom-packed 2 cm 100 mm C18 Magic 5 m particle trap column. Labeled peptides were eluted and sprayed from a custom-packed emitter (75 mm 25 cm C18 Magic 3 mm particle) with a linear gradient from 95 % solvent A (0.1 % formic acid in water) to 35 % solvent B (0.1 % formic acid in acetonitrile) over 40 min at a flow rate of 300 nL min 1 on a Waters Nano Acquity UPLC system. Data-dependent acquisitions were performed on a Q Exactive mass spectrometer (Thermo Scientific), according to an experiment in which full MS scans from 300–1750 m/z were acquired at a resolution of 70 000, followed by 12 MS/MS scans acquired under HCD fragmentation at a resolution of 35 000 with an isolation width of 1.2 Da.
Statistics: Data for each excitation wavelength were not deconvoluted by using the reference spectral signature of each culture. As such, a set of photosynthetic characteristics for each excitation light for each algal culture was used for the comparison. To compare each photosynthetic characteristic of the two K. brevis cultures, a two-way full factorial ANOVA (two algal cultures four excitation lights) was conducted with SPSS 20 by using the univariate general linear model “GLM” procedure and Fisher’s least significant difference (LSD) to test for the main factors. A significance level of 0.05 was used for all ANOVAs.
Database searching: Raw data files were processed with Proteome Discoverer (v. 1.3). Charge state deconvolution and deisotoping were not performed. All MS/MS samples were analyzed by using Mascot (Matrix Science, London, UK; v. 2.4.1). Mascot was set up to search the NCBInr_20130907 database (32 265 160 entries), assuming the digestion enzyme stricttrypsin. Mascot was searched with a fragment ion mass tolerance of 0.050 Da and a parent ion tolerance of 10.0 ppm. The carbamidomethyl of cysteine was specified in Mascot as a fixed modification. Gln- > pyro-Glu of the N terminus, oxidation of methionine, and acetylation of the N terminus were specified in Mascot as variable modifications.
ROS production by K. brevis: A solution of 5-(and-6)-chloromethyl2’,7’-dichlorodihydrofluorescein diacetate acetyl ester (CMH2DCFDA, 11 mm, Life Technologies) was prepared in DMSO/culture medium (1:4) and diluted to 50 mm in culture medium. This stock was used for both the ROS assay and for preparation of a calibration curve. Calibration standards (5, 2, 0.5, 0.2 and 0 mm CMH2DCFDA in culture medium) were prepared by dilution of a stock solution (6 mm CM-H2DCFDA; 50 mm H2O2 ; 8 U mL 1 pig liver esterase in culture medium/PBS, 3:1). K. brevis cells were concentrated tenfold by centrifugation for 5 min at 17 g, followed by gentle resuspension in one-tenth the original volume of culture medium. Concentrated K. brevis cells (100 mL), culture medium (32 mL), and CM-H2DCFDA stock (50 mm, 18 mL) were added to a 96-well microplate for a final probe concentration of 6 mm. Plates were incubated at ambient temperature under laboratory lighting. Fluorescence was monitored (lex = 485 nm, lem = 538 nm) hourly for 7 h. Samples and standards were assayed in triplicate. Because cell density and volume varied between cultures, ROS concentration was normalized to chlorophyll content. K. brevis culture (40 mL) was centrifuged for 5 min at 200 g, and the resulting pellet was homogenized in MeOH (600 mL). The MeOH homogenate was centrifuged for 10 min at 466 g to remove cell debris, and the supernatant (150 mL) was added to a 96-well microplate. Chlorophyll fluorescence was measured at lex = 485 nm and lex = 680 nm.
Criteria for protein identification: Scaffold (v. 4.1.1, Proteome Software Inc., Portland, OR) was used to validate MS/MS-based peptide and protein identifications. Peptide identifications were accepted if they could be established at greater than 80.0 % probability by the Peptide Prophet algorithm[42] with Scaffold delta-mass correction. Protein identifications were accepted if they could be established at greater than 90.0 % probability and contained at least two identified peptides. Protein probabilities were assigned by the Protein Prophet algorithm.[43] Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony. Proteins sharing significant peptide evidence were grouped into clusters. Photosynthetic characteristics: The photosynthetic characteristics of each algal culture were measured by using pulsed amplitude modulation (PAM) fluorometry.[44] This was performed with a WALZ tetrafluorometer Phyto-PAM equipped with an Emitter Detection unit (ED; http://www.walz.com/), which excited the photosynthetic pigments by using four consecutive pulses of excitation light of different wavelengths: blue (470 nm), green (520 nm), light red (645 nm), and far red (665 nm). The resulting fluorescence, recorded at room temperature (~ 20 8C), was measured subsequent to pulses of different excitation light intensities and one 0.5 s saturating pulse of actinic red light by using algae adapted to different light regimes. All fluorescence measurement routines were conducted in sextuplicate by using algal cultures standardized to 240.8 S.D. 3.0 mg Chl L 1 and 246 S.D.1.0 mg Chl L 1 for the nontoxic and toxic cultures, respectively. Additionally, the reference
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
These are not the final page numbers! ÞÞ
Acknowledgements This work was supported by Bridge funding from the FIU Division of Research and the FIU College of Arts and Sciences. Keywords: bioorganic chemistry · brevetoxin · fluorescent probes · photoaffinity labeling [1] G. Rossini, P. Hess in Molecular, Clinical and Environmental Toxicology, Vol. 100 (Ed.: A. Luch), Birkhuser, Basel, 2010, pp. 65 – 122. [2] D. G. Baden, A. J. Bourdelais, H. Jacocks, S. Michelliza, J. Naar, Environ. Health Perspect. 2005, 113, 621 – 625.
7
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers [3] a) S. M. Watkins, A. Reich, L. E. Fleming, R. Hammond, Mar. Drugs 2008, 6, 431 – 455; b) L. E. Fleming, L. C. Backer, D. G. Baden, Environ. Health Perspect. 2005, 113, 618 – 620. [4] K. Yamaoka, M. Inoue, M. Hirama, Forensic Toxicol. 2011, 29, 125 – 131. [5] A. Tubaro, V. Dell’Ovo, S. Sosa, C. Florio, Toxicon 2010, 56, 163 – 172. [6] R. M. Errera, L. Campbell, Proc. Natl. Acad. Sci. USA 2011, 108, 10597 – 10601. [7] W. G. Sunda, C. Burleson, D. R. Hardison, J. S. Morey, Z. Wang, J. Wolny, A. A. Corcoran, L. J. Flewelling, F. M. Van Dolah, Proc. Natl. Acad. Sci. USA 2013, 110, 10223 – 10228. [8] D. R. Hardison, W. G. Sunda, D. Shea, R. W. Litaker, PLoS One 2013, 8, e58545. [9] a) J. Hong, S. Talapatra, J. Katz, P. A. Tester, R. J. Waggett, A. R. Place, PLoS One 2012, 7, e36845; b) R. J. Waggett, D. Hardison, P. A. Tester, Mar. Ecol. Prog. Ser. 2012, 444, 15 – 30. [10] J. Kubanek, T. W. Snell, C. Pirkle, Limnol. Oceanogr. 2007, 52, 1026 – 1035. [11] E. K. Prince, K. L. Poulson, T. L. Myers, R. D. Sieg, J. Kubanek, Harmful Algae 2010, 10, 39 – 48. [12] D. P. Nair, M. Podgorski, S. Chatani, T. Gong, W. Xi, C. R. Fenoli, C. N. Bowman, Chem. Mater. 2014, 26, 724 – 744. [13] J. Sumranjit, S. J. Chung, Molecules 2013, 18, 10425 – 10451. [14] Z. Liu, H. Yan, K. Wang, T. Kuang, J. Zhang, L. Gui, X. An, W. Chang, Nature 2004, 428, 287 – 292. [15] M. Fuciman, M. M. Enriquez, T. Polvka, L. Dall’Osto, R. Bassi, H. A. Frank, J. Phys. Chem. B 2012, 116, 3834 – 3849. [16] J. Standfuss, A. C. T. van Scheltinga, M. Lamborghini, W. Khlbrandt, EMBO J. 2005, 24, 919 – 928. [17] R. M. Errera, A. Bourdelais, M. Drennan, E. Dodd, D. Henrichs, L. Campbell, Toxicon 2010, 55, 195 – 203. [18] L. Nikkanen, E. Rintamki, Philos. Trans. R. Soc. London Ser. B 2014, 369, 20130224. [19] M. Hall, A. Mata-Cabana, H. E. kerlund, F. J. Florencio, W. P. Schrçder, M. Lindahl, T. Kieselbach, Proteomics 2010, 10, 987 – 1001. [20] J. Kçnig, M. Baier, F. Horling, U. Kahmann, G. Harris, P. Schrmann, K.-J. Dietz, Proc. Natl. Acad. Sci. USA 2002, 99, 5738 – 5743. [21] M. Poli, T. J. Mende, D. G. Baden, Mol. Pharmacol. 1986, 30, 129 – 135. [22] M.-Y. B. Dechraoui, J. S. Ramsdell, Toxicon 2003, 41, 919 – 927. [23] M. Inoue, M. Hirama, M. Satake, K. Sugiyama, T. Yasumoto, Toxicon 2003, 41, 469 – 474. [24] E. P. Gold, H. M. Jacocks, A. J. Bourdelais, D. G. Baden, Harmful Algae 2013, 26, 12 – 19. [25] E. Cuypers, Y. Abdel-Mottaleb, I. Kopljar, J. D. Rainier, A. L. Raes, D. J. Snyders, J. Tytgat, Toxicon 2008, 51, 974 – 983. [26] A. Yokoyama, M. Murata, Y. Oshima, T. Iwashita, T. Yasumoto, J. Biochem. 1988, 104, 184 – 187.
&
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
[27] a) S. Ujihara, T. Oishi, R. Mouri, R. Tamate, K. Konoki, N. Matsumori, M. Murata, Y. Oshima, N. Sugiyama, M. Tomita, Bioorg. Med. Chem. Lett. 2010, 20, 6443 – 6446; b) S. Ujihara, T. Oishi, K. Torikai, K. Konoki, N. Matsumori, M. Murata, Y. Oshima, S. Aimoto, Bioorg. Med. Chem. Lett. 2008, 18, 6115 – 6118; c) M. Mori, T. Oishi, S. Matsuoka, S. Ujihara, N. Matsumori, M. Murata, M. Satake, Y. Oshima, N. Matsushita, S. Aimoto, Bioorg. Med. Chem. 2005, 13, 5099 – 5103. [28] K. Konoki, M. Hashimoto, K. Honda, K. Tachibana, R. Tamate, F. Hasegawa, T. Oishi, M. Murata, Heterocycles 2009, 79, 1007 – 1017. [29] a) S. Matile, N. Berova, K. Nakanishi, Chem. Biol. 1996, 3, 379 – 392; b) S. Matile, K. Nakanishi, Angew. Chem. Int. Ed. Engl. 1996, 35, 757 – 759; Angew. Chem. 1996, 108, 812 – 814. [30] F. M. Van Dolah, M. L. Zippay, L. Pezzolesi, K. S. Rein, J. G. Johnson, J. S. Morey, Z. Wang, R. Pistocchi, J. Phycol. 2013, 49, 1118 – 1127. [31] K. L. Vigna, I. Quesada, P. Verdugo, Biophys. J. 2012, 102, 319a – 320a. [32] J. Zhou, L. Fritz, Phycologia 1994, 33, 455 – 461. [33] F. M. Young, C. Thomson, J. S. Metcalf, J. M. Lucocq, G. A. Codd, J. Struct. Biol. 2005, 151, 208 – 214. [34] A. V. Ruban, M. P. Johnson, C. D. Duffy, Biochim. Biophys. 2012, 1817, 167 – 181. [35] E. Shoguchi, C. Shinzato, T. Kawashima, F. Gyoja, S. Mungpakdee, R. Koyanagi, T. Takeuchi, K. Hisata, M. Tanaka, M. Fujiwara, Curr. Biol. 2013, 23, 1399 – 1408. [36] K. B. Lidie, J. C. Ryan, M. Barbier, F. M. Van Dolah, Mar. Biotechnol. 2005, 7, 481 – 493. [37] V. Checchetto, E. Teardo, L. Carraretto, E. Formentin, E. Bergantino, G. M. Giacometti, I. Szabo, Biophys. Chem. 2013, 182, 51 – 57. [38] H. J. Jnknp, M. Frenkel, I. Zulfugarov, M. Reichelt, A. Krieger-Liszkay, Y. Mishra, J. Gershenzon, J. Moen, C.-H. Lee, S. Jansson, PLoS One 2013, 8, e53232. [39] C. Klheim, J. gren, S. Jansson, Science 2002, 297, 91 – 93. [40] M. Frenkel, C. Klheim, H. J. Jnknp, O. Skogstrçm, L. Dall’Osto, J. gren, R. Bassi, T. Moritz, J. Moen, S. Jansson, BMC Plant Biol. 2009, 9, 12. [41] M. R.-S. Vicente, J. Plasencia, J. Exp. Bot. 2011, 62, 3321 – 3338. [42] A. Keller, A. I. Nesvizhskii, E. Kolker, R. Aebersold, Anal. Chem. 2002, 74, 5383 – 5392. [43] A. I. Nesvizhskii, A. Keller, E. Kolker, R. Aebersold, Anal. Chem. 2003, 75, 4646 – 4658. [44] a) U. Schreiber in Chlorophyll a Fluorescence: A Signature of Photosynthesis (Eds.: G. C. Papageorgiou, Govinjee), Springer, Dordrecht, 2004, pp. 279 – 319; b) E. Murchie, T. Lawson, J. Exp. Bot. 2013, ert208. Received: November 19, 2014 Published online on && &&, 0000
8
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
FULL PAPERS Seeing the light: A fluorescent brevetoxin derivative localizes to the chloroplast of the Florida red tide dinoflagellate, Karenia brevis. Photoaffinity labeling of a receptor protein identified the light-harvesting complex II (LHCII) as the native brevetoxin receptor. A brevetoxin-deficient strain of K. brevis was unable to perform some of the essential functions of LHCII.
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
These are not the final page numbers! ÞÞ
R. T. Cassell, W. Chen, S. Thomas, L. Liu, K. S. Rein* && – && Brevetoxin, the Dinoflagellate Neurotoxin, Localizes to Thylakoid Membranes and Interacts with the Light-Harvesting Complex II (LHCII) of Photosystem II
9
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers
Brevetoxin, the Dinoflagellate Neurotoxin, Localizes to Thylakoid Membranes and Interacts with the LightHarvesting Complex II (LHCII) of Photosystem II Ryan T. Cassell,[a] Wei Chen,[a] Serge Thomas,[b] Li Liu,[a] and Kathleen S. Rein*[a] vesting complex II (LHCII) and thioredoxin. The LHCII is essential to non-photochemical quenching (NPQ), whereas thioredoxins are critical to the maintenance of redox homeostasis within the chloroplast and contribute to the scavenging of reactive oxygen. A culture of K. brevis producing low levels of toxin was shown to be deficient in NPQ and produced reactive oxygen species at twice the rate of the toxic culture, implicating a role in NPQ for the brevetoxins.
The brevetoxins are neurotoxins that are produced by the “Florida red tide” dinoflagellate Karenia brevis. They bind to and activate the voltage-gated sodium channels in higher organisms, specifically the Nav1.4 and Nav1.5 channel subtypes. However, the native physiological function that the brevetoxins perform for K. brevis is unknown. By using fluorescent and photoactivatable derivatives, brevetoxin was shown to localize to the chloroplast of K. brevis where it binds to the light-har-
Introduction The brevetoxins, produced by the “Florida red tide” dinoflagellate Karenia brevis, are a suite of neurotoxins that are responsible for massive fish and marine mammal mortalities in the Gulf of Mexico.[1] The brevetoxins are characteristic of a larger class of molecules known as polyether (PE) ladders (Scheme 1), which are produced principally by marine dinoflagellates. These molecules share common structural features, which include a series of trans-fused polyether rings with oxygen atoms alternating across the “top” and “bottom” of the molecules. The brevetoxins have two skeletal backbones: A-type and B-type, and a variety of side chain substituents on the rings distal to the lactone, resulting in about a dozen unique structures.[2] The most abundant of the brevetoxins, PbTx-2 (1), has the B-type skeleton with an a,b-unsaturated aldehyde side chain on the K-ring. PE ladder toxins can accumulate in fish or filter-feeding shellfish, resulting in both human and marine animal poisonings. In humans, the brevetoxins are the causative agents of a syndrome known as neurotoxic shellfish poisoning (NSP), as well as respiratory distress in beach visitors through exposure to aerosolized toxins.[3] Ciguatoxin, maitotoxin (not shown), gambieric acid, and gambierol, produced by the dinoflagellate Gambierdiscus toxicus, are responsible for ciguatera fish poisoning (CSP),[4] whereas yessotoxins from Proto-
ceratium reticulatum have been associated with diarrheic shellfish poisoning (DSP).[5] Since the discovery and structure elucidation of the brevetoxins more than 30 years ago, the intrinsic function of these and other PE ladders has remained elusive. Significant cellular resources are devoted to the production of these large and complex molecules, suggesting that they play an important role for the dinoflagellates. A recent report describes an increase in brevetoxin production in response to sudden changes in salinity, which implicates a role in osmoregulation or osmotic sensing.[6] However, other laboratories have disputed this finding.[7] More recently, an increase in cellular brevetoxin content was observed in response to phosphate limitation, leading to the suggestion that the brevetoxins play an allelopathic role for K. brevis.[8] Other studies report that copepods avoid grazing on K. brevis cells.[9] However, grazing studies that used purified brevetoxins failed to demonstrate a reduction in grazing by rotifers, even in those species that avoid grazing on K. brevis whole cells,[10] nor do purified brevetoxins exhibit allelopathic activity towards the diatom Asterionellopsis glacialis,[11] suggesting that feeding avoidance and allelopathy associated with K. brevis might be due to compounds other than brevetoxins. We reasoned that the localization of brevetoxin to an organelle and the identification of a brevetoxin receptor in K. brevis could implicate an endogenous role for the brevetoxins and other PE ladders. In an effort to better define this role in dinoflagellates, we prepared a fluorescent brevetoxin derivative (2, Figure 1 A) in order to localize brevetoxin to a specific organelle in K. brevis and a biotin-tagged, brevetoxin photoaffinity probe (3, Figure 2 A) to aide in the isolation and identification of a native brevetoxin receptor. We demonstrated that externally applied brevetoxin localizes to the chloroplast, where it binds to the light-harvesting complex II (LHCII) of photosys-
[a] R. T. Cassell,+ W. Chen,+ Dr. L. Liu, Prof. Dr. K. S. Rein Department of Chemistry and Biochemistry, Florida International University 11200 SW 8th St., Miami, FL 33199 (USA) E-mail: [email protected] [b] Prof. Dr. S. Thomas Department of Marine and Ecological Sciences Florida Gulf Coast University 10501 FGCU Blvd. S., Fort Myers, FL 33965 (USA) [+] These authors contributed equally to this work. Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/cbic.201402669.
ChemBioChem 0000, 00, 0 – 0
These are not the final page numbers! ÞÞ
1
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers
Scheme 1. Structures of representative polyether ladder toxins produced by dinoflagellates.
Figure 1. A) Synthesis of fluorescent brevetoxin probe 2. For brevity, only the K-ring of PbTx-2 is shown. a) MeOH,*100 %; b) Alexafluor-488 alkyne, tBuOH/ H2O/DMF, CuSO4·-5 H2O, sodium ascorbate, *60 %. B) K. brevis cells with Alexa Fluor 488 alkyne. C) K. brevis cells with fluorescent probe 2. Both (B) and (C) were visualized with a 450–490 nm band pass/515 nm long pass EX/EM filter set. *Yields are based on HPLC analyses.
Results and Discussion
tem II and thioredoxin (Trx). Furthermore, a comparison of photosynthetic parameters of toxic and less toxic cultures suggests that the low toxin culture is deficient in non-photochemical quenching (NPQ), or the dissipation of excess light energy as heat, a central function of LHCII.
&
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
Brevetoxin localization The fluorescent brevetoxin derivative 2 was prepared in two steps from PbTx-2 (1): a thiol-Michael reaction between PbTx-2 and azidothiol 4 provided thiazepine 5, which was subjected to a 1, 3-dipolar cycloaddition reaction with Alexa Fluor 488 2
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Papers
Figure 2. A) Synthesis of brevetoxin photoaffinity probe 3. For brevity, only the K-ring of PbTx-2 is shown. a) K2CO3, KI, DMF, 90 %; b) biotin-PEG4-alkyne, tBuOH/H2O/DMF, CuSO4·5 H2O, sodium ascorbate, 60 %; c) (NH2CH2CH2S)2, NaBH4, EtOH, 75 %; d) DTT, MeOH, *100 %; e) MeOH, *100 %. B) SDS-PAGE of protein fractions from photoaffinity experiments. K. brevis membrane fractions incubated with photoaffinity probe 3 or 8. Lane 1: molecular weight markers. Lane 2: K. brevis membrane fraction incubated with 3 (7 mg protein). Lane 3: pre-incubation membrane fraction (7 mg protein). Lane 4: K. brevis homogenate incubated with 8 (18 mg protein). Lane 5: pre-incubation membrane fraction (6 mg protein). Lane 6: spinach homogenate incubated with 3 (16 mg protein). C) Spinach membrane fractions incubated with probe 3 at different pH. *Yields are based on HPLC-MS analysis of reaction mixture.
alkyne (Figure 1 A). The Alexa Fluor dye was chosen specifically because it is cell-impermeable. The thiol-Michael reaction has been described as a “click” reaction because it is fast, typically quantitative, and can be performed under mild conditions.[12] We found that the Michael adduct spontaneously cyclized to thiazepine, 5. This probe was used to localize brevetoxin within K. brevis. Live K. brevis cells were incubated with either the fluorescent brevetoxin derivative 2 or Alexa Fluor 488 alkyne (4 mm) for 30 min in the dark. After they had been washed with culture medium, the cells were visualized by using a fluorescent microscope equipped with a 450–490 nm band pass/515 nm long pass EX/EM filter set (Figure 1 B and C). Chlorophyll autofluorescence (red) could be clearly seen in both treatments. However, treatment with the fluorescent brevetoxin derivative 2 also demonstrated localization of the fluorescent brevetoxin conjugate (yellow) to the chloroplasts. The control experiment, with Alexa Fluor 488 alkyne alone, failed to label K. brevis cells.
tion. Brevetoxin photoprobe 3 incorporated a photolabile diazirine and a biotin tag. The synthesis of the probe is illustrated in Figure 2 A. Aldehyde 6 was alkylated with 1-azido-6-bromohexane to provide azide 7. A click reaction between 7 and biotin-PEG4-alkyne provided the photolabeled 8. Reductive alkylation of 8 by using cystamine, followed by reduction of the intermediate disulfide, provided thiol 9. Thiol 9 then underwent a thiol-Michael reaction with PbTx-2 (1), followed by cyclization to thiazepine 3. This probe was used to identify a brevetoxin receptor. An homogenate of K. brevis cells was separated into soluble and membrane fractions. These fractions were incubated either with the photolabeled 8 or the brevetoxin photoprobe 3, followed by photolysis and protein isolation using streptavidin immobilized on magnetic beads. No proteins were found when the soluble protein fraction was used. However, incubation and photolysis when using the membrane fraction with 3 consistently revealed a protein with an apparent mass of 27 kD when analyzed by SDS-PAGE (Figure 2 B). A control experiment with 8 did not yield any bands by SDS-PAGE. Analysis of the excised band by in-gel protease digestion liquid chromatography electrospray ionization-tandem mass spectrometry (LC ESIMS/MS) identified two chloroplast-localized proteins: LHCII of the photosynthetic apparatus and thioredoxin. LHCII is highly conserved across species, and we found in subsequent experi-
Brevetoxin binding protein Photoaffinity labeling has become an increasingly popular technique for identification of the targets of biologically active molecules.[13] A typical photoaffinity probe incorporates three groups, a ligand, a photoreactive group, and a tag for purificaChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
These are not the final page numbers! ÞÞ
3
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers ments that our photoaffinity probe also binds to the LHCII of spinach. The light-harvesting complex LHCII is located in the thylakoid membranes of photosynthetic organisms and is composed of three 25 kD subunits with 77 % or higher sequence homology.[14] Each subunit contains three membrane-spanning a-helices. These subunits form homo- or heterotrimers. LHCII functions in the harvesting and transfer of light energy to photosystem II. It is also central to the dissipation of light energy under conditions of excess light through NPQ. NPQ is initiated by a drop in pH on the lumenal side of the thylakoid membranes in response to light. We examined the binding of brevetoxin photoaffinity probe 3 to LHCII in spinach membranes (Figure 2 C) by using a pH range from pH 2 (simulating highlight conditions) to pH 8 (simulating low-light conditions). The 27 kD band increased in intensity with a decrease in pH (or conditions of high light). However, at high pH (or low light conditions), a smaller band appeared around 12 kD. LC ESI-MS/ MS analysis of this smaller band identified five different photosystem I and six different photosystem II membrane proteins (Table S2 in the Supporting Information). Also identified were three proteins which are located in the stroma: thioredoxin, a protein kinase, and Rubisco. It could be that at high pH/low light, brevetoxin is simply released from LHCII into the thylakoid membrane, where it might interact nonspecifically with many proteins. Many of the proteins identified were considerably larger than the 12–15 kD indicated by the markers. Thus it was surprising to find them in this low MW band. We believe that because protease inhibitors were not present in our homogenization buffer, high MW proteins might have been cleaved during the incubation period.
Having determined that brevetoxin binds to LHCII under high light conditions, and knowing that LHCII is integral to light harvesting and NPQ, we sought to compare photosynthetic parameters, including NPQ, in two cultures of K. brevis (Wilson strain) whose brevetoxin content varies by tenfold. Brevetoxin content can vary significantly among K. brevis isolates.[17] The reason the low toxin culture produces so little toxin is not well understood, and we are unaware of any studies which have uncovered any metabolic differences between these two cultures. The NPQ of both cultures were determined in their log phase of growth by pulsed amplitude modulation (PAM) fluorimetry at four excitation wavelengths. Under the same culture conditions, the toxic culture reached an algal biomass of about 1590 mg Chl L 1, versus 240 mg Chl L 1 for the less toxic culture, after 20 days of growth. Both algal cultures exhibited the same fluorescence yield subsequent to the four different excitation wavelengths (470, 520, 645, and 665 nm). The fluorescence yield was always above 0.4, with a lower value at 645 nm (0.45). It peaked at 520 nm (1.0), whereas it was 0.76 and 0.86 at 470 and 665 nm, respectively. The NPQ was calculated for each excitation wavelength by measuring the saturated fluorescence at 3600 mmol photons m 2 s 1 of the 25 min dark and actinic light adapted (363 mmol photons m 2 s 1) algal cultures. NPQ of the toxic culture was 2.1–3.8 times higher than the less toxic culture (Figure 3 A). NPQ for the four excitation wavelengths were also determined at increasing levels of 1 min actinic light adaptation, ranging from 164 to 2064 mmol photons m 2 s 1 (Figure 3 B). NPQ increased with increasing actinic light intensities, but this increase was much lower for the less toxic culture and overall much less than for the toxic culture. To our surprise, a slightly negative NPQ ( 0.05 to 0.01) was observed for the less toxic culture at actinic light intensities 64–764 mmol photons m 2 s 1, whereas all NPQ values were positive for the toxic culture. To the best of our knowledge, a negative NPQ has not been reported for any photosynthetic organism. This finding was reproducible and will be the subject of further study.
Photosynthetic parameters Oxygenic photosynthesis results in the production of reactive oxygen species (ROS) when the input of light energy exceeds the saturation limit of the photosynthetic apparatus. Chloroplasts have evolved a number of mechanisms to mitigate the damaging effects of ROS, which include both the scavenging of ROS and the suppression of ROS production through NPQ. Each LHCII subunit binds 14 chlorophyll molecules and four carotenoids, including two xanthophylls: zeaxanthin (under high light conditions) or violaxanthin, the di-epoxide of zeaxanthin (under low light conditions). The xanthophyll cycle modulates light energy dissipation through the interconversion of zeaxanthin/violaxanthin.[15] Under low light conditions, LHCII selectively binds violaxanthin. However, under high light conditions, violaxanthin is converted to zeaxanthin in a process catalyzed by violaxanthin de-epoxidase (VDE). Zeaxanthin binds to LHCII, which undergoes a conformational change, facilitating the quenching of chlorophyll fluorescence.[16] Under low-light conditions, zeaxanthin is converted back to violaxanthin by the enzyme zeaxanthin epoxidase (ZE), which then replaces zeaxanthin in the LHCII xanthyophyll binding site. In dinoflagellates, zeaxanthin/violaxanthin are replaced by dinoxanthin and diadinoxanthin, respectively.
&
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
Production of ROS by K. brevis Thioredoxins (Trxs) are small oxidoreductases that function to maintain redox homeostasis within the chloroplast by thiol–disulfide exchange reactions. Numerous Trx-dependent regulatory networks have been identified within chloroplasts,[18] including VDE activity.[19] Trxs are also critical to the scavenging of excess ROS by reduction of 2-cysteine peroxiredoxins (2-Cys Prxs), enzymes whose sole function is the detoxification of peroxides.[20] Therefore, the rate of ROS production in the toxic and less toxic cultures was monitored by using the indicator dye 5-(and-6)-carboxy-2’,7’-dichlorodihydrofluorescein diacetate (carboxy-H2DCFDA). The dye is cell-permeable and, after hydrolysis of the acetate groups by esterases, might be oxidized to a fluorescent product. When normalized to chlorophyll fluorescence, the low-toxin-producing culture of K. brevis produces ROS at twice the rate of the high-toxin-producing culture (Figure 3 C). 4
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Papers as no surprise. PE ladder molecules are known to migrate to the cell membranes where they interact with membrane proteins. Brevetoxins have also been demonstrated to aggregate into artificial lipid bilayers.[29] Our findings contradict earlier localization studies: one which failed to identify brevetoxin in isolated chloroplasts by LC-MS/MS,[30] and a published abstract that reported that brevetoxins are located in secretory vesicles by using fluorescent anti-brevetoxin antibodies.[31] A follow-up paper to the latter report has not yet appeared. We also observed that the association with LHCII was reduced at neutral pH Figure 3. A) NPQ for the toxic and less toxic cultures, calculated for a sample adapted at or low light, conditions which were used for chloro363 mmol photons m 2 s 1 actinic light. The toxic culture exhibited two to four times more NPQ relative to the less toxic culture. Error bars represent standard deviation plast isolation. Our findings are consistent with the (n = 6). B) NPQs for the less toxic (^: 470, &: 520, ~: 645, *: 665 nm) and toxic (^: 470, &: localization of another dinoflagellate toxin, okadaic 520, ~: 645, *: 665 nm) algal cultures, calculated for each sample after 1 min of adaptaacid,[32] to the chloroplast and of the cyanobacterial tion to increasing levels of actinic lights. Both cultures exhibited an increase in NPQ with toxin, microcystin, to thylakoid membranes of cyanoactinic light, but a sharper increase in magnitude of the NPQs were found for the toxic culture (n = 4). C) ROS production normalized to chlorophyll fluorescence for the less bacteria.[33] toxic (^) and toxic (&) cultures of K. brevis. We also observed distinct differences in NPQ and ROS production between the toxic and low-toxin strains of K. brevis. This, coupled with the localization Conclusions of brevetoxin to the thylakoid membrane, suggests a role for In higher organisms, PE ladders exhibit extremely potent and brevetoxin in facilitating NPQ. The mechanism by which this varied biological activities. The molecular target of these toxins occurs is undetermined. However, we offer three possible sceis typically a transmembrane protein or ion channel containing narios: 1) brevetoxin interacts directly with LHCII, inducing the one or more a-helices, to which PE ladders bind with extremeconformational change associated with NPQ; 2) brevetoxin acly high affinities (nanomolar to picomolar Kd values). The bretivates an ion channel in the thylakoid membrane; 3) brevetoxvetoxins and ciguatoxin bind to and activate the voltage-gated in self-assembles into transmembrane pores, promoting cation sodium channel (VGSC).[4, 21] To date, at least ten different movement across the thylakoid membrane. LHCII is integral to light harvesting and energy transfer to sodium channel subtypes have been identified, with the brevephotosystem II as well as the dissipation of light energy as toxins and ciguatoxin specifically targeting the skeletal muscle heat, under conditions of high light, through the quenching of and cardiac channel subtypes (Nav1.4 and Nav1.5, respectively). chlorophyll fluorescence through the xanthophyll cycle. AnothThe type B brevetoxins show a lower affinity for the Nav1.5 carer transmembrane protein identified in higher plants, PsbS diac channel subtype.[22] Gambierol and gambieric acid A inhib(photosystem II subunit S), has been hypothesized to act as it binding of the brevetoxins to the VGSC.[23] Brevenal, a nona catalyst for the conformational change that occurs in LHCII toxic molecule containing only five polyether rings binds to during NPQ.[34] Although the gene for PsbS has been found in the VGSC at a site that is distinct from the brevetoxin site.[24] [25] green algae, it has yet to be identified in a dinoflagellate. It is Gambierol is also a potent potassium channel blocker, and absent from the Symbiodinium genome.[35] Furthermore, maitotoxin is a calcium channel activator.[26] A specific molecular target has yet to be identified for yessotoxin. However, tBLASTn analysis of a K. brevis EST library[36] containing 21 000 a recent study on yessotoxin, brevetoxin, and synthetic PE ladunique contigs against PsbS from Arabidopsis thaliana failed to ders has demonstrated that these molecules have strong affinireveal a transcript encoding for this protein. When brevetoxin ties for transmembrane a-helices.[27] Furthermore, a maitotoxin binds to the VGSC, it induces a conformational change from the closed to the open state of the ion channel, stabilizing the photoaffinity probe was used to label membrane proteins in open state and resulting in a depolarization of excitable memred blood cells. An unidentified 23 kD protein was labeled, and branes. Because K. brevis likely does not produce PsbS, we prothe labeling was inhibited in the presence of PbTx-2.[28] pose that brevetoxin binds to the quenched conformation of We have demonstrated that a fluorescent brevetoxin derivaLHCII and helps to stabilize it in the same way that it stabilizes tive localizes to the chloroplast of K. brevis, and photoaffinity the open conformation of the VGSC. labeling with a brevetoxin photoprobe has indicated that breLHCII is highly concentrated in the thylakoid membranes. In vetoxin interacts with LHCII. It could be argued that the labels fact, it has been described as the most abundant membrane influence the physiochemical properties of brevetoxin. The two protein on the planet. It is possible that our photoaffinity labels have distinct properties, yet each of these derivatives probe linked to LHCII simply as a result of its high density in migrate to the chloroplast. The Alexa Fluor dye, in particular, is the thylakoids and that the true brevetoxin target in chloronot membrane-permeable, and the Alexa Fluor brevetoxin conplasts is a less abundant membrane protein, such as an ion jugate (2) would seem even less likely to enter the thylakoid channel. Ion channels selective for chloride, potassium, or divamembrane than native brevetoxin. The finding that brevetoxlent cations such as calcium and magnesium have been found ins localize to the lipophilic thylakoid membranes should come ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
These are not the final page numbers! ÞÞ
5
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers gation for 5 min at 466 g, and the supernatant was discarded. The cells were resuspended in phosphate-buffered saline (PBS) buffer (400 mL) and vortexed for 1 min. This suspension was centrifuged for 10 min at 14 000 g, and the supernatant and the pellet were separated. The pellet was resuspended in PBS (500 mL). An aliquot (50 mL) of each fraction was reserved for a protein assay. Samples were quantitated against BSA standards (0, 1, 0.5, 0.25, 0.125, and 0.0625 mg mL 1) prepared in PBS by using the Coomassie Protein Assay Reagent (BioRad) according to the manufacturer’s instructions. Final protein concentrations ranged from 0.5–0.9 mg mL 1.
in outer- and inner-envelope membranes of chloroplasts, as well as from thylakoid membranes, by using the patch clamp technique and have been hypothesized to regulate photosynthesis by modulating the proton motive force.[37] Finally, brevetoxin has been demonstrated to promote selective cation transport across phosphatidylcholine vesicles containing 0.3 % PbTx-2 (1). Furthermore, the rate of transport is highly sensitive to the addition of cholesterol, which increases lateral membrane pressure and reduces lipid dynamics, thus suggesting the formation of membrane-spanning channels as opposed to carrier mechanisms. Subsequent studies on fluorescent brevetoxin derivatives revealed that brevetoxin aggregates into hexamers in lipid bilayers, forming a pore which mediates ion transport through membranes.[29b] Photosynthetic electron transport in thylakoid membranes results in the generation of transmembrane voltage Dy as well as DpH. Partial dissipation of Dy by counter ion movement could allow for an enhanced DpH and the concomitant activation of NPQ. The ecological consequences of the high variability of brevetoxin content among K. brevis isolates, and presumably a corresponding variation in capacities for NPQ and ROS production, have yet to be determined. PsbB mutants of Arabidopsis that are NPQ-deficient also produce ROS at an elevated rate.[38] These mutants demonstrate a reduced fitness in the field,[39] yet they appear to have the capacity to respond more rapidly to stress[40] and experience decreased herbivory when compared to the wild type.[38] In summary, we submit that PE ladders specifically target a transmembrane protein located in the thylakoid membrane and coincidentally associate with other transmembrane a-helices in higher organisms. However, the finding that brevetoxins might have an intrinsic biochemical role related to photosynthesis does not preclude an extrinsic function as a feeding deterrent. Indeed, secondary metabolites such as salicylic acid have been demonstrated to fill dual roles as feeding deterrents as well as hormone and growth regulators in plants.[41]
Spinach (5 g) was crushed in PBS (20 mL) and concentrated by centrifugation (5 min at 14 000 g). The supernatant was discarded. The pellet was resuspended in PBS buffer (5 mL, and an aliquot (50 mL) was reserved for a protein assay. The final protein concentration was 1.0–2.5 mg mL 1. Incubation of protein homogenates with photoaffinity probe (3) or photoaffinity label (8): Synthetic details for the preparation of 3 and 6–9 and compound characterization data can be found in the Supporting Information. The brevetoxin photoaffinity probe 3 or label 8 were incubated at room temperature with the soluble protein or membrane preparations at a probe/protein ratio of 0.2 mg probe per mg protein for 1 h. After incubation, the mixture was photolyzed (Vilber Lourmat transilluminator, Model FLX-20M) for 20 min at 312 nm. The soluble fraction was applied directly to a suspension of streptavidin beads (200 mL, Dynabeads, 2.8 mm or 1.0 mm, Life Technologies). Triton X-100 (1 %, 200 mL) was added to the membrane fractions (200 mL). The mixture was vortexed and centrifuged for 5 min at 14 000 g and applied to a suspension of streptavidin beads (200 mL). After incubation for 30 min at room temperature, the beads were separated from the solution by using a strong magnet and washed with PBS buffer (4 200 mL). Each wash was analyzed for protein content, and the final wash contained no detectable protein. Protein was eluted from the beads by using 0.1 % SDS solution (3 250 mL) at 100 8C for 5 min. An aliquot of each elution was analyzed by Bradford protein assay, concentrated to dryness by lyophilization (FreeZone, Model 2.5, 86 L min 1), and resuspended in water (10 mL) and loading buffer (2 mL, 0.25 m Tris-HCl, pH 6.8, 15 % SDS, 20 % glycerol, 25 % mercaptoethanol, 0.01 % Bromophenol Blue) for analysis by SDS-PAGE (12 % acryalamide with Tris·glycine SDS, pH 8.3) running buffer. Acrylamide gels were stained with PAGE Blue (Fermentas), according to the manufacturer’s instructions. The 25 kD band was excised from the SDS-PAGE gel. No protein was detected in the soluble fraction from the Bradford assay, nor were bands observed in the gel.
Experimental Section Incubation of K. brevis with the fluorescent brevetoxin derivative 2 or the Alexa Fluor 488 alkyne: Synthetic details for the preparation of 2 and compound characterization data can be found in the Supporting Information. Live K. brevis culture (3 mL) was incubated for 1 h in the dark with either the fluorescent brevetoxin derivative 2 or Alexa Fluor 488 alkyne (Molecular Probes). The final concentration of the Alexa Fluor 488 alkyne was 4 mm. Because the exact concentration of the brevetoxin derivative 2 was not known, the fluorescence of the solution of 2 was adjusted to be equal to that of the Alexa Fluor 488 alkyne. After incubation, the cells were centrifuged for 8 min at 130 g. The supernatant was discarded, and the cells were resuspended in RE medium (1 mL) and centrifuged again for 8 min at 130 g. The supernatant was discarded, and the cells were resuspended in RE medium (1 mL). The live cells were visualized by using a fluorescent microscope (Leica Leitz, Model DMRB) with a 450–490 nm band pass/515 nm long pass EX/EM filter.
In-gel digestion: In-gel digestion and LC-ESI MS/MS analysis was performed by the Proteomics and Mass Spectrometry Facility at the UMass Medical School. Gel slices were cut into 1 1 mm pieces and placed in 1.5 mL Eppendorf tubes with water (1 mL) for 30 min. The water was removed, and ammonium bicarbonate (50 mL, 250 mm) was added. For reduction, a solution of 1,4-dithiothreitol (DTT; 5 mL, 45 mm) was added, and the samples were incubated at 50 8C for 30 min. The samples were cooled to room temperature and, for alkylation, iodoacetamide solution (5 mL, 100 mm) was added and allowed to react for 30 min. The gel slices were washed with water (20 1 mL). The water was removed, ammonium bicarbonate/acetonitrile (50:50, 1 mL, 50 mm) was placed in each tube, and the samples were incubated at room temperature for 1 h. The solution was removed, and acetonitrile (200 mL) was added to each tube, at which point the gels slices turned opaque white. The acetonitrile was removed, and the gel slices were further dried in a Speed Vac. Gel slices were rehydrated in trypsin
Preparation of protein homogenates and membrane suspensions: K. brevis culture (200–400 mL) was concentrated by centrifu-
&
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
6
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
Full Papers (50 mL, 2 ng mL 1, Sigma) in 0.01 % ProteaseMAX Surfactant (Promega)/50 mm ammonium bicarbonate. Additional bicarbonate buffer was added to ensure complete submersion of the gel slices. Samples were incubated at 37 8C for 21 h. The supernatant of each sample was removed and placed in a separate 1.5 mL Eppendorf tube. Gel slices were further dehydrated with acetonitrile/1 % formic acid (80:20, 100 mL). The extract was combined with the supernatants of each sample. The samples were dried in a Speed Vac. Samples were dissolved in 25 mL of 5 % acetonitrile in 0.1 % trifluoroacetic acid prior to injection on LC-MS/MS.
spectral signature for each algal culture was determined at the beginning of the experiment. The maximum quantum efficiency of the photosystem II (PSII) photochemistry (referred to as “yield”) was measured by using a 25 min dark-adapted algal sample, thus allowing all reaction centers to be closed. Minimal fluorescence “F0” was recorded at 16 mmol photons m 2 s 1, and the maximal fluorescence “Fm” was determined after a 0.5 s saturating actinic red light pulse. The yield was calculated as: yield = Fv/Fm = (Fm F0)/Fm. The NPQ that estimated the rate of constant heat loss from PSII was calculated as follows: NPQ = (Fm/Fm’)/Fm’, where Fm’ is the maximum fluorescence measured after a 0.5 s saturating actinic red pulse of the lightadapted sample.
LC-MS/MS on Q Exactive: For peptide sequencing, a 3.0 mL aliquot was directly injected onto a custom-packed 2 cm 100 mm C18 Magic 5 m particle trap column. Labeled peptides were eluted and sprayed from a custom-packed emitter (75 mm 25 cm C18 Magic 3 mm particle) with a linear gradient from 95 % solvent A (0.1 % formic acid in water) to 35 % solvent B (0.1 % formic acid in acetonitrile) over 40 min at a flow rate of 300 nL min 1 on a Waters Nano Acquity UPLC system. Data-dependent acquisitions were performed on a Q Exactive mass spectrometer (Thermo Scientific), according to an experiment in which full MS scans from 300–1750 m/z were acquired at a resolution of 70 000, followed by 12 MS/MS scans acquired under HCD fragmentation at a resolution of 35 000 with an isolation width of 1.2 Da.
Statistics: Data for each excitation wavelength were not deconvoluted by using the reference spectral signature of each culture. As such, a set of photosynthetic characteristics for each excitation light for each algal culture was used for the comparison. To compare each photosynthetic characteristic of the two K. brevis cultures, a two-way full factorial ANOVA (two algal cultures four excitation lights) was conducted with SPSS 20 by using the univariate general linear model “GLM” procedure and Fisher’s least significant difference (LSD) to test for the main factors. A significance level of 0.05 was used for all ANOVAs.
Database searching: Raw data files were processed with Proteome Discoverer (v. 1.3). Charge state deconvolution and deisotoping were not performed. All MS/MS samples were analyzed by using Mascot (Matrix Science, London, UK; v. 2.4.1). Mascot was set up to search the NCBInr_20130907 database (32 265 160 entries), assuming the digestion enzyme stricttrypsin. Mascot was searched with a fragment ion mass tolerance of 0.050 Da and a parent ion tolerance of 10.0 ppm. The carbamidomethyl of cysteine was specified in Mascot as a fixed modification. Gln- > pyro-Glu of the N terminus, oxidation of methionine, and acetylation of the N terminus were specified in Mascot as variable modifications.
ROS production by K. brevis: A solution of 5-(and-6)-chloromethyl2’,7’-dichlorodihydrofluorescein diacetate acetyl ester (CMH2DCFDA, 11 mm, Life Technologies) was prepared in DMSO/culture medium (1:4) and diluted to 50 mm in culture medium. This stock was used for both the ROS assay and for preparation of a calibration curve. Calibration standards (5, 2, 0.5, 0.2 and 0 mm CMH2DCFDA in culture medium) were prepared by dilution of a stock solution (6 mm CM-H2DCFDA; 50 mm H2O2 ; 8 U mL 1 pig liver esterase in culture medium/PBS, 3:1). K. brevis cells were concentrated tenfold by centrifugation for 5 min at 17 g, followed by gentle resuspension in one-tenth the original volume of culture medium. Concentrated K. brevis cells (100 mL), culture medium (32 mL), and CM-H2DCFDA stock (50 mm, 18 mL) were added to a 96-well microplate for a final probe concentration of 6 mm. Plates were incubated at ambient temperature under laboratory lighting. Fluorescence was monitored (lex = 485 nm, lem = 538 nm) hourly for 7 h. Samples and standards were assayed in triplicate. Because cell density and volume varied between cultures, ROS concentration was normalized to chlorophyll content. K. brevis culture (40 mL) was centrifuged for 5 min at 200 g, and the resulting pellet was homogenized in MeOH (600 mL). The MeOH homogenate was centrifuged for 10 min at 466 g to remove cell debris, and the supernatant (150 mL) was added to a 96-well microplate. Chlorophyll fluorescence was measured at lex = 485 nm and lex = 680 nm.
Criteria for protein identification: Scaffold (v. 4.1.1, Proteome Software Inc., Portland, OR) was used to validate MS/MS-based peptide and protein identifications. Peptide identifications were accepted if they could be established at greater than 80.0 % probability by the Peptide Prophet algorithm[42] with Scaffold delta-mass correction. Protein identifications were accepted if they could be established at greater than 90.0 % probability and contained at least two identified peptides. Protein probabilities were assigned by the Protein Prophet algorithm.[43] Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony. Proteins sharing significant peptide evidence were grouped into clusters. Photosynthetic characteristics: The photosynthetic characteristics of each algal culture were measured by using pulsed amplitude modulation (PAM) fluorometry.[44] This was performed with a WALZ tetrafluorometer Phyto-PAM equipped with an Emitter Detection unit (ED; http://www.walz.com/), which excited the photosynthetic pigments by using four consecutive pulses of excitation light of different wavelengths: blue (470 nm), green (520 nm), light red (645 nm), and far red (665 nm). The resulting fluorescence, recorded at room temperature (~ 20 8C), was measured subsequent to pulses of different excitation light intensities and one 0.5 s saturating pulse of actinic red light by using algae adapted to different light regimes. All fluorescence measurement routines were conducted in sextuplicate by using algal cultures standardized to 240.8 S.D. 3.0 mg Chl L 1 and 246 S.D.1.0 mg Chl L 1 for the nontoxic and toxic cultures, respectively. Additionally, the reference
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
These are not the final page numbers! ÞÞ
Acknowledgements This work was supported by Bridge funding from the FIU Division of Research and the FIU College of Arts and Sciences. Keywords: bioorganic chemistry · brevetoxin · fluorescent probes · photoaffinity labeling [1] G. Rossini, P. Hess in Molecular, Clinical and Environmental Toxicology, Vol. 100 (Ed.: A. Luch), Birkhuser, Basel, 2010, pp. 65 – 122. [2] D. G. Baden, A. J. Bourdelais, H. Jacocks, S. Michelliza, J. Naar, Environ. Health Perspect. 2005, 113, 621 – 625.
7
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
Full Papers [3] a) S. M. Watkins, A. Reich, L. E. Fleming, R. Hammond, Mar. Drugs 2008, 6, 431 – 455; b) L. E. Fleming, L. C. Backer, D. G. Baden, Environ. Health Perspect. 2005, 113, 618 – 620. [4] K. Yamaoka, M. Inoue, M. Hirama, Forensic Toxicol. 2011, 29, 125 – 131. [5] A. Tubaro, V. Dell’Ovo, S. Sosa, C. Florio, Toxicon 2010, 56, 163 – 172. [6] R. M. Errera, L. Campbell, Proc. Natl. Acad. Sci. USA 2011, 108, 10597 – 10601. [7] W. G. Sunda, C. Burleson, D. R. Hardison, J. S. Morey, Z. Wang, J. Wolny, A. A. Corcoran, L. J. Flewelling, F. M. Van Dolah, Proc. Natl. Acad. Sci. USA 2013, 110, 10223 – 10228. [8] D. R. Hardison, W. G. Sunda, D. Shea, R. W. Litaker, PLoS One 2013, 8, e58545. [9] a) J. Hong, S. Talapatra, J. Katz, P. A. Tester, R. J. Waggett, A. R. Place, PLoS One 2012, 7, e36845; b) R. J. Waggett, D. Hardison, P. A. Tester, Mar. Ecol. Prog. Ser. 2012, 444, 15 – 30. [10] J. Kubanek, T. W. Snell, C. Pirkle, Limnol. Oceanogr. 2007, 52, 1026 – 1035. [11] E. K. Prince, K. L. Poulson, T. L. Myers, R. D. Sieg, J. Kubanek, Harmful Algae 2010, 10, 39 – 48. [12] D. P. Nair, M. Podgorski, S. Chatani, T. Gong, W. Xi, C. R. Fenoli, C. N. Bowman, Chem. Mater. 2014, 26, 724 – 744. [13] J. Sumranjit, S. J. Chung, Molecules 2013, 18, 10425 – 10451. [14] Z. Liu, H. Yan, K. Wang, T. Kuang, J. Zhang, L. Gui, X. An, W. Chang, Nature 2004, 428, 287 – 292. [15] M. Fuciman, M. M. Enriquez, T. Polvka, L. Dall’Osto, R. Bassi, H. A. Frank, J. Phys. Chem. B 2012, 116, 3834 – 3849. [16] J. Standfuss, A. C. T. van Scheltinga, M. Lamborghini, W. Khlbrandt, EMBO J. 2005, 24, 919 – 928. [17] R. M. Errera, A. Bourdelais, M. Drennan, E. Dodd, D. Henrichs, L. Campbell, Toxicon 2010, 55, 195 – 203. [18] L. Nikkanen, E. Rintamki, Philos. Trans. R. Soc. London Ser. B 2014, 369, 20130224. [19] M. Hall, A. Mata-Cabana, H. E. kerlund, F. J. Florencio, W. P. Schrçder, M. Lindahl, T. Kieselbach, Proteomics 2010, 10, 987 – 1001. [20] J. Kçnig, M. Baier, F. Horling, U. Kahmann, G. Harris, P. Schrmann, K.-J. Dietz, Proc. Natl. Acad. Sci. USA 2002, 99, 5738 – 5743. [21] M. Poli, T. J. Mende, D. G. Baden, Mol. Pharmacol. 1986, 30, 129 – 135. [22] M.-Y. B. Dechraoui, J. S. Ramsdell, Toxicon 2003, 41, 919 – 927. [23] M. Inoue, M. Hirama, M. Satake, K. Sugiyama, T. Yasumoto, Toxicon 2003, 41, 469 – 474. [24] E. P. Gold, H. M. Jacocks, A. J. Bourdelais, D. G. Baden, Harmful Algae 2013, 26, 12 – 19. [25] E. Cuypers, Y. Abdel-Mottaleb, I. Kopljar, J. D. Rainier, A. L. Raes, D. J. Snyders, J. Tytgat, Toxicon 2008, 51, 974 – 983. [26] A. Yokoyama, M. Murata, Y. Oshima, T. Iwashita, T. Yasumoto, J. Biochem. 1988, 104, 184 – 187.
&
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
[27] a) S. Ujihara, T. Oishi, R. Mouri, R. Tamate, K. Konoki, N. Matsumori, M. Murata, Y. Oshima, N. Sugiyama, M. Tomita, Bioorg. Med. Chem. Lett. 2010, 20, 6443 – 6446; b) S. Ujihara, T. Oishi, K. Torikai, K. Konoki, N. Matsumori, M. Murata, Y. Oshima, S. Aimoto, Bioorg. Med. Chem. Lett. 2008, 18, 6115 – 6118; c) M. Mori, T. Oishi, S. Matsuoka, S. Ujihara, N. Matsumori, M. Murata, M. Satake, Y. Oshima, N. Matsushita, S. Aimoto, Bioorg. Med. Chem. 2005, 13, 5099 – 5103. [28] K. Konoki, M. Hashimoto, K. Honda, K. Tachibana, R. Tamate, F. Hasegawa, T. Oishi, M. Murata, Heterocycles 2009, 79, 1007 – 1017. [29] a) S. Matile, N. Berova, K. Nakanishi, Chem. Biol. 1996, 3, 379 – 392; b) S. Matile, K. Nakanishi, Angew. Chem. Int. Ed. Engl. 1996, 35, 757 – 759; Angew. Chem. 1996, 108, 812 – 814. [30] F. M. Van Dolah, M. L. Zippay, L. Pezzolesi, K. S. Rein, J. G. Johnson, J. S. Morey, Z. Wang, R. Pistocchi, J. Phycol. 2013, 49, 1118 – 1127. [31] K. L. Vigna, I. Quesada, P. Verdugo, Biophys. J. 2012, 102, 319a – 320a. [32] J. Zhou, L. Fritz, Phycologia 1994, 33, 455 – 461. [33] F. M. Young, C. Thomson, J. S. Metcalf, J. M. Lucocq, G. A. Codd, J. Struct. Biol. 2005, 151, 208 – 214. [34] A. V. Ruban, M. P. Johnson, C. D. Duffy, Biochim. Biophys. 2012, 1817, 167 – 181. [35] E. Shoguchi, C. Shinzato, T. Kawashima, F. Gyoja, S. Mungpakdee, R. Koyanagi, T. Takeuchi, K. Hisata, M. Tanaka, M. Fujiwara, Curr. Biol. 2013, 23, 1399 – 1408. [36] K. B. Lidie, J. C. Ryan, M. Barbier, F. M. Van Dolah, Mar. Biotechnol. 2005, 7, 481 – 493. [37] V. Checchetto, E. Teardo, L. Carraretto, E. Formentin, E. Bergantino, G. M. Giacometti, I. Szabo, Biophys. Chem. 2013, 182, 51 – 57. [38] H. J. Jnknp, M. Frenkel, I. Zulfugarov, M. Reichelt, A. Krieger-Liszkay, Y. Mishra, J. Gershenzon, J. Moen, C.-H. Lee, S. Jansson, PLoS One 2013, 8, e53232. [39] C. Klheim, J. gren, S. Jansson, Science 2002, 297, 91 – 93. [40] M. Frenkel, C. Klheim, H. J. Jnknp, O. Skogstrçm, L. Dall’Osto, J. gren, R. Bassi, T. Moritz, J. Moen, S. Jansson, BMC Plant Biol. 2009, 9, 12. [41] M. R.-S. Vicente, J. Plasencia, J. Exp. Bot. 2011, 62, 3321 – 3338. [42] A. Keller, A. I. Nesvizhskii, E. Kolker, R. Aebersold, Anal. Chem. 2002, 74, 5383 – 5392. [43] A. I. Nesvizhskii, A. Keller, E. Kolker, R. Aebersold, Anal. Chem. 2003, 75, 4646 – 4658. [44] a) U. Schreiber in Chlorophyll a Fluorescence: A Signature of Photosynthesis (Eds.: G. C. Papageorgiou, Govinjee), Springer, Dordrecht, 2004, pp. 279 – 319; b) E. Murchie, T. Lawson, J. Exp. Bot. 2013, ert208. Received: November 19, 2014 Published online on && &&, 0000
8
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!
FULL PAPERS Seeing the light: A fluorescent brevetoxin derivative localizes to the chloroplast of the Florida red tide dinoflagellate, Karenia brevis. Photoaffinity labeling of a receptor protein identified the light-harvesting complex II (LHCII) as the native brevetoxin receptor. A brevetoxin-deficient strain of K. brevis was unable to perform some of the essential functions of LHCII.
ChemBioChem 0000, 00, 0 – 0
www.chembiochem.org
These are not the final page numbers! ÞÞ
R. T. Cassell, W. Chen, S. Thomas, L. Liu, K. S. Rein* && – && Brevetoxin, the Dinoflagellate Neurotoxin, Localizes to Thylakoid Membranes and Interacts with the Light-Harvesting Complex II (LHCII) of Photosystem II
9
0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
