In Vitro Hepatotoxicity of the Cyanobacterial Alkaloid Cylindrospermopsin and Related Synthetic Analogues

Maria T. Runnegar*,1, Chaoyu Xie{dagger}, Barry B. Snider{dagger}, Grier A. Wallace{ddagger}, Steven M. Weinreb{ddagger} and John Kuhlenkamp*

* USC Research Center for Liver Diseases, University of Southern California, Los Angeles, California 90033; {dagger} Department of Chemistry, Brandeis University, Waltham, Massachusetts 02454; and {ddagger} Department of Chemistry, Pennsylvania State University, University Park, Pennsylvania 16802

Received September 28, 2001; accepted January 11, 2002


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Cylindrospermopsin (CY), a sulfate ester of a tricyclic guanidine substituted with a hydroxymethyluracil, is a cyanobacterial toxin of increasing environmental import as it frequently occurs in drinking water reservoirs. As a toxin, CY mainly targets the liver but also involves other organs. In hepatocytes CY inhibits the synthesis of protein and of glutathione, leading to cell death. The total chemical synthesis of CY has recently been reported (Xie et al., 2000Go, J. Am. Chem. Soc. 22, 5017–5024). The synthesis has provided analogues of CY to study aspects of the relationship between chemical structure and activity that contribute to toxicity. Protein synthesis inhibition was measured in vitro using a rabbit reticulocyte system. Primary cultures of rat hepatocytes were used to determine the biological activity of CY and analogues in intact cells. Protein synthesis and cell glutathione levels were measured. We could distinguish between CY transport and biological activity by comparing the results in vitro to those in intact cells. The role of the sulfate group in CY toxicity was examined by comparing biological effects of CY with that of CY-DIOL (synthetic CY lacking the sulfate group). The sulfate group was found not to play a role in CY activity or in its uptake into cells, since there was no significant difference in biological activity in vitro or in cells between natural CY and CY-DIOL. The orientation of the hydroxyl group at C7 also had no impact on biological activity or transport of CY, since the C7 epimer of CY (EPI-CY) and the corresponding diol (EPI-DIOL) had activity similar to RAC-CY in vitro and in intact cells. AB-MODEL, the analogue lacking an intact C ring, and the methyl and hydroxyl groups of ring A could inhibit protein synthesis (but at concentrations 500–1000-fold higher than natural CY). Other structurally simpler synthetic analogues lacked biological activity.

Key Words: cylindrospermopsin; GSH synthesis inhibition; hepatocyte toxicity; protein synthesis inhibition; toxic cyanobacteria.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
A bloom of the cyanobacterium Cylindrospermopsis raciborskii contaminating the water supply is believed to have caused an outbreak of hepatoenteritis in 1979 that resulted in the hospitalization of 140 children and some adults on Palm Island, Queensland, Australia (Bourke et al., 1983Go; Byth, 1980Go). Cultures of the cyanobacterium isolated from the same water body were shown to be toxic in mice, causing mainly hepatocellular degeneration, necrosis, and lipidosis but also affecting other organs (Hawkins et al., 1985Go). More recent studies have shown that C. raciborskii is commonly found in water reservoirs, originally reported from Australia but since found in water bodies worldwide, including Europe and the United States (Padisak, 1997Go).

Cylindrospermopsin (CY) was shown to be the toxic principle of C. raciborskii (see Fig. 1Go for the structures of CY and analogues). It is a stable compound not removed by boiling (Norris et al., 1999Go). CY is a novel alkaloid of polyketide origin: a sulfate ester of a tricyclic guanidine substituted with a hydroxymethyluracil (Ohtani et al., 1992Go). Feeding studies of cultures have shown that CY is an acetogenin with guanidinoacetic being the starter unit for the polyketide chain (Burgoyne et al., 2000Go).



View larger version (18K):
[in this window]
[in a new window]
 
FIG. 1. Structures of cylindrospermopsin and analogues.

 
Different groups have investigated, in mice, the parenteral and oral toxicity of CY-containing cyanobacterial extracts, as well as purified CY (Falconer et al., 1999Go; Hawkins et al., 1985Go; Seawright et al., 1999Go; Shaw et al., 2000Go; Terao et al., 1994Go). In all reports, the liver is the main target organ of CY toxicity but CY also causes significant lesions in the kidney, thymus, and spleen.

In mammalian cell cultures, CY caused significant dose-dependent cell death. Cell death was always preceded by profound decreases in glutathione (GSH) levels in hepatocytes (Runnegar et al., 1994Go). The mechanism of the fall in GSH levels by CY was investigated (Runnegar et al., 1995Go), and CY was found to be a very potent inhibitor of the synthesis of GSH. CY has also been shown to inhibit protein synthesis in vitro, using a rabbit reticulocyte system (Terao et al., 1994Go).

The relationship between these biological effects and the in vivo toxicity of CY has not been fully elucidated. Changes found in liver by electron microscopy of mice dosed with CY have several features in common with those found in mice dosed with the protein synthesis inhibitor cycloheximide (Terao et al., 1994Go). In both cases there was a dissociation of microsomes from the endoplasmic reticulum, leading the authors to propose that protein synthesis inhibition plays a role in CY toxicity in vivo. The liver of CY-dosed mice, but not that of cycloheximide-dosed mice, showed membrane proliferation and fat droplet accumulation, indicating that mechanisms other than protein synthesis inhibition must contribute to CY toxicity.

The complexity of the biological activity of CY is matched by the complexity of its chemical structure. The synthesis of CY has been a challenge for a number of laboratories (Djung et al., 2000Go; Harvey, 1996Go; Keen and Weinreb, 2000Go; Looper and Williams, 2001Go; McAlpine and Armstrong, 2000Go; Snider and Harvey, 1995Go; Snider and Xie, 1998Go). The total synthesis of CY (RAC-CY) from 4-methoxy-3-methylpyridine was recently completed (Xie et al., 2000Go). The complete synthesis of epi-cylindrospermopsin (EPI-CY) has also been recently accomplished (Heintzelman et al., 2001Go). The latter work indicated that the stereochemistry reported for cylindrospermopsin and 7-EPI-CY at C-7 should be switched.

The steps in the synthesis of CY provide analogues and intermediates that can be tested for biological activity in an effort to better understand the different biological components of CY toxicity and their relationship to structural features of CY. In this study, we compare the toxicity of natural CY with synthetic analogues and intermediates.

Because of its hydrophilic character, CY (Fig. 1Go) is very unlikely to be cell-permeant and therefore would need to be transported across the cell membrane to result in toxicity. We hypothesized that the bulky, charged sulfate group at position C-12 could play a role in uptake, with cellular uptake mediated by one or more members of the solute carrier family (Waldegger et al., 2001Go) or other transporters/exchangers.

We used CY-DIOL, an intermediate from the CY synthesis (Xie et al., 2000Go) that lacks the sulfate group, to test this in cultured rat hepatocytes. We examined the toxicity of EPI-CY and EPI-DIOL (epimers of CY at C-7) to determine whether the stereochemistry of the hydroxyl group at position C-7 plays a role in the biological activity of CY. We also examined the toxicity of the chemically simpler analogues AB-MODEL, AC-MODEL, and URACIL-MODEL to determine what parts of the chemical structure of CY are important for biological activity.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Materials.
The culture medium for isolated rat hepatocytes, DME/F12 and sulfur amino acid-free DME, were purchased from Gibco BRL Life Technologies (Rockville, MD). L-[35S]methionine and [35S]protein labeling mix were purchased from Amersham Pharmacia Biotech (Piscataway, NJ) and NEN Life Sciences (Boston, MA), respectively. LDH was measured using a kit from Sigma Chemicals (Kit DG1340-K). All other reagents were from Sigma Chemicals (St. Louis, MO) or from other routine commercial sources.

C. raciborskii cultures.
The cyanobacterium C. raciborskii was isolated from Solomon Dam, Palm Island in Northern Queensland (Hawkins et al., 1985Go). Unialgal cultures were grown in modified CHU-10 medium buffered with 0.02 M HEPES (N-2`-hydroxyethylpiperazine-N`-2-ethanesulfonic acid), pH adjusted to 7.5 and containing NaNO3, 0.17 g/l. All cultures were incubated at 25 ± 1°C with continuous aeration and illumination from cool-white fluorescent light at an intensity of about 90 µE.s–1.m–2.

Isolation of natural cylindrospermopsin (CY).
C. raciborskii from cultures was harvested by centrifugation and freeze-dried. This freeze-dried material was extracted twice with distilled water for 1 h at 4°C. The combined extracts were evaporated to dryness in a vacuum, and the residue was suspended in a minimal volume of 1:1 MEOH/H2O. This was applied to a Toyopearl HW40F pre-column connected to a Toyopearl HW40F column. Fractions were eluted with 1:1 MEOH/H2O. Toxicity of fractions in the original publication that first reported the structure of the alkaloid CY was determined by mouse bioassay (ip injection in male CH3 mice) (see Ohtani et al., 1992Go). Toxic fractions were purified further by reverse-phase flash chromatography on ODS YMC GEL (120A) by sequential elution with H2O, 1:9 MEOH/H2O, 1:1 MEOH/H2O, and MEOH. Final purification of CY from the 10% MEOH in H2O fraction was by reversed-phase HPLC with an Econosil C8 column. The LD50 of the purified alkaloid CY was 2.1 mg/kg (ip in CH3 mice) at 24 h after dosing, with symptoms and histology that could not be distinguished from those described in mice dosed with cyanobacterial extracts (Hawkins et al., 1985Go). For subsequent isolations of natural CY, the chemical structure was determined by NMR, while biological activity was determined either by ip injection in mice or by the effect of CY on isolated rat hepatocyte incubations (Runnegar et al., 1994Go).

The isolation and chemical characterization of natural CY used in this work were done by Drs. David Burgoyne and Thomas Hemscheidt in the laboratory of Dr. Richard Moore in the Chemistry Department of the University of Hawaii (Burgoyne et al., 2000Go; Ohtani et al., 1992Go).

Synthetic CY and analogues.
The syntheses of RAC-CY, CY-DIOL, and AB-MODEL are described in Snider and Xie (1998) and in Xie et al. (2000), the synthesis of AC-MODEL is described in Snider and Harvey (1995), and the synthesis of URACIL-MODEL is described in Harvey (1996). The syntheses of EPI-CY and EPI-DIOL are described in Heintzelman et al. (2001). Natural CY is optically active, while synthetic CY (RAC-CY) is racemic, as are all other synthetic compounds used in this study, including EPI-CY.

All test compounds were dissolved in water with the exception of URACIL-MODEL, which was dissolved in DMSO. Because of the small amounts available of the various synthetic compounds, their concentrations are approximate. Weights of necessity were not exact and the presence of trace amounts of salts, etc., used in the synthesis would also contribute to the uncertainty. Single stock solutions were prepared for each compound, assuming that the nominal weight was accurate. For all of the measurements reported here, the working solutions of the synthetic compounds were diluted from the one original stock. Therefore, the effects of a particular analogue in vitro and in experiments with intact hepatocytes are directly comparable.

In vitroprotein synthesis.
This was measured using the Rabbit Reticulocyte Lysate System of Promega (Madison, WI; catalog number L4960). The incorporation of [35S]-methionine (1.175 Ci/mmol) into luciferase protein was used to measure protein synthesis. One µCi of [35S]-methionine was added to each incubation. The effect of CY and related compounds was determined by comparing the incorporation of label into luciferase protein with that of control incubations.

Hepatocyte cell culture.
Isolation of rat hepatocytes was done aseptically according to the method of Moldeus et al. (1978). Initial cell viability was >=90% as determined by 0.2% Trypan blue exclusion. The plating medium was DME/F12 containing high glucose, 10% fetal bovine serum, insulin (1 µg/ml), and hydrocortisone (50 nM) supplemented with 1 mM methionine. Cells (2 ml suspension of 0.8–1.0 x 106 cells) were plated in six-well cluster plates (35 mm), precoated with rat tail collagen, and incubated at 37°C in 5% CO2 and 95% air. Cells were allowed to attach for 2 to 3 h, and the medium was changed to remove the fetal bovine serum and any unattached cells. Natural CY, synthetic CY, synthetic EPI-CY, and intermediates were added at the concentrations stated in the Results section. Cells were incubated for 17 h followed by a 2-h incubation in sulfur amino acid-free medium containing about 2–3 µCi of [35S] methionine/ml (1 Ci/µmol) to determine the effect of CY on protein synthesis. Aliquots of medium at the end of the 17 h and the 2 h incubations were taken for measurement of lactate dehydrogenase (LDH) activity and for counting of radioactivity.

Hepatocyte extraction.
At the end of the incubation, hepatocytes were washed in phosphate-buffered saline (PBS) followed by further washing in PBS containing 1 mM methionine. The cells were then scraped in 0.5 ml of PBS. LDH activity and protein levels were measured. Ten percent trichloroacetic acid (TCA) was added to an equivalent volume of hepatocyte extract to precipitate protein.

Measurement of the toxicity to hepatocytes of CY.
Toxicity (cell lysis) due to CY was measured by the release of LDH from the cytosol into the medium (Runnegar et al., 1994Go). LDH was measured in the medium and in the cell extract. Percentage LDH release (cell death) was the LDH activity in the medium as a percentage of total LDH (cellular + medium).

Effect of CY on protein synthesis in hepatocytes.
The protein precipitate obtained by centrifugation, following the addition of 10% TCA to the hepatocyte extract from cells that had been incubated with [35S] methionine after pretreatment with and without the test compounds, was resuspended in 5% TCA and centrifuged again. The cell pellet was then dissolved in 0.5 ml of 0.2 N NaOH. A 0.1-ml aliquot of the NaOH solution was used to measure radioactivity by scintillation counting. The radioactivity of the samples reflected the relative incorporation of [35S] methionine into the protein fraction of hepatocytes during incubations and is a measure of the rate of protein synthesis. The effect of CY and related compounds on protein synthesis was determined by comparing the [35S] methionine incorporation in treated incubations with that of parallel control incubations.

Measurement of reduced glutathione (GSH).
Cellular GSH was measured in the 10% TCA supernatant of the cell extract (see previous section) by the method of Tietze (1969).

Statistical analysis.
For cultured hepatocytes, each cell preparation was derived from one animal, and duplicate plates were used for each condition. The mean of each duplicate from one experiment was considered n = 1. Comparison between controls and dose groups was done by one-way analysis of variance (ANOVA) with Dunnett's multiple comparison test at the significance level of alpha = 0.05. For most treatments, n was between 3 and 6. Because of the very limited amounts of the synthetic compounds for some incubations that required relatively higher concentrations or were expected to only corroborate other findings, n, the number of independent experiments, was only 1 or 2. IC50 values were calculated with a model that assumes the linear relationship between measurements and dose with random (experiments) effect. The significance level, alpha, was set at 0.1 to estimate the 90% confidence limits and standard error. All statistical analyses were performed on raw data, which were then converted to percentage of control values for graphical representation of mean ± SE.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The effect of CY on in vitro protein synthesis.
Natural CY and racemic synthetic CY (RAC-CY) inhibited protein synthesis dose-dependently in the rabbit reticulocyte lysate system (Figs. 2A and 2BGo). IC50 values for natural CY and RAC-CY were 0.21 and 0.57 µM, respectively. To determine whether the sulfate group is necessary for protein-synthesis inhibition, the corresponding diol (CY-DIOL), which lacked the sulfate, was tested. CY-DIOL was found to be as potent as CY in inhibiting protein synthesis in vitro (IC50 = 0.20 µM) (Fig. 2CGo). Concentrations of 1 µM of any of the three compounds inhibited protein synthesis completely.



View larger version (39K):
[in this window]
[in a new window]
 
FIG. 2. In vitro protein-synthesis inhibition of translation in a rabbit reticulocyte lysate system with luciferase mRNA as the message. Concentration-dependence on inhibition for A: natural CY; B: RAC-CY; C: CY-DIOL; D:EPI-CY; and E: EPI-DIOL. Concentrations of CY and analogues are micromolar in the complete incubations. [35S]Methionine incorporation for control incubations was 6.96 ± 0.62% of added label. Each bar represents the rate of protein synthesis as percentage of control incubations ± SE for n = 3–6 incubations; *p < 0.05 when compared to control.

 
The stereochemistry at C-7 is unlikely to play a role in the biological activity of natural optically active CY, since racemic EPI-CY and the corresponding EPI-DIOL inhibited protein synthesis at concentrations micromolar or less (Figs. 2D and 2EGo). IC50 were 0.48 and 1.65 µM, respectively.

The effect of CY on protein synthesis of cultured hepatocytes.
Natural CY dose-dependently inhibited protein synthesis in hepatocytes (Fig. 3AGo). CY-DIOL was as, or slightly more effective than CY in inhibiting protein synthesis in hepatocytes (Fig. 3BGo). The IC50 for CY and CY-DIOL were 1.28 and 0.76 µM, respectively. These results indicate that the sulfate group in CY is not necessary for cellular uptake of CY.



View larger version (55K):
[in this window]
[in a new window]
 
FIG. 3. Protein synthesis inhibition in cultured rat hepatocytes, measured as the decrease in incorporation of [35S]methionine in TCA-precipitated protein after a 19-h incubation. Concentration-dependence of inhibition for A: natural CY; B: CY-DIOL; C: EPI-CY; and D: EPI-DIOL. Concentrations of CY and analogues are micromolar in the final incubations. [35S]Methionine incorporation for control incubations was 2.52 ± 0.25% of added label. Each bar represents the rate of protein synthesis as a percentage of control incubations ± SE for n = 3–6 cell preparations for most incubations; *p < 0.05 when compared to control.

 
The EPI-CY and EPI-DIOL inhibited protein synthesis in hepatocytes dose-dependently (Figs. 3C and 3DGo) with IC50s of 2.66 and 3.50 µM, respectively. When the racemic nature and the uncertainty in the original amounts of the synthetic analogues are taken in account, it is most likely that the stereochemistry at C-7 also does not have a major effect on cellular uptake of CY.

The effect of CY on GSH levels of hepatocytes and on cell lysis.
We had shown previously that natural CY causes dose- and time-dependent loss of the cellular antioxidant GSH by inhibiting its synthesis (Runnegar et al., 1994Go, 1995Go). The effect of natural CY on cell GSH was compared to that of synthetic RAC-CY (Figs. 4A and 4BGo). IC50 values were 2.38 and 8.99 µM respectively. When the racemic nature of RAC-CY and the uncertainty in the original amounts of the synthetic analogues are taken into account, the decrease in GSH by RAC-CY is almost equivalent to that of natural CY. CY-DIOL was as potent as CY in lowering cell GSH levels, with the IC50 at 2.33 µM (Fig. 4CGo).



View larger version (28K):
[in this window]
[in a new window]
 
FIG. 4. Concentration-dependence for the decrease in cellular levels of GSH 19 h after the addition of CY and derivatives to cultured rat hepatocytes. For A: natural CY; B: RAC-CY; and C: CY-DIOL. Concentrations of CY and analogues are micromolar in the final incubations. Cell GSH for control incubations was 60.52 ± 5.56 nmol/mg protein. Each bar represents the GSH level as a percentage of control incubations ± SE for n = 3–7 cell preparations for most incubations; *p < 0.05 when compared to control.

 
EPI-CY and EPI-DIOL also decreased GSH levels in hepatocytes. Hepatocytes incubated with 6.25 µM EPI-CY or EPI-DIOL had cell GSH levels of 39 ± 2.5 and 66 ± 14% of control, respectively.

We had shown that cell death of hepatocytes occurred at higher doses of CY following decreases in GSH (Runnegar et al., 1994Go, 1995Go). We confirmed that cell death followed the inhibition of protein and GSH synthesis in hepatocytes for the synthetic compounds. RAC-CY (20 µM) increased cell death in hepatocytes from 14% (controls) to 23%, while CY-DIOL (10 µM) increased cell death to 38%. In the same experiment, cell death in hepatocytes incubated with natural CY (10 µM) was 33.4%. Similarly, the C-7 epimers EPI-CY (12.5 µM) and EPI-DIOL (50 µM) also increased cell death of hepatocytes to 38.5% and 35%, respectively (control = 23%).

Biological activity of simpler intermediates or analogues of CY.
Synthetic analogues AC-MODEL (Snider and Harvey, 1995Go) or MODEL-URACIL (Harvey, 1996Go) had no effect on in vitro protein synthesis at concentrations of 800 µM and 2000 µM, respectively.

We were able to show inhibition of in vitro protein synthesis by AB-MODEL (Xie et al., 2000Go). This inhibition was dose-dependent (Fig. 5AGo). A similar concentration was needed for protein synthesis inhibition in hepatocytes (Fig. 5BGo). Two hundred fifty to 500 µM AB-MODEL had no detectable effect on cell GSH. The amount of AB-MODEL available precluded incubations at sufficiently high concentrations (mM or more) to show significant decreases in cell GSH or increases in cell lysis.



View larger version (34K):
[in this window]
[in a new window]
 
FIG. 5. AB-MODEL inhibits protein synthesis. Concentration dependence: (A) In vitro protein synthesis inhibition of translation in a rabbit reticulocyte lysate system, with luciferase mRNA as the message. [35S]Methionine incorporation for control incubations was 6.96 ± 0.62% of added label. (B) Protein synthesis inhibition in cultured rat hepatocytes measured as the decrease in incorporation of [35S]methionine in TCA-precipitated protein after a 19-h treatment. [35S]Methionine incorporation for control incubations was 2.52 ± 0.25% of added label. Concentrations of CY and analogues are micromolar in the final incubations. Each bar represents the rate of protein synthesis as a percentage of control incubations ± SE for n = 3–4 cell preparations; *p < 0.05 when compared to control.

 

    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
CY is a complex sulfated alkaloid. Biological effects of natural CY have been described, but little is known of how the different functional groups influence biological activity and hence toxicity. For toxicity to occur in cells or animals, CY needs to be able to cross the cell membrane to reach its intracellular target(s). Because of its hydrophilic nature, CY is unlikely to be cell-permeant and would therefore require carrier transport into cells. We postulated that the bulky sulfate group would be necessary for transport across the cell membrane and that CY transport would perhaps be mediated by a member of the solute-carrier family (Waldegger et al., 2001Go). In this study we examined whether the determinants of uptake could be separated from the intracellular biological effects of CY.

If we could show that a synthetic intermediate lacking the sulfate group retained in vitro biological activity, we could then use the same compound to explore the role of the sulfate group in the transport of CY across the cell membrane of hepatocytes.

The in vitro inhibition of protein synthesis is independent of transport. We established that the diol (CY-DIOL), lacking the sulfate group, was essentially as potent as native CY in inhibiting the rate of protein synthesis in the rabbit reticulocyte lysate system, thus proving that the sulfate group is not necessary for inhibition of translation. Since the C-7 epimer of CY (EPI-CY) and the corresponding diol (EPI-DIOL) also inhibited protein synthesis at similar concentrations, we can conclude that the orientation of the hydroxyl group at C-7 has no effect on the inhibition of translation.

To determine whether the sulfate group is important in the cellular uptake of CY, we used the cultured rat hepatocyte model to determine the activity of the various CY compounds. The rat hepatocyte model was chosen because the liver is the major target organ of CY toxicity, and we already had extensively characterized the toxic effect of CY in this system in our previous work (Runnegar et al., 1994Go, 1995Go).

We found that neither the sulfate group at C-12 nor the hydroxyl group orientation at C-7 affected toxicity to any significant degree, indicating that transport into the cell must be mediated through some other structural component(s) of CY. Inhibition of protein synthesis, cellular GSH depletion (through inhibition of its synthesis), and cell death resulted from addition of any of the CY compounds to hepatocytes.

Many groups have described the toxicity in mice of natural CY (Falconer et al., 1999Go; Hawkins et al., 1985Go; Seawright et al., 1999Go; Shaw et al., 2000Go; Terao et al., 1994Go). Natural epi-cylindrospermopsin, a minor component isolated from CY-producing Aphanizomenon ovalisporum, was shown to be toxic in a mouse bioassay (Banker et al., 2000Go, 2001Go). Our findings indicate that the mechanism of toxicity is most likely common for all the compounds.

Of the synthetic intermediates we tested, only the diols had biological activity comparable to natural CY. AC-MODEL and URACIL-MODEL showed no detectable biological activity. AB-MODEL had measurable dose-dependent biological activity, although concentrations from 200- to more than 1000-fold greater than for natural CY were required. AB-MODEL inhibited protein synthesis both in vitro and in hepatocytes. The intact C ring and functionality on the A ring are therefore important for efficient inhibition of translation by CY.

Because of the limited amounts of the synthetic AB-MODEL that were available, we could not increase the concentrations of this analogue sufficiently (to 1–5 mg/incubation) to show unequivocally whether the AB-MODEL also would have inhibited GSH synthesis and caused cell death in hepatocytes. Hepatocytes incubated with 250 and 500 µM AB-MODEL (single incubations) did not show any detectable decrease in cell GSH or increased cell lysis. At longer incubation times (42 instead of 19 h), 100 µM compound AB-MODEL did not decrease GSH and did not cause cell death. Under these experimental conditions, additions of natural, RAC-CY, EPI-CY, and the corresponding diols at low micromolar concentrations caused total cell death.

Deoxycylindrospermopsin (CY or EPI-CY lacking the C-7 hydroxyl group) has been purified as a minor component of Cylindrospermopsis extracts. By mouse bioassay, this compound was found not to be toxic at four times the median lethal dose for CY (Norris et al., 1999Go). One of the oxidized by-products of chlorination of CY is 5-chloro-cylindrospermopsin, in which a chlorine is added at C-5 of the uracil ring. This compound was shown to be at least 50-fold less toxic by mouse bioassay than CY (Banker et al., 2001Go).

Our results, together with the findings of others, indicate that the lack of the C ring and A ring functionality in CY decreases more than 100-fold the potency of CY. The uracil ring is also important, at least in the biological activity that requires transport into the cell. As far as the authors are aware, 5-chloro-cylindrospermopsin has not yet been tested in an in vitro assay (protein synthesis inhibition); therefore, it is not possible to conclude whether the uracil moiety of CY plays a major role in transport only or in in vitro activity also.

Somewhat more unexpected is the finding that lack of a hydroxyl group at C-7 leads to loss of in vivo activity when the orientation of the group has no effect. As proposed by the authors (Norris et al., 1999Go), a possible explanation is that the charge distribution of the adjacent uracil ring is altered. Alternatively, perhaps activity was not completely lost but rather decreased, since in the mouse bioassay, only a fourfold greater dose than the median lethal dose for natural CY was tested. As in the case of 5-chloro-CY, it is not possible to decide at this stage whether the charge distribution in the uracil moiety is a determinant in both transport and biological activity.

In summary, our results show that the sulfate group of the alkaloid CY plays no role in its biological activity or in the transport across the cell membrane of hepatocytes. This result is unexpected, given the large size and charge of the sulfate group. As far as the authors are aware, the diols have not yet been shown to occur naturally in cyanobacterial blooms or cultures that produce CY. Our results also show that orientation of the hydroxyl group at C-7 plays no significant role in transport or biological activity of CY.


    ACKNOWLEDGMENTS
 
The University of Southern California Center for Liver Disease Cell Culture Core provided the rat hepatocytes used in these studies. We thank the Statistical Consultation and Research Center of the USC Keck School of Medicine for its help. We are grateful to the National Institutes of Health for financial support: grant DK-51788 to M.T.R. and grant GM-46470 to B.B.S. The work at Penn State was supported by National Science Foundation grants CHE-9732038 and CHE-0102402 to S.M.W. and by a National Institutes of Health postdoctoral fellowship (IF32GM-20664) to G.A.W.


    NOTES
 
1 To whom correspondence should be addressed at the University of Southern California, Hoffman 602a, 2011 Zonal Avenue, Los Angeles, CA, 90089–9141. Fax: (323) 442-3236. E-mail: runnegar{at}hsc.usc.edu. Back

Portions of this paper were presented at the 5th International Conference on Toxic Cyanobacteria (ICTC V), Noosa, Australia, July 2001.


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Banker, R., Carmeli, S., Werman, M., Teltsch, B., Porat, R., and Sukenik, A. (2001). Uracil moiety is required for toxicity of the cyanobacterial hepatotoxin cylindrospermopsin. J. Toxicol. Environ. Health A 62, 281–288.[ISI][Medline]

Banker, R., Teltsch, B., Sukenik, A., and Carmeli, S. (2000).7-Epicylindrospermopsin, a toxic minor metabolite of the cyanobacterium Aphanizomenon ovalisporum from lake Kinneret, Israel. J. Nat. Prod. 63, 387–389.[ISI][Medline]

Bourke, A. T. C., Hawes, R. B., Nielson, A., and Stollman, N. D. (1983). An outbreak of hepatoenteritis (the Palm Island Mystery Disease) possibly caused by algal intoxication. Toxicon 3(Suppl.), 45–48.

Burgoyne, D. L., Hemscheidt, T. K., Moore, R. E., and Runnegar, M. T. C. (2000). Biosynthesis of cylindrospermopsin. J. Org. Chem. 65, 152–156.[ISI][Medline]

Byth, S. (1980). Palm Island mystery disease. Med. J. Aust. 2, 40–42.[ISI][Medline]

Djung, J. F., Hart, D. J., and Young, E. R. R. (2000). Vinyl and alkynyl pyrimidines as Michael acceptors: An approach to a cylindrospermopsin substructure. J. Org. Chem. 65, 5668–5675.[ISI][Medline]

Falconer, I. R., Hardy, S. J., Humpage, A. R., Froscio, S. M., Tozer, G. J., and Hawkins, P. R. (1999). Hepatic and renal toxicity of the blue-green alga (cyanobacterium) Cylindrospermopsis raciborskii in male Swiss albino mice. Environ. Toxicol. 14, 143–150.[ISI]

Harvey, T. C. (1996). Studies toward the synthesis of cylindrospermopsin. Ph.D. Dissertation, Brandeis University (abstract), Waltham, MA.

Hawkins, P. R., Runnegar, M. T. C., Jackson, A. R. B., and Falconer, I. R. (1985). Severe hepatotoxicity caused by the tropical cyanobacterium (blue-green alga) Cylindrospermopsis raciborskii (Woloszynska) Seenaya, and Subba Raju isolated from a domestic water supply reservoir. Appl. Environ. Microbiol. 50, 1292–1295.[ISI][Medline]

Heintzelman, G. R., Fang, W.-K., Keen, S. P., Wallace, G. A., and Weinreb, S. M. (2001). Stereoselective total synthesis of the cyanobacterial hepatotoxin 7-epicylindrospermopsin: Revision of the stereochemistry of cylindrospermopsin J. Am. Chem. Soc. 123, 8851–8853.[ISI][Medline]

Keen, S. P., and Weinreb, S. M. (2000). Studies on total synthesis of cylindrospermopsin: New constructions of uracils from {alpha},ß-unsaturated esters. Tetrahedron Lett. 41, 4307–4310.[ISI]

Looper, R. E., and Williams, R. M. (2001). Construction of the A-ring of cylindrospermopsin via an intramolecular oxazinone-N-oxide dipolar cycloaddition. Tetrahedron Lett. 42, 769–772.[ISI]

McAlpine, I. M., and Armstrong, R. W. (2000). Stereoselective synthesis of a tricyclic guanidinium model of cylindrospermopsin. Tetrahedron Lett. 41, 1849–1853.[ISI]

Moldeus, P., Hogberg, J., and Orrenius, S. (1978). Isolation and use of liver cells. Methods Enzymol. 51, 60–71.

Norris, R. L., Eaglesham, G. K., Pierens, G., Shaw, G. R., Smith, M. J., Chiswell, R. K., Seawright, A. A., and Moore, M. R. (1999). Stability of cylindrospermopsin, the toxin from the cyanobacterium Cylindrospermopsis raciborskii: Effect of pH, temperature, and sunlight on decomposition. Environ. Toxicol. 14, 163–165.[ISI]

Ohtani, I., Moore, R. E., and Runnegar, M. T. C. (1992). Cylindrospermopsin: A potent hepatotoxin from the blue-green alga Cylindrospermopsis raciborskii. J. Am. Chem. Soc. 114, 7941–7942.[ISI]

Padisak, J. (1997). Cylindrospermopsis raciborskii Seenaya and Subba Raju, an expanding, highly adaptive cyanobacterium: Wordwide distribution and review of its ecology. Arch. Hydrobiologie 107(Suppl.), 563–593.

Runnegar, M. T., Kong, S.-M., Zhong, Y.-Z., Ge, J.-L., and Lu, S. C. (1994). The role of glutathione in the toxicity of a novel cyanobacterial alkaloid cylindrospermopsin in cultured rat hepatocytes. Biochem. Biophys. Res. Commun. 201, 235–241.[ISI][Medline]

Runnegar, M. T., Kong, S. M., Zhong, Y. Z., and Lu, S. C. (1995). Inhibition of reduced glutathione synthesis by cyanobacterial alkaloid cylindrospermopsin in cultured hepatocytes. Biochem. Pharmacol. 49, 219–225.[ISI][Medline]

Seawright, A. A., Nolan, C. C., Shaw, G. R., Chiswell, R. K., Norris, R. L., Moore, M. R., and Smith, M. J. (1999). The oral toxicity for mice of the tropical cyanobacterium Cylindrospermopsis raciborskii (Woloszynska). (1999). Environ. Toxicol. 14, 135–142.[ISI]

Shaw, G. R., Seawright, A. A., Moore, M. R., and Lam, P. K. S. (2000). Cylindrospermopsin, a cyanobacterial alkaloid: Evaluation of its toxicologic activity. Ther. Drug Monit. 22, 89–92.[ISI][Medline]

Snider, B. B., and Harvey, T. C. (1995). Synthesis of a bicyclic model for the marine hepatotoxin cylindrospermopsin. Tetrahedron Lett. 36, 4587–4590.[ISI]

Snider, B. B., and Xie, C. (1998). Model studies for the synthesis of the marine hepatotoxin cylindrospermopsin: Preparation of a bicyclic guanidine with the hydroxymethyluracil side chain. Tetrahedron Lett. 39, 7021–7024.[ISI]

Terao, K., Ohmori, S., Igarashi, K., Ohtani, I., Watanabe, M. F., Harada, K. I., Ito, E., and Watanabe, M. (1994). Electron microscopic studies on experimental poisoning in mice induced by cylindrospermopsin isolated from blue-green alga Umezakia natans. Toxicon 32, 833–843.[ISI][Medline]

Tietze, F. (1969). Enzymic method for quantitative determination of nanogram amounts of total and oxidized glutathione: Applications to mammalian blood and other tissues. Anal. Biochem. 27, 502–522.[ISI][Medline]

Waldegger, S., Moschen, I., Ramirez, A., Smith, R. J., Ayadi, H., Lang, F., and Kubisch, C. (2001). Cloning and characterization of SLC26A6, a novel member of the solute carrier 26-gene family. Genomics 72, 43–50.[ISI][Medline]

Xie, C., Runnegar, M. T. C., and Snider, B. B. (2000). Total synthesis of (±) cylindrospermopsin. J. Am. Chem. Soc. 122, 5017–5024.[ISI]