Genotoxicity Studies of PolyGlycopleX (PGX)

Abstract

PolyGlycopleX (PGX), a novel dietary fiber, produces no mutagenic effects in bacterial tester strains Salmonella typhimurium TA 98, TA 100, TA 1535, and TA 1537 and Escherichia coli WP2 uvrA at concentrations of 0.316, 1.00, 3.16, 10.0, 31.6, and 100 μg/plate. No biologically relevant increases in revertant colonies of any of the 5 strains are observed at any concentration; however, a reduction at 100 μg/plate in TA 1537 is noted. PGX, analyzed for polychromatic erythrocyte micronuclei induction in mice following a single 1×, 0.5×, and 0.2× maximum tolerable dose intraperitoneal treatment, produces no biologically relevant increase in any dose group. Males at 1× maximum tolerable dose show a reduction of micronuclei-containing cells. High-dose animals show signs of systemic toxicity, including a reduction of spontaneous activity, rough fur, palpebral closure, prone position, and constricted abdomen. These genotoxicity studies show PGX to be nonmutagenic in both the Ames bacterial reverse mutation assay and the mammalian erythrocyte micronucleus test.

Obesity, diabetes, coronary heart disease, colorectal cancer, and gastrointestinal irregularities are significant causes of morbidity worldwide and are just some of the diseases for which dietary fiber has been shown to have protective effects. Dietary fiber is composed of nonstarch plant polysaccharides and lignin and is incapable of being digested in the human gastrointestinal tract. Consumption of dietary fiber has been known to alter transit time and gut microflora, expand stool weight, and alter absorption of fats, carbohydrates, toxins, vitamins, and minerals. ¹ The extent, rate, and site of dietary fiber action depend on its physicochemical structure, solubility, and fermentability and the intestinal environment, among other factors. Furthermore, viscosity, through thickening upon mixing with fluids, is a property of soluble dietary fiber that has a great influence on its action and potential use. ² Increased viscosity has been reported to alter transit time and absorption, leading to a reduction in postprandial peak blood glucose level ² and induction of satiety by providing bulk, in turn leading to reduction in body weight, ³ changes in intestinal enzymatic activity and laxation, ⁴ and changes in blood lipids. ⁵ It has been proposed that the prospective beneficial effects on colonic health and coronary heart disease reside in direct proportion to the viscosity of the fiber in normal ² and diabetic ⁶ subjects.

PolyGlycopleX (PGX, Inovobiologic Inc, Calgary, Canada) is a novel, highly viscous, nonstarch polysaccharide manufactured by reacting konjac (glucomannan) with sodium alginate and xanthan gum. The proprietary processing leads to a synergistic molecular interaction and the formation of a highly fermentable fiber with a high viscosity and water-holding capacity. Glucomannan, a β-D-(1–4)-linked linear polymer of glucose and mannose ⁷ that is prepared from konjac powder of the tuber root Amorphophallus konjac, produces a highly viscous fiber that is depolymerized to a variety of glucomanno-oligosaccharides in the large intestine as a result of the intestinal microflora. ⁸ Sodium alginate, the sodium salt of alginic acid and a natural polyuronide constituent of certain brown algae, is a polysaccharide that is also used in the food industry as a thickening agent. Xanthan gum, a polysaccharide produced by the bacterium Xanthomonas campestris, has a therapeutically glutinous or viscous property that accounts for its use as a bulking agent because it increases stool water retention and reduces intestinal transit time. The microbe-derived gum and alginate are both considered indigestible dietary fibers, shown to lower the rate of small intestinal absorption ^9–11 in part by influencing bile acid metabolism, thereby reducing the formation of potential tumor promoters in the colon as well as reducing cholesterol. The 3 polysaccharides act synergistically with a resultant viscosity that is higher than any of the individual components. PGX possesses such high viscosity because it uses a form of konjac (glucomannan) that is highly refined, and it is uniquely processed with its other ingredients resulting in molecular interactions and increasing its hygroscopic properties. Because of this, it is 3 times more viscous than guar and approximately 7 times as viscous as psyllium gums. ^12,13 By combining konjac (glucomannan) with the other soluble fibers, sodium alginate and xanthan gum, the viscosity of PGX is amplified further and has a viscosity 3 to 5 times higher than konjac (glucomannan) alone.

Xanthan gum, sodium alginate, and konjac (glucomannan) have been extensively studied toxicologically, and much has been documented on the safety of these individual components. ^14–16 Xanthan gum and sodium alginate are generally regarded as safe (GRAS), according to the US Food and Drug Administration (FDA), with limited use. A GRAS determination for konjac flour has been submitted to the FDA but remains to be resolved. ¹⁷ Furthermore, limitations of use have been issued for konjac. ¹⁸ All 3 components can be used as stabilizers, thickeners, emulsifiers, or suspending agents, although sodium alginate, depending on its application, has been used to a maximum level of 10%, whereas konjac (glucomannan) is acceptable to 1% in food. ¹⁹

Data on the mutagenic potential of the 3 components of PGX are less complete. Konjac (glucomannan) and sodium alginate have been tested individually for their genotoxic safety. Konjacmannan was determined to be nonmutagenic in 5 tester strains, TA 98, TA 100, TA 1535, TA 1537, and TA 1538, of Salmonella typhimurium in the presence or absence of liver microsomal metabolic activation. ²⁰ Similarly, in a mouse lymphoma thymidine kinase gene mutation assay, konjac flour was found noncytotoxic during the course of 4 hours at concentrations of 7.81 to 1000 μg/mL or 15.6 to 997 μg/mL in the presence or absence of rat liver S9 enzymes, respectively. ²¹ In an in vivo mouse micronucleus test, no increase in micronucleus formation in bone marrow cells harvested at 24, 48, or 96 hours was observed in mice given 5000 mg per kilogram of body weight suspended in corn oil by gavage. ²² Similarly, sodium alginate has been deemed nonmutagenic in the in vitro reverse bacterial mutation assay (maximum dose 10 mg/plate; no metabolic activation) and chromosome aberration tests (maximum dose 1.0 mg/mL) in Chinese hamster lung cells at 24 and 48 hours. ²³ Nomutagenicity data have been reported for xanthan gum.

PGX has been examined in multiple clinical studies for its effectiveness in managing postprandial glycemia, cholesterol, appetite cravings (weight), and colonic health. High-fiber diets have been recommended for the nutritional management of patients with diabetes mellitus. ²⁴ PGX or its prototypes have been studied for their effectiveness in lowering postprandial glycemia and promoting satiety in both healthy ²⁵ and diabetic individuals. ²⁶ In multiple, currently unpublished clinical reports, the glycemic index was found to be significantly reduced by 70% to 75% ²⁷ or 7.0/g ²⁸ in both normal and diabetic individuals following consumption of various foods and beverages. Furthermore, food intake was significantly reduced following administration of the high-viscosity PGX as preload drink in normal adolescents. ²⁹ A PGX prototype, examined for its use in the treatment of cardiovascular disease, significantly improved serum lipid profile ³⁰ and reduced total/high-density lipoprotein cholesterol ratio and systolic blood pressure ³¹ in people with insulin resistance syndrome.

PGX was recently tested in a 90-day feeding study in Sprague-Dawley rats ³² and was found to cause no test substance–related changes in body weight, food consumption, organ weights, histopathology, or clinical pathology compared with control diet, and the “no observable adverse affect level” (NOAEL) for PGX has been determined as 5% of the diet (3219 and 3799 mg per kilogram of body weight per day for male and female rats, respectively), the highest dose tested.

The present study evaluates the genotoxic potential of PGX to promote a mutagenic response in the bacterial reverse mutation assay (Ames test) and a clastogenic and/or aneugenic response in the mammalian erythrocyte micronucleus (MMA) test.

Materials and Methods

These studies, conducted at Bioservice Scientific Laboratories (BSL) GmbH in Planegg, Germany, were in compliance with the Organization for Economic Co-operation and Development (OECD) Principles of Good Laboratory Practices (ENV/MC/CHEM [98] 17 OECD, Paris, 1998) and the Chemikaliengesetz (Chemicals Act) of the Federal Republic of Germany (Appendix 1 to §19a as amended and promulgated on June 20, 2002 [BGB1.I Nr. 40 SA. 2090], revised October 31, 2006 [BGB1.I Nr. 50 S. 2407]). All work undertaken by the testing laboratory was in accordance with the most recent Guide for the Care and Use of Laboratory Animals by the National Institutes of Health (1996), operated under the surveillance of the Regierung von Oberbayern (German regulatory authority) according to the Association for Assessment and Accreditation for Laboratory Animal Care standards and accreditation.

Test Substance

The test substance, a granular, off-white powder, marketed under the trade name PGX (InovoBiologic, Inc, Calgary, Canada; lot no. 1284060330R; expiration date March 21, 2009) was used for both the in vitro bacterial reverse mutation (Ames) assay and the in vivo micronucleus test. The test substance was stored in a cool, dry environment at room temperature, away from heat and light; it has a high gelling capacity when mixed in water.

Positive and Negative Controls

Positive and negative controls were included in each experiment for both bacterial reverse mutation assay and the in vivo micronucleus test.

Negative and solvent control for the in vitro test, both with and without S9 mix, consisted of distilled water (solvent). Cottonseed oil (batch no. 067K0116, Sigma-Aldrich, Munich, Germany) was selected as the vehicle (negative control) for the in vivo MMA test and was administered via a single intraperitoneal injection at 10 mL per kilogram of body weight and sampled at 44 and 68 hours after treatment. Solvent control, consisting of solvent or vehicle alone, was treated in the same way as the treatment groups.

Positive control substances used for the Ames testing (strain-specific) without metabolic activation included sodium azide (cat no. 106688, lot no. K 28585088, Merck, Darmstadt, Germany) for tester strains S typhimurium TA 100 and TA 1535, dissolved in distilled water at a concentration of 10 μg/plate; 4-nitro-o-phenylene-diamine (4-NOPD) (cat no. 73630, lot no. 416325, Fluka Chemical GmbH, Deisenhofen, Germany) for tester strains S typhimurium TA 98 and TA 1537, dissolved in dimethyl sulfoxide (DMSO) at a concentration of 10 μg/plate and 40 μg/plate for TA 98 and TA 1537, respectively; methyl methane sulfonate (MMS; cat no. M4016, lot no. 126K3721, Sigma-Aldrich, Munich, Germany) for tester strain E coli WP2 uvrA dissolved in distilled water at a concentration of 1 μL/plate. Assays conducted with metabolic activation included 2-aminoanthracene (2-AA) (cat no. A3, 880-0, lot no. S11804-1124, Sigma-Aldrich, Munich, Germany) for tester strains S typhimurium TA 98, TA 100, TA 1535, and TA 1537 and E coli WP2uvrA dissolved in DMSO at concentrations of 2.5 μg/plate for TA 98, TA 100, TA 1535, and TA 1537 and 10 μg/plate for E coli WP2 uvrA. The positive control for the MMA test was cyclophosphamide (cat no. C0768, batch no. 076K1050, Sigma-Aldrich, Taufkirchen, Germany) dissolved in physiologic saline, dosed at 40 mg per kilogram of body weight at 10 mL per kilogram of body weight, administered by the same route (intraperitoneal) as the test substance, and sampled. Blood was collected via tail vein incision (approximately 100 μL) at 44 hours after treatment.

Mammalian Microsomal Fraction S9 Mix

The bacteria most commonly used in the bacterial reverse mutation assay (Ames test) do not possess the enzyme system, which in mammals is known to convert promutagens into active DNA damaging metabolites. To overcome this major drawback, an exogenous metabolic system is added in form of mammalian microsome enzyme activation mixture. Rat liver (S9) homogenate ³³ for the Ames test was prepared at BSL Bioservice GmbH. Male Wistar rats were induced with phenobarbital (80 mg per kilogram of body weight) and β-naphthoflavone (100 mg per kilogram of body weight) for 3 consecutive days by the oral route of administration, according to the method of Maron and Ames. ³⁴ The protein concentration in the S9 preparation (lot no. 150508) was 33 mg/mL with quality controls for biologic activity and sterility. The applied S9 mix preparation consists of 95% co-factor solution (Na-phosphate buffer, MgCl₂, KCl, glucose-6-phosphate, NADP, and 5% rat liver (S9) homogenate (Table 1).

Bacterial Reverse Mutation Assay (Ames Test)

This study was conducted in conformance with the following internationally accepted guidelines and recommendations: Ninth Addendum to OECD Guidelines for the Testing of Chemicals (Section 4, No. 471, Bacterial Reverse Mutation Test, adopted July 21, 1997); EEC Directive 2000/32 (L 136, Annex 4C, B 12, Mammalian Erythrocyte Micronucleus Test, dated May 19, 2000); EPA Health Effects Test Guidelines (OPPTS 870.5100 Bacterial Reverse Mutation Assay EPA 712-C-98-247, August 1998).

The study was conducted according to the plate incorporation protocol ^33,35 and the preincubation method. ³⁴ The principle of this bacterial reversion mutation assay is that it detects mutations, which functionally reverse mutations present in the tester strains and restore the capability to synthesize an essential amino acid. ^33–36 S typhimurium tester strains TA 98, TA 100, TA 1535, and TA 1537 were obtained from MOLTOX, Inc (Vilas, NC). The E coli strain (WP2 uvrA) was obtained from DSMZ Sales (Braunschweig, Germany). Cultures were stored as stock ampoules in nutrient broth and Luria Bertani medium for S typhimurium and E coli, respectively, supplemented with DMSO (approximately 8% vol/vol) over liquid nitrogen. For the experiments, cultures inoculated from the stock ampoules were grown overnight to the late exponential or early stationary phase of growth (approximately 10⁹ cells/mL) for 12 hours at 38.5^°C. A solution of 125 μL of ampicillin (10 mg/mL) for TA 98 and TA 100 was added to retain the phenotypic characteristics of these strains. The highest concentration prepared was a suspension of 1 mg/mL in distilled water (see below). This suspension was diluted prior to treatment. The solvent was compatible with the survival of the bacteria and the S9 activity.

Experiment for Toxicity

The toxicity of PGX was determined with tester strains TA 98 and TA 100 in a preliminary experiment. Eight concentrations were tested for toxicity and induction of mutations with 3 plates each. The experimental conditions in the preliminary experiment were the same as described for the main experiment (plate incorporation, experiment 1). The test substance concentrations to be applied in the main experiments were chosen according to the results of the preliminary experiment. Given the insolubility of PGX in organic solvents and its high water-gelling capacity, the highest possible concentration of the test article that could be prepared was a suspension of 1 mg/mL in distilled water necessitating a maximum concentration of 100 μg/plate for both tests. The concentration range covered 2 logarithmic decades. Two independent experiments were performed. For the preliminary experiment, concentrations of 0.0316, 0.100, 0.316, 1.00, 3.16, 10.0, 3.16, and 100 μg/plate were selected. For the main experiments, concentrations of 0.316, 1.00, 3.16, 10.0, 3.16, and 100 μg/plate were selected. Because the results of the preliminary experiment were in accordance with the validity criteria set for the experiments, these were reported as a part of the main experiment 1.

Experimental Performance

For the plate incorporation method, the following materials were mixed in a test tube and poured over the surface of an agar plate containing a mixture of sterilized agar-agar, Vogel-Bonner salts, and a 2% glucose-solvent solution: 100 μL of test solution at each dose level, solvent or negative control, or reference mutagen solution (positive control); 500 μL of S9 mix (for testing with metabolic activation) or S9 mix substitution buffer (for testing without metabolic activation); 100 μL of bacteria suspension (S typhimurium or E coli); 2000 μL of overlay agar containing agar-agar, NaCl, histidine and biotin, and tryptophane, respectively.

For the pre-incubation method, 100 μL of the test substance preparation was pre-incubated with the tester strains (100 μL) and the metabolic activation system or substitution buffer (500 μL) for 60 minutes at 37^°C prior to adding the overlay agar (2000 μL) and pouring onto the surface of a minimal agar plate. The plates were incubated at 37^°C for at least 48 hours in the dark.

Data Recording

The colonies were counted using a ProtoCOL counter (Meintrup DWS Laborgerate GmbH, Laehden, Germany). If precipitation of the test item precluded automatic counting, the revertant colonies were counted by hand. Tester strains with a low spontaneous mutation frequency, such as TA 1535 and TA 1537, were counted manually.

Evaluation of Cytotoxicity

Toxicity was defined by a clearing or diminution of the background lawn or a reduction in the number of revertants down to a mutation factor of approximately ≤0.5 in relation to the solvent control.

Criteria of Validity

A test is considered acceptable if, for each strain, (1) the bacteria demonstrate their typical responses to ampicillin (TA 98, TA 100), ³³ (2) the control plates with and without S9 mix are within the historical control data range (Table 2), (3) corresponding background growth on both negative control and test plates is observed, and (4) the positive controls show a distinct enhancement of revertant rates over the control plate.

Evaluation of Mutagenicity

The mutation factor is calculated by dividing the mean value of the revertant counts by the mean values of the solvent control (the exact and not the rounded values are used for calculation). The test substance is considered mutagenic if a clear and dose-related increase in the number or revertants occurs and/or a biologically relevant positive response for at least 1 of the dose groups occurs in at least 1 tester strain with or without metabolic activation. A biologically relevant increase is described as if, in tester strains TA 100 and E coli WP2 uvrA, the number of reversions is at least twice as high and/or if in tester strains TA 1535, TA 1537, and TA 98 the number of reversions is at least 3 times higher than the reversion rate of the solvent control. ³⁷

Mammalian Erythrocyte Micronucleus Test

A mouse micronucleus test was performed to identify the potential of PGX to cause cytogenetic damage, through detection of the formation of micronuclei containing lagging chromosome fragments or whole chromosomes of erythroblasts, by analysis of the erythrocytes of bone marrow. The first appearance of micronuclei in polychromatic erythrocytes (PCE) is at least 10 to 12 hours after a clastogenic exposure, the time required for the affected erythroblast to differentiate into a PCE. The assessment of clastogenic activity was carried out with a dose level at the maximum tolerable dose (MTD) or that producing some indication of cytotoxicity (changes in the relative PCE in peripheral blood). Two additional doses (a middle and a low dose) were also assayed.

This study was conducted in conformance with the following internationally accepted guidelines and recommendations: BSL Bioservice accreditation scope guidelines 90/385/EWG, 93/42/EWG, and DIN EN ISO/IEC 17025 for testing of medical devices; Ninth Addendum to OECD Guidelines for the Testing of Chemicals (Section 4, No. 474, Mammalian Erythrocyte Micronucleus Test, adopted July 21, 1997); EEC Directive 2000/32 (L 136, Annex 4C, B 12, Mammalian Erythrocyte Micronucleus Test, dated May 19, 2000); EPA Health Effects Test Guidelines (OPPTS 870.5395 Mammalian Erythrocyte Micronucleus Test, EPA 712-C-98-226, August 1998); ISO 10993-1, 2003, Evaluation and Testing, ISO 10993-3, 2003, Tests for Genotoxicity, Carcinogenicity and Reproductive Toxicity, and ISO 10993-12, 2007, Sample Preparation and Reference Materials.

Test System

Given the ability of the test substance to gelatinize, for these experiments PGX was extracted via agitation in clear, chemically inert closed tubes in cottonseed oil for 72 hours (±2 hours) at 37^°C (±1^°C) with a mass/volume ratio of 0.2 g/mL according to medical device and chemicals guidelines ISO 10993-3, 2003, and ISO 10993-12, 2007. The supernatant extract was carefully isolated by pipetting without centrifugation or filtration and diluted with cottonseed oil (0.5× MTD and 0.2× MTD) within 1 hour before treatment and administered intraperitoneally at 10 mL per kilogram of body weight (maximum volume 10 mL nonaqueous solution per kilogram of body weight) to 7- to 12-weekold NMRI mice obtained from Harlan Winkelmann GmbH (Borchen, Germany) according to the following design ^38,39 : 100% extract was considered 1× MTD, 50% extract was considered 0.5× MTD, and 20% extract was considered 0.2× MTD, where the maximum tolerable dose is defined as the dose producing signs of toxicity according to a preliminary experiment using 3 males and 3 females receiving a single 100% extract dose. In the main experiment, for each dose group, 5 male and 5 female mice were acclimated (at least 5 days) and randomly distributed into treatment groups and tail tagged. The volume administered for PGX and positive and negative controls was 10 mL per kilogram of body weight in a single intraperitoneal injection. Administration by intraperitoneal injection was selected according to guideline and for ease of administration for a viscous product. In addition, the expected bioavailability of the extract by the intraperitoneal route provided a more rigorous test of the oil-soluble components and suspended small particles of PGX, which may play a role in genotoxic potential. Sampling of the peripheral blood was carried out on animals 44 hours following treatment for each treatment group evaluated and for the positive control group and 68 hours for the additional negative control and 1× MTD dose group. During the experiment, animals were group housed (5/sex/cage) in a polysulphone cage with Altromin saw fiber bedding (Altromin Spezialfutter GmbH & Co KG, Lage, Germany) and a 12-hour light/dark cycle with free access to tap water and Altromin 1324 maintenance diet (Altromin Spezialfutter GmbH & Co KG) for rodents with temperature and relative humidity of 22^°C ± 3^°C and 55% ± 10%, respectively, with air changes of 10 times per hour.

Blood Preparation

Blood was obtained from the tail vein. Blood cells were immediately fixed in ultra-cold methanol. Before analysis (at least 16 hours after fixation), fixed blood cells were washed in Hanks’ balanced salt solution and centrifuged at 600 × g for 5 minutes, and the supernatant was discarded. Blood cell populations were discriminated using specific antibodies against CD71 (expressed only at the surface of immature erythrocytes) and CD61 (expressed at the surface of platelets), and DNA content of micronuclei was determined by the use of a DNA specific stain (propidium iodide). ⁴⁰

Analysis

Evaluations of all samples, including those of positive and negative controls, were performed using a flow cytometer (FACScan, BD Biosciences Clontech GmbH, Heidelberg, Germany). Anti-CD71 antibodies were labeled with fluorescein isothiocyanate (AbD Serotec, Düsseldorf, Germany), and anti-CD61 antibodies were labeled with phycoerythrin (AbD Serotec, Düsseldorf, Germany). Particles were differentiated using forward scatter and side scatter parameters of the flow cytometer. Fluorescence intensity was recorded on the FL1, FL2, and FL3 channels for fluorescein isothiocyanate, phycoerythrin, and propidium iodide, respectively. At least 10 000 immature erythrocytes per animal were scored for the incidence of micronucleated immature erythrocytes. To detect an eventually occurring cytotoxic effect of the test substance extracts, the ratio between immature and mature erythrocytes was determined. The results are expressed as relative PCE (relative PCE = proportion of polychromatic [immature] erythrocytes among total erythrocytes). The relative PCE is the supportive end point to assess cytotoxicity, an indicator of target cell exposure to the test substance. A decrease of PCE values in treated animals compared with control animals gives an indication of target cell exposure to the test substance extract.

Evaluation of Results and Acceptance Criteria

The test substance was considered to be negative if there was no biologically relevant and/or statistically significant increase in the number of micronucleated cells at any dose level (dose-related increase). For statistical analysis, the nonparametric Mann-Whitney test was used. However, biological relevance and statistical significance were considered together. ^41,42 The data generated were considered acceptable if at the commencement of the study, the weight variation of animals does not exceed ± 20% of the mean weight of each sex, the background frequency of micronucleated cells is in the normal historical range, and the test system is sensitive to the known mutagen as judged by the results in the concurrent positive control animals.

The overall body weight ranges of all animals differentiated by sex were as follows: males with a mean body weight of 30.1 ± 1.0 g (range 27.4–31.7 g) and a variation of ± 7.2%; females with a mean body weight of 26.2 ± 1.5 g (range 23.9–29.8 g) and a variation of ± 11.3%. The weight variation of the animals prior to treatment did not exceed ± 20% and conformed to acceptance criteria and the regulatory requirement for the MMA test.

Results

Bacterial Reverse Mutation Assay (Ames Test)

Toxicity of the test compound in the Ames test can be detected by a clearing of the background lawn, or a reduction in the number of revertants down to a mutation factor of approximately ≤0.5 in relation to the solvent control. No cytotoxic effects of PGX were observed at any concentration in either of the 2 tester strains used in the preliminary experiment (Table 3). No cytotoxic effects of the test item were noted in any of the 5 tester strains used up to the highest dose group evaluated (with and without metabolic activation) in experiment 1 (Table 4). In experiment 2 (Table 5), a weak cytotoxic effect of PGX, indicated by a reduction of the number of revertant colonies down to a mutation factor of 0.5, was noted only in tester strain TA 1537 at a concentration of 100 μg/plate with and without metabolic activation. No biologically relevant increases in revertant colony numbers of any of the 5 tester strains were observed following treatment with PGX at any concentration level, either in the presence or absence of metabolic activation in experiments 1 and 2. The reference mutagens induced a distinct increase of revertant colonies, indicating the validity of the experiments. Furthermore, no precipitation of PGX was observed in any of the 5 tester strains used in experiments 1 and 2 with and without metabolic activation. By these criteria, PGX did not cause gene mutations by base pair changes or frame shifts in the genome of the tester strains used.

Mammalian Micronucleus Test of Murine Peripheral Blood Cells

For the assessment of acute toxicity, a preliminary experiment was performed using 100% extract as the highest dose group. This dose was selected as the MTD in the micronucleus assay main experiment as well. In the main experiment, 3 dose levels were used covering a range from the maximum tolerable dose to slight toxicity. Based on the toxicity observed in the preliminary experiment, dose levels of 1 (100%), 0.5 (50%), and 0.2 (20%) PGX extract were tested.

Toxicity

In the preliminary experiment, 3 male mice and 3 female mice received a single dose of 100% extract (1× MTD) intraperitoneally and showed toxic symptoms, such as reduction of spontaneous activity, rough fur, prone position, palpebral closure, and constricted abdomen (Table 6), signs that appeared to abate with time. All animals survived 72 hours after the treatment. In the main experiment, all animals treated with the highest dose group (1× MTD) showed toxic effects. Signs of systemic toxicity, including reduction of spontaneous activity, prone position, rough fur, palpebral closure, and constricted abdomen, were observed at 4 hours after treatment in all 5 male and 5 female mice. By 44 hours after treatment, evidence of toxicity was limited to mild rough fur in all the males and females, and signs for all animals were cleared by 68 hours. The 5 male and 5 female mice treated with a dose of 0.2× MTD or 0.5× MTD showed signs of systemic toxicity after the treatment with the test item extracts, but these toxic symptoms were moderately or weakly developed, only mild instances of rough fur being manifested in both males and females after 4 hours. All signs in all animals were cleared by 48 hours with the exception of mild rough fur in all female animals of 0.5× MTD. There were no clinical signs in to any of the animals administered cottonseed oil alone (negative controls).

Relative PCE

The relative PCE was determined for each animal (Table 7). The negative controls (44 hours, 68 hours) were within the range of the historical laboratory control data of the negative control (1.19–3.97). The values noted for the 44-hour negative control were 2.79 for the male and 2.80 for the female mice. The values detected for the 68-hour negative control were 2.86 for the male and 4.35 for the female mice. The animals that received 1× MTD (44-hour treatment) showed mean values of 2.09 (male) and 2.56 (female). The value observed for the male group was significantly reduced compared with the corresponding negative control (2.44), whereas in the female group the value was reduced compared with the corresponding negative control (1.96), but the reduction was not statistically significant. The dose groups that were treated with 0.5× and 0.2× MTD showed mean values of the relative PCE of 2.44 (male) and 1.99 (female) and 2.25 (male) and 2.30 (female), respectively. The values observed in the male and female group were reduced compared with the corresponding negative control, but the reduction was not statistically significant. Males and females treated with 1× MTD PGX and sampled 68 hours afterward showed mean values for relative PCE of 2.58 (male) and 3.01 (female), again slightly reduced from the corresponding negative control but not statistically significant.

Micronucleated Polychromatic Erythrocytes

For all dose groups, including positive and negative controls, at least 10 000 immature erythrocytes per animal were scored for the incidence of micronucleated immature erythrocytes. The data are presented in Table 7.

The negative controls (44 hours and 68 hours) evaluated were within the range of the historical laboratory control data of the negative control (0.08%–0.43%). Results of the present study reveal mean values of micronuclei of 0.15% and 0.09% for male and female mice for the negative control at 44 hours, respectively. The mean values for the 68-hour negative control were 0.09% for the male and 0.06% for the female mice. The dose group treated with 1× MTD (44 hours treatment) showed mean values of 0.08% for male and female mice. These values were decreased compared with the corresponding negative control but were within the range of historical values. The mean values noted for the 0.5× and 0.2× MTD dose groups were 0.11% for male and 0.10% for female mice and 0.09% for male and 0.10% for female mice, respectively. These values were within the range of the corresponding negative control. The mean values observed for the 1× MTD 68-hour treatment were 0.08% and 0.06% for male and female mice, respectively. The mean values observed in both sexes were slightly decreased compared with the range of the corresponding negative control, but this decrease was not statistically significant. Cyclophosphamide (40 mg per kilogram of body weight), administered intraperitoneally and used as a positive control, induced a statistically significant increase in micronucleus frequency of 0.93% for male and 0.98% for female mice (Table 7), thus demonstrating the validity of the assay.

Discussion

In the present study, the genotoxicity of PGX was investigated using a bacterial reverse mutation and mouse micronucleus assay. The test item PGX was investigated for its potential to induce gene mutations according to the plate incorporation test (experiment 1) and the pre-incubation test (experiment 2) using S typhimurium strains TA 98, TA 100, TA 1535, and TA 1537 and tester strain E coli WP2 uvrA. A weak cytotoxic effect of PGX was only noted in experiment 2 in tester strain TA 1537 at a concentration of 100 μg/plate with and without metabolic activation because a reduction of the number of revertants down to a mutation factor of 0.5 was observed. Cytotoxic effects of PGX were evaluated, because cytotoxicity is among the criteria to be taken into consideration when determining the highest amount of the test substance to be used in the bacterial reverse mutation assay. Given insolubility of PGX in organic solvents and its high water-gelling capacity, the highest possible concentration that could be prepared was a suspension of 1 mg/mL in distilled water. Dilution to a maximum dosage of 100 μg/plate allowed proper growth conditions without increased plate viscosity. OECD guidelines allow investigators to consider the solubility of the test product in determining the highest dose to be tested. Results under these conditions indicated no biologically relevant increases in revertant colony numbers of any of the 5 tester strains observed following treatment with PGX at any concentration level, in either the presence or the absence of metabolic activation in experiments 1 and 2. In the Ames test under the experimental conditions reported, PGX did not cause gene mutations by base pair changes or frame shifts in the genome of the tester strains used, because neither a dose-related increase in the number of revertants nor a reproducible, biologically relevant, positive response in any of the dose groups was produced by this compound.

Genotoxicity of PGX was also assessed in the micronucleus assay for its potential to induce micronuclei in polychromatic erythrocytes in the bone marrow of the mouse. All mean values of micronuclei formation noted after treatment with PGX were within the range of the corresponding negative control or slightly decreased compared with the corresponding negative control. No biologically relevant increase of micronuclei was found. No statistically significant change (P < .05) in the number of cells with micronuclei was noted in the PGX-dosed groups. However, a statistically significant reduction from negative control was noted for the 1× MTD male group, which was within historical control range for this parameter. Furthermore, PGX did not induce structural or numerical chromosomal damage in the immature erythrocytes of the mouse under the experimental conditions reported and is considered to be nonmutagenic with respect to clastogenicity and aneugenicity in the mammalian erythrocyte micronucleus test. The current genotoxicity studies showed PGX to be nonmutagenic in both the in vitro bacterial reverse mutation and the in vivo mammalian erythrocyte micronucleus assay. These results compare favorably with the available nonmutagenic data of many gels and thickeners currently in use within the food industry. Genotoxic data available for other common stabilizers and thickening gums, such as acacia, guar, carrageenan, and others, have generally been as nonmutagenic. ^43–48

The only other available toxicity data for PGX include a recent OECD/FDA guidelined preclinical subchronic study wherein Sprague-Dawley rats consumed 0%, 1.25%, 2.5%, or 5.0% PGX in the diet for 90 days and were evaluated for toxicological effects. ³² There were no test substance–related changes in viability, behavior, clinical signs, body weight, food consumption, mean organ weight, organ-to-body weight, or organ-to-brain weight values or histopathology between controls and treated animals. Minimal clinical pathology changes in high-dose female rats were considered adaptive in nature and may be related to the increased urine volume observed, which could also account for the increase in urine pH and decreased specific gravity found in both the male and female rats. The high oral viscosity of PGX may have increased water intake, thereby resulting in the lower red blood cell count found in high-dose males and lower plasma mineral levels seen in high-dose females. Changes in the high-dose group were within historical control values and may represent a nutritional displacement of the diet. As such, the NOAEL for PGX was 5% of the diet, corresponding to 3219 and 3799 mg per kilogram of body weight per day in male and female rats, respectively.

In summary, in confirmation of the absence of genotoxic activity of PGX in the bacterial reverse mutation (Ames) assay and mammalian erythrocyte micronucleus test of murine peripheral blood cells, adaptations for the viscosity of PGX in both procedure and results were noted and were considered within normal range for use in the food industry.