Dietary safety of cycloastragenol from Astragalus spp.: Subchronic toxicity and genotoxicity studies

Food and Chemical Toxicology 64 (2014) 322–334

Contents lists available at ScienceDirect

Food and Chemical Toxicology journal homepage: www.elsevier.com/locate/foodchemtox

Dietary safety of cycloastragenol from Astragalus spp.: Subchronic toxicity and genotoxicity studies Nancy J. Szabo ⇑ Burdock Group, 859 Outer Road, Orlando, FL 32814, United States

a r t i c l e

i n f o

Article history: Received 22 August 2013 Accepted 27 November 2013 Available online 4 December 2013 Keywords: Astragalus Clastogenic Cycloastragenol Genotoxic Mutagenic Subchronic

a b s t r a c t Extracts, teas, and other preparations of Astragalus roots (e.g., Radix Astragali) are historically recognized traditional medicines and foods. Cycloastragenol (CAG), a bioactive triterpene aglycone from Astragalus root extracts, is being developed as a modern dietary ingredient. To this end, studies assessing subchronic toxicity and genotoxic potential were conducted. In the subchronic study with recovery component, rats ingested 0, 40, 80, or 150 mg/kg/d CAG by oral gavage for P91 consecutive days. No treatment-related mortalities occurred and no cardiac effects were identified. Although several endpoints among those monitored (i.e., clinical observations, body weight, food consumption, ophthalmology, urinalysis, hematology, clinical chemistry, gross pathology, organ weights, or histopathology) exhibited statistically significant effects, none was adverse. The oral no-observed-adverse-effect level (NOAEL) for CAG was >150 mg/kg/d in male and female rats. CAG (65000 lg/plate) did not induce mutagenicity in Salmonella typhimurium or Escherichia coli tester strains. Although the in vitro chromosome aberration assay gave a moderately positive response (likely due to poor solubility) for one intermediate concentration (1.50 mM) with metabolic activation, responses were negative in all other test groups. Finally, in the in vivo micronucleus assay no clastogenicity was observed in peripheral erythrocytes from mice administered 2000 mg/kg CAG by intraperitoneal injection. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Historical human usage of Astragalus spp.,1 perennial flowering shrubs of the Fabaceae (legume) family indigenous to Europe, the

Abbreviations: 9-AA, 9-aminoacridine; 2-AA, 2-aminoanthracene; ANOVA, Analysis of Variance; AST, aspartate aminotransferase; AAALAC, Association for Assessment and Accreditation of Laboratory Animal Care International; AST, astragaloside; ATCC, American Type Culture Collection; CK, creatine phosphokinase; CAG, cycloastragenol; CPA, cyclophosphamide; DMSO, dimethyl sulfoxide; EMS, ethylmethanesulfonate; FOB, functional observational battery; GLP, good laboratory practice; GRAS, generally recognized as safe; IACUC, Institutional Animal Care and Use Committee; ICH, International Conference on Harmonisation; i.p., intraperitoneal; LDH, lactate dehydrogenase; MTD, maximum tolerated dose; MCH, mean corpuscular hemoglobin; MMS, methylmethanesulfonate; MEM, minimal essential medium; NOAEL, no-observed-adverse-effect level; 2-NF, 2-nitrofluorene; NaN3, sodium azide; OECD, Organisation for Economic Co-operation and Development; PCE, polychromatic erythrocytes; rel. PCE, relative PCE; TCM, traditional Chinese medicine; WHO, World Health Organization. ⇑ Tel.: +1 407 802 1400; fax: +1 407 802 1405. E-mail address: [email protected] 1 Several Astragalus species out of the approximately 2200 that have been named and described are known to be toxic. Loosely grouped as selenium hyper-accumulators, nitroglycoside producers, and swainsonine alkaloid carriers (Castillo et al., 1993; Rios and Waterman, 1997; Ralphs et al., 2008; Sors et al., 2009), these species were not used historically for food or forage and are not included among the Astragalus spp. from which CAG is derived. 0278-6915/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.fct.2013.11.041

Middle East, and Asia, has included traditional medicines, coffee and tea substitutes, food, cosmetics, gum tragacanth (a vegetable gum composed of the dried sap collected from Astragalus gummifer and related species that is an approved food additive [21 CFR §184.1351] and pharmaceutical excipient), and forage for livestock (Anderson, 1989; Gregory, 1819; Movafeghi et al., 2010; Rios and Waterman, 1997; Verotta and El-Sebakhy, 2001; Yang et al., 2005). The dried root of Astragalus membranaceus, commonly called Radix Astragali, ‘Huang Qi’or ‘Huangqi’, has been used in traditional Chinese medicine (TCM) for hundreds to thousands of years and in Western medicine since the 1800s (WHO, 1999; McKenna et al., 2002). Authentic botanic sources of Radix Astragali, for which there is a WHO monograph (1999), are limited to the roots of A. membranaceus (Fisch.) Bunge and A. membranaceus var. Astragalus mongholicus (Bge.) Hsiao (syn. A. mongholicus Bge.), species indigenous to China, northern Korea, Mongolia, and Siberia and commercially cultivated in China and northern Korea (Ma et al., 2002; McKenna et al., 2002; Xiao et al., 2011; WHO, 1999). Currently available in US markets as a dietary supplement (Ganzera et al., 2001; Xiao et al., 2011), Radix Astragali is among the most popular of the Chinese herbs (Ma et al., 2002; Yu et al., 2007a,b) where it is used alone and in combination with other ingredients. In traditional preparations, a decoction or soup made from the sliced or shredded root is taken to treat deficiencies in

N.J. Szabo / Food and Chemical Toxicology 64 (2014) 322–334

323

sets of treatment and control animals assigned to the recovery study. In addition to the subchronic oral exposure study, bacterial reverse mutation, in vitro chromosome aberration, and in vivo mammalian erythrocyte micronucleus assays were conducted to assess any mutagenic and/or clastogenic potential of CAG. 2. Materials and methods 2.1. Test articles

Fig. 1. Structure of cycloastragenol.

organs such as the spleen or lungs. Current preparations include teas and extracts readied from the ground or powdered root taken to enhance the immune system, increase stamina, and as a cardiotonic agent to strengthen or increase cardiac output (McKenna et al., 2002; WHO, 1999). In China the officially accepted daily dose of Astragalus root is 9–30 g (McKenna et al., 2002). Cycloastragenol (CAG) is a secondary metabolite isolated from Radix Astragali. Present in all known Astragalus spp., CAG (9,19cycloanostane-3,6,16,25-tetrol,20,24-epoxy-(3b,6a,16b,20R,24S); Fig. 1) is both a triterpene aglycone2 and the most common genuine aglycone3 in the bioactive triterpenoid saponins called astragalosides (ASTs) (Rios and Waterman, 1997). Although CAG and the ASTs are found in all tissues of the Astragalus shrubs, the highest concentrations are localized in the roots (Yu et al., 2007b). Ten out of the eleven ASTs found in the root of A. membranaceus contain CAG as the aglycone, including AST IV (also called astramembrainnin I), the primary characteristic and main bioactive AST present in the roots of this species (Kitagawa et al., 1983; McKenna et al., 2002; Sevimli-Gür et al., 2011; Verotta and El-Sebakhy, 2001; Yu et al., 2007a; Zhou et al., 2012). In the course of isolating, identifying, and testing the components of the major bioactive fractions (i.e., saponins, polysaccharides and isoflavones), cardiovascular effects such as hypotension induced by Astragalus extracts, have been attributed primarily to the activities of AST IV, a molecule which has been demonstrated to exert a positive inotropic effect (Verotta and El-Sebakhy, 2001; Li and Cao, 2002; Rios and Waterman, 1997; Yu et al., 2007a). CAG, AST IV, and other related molecules isolated from Astragalus spp. have also been identified as small-molecule telomerase activators, substances that can induce the elongation of telomeres, the protective DNA sequences at the terminal ends of chromosomes (de Jesus et al., 2011; Fauce et al., 2008; Harley et al., 2011; Yang et al., 2012; Yung et al., 2012; Zhou et al., 2012). To support the development of CAG as a modern dietary ingredient for human consumption, the ingested safety of CAG was evaluated in the rat model via a 13-week repeated dose study with 4-week recovery period. Clinical observations, blood and urine assays, and gross and microscopic pathology were conducted to monitor for any sign of induced toxicity in the rat model. Although cardiotonic effects associated with Astragalus root extracts have not been attributed to CAG, a cardio-component consisting of blood pressure measurement and selected serum enzymes was also conducted. Reversibility, progression and/or the appearance of delayed changes were evaluated in additional 2 Triterpene glycosides are composed of a polycyclic aglycone (choline steroid or triterpenoid) having oxygenated positions to which sugar side chains are frequently attached by ether or acetal linkages (Verotta and El-Sebakhy, 2001). 3 Astragenol, an artifactual aglycone, can be produced if triterpenoid saponins are subjected to acidic hydrolysis (Kitagawa et al., 1983).

Cycloastragenol (CAS Nos. 78574-94-4 and 84605-18-5), a white to off-white loose powder, was derived from the dried ground roots of Astragalus trojanus that had been wild-harvested from the mountainous steppes of Turkey. Prior to extraction, the collected plants were authenticated by scientific faculty of the Department of Biology, Ege University, Izmir, Turkey, via visual and microscopic comparison to authentic standards.4 The final material contained 98% (w/w) CAG with no more than 2% water. In support of the current studies evaluating the safety of CAG, the test material was stored at room temperature (15–25 °C) and suspended with continuous stirring in (1) deionized reverse osmosis water containing 0.5% (w/v) methylcellulose and 2% (v/v) Tween 80Ò for the oral subchronic toxicity trial in rodents, (2) dimethyl sulfoxide (DMSO) and minimal essential medium (MEM) for the in vitro chromosome aberration assay, and (3) cottonseed oil for the in vivo mouse micronucleus assay. For the bacterial reverse mutation assay, the CAG test article (90% CAG) from acid-hydrolyzed AST IV isolated from A. membranaceus root was also stored at room temperature and was suspended in DMSO for the assay. Because CAG is a discrete, well-defined small molecule, isolation from the different Astragalus sources does not affect its structure or chemical or physical properties. 2.2. Chemicals and materials 9-Aminoacridine (9-AA), 2-aminoanthracene (2-AA), cottonseed oil, cyclophosphamide (CPA), ethylmethanesulfonate (EMS), methylmethanesulfonate (MMS), 2nitrofluorene (2-NF) and sodium azide (NaN3) were purchased from Sigma–Aldrich Corp. (St. Louis, MO). DMSO used in the bacterial reverse assay was purchased from Fisher Scientific (Pittsburgh, PA). The S9 metabolic activation mix used in the bacterial reverse mutation assay consisted of the supernatant fraction of rat liver homogenate (10%) (derived from male Sprague–Dawley (SD) rats pretreated with Aroclor 1254), 5 mM glucose-6-phosphate, 4 mM b-nicotinamide-adenine dinucleotide phosphate, 8 mM magnesium chloride, and 33 mM potassium chloride in 100 mM sodium phosphate buffer at pH 7.4. Nutrient Broth in the bacterial reverse mutation assay was Vogel–Bonner salt solution (Vogel and Bonner, 1956) supplemented with 2.5% (w/v) Oxoid Nutrient Broth No. 2 (Oxoid Ltd., Basingstoke, Hampshire, England). DMSO used in the in vitro chromosome aberration assay was supplied by AppliChem GmbH (Darmstadt, Germany). The S9 metabolic activation mix used in the in vitro chromosome aberration assay consisted of the supernatant fraction of rat liver homogenate (derived from male Wistar rats pretreated with phenobarbital and b-naphthoflavone) and mixed with cofactors as described for the bacterial reverse mutation assay. Water was in-house and distilled, unless otherwise specified. 2.3. Animals and organisms For the subchronic repeated dose assay performed at Product Safety Labs (Dayton, NJ), Hsd:SDÒrats were obtained from Harlan, Haslett, MI. All rats were inspected at delivery and during acclimation (8 days). At study initiation, the rats were 8 weeks old. Initial body weights for the males ranged from 220 to 245 g (mean, 230.2 ± 5.0 g) with females weighing from 165–191 g (mean, 177.0 ± 6.4 g); body weight variations were ±2.2% and ±3.6%, respectively. The rats were individually caged (National Research Council of the National Academies, 2011) in rooms where photoperiod (6:00 AM to 6:00 PM), room temperature (19–23 °C), humidity (43–75%) and air exchange (P10 air changes/h) were all controlled. Filtered tap water and 2016CM Harlan Teklad Global Rodent DietÒ (Harlan Laboratories, Indianapolis, IN) were freely available. The Salmonella typhimurium and Escherichia coli tester strains for the bacterial reverse mutation assay performed at BioReliance (Rockville, MD) were obtained, respectively, from Discovery Partners International, Inc. (San Diego, CA) and the National Collection of Industrial and Marine Bacteria (Aberdeen, Scotland). Chinese hamster V79 cells (CCL-93) were obtained from American Type Culture Collection (ATCC; Manassas, VA) for the in vitro chromosome aberration assay performed at BSL Bioservice Scientific Laboratories GmbH (Planegg, Germany). For the in vivo erythrocyte micronucleus assay performed at BSL Bioservice, NMRI mice were obtained from Charles River GmbH (Sulzfeld Germany). All mice

4 In addition to Ege University, voucher specimens of local Astragalus spp. have been deposited in collections including the Herbarium of the Department of Pharmacognosy, Hacettepe University, Ankara, Turkey (Bedir et al., 1998a,b, 1999a,b, 2001a,b).

324

N.J. Szabo / Food and Chemical Toxicology 64 (2014) 322–334

were visually inspected at delivery and during acclimation (P7 days). At dosing, the mice were at least 7 weeks old (range, 7–13 weeks); male weights ranged from 25.0 to 29.1 g (mean, 27.4 ± 1.0 g) with females weighing 22.3–26.2 g (mean, 24.5 ± 1.1 g); body weights varied by ± 7.5% and ± 8.0%, respectively. The mice were caged in groups (5 same sex/cage) in rooms where photoperiod (6:00 AM to 6:00 PM), room temperature (22 ± 3 °C), humidity (55 ± 10%), and air exchange (P10 air changes/h) were all controlled. Potable water (tap water, sulfur-acidified to pH 2.8) and Altromin 1324 maintenance diet (Altromin Spezialfutter GmbH & Co.) were freely available.

2.4. Guidelines The subchronic toxicity study in rats complied with OECD Guidelines for Testing of Chemicals and Food Ingredients, Section 4, No. 408, ‘‘Health Effects, Repeated Dose 90-Day Oral Toxicity Study in Rodents’’, adopted September 21, 19985 and was conducted under GLP in compliance with OECD Principles of Good Laboratory Practice (as revised in 1997)6 and US FDA GLP: 21 CFR 58 (1987). The protocol was reviewed by the Institutional Animal Care and Use Committee (IACUC) of PSL which is fully accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC) (Accredited Unit No. 000939). The mutagenic assay was conducted under GLP conditions in accordance with US EPA GLP: 40 CFR §160 and 40 CFR §792 (1989) and US FDA GLP: 21 CFR §58 (1987) and complied with the international guidelines of (1) the Ninth Addendum to OECD Guidelines for Testing of Chemicals, Section 4, No. 471, ‘‘Bacterial reverse mutation test’’, adopted July 21, 1997, (2) International Conference on Harmonisation (ICH) Technical requirements for registration of pharmaceuticals for human use. Guidance on Specific Aspects of Regulatory Genotoxicity Tests for Pharmaceuticals. S2A document, April 24, 1996, and (3) ICH Technical requirements for registration of pharmaceuticals for human use. Genotoxicity: A standard battery for genotoxicity testing of pharmaceuticals. S2B document, November 21, 1997. The in vitro clastogenic assay followed the international guidelines of (1) the Ninth Addendum to OECD Guidelines for Testing of Chemicals, Section 4, No. 473, ‘‘In vitro Mammalian Chromosome Aberration Test’’, adopted July 21, 1997 and (2) Commission Regulation (EC) No. 440/2008 B.10, ‘‘Mutagenicity – In vitro Mammalian Chromosome Aberration Test’’, dated May 30, 2008. The in vivo clastogenic assay followed the international guidelines of (1) the Ninth Addendum to OECD Guidelines for Testing of Chemicals, Section 4, No. 474, ‘‘Mammalian Erythrocyte Micronucleus Test’’, adopted July 21, 1997 and (2) Commission Regulation (EC) No. 440/2008 B.12, ‘‘Mammalian Erythrocyte Micronucleus Test’’, dated May 30, 2008. Both clastogenic assays were conducted under GLP in compliance with Chemikaliengesetz (‘‘Chemicals Act’’) of the Federal Republic of Germany, Appendix 1 to §19a as amended and promulgated on June 20, 2002 (BGBl.7 I Nr. 40 S. 2090) and revised October 31, 2006 (BGBl I Nr. 50 S. 2407) and with OECD (1998) Principles of Good Laboratory Practice (as revised in 1997).7 The rodent phase of the in vivo micronucleus study was conducted in accordance with the protocol reviewed by the Animal Welfare Officer of BSL Bioservice who is fully accredited by the competent authority Regierung von Oberbayern, Accreditation No 5.1-568-Bg/BSL, Landratsamt München, last revised, September 7, 2009.

2.5. Experimental design 2.5.1. 13-Week repeated dose study In the main toxicity study, rats were assigned to groups (10/sex/group) that received 0 (Group 1, vehicle control), 40 (Group 2), 80 (Group 3) or 150 (Group 4) mg/ kg bw/day CAG5 in aqueous-based vehicle8 via oral gavage (10 ml/kg bw/day) for 91 consecutive days. In the recovery study, additional groups (5/sex/group) received 0 (Group 5, vehicle control) or 150 (Group 6) mg/kg bw/day CAG for 91 consecutive days, then were observed post-treatment for 28 days. Surviving rats were terminated on Day 92 (male) or Day 93 (female) in the main toxicity study and on Day 120 in the recovery study. The stability, homogeneity and dose concentration of CAG in the test article and test formulations were confirmed using a validated high performance liquid chromatography method with evaporative light scattering detection. Clinical signs were reported twice daily during study and recovery periods and on the morning of necropsy. Detailed observations were made on Day 1 before initial dosing and weekly thereafter. Body weights and food consumption were also recorded on Day 1, weekly during the dosing and recovery periods, and on the day of necropsy. During Week 12 at the end of the exposure period, a blind func5 Concentrations of the test substance at up to the maximum tolerated dose (MTD) were not tested as part of this study. Dose levels were based on a previously conducted pilot study in rodents (up to 1000 mg/kg bw/d via oral gavage for 7 d with no mortalities and no adverse effects). 6 OECD Environmental Health and Safety Publication, Series on Principles of Good Laboratory Practice and Compliance Monitoring – Number 1. Environment Directorate Organisation for Economic Co-operation and Development, Paris, 1998. 7 Bundesgesetzblatt (Federal Law Gazette). 8 0.5% (w/v) methylcellulose and 2% (v/v) Tween 80Ò in deionized reverse-osmosis water.

tional observational battery9 (FOB) was performed on all surviving animals in the main toxicity study. Ophthalmological examinations (both eyes) were conducted on all rats during the acclimation period and surviving main toxicity animals on Day 90. The cardiac assessment consisted of selected serum biochemistry parameters and tail-cuff blood pressure measurements. Blood for aspartate aminotransferase (AST), total creatine phosphokinase (CK), and lactate dehydrogenase (LDH) was collected on Day 86 (main toxicity) and Day 118 (recovery). After a two-week procedural acclimation, tail-cuff blood pressures were also collected (Week 12 for main study and Week 16 for recovery). Each animal was evaluated five times with each 30-s evaluation consisting of an average of five data points. All animals were fasted P15 h prior to sample collection for clinical pathology measures. Urine, collected on Day 86 (main study) and Day 118 (recovery study), was examined for quality, color, clarity, volume, microscopic urine sediments, pH, glucose, specific gravity, protein, ketone, bilirubin, blood, and urobilinogen. Blood for hematology and clinical chemistry (Tables 1 and 2) was collected via sublingual bleeding under isoflurane anesthesia on the same days. On Day 92 (males) or 93 (females), the main study rats, having been fasted for at least 15 h, were euthanized by exsanguination from the abdominal aorta under isoflurane anesthesia and necropsied; on Day 120 all rats in the recovery study were euthanized in the same manner. At termination blood for coagulation parameters (prothrombin time, activated partial thromboplastin time) was collected via the inferior vena cava under isoflurane anesthesia from all rats. Organs and tissues10 were also removed, weighed and examined macroscopically from all rats. Organs and tissues from main study animals (but not recovery animals) were examined microscopically.

2.5.2. Bacterial reverse mutation assay The mutagenic potential of CAG was assessed in S. typhimurium tester strains TA98, TA100, TA1535 and TA1537 and E. coli WP2 uvrA in the presence and absence of S9 metabolic activation system mix using a two-phase plate incorporation method. The initial toxicity-mutation assay (first phase) provided a preliminary mutagenicity evaluation and established the dose-range for the second phase which then confirmed mutagenic potential. Based on pre-experiments to evaluate cytotoxicity, 5000 lg/plate CAG in DMSO was selected as the highest dose for all tester strains and conditions. For the initial toxicity-mutation assay, eight dose levels (5000, 1500, 500, 150, 50, 15, 5.0 and 1.5 lg/plate) were prepared by serial dilution in DMSO; for the second confirmatory phase, five dose levels (5000, 1500, 500, 150 and 50 lg/plate) were prepared. In the absence of S9 mix, the positive controls were NaN3 for S. typhimurium tester strains TA100 and TA1535, 2-NF for tester strain TA98, 9-AA for tester strain TA1537, and MMS for E. coli WP2 uvrA. In the presence of S9 mix, the positive control for all tester strains was 2-AA. DMSO was the vehicle (negative) control for all strains whether or not S9 mix was present. Plates were prepared by adding 0.05 ml of positive or negative control solution or test formulation, 0.1 ml bacterial suspension (109 cells/ml), and 0.5 ml of either the S9 mix (for metabolic activation) or the 100 mM phosphate sham buffer (pH 7.4) to 2.0 ml of molten selective top agar at 45 ± 2 °C in sterile test tubes and mixing. (Top agar contained 0.05 mM histidine, 0.05 mM biotin and 0.05 mM tryptophan.) After mixing and without incubation, the tube contents were poured immediately onto minimal glucose agar plates. The plates were inverted and incubated at 37 ± 2 °C for 48–72 h before counting. Plates for the first phase were prepared and tested in duplicate; plates for the second phase were prepared and tested in triplicate. Growth inhibition (evidence of test article toxicity) and the presence of precipitate were assessed in all test plates. In the absence of toxicity, revertant colonies for a given tester strain and activation condition were counted. The test article would be considered mutagenic if the number of revertant colonies in the test article groups increased in a dose-dependent manner over at least two increasing concentrations and/or the number of revertants were biologically relevant (i.e., at least equal to two times the TA 98, TA 100 and/or E. coli WP2 uvrA negative control means or three times the TA 1535 and/or TA 1537 negative control means). Data were not analyzed statistically. 2.5.3. In vitro chromosome aberration assay

9 The rats were evaluated in random order during handling and in an open field for excitability, autonomic function, gait and sensorimotor coordination (open field and manipulative evaluations), reactivity and sensitivity (elicited behavior) and other abnormal clinical signs including convulsions, tremors, unusual or bizarre behavior, emaciation, dehydration, and general appearance. 10 Accessory genital organs (prostate and seminal vesicles); adrenal glands⁄; all gross lesions; aorta; bone (femur); brain⁄ (cerebellar cortex, cerebral cortex, medulla/ pons); cecum; colon; duodenum; epididymides⁄; esophagus; eyes (with optic nerves); Harderian gland; heart⁄; ileum (with Peyer’s patches); jejunum; kidneys⁄; larynx; liver⁄; lungs; lymph nodes (mesenteric and mandibular); mammary gland; nose and nasal turbinates; ovaries⁄ (with oviducts); pancreas; peripheral nerve (sciatic); pharynx; pituitary; prostate; rectum; salivary glands (sublingual, submanidibular, parotid); skeletal muscle; skin; spinal cord (cervical, mid-thoracic, lumbar); spleen⁄; sternum (with bone marrow); stomach; testes⁄; thymus⁄; thyroid/parathyroid; trachea; urinary bladder; uterus with oviducts⁄; vagina. (⁄Weighed organs).

325

N.J. Szabo / Food and Chemical Toxicology 64 (2014) 322–334 Table 1 Hematology, coagulation, and clinical chemistry parameters in male rats following 86 or 91 days of treatment with CAGa. Parameter

Main toxicity study Group 1 0 mg/kg bw/d

Recovery study Group 2 40 mg/kg bw/d

Group 3 80 mg/kg bw/d

Group 4 150 mg/kg bw/d

Group 5 0 mg/kg bw/d

Group 6 150 mg/kg bw/d

Hematology – Day 86 Main Toxicity/Day 118 Recovery RBC (106/ll) 9.16 ± 0.47b Hemoglobin (g/dl) 15.6 ± 0.8b Hematocrit (%) 46.7 ± 2.6b MCV (fl) 51.0 ± 1.0b MCH (pg) 17.1 ± 0.4b MCHC (g/dl) 33.5 ± 0.4b RDW (%) 12.3 ± 0.4b Platelet count (103/ll) 743 ± 148b WBC (103/ll) 11.28 ± 1.32b 171.2 ± 24.4b ARET (103/ll) ANEU (103/ll) 1.74 ± 0.49b ALYM (103/ll) 8.87 ± 0.95b AMON (103/ll) 0.31 ± 0.08b AEOS (103/ll) 0.16 ± 0.05b ABAS (103/ll) 0.09 ± 0.03b 0.10 ± 0.03b ALUC (103/ll) AATL (103/ll) – AIL (103/ll) –

8.86 ± 0.42 15.4 ± 0.6 46.1 ± 1.8 52.1 ± 1.3 17.4 ± 0.4 33.5 ± 0.5 12.6 ± 1.0 788 ± 128 10.92 ± 1.34 181.5 ± 75.8 1.61 ± 0.57 8.66 ± 1.07 0.32 ± 0.09 0.16 ± 0.05 0.08 ± 0.03 0.09 ± 0.03 – –

9.06 ± 0.25 15.8 ± 0.5 47.2 ± 1.6 52.1 ± 0.7 17.4 ± 0.3 33.5 ± 0.6 12.1 ± 0.4 800 ± 156 10.68 ± 1.47 164.4 ± 17.3 1.42 ± 0.49 8.64 ± 1.08 0.30 ± 0.06 0.17 ± 0.04 0.07 ± 0.03 0.09 ± 0.03 – –

8.91 ± 0.30 15.6 ± 0.7 45.9 ± 2.0 51.5 ± 1.2 17.5 ± 0.4* 34.0 ± 0.4 12.2 ± 0.3 754 ± 79 11.22 ± 2.38 162.8 ± 22.1 1.55 ± 0.62 8.99 ± 2.18 0.35 ± 0.11 0.17 ± 0.04 0.08 ± 0.04 0.10 ± 0.04 – –

9.48 ± 0.35c 16.2 ± 0.3c 49.1 ± 1.3c 51.8 ± 2.0c 17.0 ± 0.6c 32.9 ± 0.2c 12.2 ± 0.2c 855 ± 61c 11.41 ± 2.77c 194.6 ± 38.3 1.55 ± 0.78c 9.28 ± 2.18c 0.26 ± 0.07c 0.14 ± 0.02c 0.10 ± 0.05c 0.09 ± 0.02c – –

9.25 ± 0.36 16.1 ± 0.7 48.8 ± 2.6 52.7 ± 1.8 17.5 ± 0.3 33.1 ± 0.5 12.2 ± 0.5 922 ± 72 11.01 ± 0.58 192.6 ± 10.0 1.40 ± 0.16 9.00 ± 0.51 0.30 ± 0.09 0.14 ± 0.04 0.08 ± 0.03 0.09 ± 0.04 – –

Coagulation – Day 92 Main Toxicity/Day 120 Recovery PT (s) 10.5 ± 0.3b APTT (s) 17.7 ± 1.2b

10.6 ± 0.2b 17.7 ± 2.2b

10.7 ± 0.2 18.6 ± 1.6

10.6 ± 0.2b 18.4 ± 0.9b

10.3 ± 0.2 17.5 ± 0.6

10.3 ± 0.2c 17.0 ± 2.5c

72 ± 10b 41 ± 4 7.1 ± 2.1 79 ± 11* 0.10 ± 0.0* 17 ± 2 0.31 ± 0.03 89 ± 9* 40 ± 10* 151 ± 24 6.1 ± 0.2 3.2 ± 0.1 3.0 ± 0.2 10.0 ± 0.3 6.3 ± 0.5 142.2 ± 1.2 5.59 ± 0.31 103.5 ± 1.2 187 ± 65 529 ± 205

73 ± 7 40 ± 9 7.0 ± 1.4d 85 ± 11 0.10 ± 0.02b,* 17 ± 2 0.29 ± 0.03 93 ± 14 45 ± 10 146 ± 24b 6.0 ± 0.3b 3.1 ± 0.1 3.0 ± 0.3b 9.8 ± 0.6b 6.8 ± 0.5 142.3 ± 1.5 5.66 ± 0.39 103.1 ± 1.2 176 ± 44 549 ± 120b

99 ± 19 43 ± 5 6.1 ± 2.1c 79 ± 26 0.14 ± 0.02 17 ± 0 0.35 ± 0.03 74 ± 14 53 ± 11 132 ± 17 6.0 ± 0.1c 3.3 ± 0.1 2.7 ± 0.2c 9.1 ± 0.4c 6.2 ± 0.2 140.2 ± 1.3 4.91 ± 0.13 101.4 ± 1.3 151 ± 72 339 ± 118

104 ± 25 44 ± 15 5.7 ± 0.9e 64 ± 18 0.15 ± 0.03 17 ± 1 0.36 ± 0.03 78 ± 19 55 ± 6 138 ± 20 6.0 ± 0.2e 3.2 ± 0.3 e 2.6 ± 0.1 e 9.2 ± 0.2 6.1 ± 0.3 140.1 ± 1.9 4.92 ± 0.12 101.0 ± 1.8 153 ± 57 363 ± 176

Clinical Chemistry – Day 86 Main Toxicity/Day 118 Recovery AST (U/l) 79 ± 10b 78 ± 16 ALT (U/l) 46 ± 7d 44 ± 5b SDH (U/l) 6.0 ± 2.2e 8.0 ± 0.8b ALKP (U/l) 98 ± 18b 89 ± 17 BILI (mg/dl) 0.12 ± 0.02b 0.10 ± 0.01* BUN (mg/dl) 18 ± 2b 18 ± 2 Creatinine (mg/dl) 0.31 ± 0.03b 0.30 ± 0.03 b Total cholesterol (mg/dl) 104 ± 10 98 ± 10 b Triglycerides (mg/dl) 56 ± 10 43 ± 10* Glucose, fasting (mg/dl) 148 ± 27b 141 ± 20 Total protein (g/dl) 6.1 ± 0.2d 5.9 ± 0.2b Albumin (g/dl) 3.1 ± 0.1b 3.2 ± 0.1 Globulin (g/dl) 3.0 ± 0.2d 2.8 ± 0.2b d Calcium (mg/dl) 9.8 ± 0.3 9.7 ± 0.5 Inorganic phosphorus (mg/dl) 6.5 ± 0.4 b 6.5 ± 0.4 b Sodium (mmol/l) 142.1 ± 1.3 142.8 ± 2.1 Potassium (mmol/l) 5.59 ± 0.36b 5.54 ± 0.20 Chloride (mmol/l) 103.7 ± 1.2b 103.9 ± 1.3 CK (U/l) 207 ± 96b 176 ± 71 LDH (mmol/l) 610 ± 312b 539 ± 263 *

P < 0.05, compared to control. All data are presented as mean values ± standard deviations with N = 10 for main toxicity groups and N = 5 for recovery groups, except when otherwise indicated. b N = 9. c N = 4. d N = 8. e N = 7. a

The potential of CAG to induce structural changes in a cell’s genetic material was evaluated in an in vitro mammalian chromosome aberration test using Chinese hamster V79 cells. Test solutions were prepared 1 h before treatment by dissolving CAG in DMSO (ultrasound, 5 min) then diluting in culture medium until the DMSO concentration was 1% (ultrasound, 5 min). At final dilution, precipitation of the test material was observed. Due to a lack of solubility, CAG was applied as suspensions of selected concentrations to V79 cells. Positive controls were prepared in nutrient medium and consisted of EMS (400 and 600 lg/ml) in the absence of metabolic activation and CPA (0.83 lg/ml) in the presence of metabolic activation. Negative controls consisted of 1% DMSO in MEM. All cultures (control and test) were prepared in duplicate. At least 200 metaphases per test concentration and validity control sample were scored for cytogenic damage to determine the incidence of structural chromosomal aberrations (i.e., breaks, fragments, deletion exchanges, chromosomal disintegrations and gaps). In addition, 1000 cells per test culture and validity control were evaluated for cytotoxicity to determine the mitotic index. As an additional measure of cytotoxicity, relative cell density was calculated as the mean of 20 cell counts per test group (cells within the visual field at a 400-fold magnification). The test article would be considered clastogenic if the aberration rate in the dose groups increased significantly and in a dose-dependent manner compared to the negative

control groups and the increase was also biologically relevant (i.e., greater than the laboratory negative control range of up to 4.0% aberrant cells) for at least one dose group. Data were not analyzed statistically.

2.5.4. In vivo erythrocyte micronucleus assay The clastogenic potential of CAG was also assessed using the in vivo erythrocyte micronucleus assay in the mouse. In the range-finding study, 2000 mg/kg bw CAG was prepared as a suspension in cottonseed oil and administered via intraperitoneal (i.p.) injection to Crl:NMRI mice (n = 3/sex) in two injections (each equivalent to 1000 mg/kg bw) over 3 h. Symptoms of systemic toxicity (i.e., reduced spontaneous activity, constricted abdomen, piloerection, bradykinesia, half-closed eyes, weight loss, kyphosis and recumbency) were observed. In accordance with OECD guideline 474, the MTD was determined to be 2000 mg/kg bw and was used in the main study. In the main experiment, CAG in cottonseed oil at levels of 400 (0.2 MTD), 1000 (0.5 MTD) and 2000 (1 MTD) mg/kg bw was administered via i.p. injection (10 ml/kg bw) to Crl:NMRI mice (n = 5/sex/group). Administration of the low- and mid-level doses was provided in a single injection (10 ml/kg bw). Administration of the highest dose was provided as described in the range finding study. Negative

326

N.J. Szabo / Food and Chemical Toxicology 64 (2014) 322–334

Table 2 Hematology, coagulation, and clinical chemistry parameters in female rats following 86 or 92 days of treatment with CAGa. Parameter

Main toxicity study Group 1 0 mg/kg bw/d

Recovery study Group 2 40 mg/kg bw/d

Group 3 80 mg/kg bw/d

Group 4 150 mg/kg bw/d

Group 5 0 mg/kg bw/d

Group 6 150 mg/kg bw/d

Hematology – Day 86 Main Toxicity/Day 118 Recovery RBC (106/ll) 8.04 ± 0.26 Hemoglobin (g/dl) 14.7 ± 0.6 Hematocrit (%) 42.8 ± 2.1 MCV (fl) 53.3 ± 1.4 MCH (pg) 18.2 ± 0.6 MCHC (g/dl) 34.3 ± 0.9 RDW (%) 11.2 ± 0.3 Platelet count (103/ll) 807 ± 137 3 WBC (10 /ll) 6.74 ± 1.48 ARET (103/ll) 138.8 ± 22.5 ANEU (103/ll) 0.75 ± 0.24 ALYM (103/ll) 5.62 ± 1.43 AMON (103/ll) 0.15 ± 0.04 AEOS (103/ll) 0.12 ± 0.05 3 ABAS (10 /ll) 0.02 ± 0.01 0.05 ± 0.04 ALUC (103/ll) AATL (103/ll) 0.06b AIL (103/ll) 0.12b

8.22 ± 0.25 14.8 ± 0.5 43.5 ± 1.2 53.0 ± 1.4 18.0 ± 0.5 33.9 ± 0.6 11.3 ± 0.3 837 ± 133 7.07 ± 2.18 170.3 ± 32.9 1.02 ± 0.55 5.66 ± 1.62 0.16 ± 0.07 0.15 ± 0.06 0.03 ± 0.01 0.05 ± 0.03 – –

7.97 ± 0.10 14.7 ± 0.4 43.0 ± 0.7 54.0 ± 1.0 18.5 ± 0.5 34.3 ± 0.8 11.3 ± 0.3 775 ± 115 6.79 ± 1.32 182.8 ± 30.1* 0.85 ± 0.29 5.54 ± 1.16 0.16 ± 0.03 0.15 ± 0.05 0.03 ± 0.01 0.06 ± 0.01 – –

8.23 ± 0.20 14.8 ± 0.2 43.7 ± 1.2 53.1 ± 1.1 18.0 ± 0.3 33.9 ± 0.5 11.3 ± 0.3 880 ± 212 7.21 ± 1.41 169.1 ± 40.4 1.02 ± 0.38 5.79 ± 1.29 0.16 ± 0.05 0.16 ± 0.06 0.03 ± 0.02 0.06 ± 0.02 – –

8.37 ± 019 15.1 ± 0.4 45.0 ± 0.9 53.7 ± 0.8 18.1 ± 0.4 33.7 ± 0.2 10.8 ± 0.3 954 ± 208 6.09 ± 0.30 166.2 ± 20.4 0.67 ± 0.15 5.09 ± 0.16 0.14 ± 0.01 0.12 ± 0.02 0.03 ± 0.01 0.04 ± 0.01 – –

8.53 ± 0.29 15.0 ± 0.6 45.0 ± 1.4 52.8 ± 1.8 17.6 ± 0.7 33.3 ± 0.5 11.2 ± 0.5 947 ± 68 7.79 ± 1.22* 170.6 ± 53.2 0.86 ± 0.36 6.54 ± 0.93* 0.16 0.06 0.12 ± 0.02 0.04 ± 0.02 0.07 ± 0.04 – –

Coagulation – Day 93 Main Toxicity/Day 120 Recovery PT (sec) 10.3 ± 0.2 APTT (s) 16.3 ± 1.2

10.3 ± 0.2 16.7 ± 1.3

10.3 ± 0.2 16.0 ± 1.2

10.3 ± 0.4 16.2 ± 1.9

9.9 ± 0.1 15.6 ± 1.6

10.0 ± 0.1 15.2 ± 1.0

78 ± 7 35 ± 3 5.1 ± 1.4 64 ± 12 0.16 ± 0.02 19 ± 3 0.39 ± 0.04 105 ± 16 31 ± 6 114 ± 14 5.9 ± 0.4 3.3 ± 0.2 2.6 ± 0.2 9.7 ± 0.4 5.5 ± 0.4 135.9 ± 1.9 4.93 ± 0.33 99.6 ± 1.6 204 ± 49 591 ± 177

73 ± 10 33 ± 4 6.4 ± 1.1 71 ± 14 0.13 ± 0.02 19 ± 2 0.40 ± 0.03 115 ± 22 35 ± 5 102 ± 6 6.0 ± 0.3 3.3 ± 0.1 2.7 ± 0.2 9.6 ± 0.4 5.4 ± 0.4 130.9 ± 10.6 4.48 ± 0.48 96.2 ± 8.2 152 ± 57 425 ± 210

75 ± 7 37 ± 3 3.5 ± 1.2d 71 ± 14 0.16 ± 0.01e 20 ± 1 0.45 ± 0.03 118 ± 12 35 ± 7 109 ± 11d 5.5 ± 0.4d 3.2 ± 0.3 2.4 ± 0.1d 8.5 ± 0.2d 4.7 ± 0.1 141.3 ± 0.8 4.39 ± 0.14 103.9 ± 1.1 171 ± 87 425 ± 312

68 ± 4 35 ± 3 3.5 ± 2.6d 66 ± 14 0.15 ± 0.02 18 ± 2 0.37 ± 0.06* 89 ± 23* 33 ± 5 113 ± 8 5.7 ± 0.2d 3.1 ± 0.1 2.5 ± 0.2d 8.9 ± 0.1d,* 5.4 ± 0.5* 140.7 ± 0.8 4.74 ± 0.25* 103.5 ± 1.1 128 ± 44 265 ± 159

Clinical Chemistry – Day 86 Main Toxicity/Day 118 Recovery AST (U/l) 75 ± 14 77 ± 8 ALT (U/l) 33 ± 5 33 ± 4 SDH (U/l) 6.0 ± 1.9c 5.3 ± 1.2c ALKP (U/l) 69 ± 12 71 ± 22 BILI (mg/dl) 0.14 ± 0.02 0.14 ± 0.01 BUN (mg/dl) 21 ± 3 20 ± 3 Creatinine (mg/dl) 0.41 ± 0.02 0.42 ± 0.05 Total cholesterol (mg/dl) 113 ± 23 103 ± 21 Triglycerides (mg/dl) 34 ± 9 36 ± 7 Glucose, fasting (mg/dl) 120 ± 32 106 ± 12c Total protein (g/dl) 5.8 ± 0.3 5.8 ± 0.3c Albumin (g/dl) 3.2 ± 0.2 3.2 ± 0.2 Globulin (g/dl) 2.6 ± 0.2 2.6 ± 0.2c Calcium (mg/dl) 9.8 ± 0.2 9.6 ± 0.3c Inorganic phosphorus (mg/dl) 5.5 ± 0.7 5.3 ± 0.7 Sodium (mmol/l) 136.1 ± 2.1 135.5 ± 2.5c Potassium (mmol/l) 4.89 ± 0.40 4.76 ± 0.35c Chloride (mmol/l) 100.6 ± 1.8 99.5 ± 2.2c CK (U/l) 182 ± 105 190 ± 68 LDH (mmol/l) 494 ± 293 547 ± 207 *

P < 0.05, compared to control. All data are presented as mean values ± standard deviations with N = 10 for main toxicity groups and N = 5 for recovery groups, except when otherwise indicated. b N = 1. c N = 9. d N = 3. e N = 4. a

control animals (n = 5/sex) received a single dose of cottonseed oil vehicle. Positive control animals (n = 5/sex) received a single dose of 40 mg/kg bw CPA. Peripheral blood was collected from all animals at 44 h and a second time from mice in the negative control and high dose groups at 68 h. A minimum of 16 h after fixing in ultracold methanol, blood cells were washed, labeled and/or stained (anti-CD71 antibodies with fluorescein-isothiocyanate for immature erythrocytes, anti-CD61 antibodies with phycoerythrin for platelets, DNA specific stain propidium iodide micronuclei DNA content) in preparation for evaluation by flow cytometry. To determine the incidence (percentage) of polychromatic (micronucleated immature) erythrocytes (PCE) in peripheral blood, a minimum of 10,000 immature erythrocytes from each mouse were examined and scored. To assess cytotoxicity, relative (rel.) PCE11 was determined for each mouse. The test material would be considered clastogenic if the incidence of micronucleated cells in the test groups increased in a dose-dependent manner and/or the increase was also biologically relevant for at least one dose group.

11

The proportion of PCEs among total erythrocytes.

2.6. Statistical analyses 2.6.1. Subchronic repeated dose study Data from male and female rats were analyzed separately. Differences from the control group were statistically significant at a level of 5% (P < 0.05). Group means and standard deviations were calculated for body weight, daily body weight gain, daily food consumption, food efficiency, functional observation, organ weight, and organ-to-body/brain weight ratios. Data from Groups 1 through 4 were analyzed for homogeneity of variances and normality using Bartlett’s test. When variances were homogeneous, treatment and control groups were compared using a One-way Analysis of Variance (ANOVA); Dunnett’s test for multiple comparisons was used when One-way ANOVA was significant. When variances were significantly different, treatment and control groups were compared using the Kruskal– Wallis nonparametric ANOVA; Dunn’s test was used when non-parametric ANOVA was significant. Data for Groups 5 and 6 were analyzed using an unpaired T-test; when the difference between the two standard deviations was significant, the Mann–Whitney test was used. Clinical pathology data were analyzed for homogeneity by Levine’s test and for normality by the Shapiro–Wilk test. When preliminary testing was not significant, One-way ANOVA was used, followed by Dunnett’s test. When preliminary testing

N.J. Szabo / Food and Chemical Toxicology 64 (2014) 322–334 was significant, the data were transformed12 to achieve normality and variance homogeneity then analyzed by One-way ANOVA and Dunnett’s test. If an individual observation was recorded as being less than a certain value, calculations were performed using half the recorded value. When the Shapiro–Wilk test was not significant but Levine’s test was significant, a robust version of Dunnett’s test was used. 2.6.2. In vivo erythrocyte micronucleus assay The nonparametric Mann–Whitney test was used to analyze the 44-h and 68-h post-treatment PCE incidence and rel. PCE data. Mean values for the induction of immature erythrocytes and rel. PCE were calculated for each group with differences from the negative control being statistically significant at the 5% level. In accordance with OECD guidelines for the assay, biological relevance was the primary criterion for interpreting results, although statistical evaluation aided the interpretation.

3. Results 3.1. 13-Week repeated dose study 3.1.1. Test article and gavage formulations The stability, homogeneity and dose concentrations of CAG were confirmed in the test substance and dose formulations. Analysis of the test substance on Days 1 (94.30%) and 91 (95.26%) indicated stability over the course of the study. Similarly, CAG concentrations ranged from 99.3–100.3% of the targeted concentrations for the 0.4% (4000 ppm), 0.8% (8000 ppm), and 1.5% (15,000 ppm) dose formulations during 91 days of storage. CAG was also shown to be homogenously distributed in the dose formulations; Day 0 samples for the top, middle, and bottom of the dose formulations resulted in average CAG target concentrations of 97.5% (low-dose), 92.9% (mid-dose), and 96.9% (high-dose). Averaging 89.0%, 91.3%, and 87.9%, respectively, of the theoretical values for nominal concentrations of 0.4%, 0.8%, and 1.5% CAG, the dose formulation concentrations overall were consistent with the expected concentrations for CAG. Because CAG was stable in the test substance and the test substance was homogenously mixed into the dose formulations at concentrations that were confirmed to be acceptable and stable over the course of the study, the animals received the targeted concentrations of 0.4%, 0.8%, and 1.5% of CAG via oral gavage throughout the study. 3.1.2. Animal-related endpoints No treatment-related mortalities13 occurred during main toxicity or recovery phases, nor were any treatment-related changes in physical condition or behavior observed. Further, no statistically significant changes in body weight were noted for any treatment group compared with the control group (Fig. 2). On Days 22–29, both mean daily body weight gain and mean food efficiency in Group 2 males were significantly decreased compared to Group 1 control males (P < 0.01 and P < 0.05, respectively). In all other male treatment groups mean daily body weight gains and food consumption were comparable to the control group. Because decreased food efficiency in Group 2 males was sporadic, not dose-dependent and not accompanied by decreases in body weight or food consumption, the findings were incidental. Among females, mean daily food consumption in Group 2 on Days 36–43 (P < 0.01) and Group 4 on Days 36–43 (P < 0.01), Days 50–57 (P < 0.05) and overall (Days 1– 91) (P < 0.05) was significantly increased compared to Group 1 controls. Because the increased food consumption observed in Group 2 and 4 females was sporadic, not dose-dependent, and not accompanied by corresponding changes in body weight gain or food efficiency, the findings were incidental. No ophthalmological

12

Order of attempted transformation was log, square-root, and rank-order. One main toxicity control male was euthanized for humane reasons on Day 65. The animal had sustained internal injury from a gavage needle while escaping from the technician during dosing; macroscopic and microscopic findings were consistent with injury incurred from a gavage needle. 13

327

abnormalities were identified in any animals prior to study initiation or on Day 90. No statistically significant effects were reported for urinalysis parameters (data not shown) in any surviving rats on Day 86 (main toxicity) or Day 118 (recovery). The hematology results (Tables 1 and 2) included several statistically significant changes: increased mean corpuscular hemoglobin (MCH) in Group 4 males on Day 86 compared to the control group (P < 0.05), increased absolute reticulocyte count in Group 3 females compared to corresponding Group 1 females (P < 0.05), and increased total white blood cell count and absolute lymphocyte concentration in Group 6 females on Day 118, compared to corresponding Group 5 control females (P < 0.05 for both). Each of these findings was judged to be incidental and unrelated to the test article because the magnitude of change was low and not clinically significant, the change was not correlated to other clinical or histopathologic changes, and/or the change was within the historical ranges observed in the age and strain of rat used (Pettersen et al., 1996; Derelanko, 2000). Absolute number of immature lymphocytes and absolute atypical lymphocytes were also identified in one Group 1 main study female. These changes were incidental; occurring in a control animal, they were not treatment-related. No effect on blood cell morphology was reported. Clinical chemistry results (Tables 1 and 2) on Day 86 revealed statistically significant decreases in alkaline phosphatase and total cholesterol in Group 3 males, triglycerides in Group 2 and Group 3 males, and bilirubin in Group 2, 3 and 4 males, when compared to Group 1 control males (P < 0.05 for all). On Day 118 in Group 6 females, statistically significant decreases were noted in creatinine and cholesterol, along with statistically significant increases in calcium, potassium, and inorganic phosphorus concentrations compared to Group 5 control females (P < 0.05 for all). None of these findings was judged to be toxicologically significant, as none demonstrated a dose-dependent relationship, was accompanied by any other clinical or histopathologic changes, and/or was outside the historical range observed in the age and strain of rat used (Pettersen et al., 1996; Derelanko, 2000).None of the coagulation parameters in the main toxicity (Day 92/93) or recovery (Day 120) studies was affected by treatment (Table 1 and 2). In the cardiac assessment, no effect on blood pressure (i.e., systolic, diastolic, or mean arterial pressure values) (data not shown) was identified between the dose groups and the corresponding control groups for either sex in the main toxicity or recovery study. In addition, the Week 16 blood pressure measures for Groups 5 and 6 and Week 12 values for Groups 1 and 4 were comparable. No statistically significant changes in the serum biochemistry parameters, AST, CK or LDH were identified in male or female rats in the main toxicity (Day 86) or recovery (Day 118) studies (Tables 1 and 2). Analysis of absolute and relative organ weights revealed one statistically significant change in each of two male treatment groups when compared with their respective controls at termination (Table 3), increased absolute liver weight in Group 4 males (P < 0.05) and increased heart-to-brain weight ratio in Group 6 males (P < 0.05). Neither finding was not dose-dependent or had any clinical or histopathologic correlate; the observations were therefore determined to be incidental and not related to treatment. In Group 3 and Group 4 females (Table 4), statistically significant changes included increased absolute heart weight (P < 0.05 for both groups), heart-to-body weight (P < 0.01, Group 3; P < 0.05, Group 4) and heart-to-brain weight ratios (P < 0.01, Group 3; P < 0.05, Group 4). Although there appeared to be a dose-dependent trend across the control, low-dose, and intermediate-dose groups for these parameters (Table 4), only the intermediate-dose group differed from the control to a statistically significant degree. In addition, these same parameters in high-dose Group 4 – although significantly increased over the control – did not continue

328

N.J. Szabo / Food and Chemical Toxicology 64 (2014) 322–334 450

400

Body weight (g)

350

300

250

200

150 1

8

15

22

29

36

43

50

57

64

71

78

85

91

98

105

112

119

Period (Day) Male -Group 1 (0 mg/kg) Male -Group 4 (150 mg/kg) Female -Group 1 (0 mg/kg) Female -Group 4 (150 mg/kg)

Male -Group 2 (40 mg/kg) Male -Group 5 (0 mg/kg) Female -Group 2 (40 mg/kg) Female -Group 5 (0 mg/kg)

Male -Group 3 (80 mg/kg) Male -Group 6 (150 mg/kg) Female -Group 3 (80 mg/kg) Female -Group 6 (150 mg/kg)

Fig. 2. Mean body weights of male and female rats during 13-week treatment and 4-week recovery periods.

Table 3 Terminal body weights and absolute and relative organ weights of male rats following 91 days of treatment with CAGa. Parameter

Main toxicity study – Day 92

Recovery study – Day 120

Group 1b 0 mg/kg bw/d

Group 2 40 mg/kg bw/d

Group 3 80 mg/kg bw/d

Group 4 150 mg/kg bw/d

Group 5 0 mg/kg bw/d

Group 6 150 mg/kg bw/d

Absolute weights Terminal body weight (g) Adrenals (g) Brain (g) Epididymides (g) Heart (g) Kidneys (g) Liver (g) Spleen (g) Testes (g) Thymus (mg)

377.3 ± 16.1 0.0633 ± 0.0064 1.971 ± 0.063 1.368 ± 0.167 1.280 ± 0.099 2.581 ± 0.172 9.867 ± 0.498 0.773 ± 0.041 3.614 ± 0.568 0.2528 ± 0.0491

380.8 ± 25.7 0.0643 ± 0.0099 1.963 ± 0.097 1.437 ± 0.093 1.234 ± 0.085 2.619 ± 0.205 9.859 ± 0.925 0.745 ± 0.104 3.724 ± 0.309 0.2736 ± 0.0909

382.6 ± 18.8 0.0655 ± 0.0064 2.032 ± 0.059 1.418 ± 0.064 1.269 ± 0.114 2.600 ± 0.162 10.393 ± 0.820 0.771 ± 0.074 3.839 ± 0.218 0.2710 ± 0.0628

395.4 ± 30.1 0.0598 ± 0.0073 2.011 ± 0.077 1.417 ± 0.110 1.280 ± 0.102 2.670 ± 0.195 10.918 ± 1.130* 0.811 ± 0.085 3.959 ± 0.225 0.2908 ± 0.0262

393.4 ± 21.1 0.0564 ± 0.0068 1.940 ± 0.097 1.428 ± 0.124 1.202 ± 0.083 2.454 ± 0.244 10.060 ± 1.254 0.766 ± 0.090 3.836 ± 0.379 0.1806 ± 0.0382

407.0 ± 28.1 0.0560 ± 0.0058 1.906 ± 0.078 1.440 ± 0.122 1.320 ± 0.131 2.566 ± 0.301 10.416 ± 1.054 0.700 ± 0.035 4.074 ± 0.336 0.2054 ± 0.0314

Organ-to-body weight ratios Adrenals/BW Brain/BW Epididymides/BW Heart/BW Kidneys/BW Liver/BW Spleen/BW Testes/BW Thymus/BW

0.1679 ± 0.0161 5.229 ± 0.196 3.6355 ± 0.5060 3.390 ± 0.185 6.841 ± 0.357 26.162 ± 1.115 2.055 ± 0.176 9.615 ± 1.701 0.6707 ± 0.1321

0.1695 ± 0.0282 5.172 ± 0.379 3.7869 ± 0.3316 3.245 ± 0.178 6.884 ± 0.431 25.888 ± 1.645 1.953 ± 0.198 9.790 ± 0.671 0.7158 ± 0.2251

0.1717 ± 0.0202 5.322 ± 0.297 3.7160 ± 0.2796 3.314 ± 0.200 6.799 ± 0.356 27.142 ± 1.210 2.014 ± 0.139 10.053 ± 0.730 0.7087 ± 0.1659

0.1511 ± 0.0122 5.116 ± 0.483 3.5876 ± 0.1849 3.246 ± 0.256 6.762 ± 0.336 27.621 ± 2.031 2.058 ± 0.225 10.064 ± 0.935 0.7377 ± 0.0709

0.1431 ± 0.0115 4.934 ± 0.167 3.6294 ± 0.2365 3.056 ± 0.144 6.226 ± 0.293 25.506 ± 1.879 1.955 ± 0.285 9.734 ± 0.478 0.4601 ± 0.1010

0.1378 ± 0.0147 4.692 ± 0.179 3.5571 ± 0.4427 3.244 ± 0.234 6.290 ± 0.377 25.558 ± 1.220 1.723 ± 0.088 10.020 ± 0.662 0.5054 ± 0.0747

Organ-to-brain weight ratios Adrenals/BrW Epididymides/BrW Heart/BrW Kidneys/BrW Liver/BrW Spleen/BrW Testes/BrW Thymus/BrW

0.0321 ± 0.0028 0.6943 ± 0.0854 0.650 ± 0.051 1.310 ± 0.083 5.009 ± 0.268 0.393 ± 0.026 1.835 ± 0.294 0.1280 ± 0.0226

0.0329 ± 0.0056 0.7344 ± 0.0688 0.629 ± 0.037 1.334 ± 0.082 5.029 ± 0.482 0.380 ± 0.057 1.901 ± 0.178 0.1408 ± 0.0510

0.0323 ± 0.0032 0.6987 ± 0.0450 0.626 ± 0.065 1.279 ± 0.069 5.117 ± 0.409 0.380 ± 0.040 1.891 ± 0.125 0.1332 ± 0.0299

0.0298 ± 0.0041 0.7059 ± 0.0641 0.637 ± 0.053 1.330 ± 0.119 5.442 ± 0.663 0.403 ± 0.040 1.970 ± 0.121 0.1447 ± 0.0132

0.0290 ± 0.0024 0.7369 ± 0.0638 0.620 ± 0.040 1.264 ± 0.092 5.179 ± 0.485 0.395 ± 0.045 1.977 ± 0.156 0.0931 ± 0.0189

0.0294 ± 0.0030 0.7565 ± 0.0713 0.691 ± 0.044* 1.343 ± 0.109 5.460 ± 0.450 0.367 ± 0.019 2.137 ± 0.151 0.1077 ± 0.0152

BW = body weight; BrW = brain weight; CAG = cycloastragenol. P < 0.05. a All data are presented as mean values ± standard deviations with N = 10 for main toxicity groups and N = 5 for recovery groups, except when otherwise indicated; b N = 9.

*

the trend; the values fell between those of the low- and intermediate-dose groups (Table 4).The observed changes, which did not correlate to any clinical or histopathologic findings, were not

considered to be toxicologically relevant. The changes could potentially be considered adaptive and fully reversible as borne out by the recovery phase; no absolute or relative organ weights

329

N.J. Szabo / Food and Chemical Toxicology 64 (2014) 322–334 Table 4 Terminal body weights and absolute and relative organ weights of female rats following 92 days of treatment with CAGa. Parameter

Main toxicity study – Day 93 b

Recovery study – Day 120

Group 1 0 mg/kg bw/d

Group 2 40 mg/kg bw/d

Group 3 80 mg/kg bw/d

Group 4 150 mg/kg bw/d

Group 5 0 mg/kg bw/d

Group 6 150 mg/kg bw/d

Absolute weights Terminal body weight (g) Adrenals (g) Brain (g) Heart (g) Kidneys (g) Liver (g) Ovaries (g) Spleen (g) Thymus (mg) Uterus-Oviduct (g)

240.5 ± 9.8 0.0744 ± 0.0088 1.902 ± 0.097 0.870 ± 0.051 1.658 ± 0.120 6.156 ± 0.396 0.0851 ± 0.0153 0.643 ± 0.089 0.2114 ± 0.0461 0.754 ± 0.237

245.1 ± 8.9 0.0652 ± 0.0092 1.871 ± 0.081 0.916 ± 0.051 1.607 ± 0.064 6.131 ± 0.282 0.0968 ± 0.0314 0.664 ± 0.087 0.2254 ± 0.0299 0.681 ± 0.162

241.7 ± 9.5 0.0698 ± 0.0101 1.838 ± 0.049 0.941 ± 0.032* 1.665 ± 0.107 6.208 ± 0.496 0.0796 ± 0.0248 0.624 ± 0.072 0.2281 ± 0.0368 0.814 ± 0.274

243.9 ± 15.6 0.0670 ± 0.0073 1.878 ± 0.062 0.939 ± 0.079* 1.640 ± 0.112 6.454 ± 0.660 0.0925 ± 0.0175 0.651 ± 0.081 0.2127 ± 0.0482 0.558 ± 0.150

243.0 ± 9.3 0.0644 ± 0.0079 1.828 ± 0.064 0.890 ± 0.047 1.566 ± 0.076 5.898 ± 0.308 0.0806 ± 0.0131 0.618 ± 0.036 0.1338 ± 0.0408 0.786 ± 0.288

241.0 ± 6.4 0.0608 ± 0.0029 1.840 ± 0.050 0.920 ± 0.076 1.606 ± 0.106 5.858 ± 0.434 0.0970 ± 0.0393 0.600 ± 0.070 0.1528 ± 0.0316 0.808 ± 0.334

Organ-to-body weight ratios Adrenals/BW Brain/BW Heart/BW Kidneys/BW Liver/BW Ovaries/BW Spleen/BW Thymus/BW Uterus-Oviduct/BW

0.3094 ± 0.0354 7.914 ± 0.379 3.621 ± 0.235 6.900 ± 0.496 25.590 ± 1.051 0.3543 ± 0.0638 2.667 ± 0.279 0.8771 ± 0.1803 3.152 ± 1.038

0.2668 ± 0.0430 7.642 ± 0.419 3.740 ± 0.213 6.561 ± 0.275 25.031 ± 1.212 0.3968 ± 0.1302 2.714 ± 0.376 0.9217 ± 0.1324 2.775 ± 0.633

0.2888 ± 0.0404 7.613 ± 0.301 3.896 ± 0.118** 6.899 ± 0.529 25.705 ± 2.117 0.3294 ± 0.0989 2.584 ± 0.293 0.9433 ± 0.1448 3.388 ± 1.210

0.2750 ± 0.0275 7.724 ± 0.506 3.849 ± 0.193* 6.728 ± 0.287 26.420 ± 1.424 0.3780 ± 0.0621 2.668 ± 0.290 0.8777 ± 0.2149 2.302 ± 0.650

0.2655 ± 0.0345 7.529 ± 0.325 3.668 ± 0.250 6.443 ± 0.130 24.265 ± 0.598 0.3305 ± 0.0417 2.542 ± 0.078 0.5521 ± 0.1753 3.226 ± 1.152

0.2525 ± 0.0162 7.639 ± 0.283 3.816 ± 0.270 6.669 ± 0.498 24.290 ± 1.329 0.4018 ± 0.1591 2.487 ± 0.250 0.6366 ± 0.1443 3.367 ± 1.453

Organ-to-brain weight ratios Adrenals/BrW Heart/BrW Kidneys/BrW Liver/BrW Ovaries/BrW Spleen/BrW Thymus/BrW Uterus-Oviduct/BrW

0.0393 ± 0.0055 0.458 ± 0.031 0.872 ± 0.058 3.238 ± 0.157 0.0450 ± 0.0092 0.337 ± 0.035 0.1116 ± 0.0257 0.399 ± 0.131

0.0349 ± 0.0052 0.491 ± 0.038 0.861 ± 0.062 3.284 ± 0.223 0.0517 ± 0.0165 0.355 ± 0.047 0.1204 ± 0.0138 0.364 ± 0.085

0.0380 ± 0.0058 0.512 ± 0.023** 0.907 ± 0.070 3.380 ± 0.293 0.0433 ± 0.0132 0.340 ± 0.046 0.1245 ± 0.0220 0.443 ± 0.151

0.0357 ± 0.0043 0.500 ± 0.041* 0.874 ± 0.057 3.436 ± 0.314 0.0493 ± 0.0094 0.348 ± 0.049 0.1134 ± 0.0261 0.298 ± 0.083

0.0353 ± 0.0051 0.487 ± 0.030 0.857 ± 0.029 3.228 ± 0.161 0.0440 ± 0.0065 0.338 ± 0.017 0.0728 ± 0.0205 0.429 ± 0.159

0.0330 ± 0.0009 0.500 ± 0.035 0.872 ± 0.040 3.183 ± 0.201 0.0527 ± 0.0211 0.326 ± 0.032 0.0833 ± 0.0183 0.438 ± 0.177

BW = body weight; BrW = brain weight; CAG = cycloastragenol. P < 0.05. ** P < 0.01. a All data are presented as mean values ± standard deviations with N = 10 for main toxicity groups and N = 5 for recovery groups, except when otherwise indicated. b N = 9. *

for Group 6 females in the recovery study, including those related to the heart, showed any effect. Macroscopic findings in main toxicity males (Day 92) included small, bilateral testes, which histopathology later confirmed as due to slight germ cell atrophy, and a small, bilateral epididymis that was later determined to correspond to slight hypospermia in one Group 1 control male; a soft mass in the left epididymis due to the presence of a sperm granuloma in one Group 2 male; and a discolored liver with a supernumerary lobe in one Group 4 male that corresponded to chronic liver infarction consistent with liver lobe torsion. Each of these findings was incidental and not treatment-related. No macroscopic observations were reported in recovery males (Day 120). Incidental macroscopic findings in females included fluid-filled uteri/oviducts consistent with variations in the estrous cycle; an endometrial cyst in one Group 1 female; and multiple mass-like lesions embedded in fat and connective tissue in close proximity to the thymus and lungs in one control female. Microscopic evaluation of the latter finding was not confirmed, but was toxicologically insignificant, as it occurred in a control animal. Microscopic findings in control and high dose animals of the main study consisted of only incidental findings (i.e., minimal chronic progressive neuropathy commonly observed in SD rats; trauma to the esophageal walls consistent with daily gavage procedures; a mesenchymal tumor in one kidney of one Group 4 female). No adverse macroscopic or microscopic observations related to the test substance were noted for the heart. Under the conditions of this study, the no-observed-adverse-effect level (NOAEL) for orally administered CAG in the rat was 150 mg/kg bw/day (equivalent to 10,500 mg/day in a 70-kg individual), the highest dose tested.

3.3. Bacterial reverse mutation assay The assay was considered valid because the means of the revertant colonies on the negative control groups remained within the historical data ranges for negative controls with or without metabolic activation, background growth in the negative control and test groups was comparable, and revertant colony counts in the positive control groups were greater than three times the counts in the corresponding negative control groups. In the initial toxicity-mutation assay, no cytotoxicity14 and no positive mutagenic response was observed at concentrations of CAG of up to 5000 lg/plate in the presence or absence of S9 metabolic activation (data not shown). In the confirmatory mutagenicity assay, dose-dependent increases in the number of revertant colonies of greater than two times (TA98, TA100, WP2 uvrA) or three times (TA1535, TA1537) the negative control values were not observed in any strain treated with CAG regardless of whether or not S9 mix was present (data not shown). Precipitation at CAG concentrations of 1500 lg/plate and higher was noted in all tester strains in the presence and absence of S9 mix in assays of both the initial and secondary phases, but no positive mutagenic response and no cytotoxicity were observed. Under the conditions of this assay, the CAG test material was not a mutagen. 3.4. In vitro chromosome aberration assay In accordance with the guidelines and as permitted by the solubility of the test substance, concentrations from 0.016 mM 14

The background lawn was unaffected.

330

N.J. Szabo / Food and Chemical Toxicology 64 (2014) 322–334

Table 5 Summary of results for in vitro chromosome aberration assay in Chinese hamster V79 cells including the number of metaphases examined, the total and percentage of aberrant cells, the mitotic index, and cell density.

a b

Dose group

Metaphases

Aberrant cells

Experiment I – with S9 Negative control Solvent control 0.05 mM 1.25 mMb 1.50 mMb 1.75 mMb 2.00 mMb CPA, 0.83 lg/ml

200 200 200 200 400 200 200 200

Experiment I – without S9 Negative control Solvent control 0.05 mM 0.10 mMb 0.125 mMb EMS, 600 lg/ml

a

(Total/Mean%)

Mitotic index (% rel. index)

Cell density (% rel. density)

3/1.5 3/1.5 5/2.5 3/1.5 21/5.3 4/2.0 2/1.0 17/8.5

102 100 105 75 53 54 39 72

99 100 97 78 30 24 20 96

200 200 200 200 200 200

5/2.5 5/2.5 1/0.5 4/2.0 5/2.5 16/8.0

96 100 91 80 47 95

105 100 99 53 47 90

Experiment II – with S9 Negative control Solvent control 0.4 mMb 1.0 mMb 1.6 mMb 2.1 mMb CPA, 0.83 lg/ml

200 200 200 200 400 400 200

4/2.0 5/2.5 2/1.0 4/2.0 7/1.8 9/2.3 23/11.5

97 100 72 63 63 44 78

100 100 97 86 62 30 100

Experiment II – without S9 Negative control Solvent control 0.065 mM 0.070 mM 0.075 mM EMS, 400 lg/ml

200 200 200 200 200 200

2/1.0 6/3.0 3/1.5 1/0.5 3/1.5 16/8.0

106 100 86 82 45 99

102 100 84 81 51 93

Range of historical laboratory control data: 0.0–4.0% (with and without metabolic activation). Precipitation observed.

up to 10 mM were tested in the presence and absence of metabolic activation in the pre-experiment. Precipitation was observed at concentrations of 0.125 mM and higher without metabolic activation and at concentrations of 0.25 mM and higher with activation. Toxicity, as evaluated by a biologically relevant decrease of the relative (rel.) mitotic index (<70%) and cell density (<50%) was observed at concentrations of 0.125 mM (63% rel. mitotic index) and 0.25 mM (12% rel. cell density), respectively, and higher without metabolic activation and at concentrations of 2.5 mM (24% rel. mitotic index; 11% rel. cell density) and higher with activation. In Experiment I, V79 cells were exposed to selected concentrations of CAG for 4 h with metabolic activation (0.05–2.50 mM) and without (0.025–0.250 mM). In the presence of activation, 0.05, 1.25, 1.50, 1.75, and 2.00 mM were selected for microscopic evaluation; in the absence of activation, the concentrations 0.05, 0.10, and 0.125 mM concentrations were selected. The findings for Experiment I are summarized in Table 5. Precipitation of the test substance was noted with metabolic activation at concentrations of 1.25 mM and higher and without metabolic activation at concentrations of 0.10 mM and higher. Toxicity was observed at concentrations of 1.50 mM (53% rel. mitotic index; 30% rel. cell density) and higher with activation and at concentrations of 0.125 mM (47% rel. mitotic index) and 0.10 mM (53% rel. cell density) and higher without metabolic activation. Clastogenicity as indicated by the aberration rates of the cells15 remained within

15 A total of 200 metaphases were evaluated (100 per duplicate) for all controls and most dose groups. A total of 400 metaphases (200 per duplicate) were scored for all dose groups that gave inconsistent results, Experiment I test concentration 1.50 mM and Experiment II test concentrations 1.6 and 2.1 mM (all with metabolic activation). All aberration rates were normalized to 100 metaphases.

the historical control data range (0.0–4.0%) for the negative and solvent controls and all dose groups in the absence of metabolic activation. With activation, the aberration rates for the negative and solvent controls and for all but one test group, the 1.50 mM group (5.3% from a total of 400 scored metaphases), were within the historical control data range. The inconsistency observed in the 1.50 mM test group may be an effect of the limited solubility of the test substance. No dose–response relationship was observed, as the aberration rates for the 1.75 and 2.00 mM dose groups both fell within the range of the historical control data. In Experiment II, V79 cells were exposed to selected concentrations of CAG for 4 h with metabolic activation (0.026–2.1 mM) and for 20 h without activation (0.01–0.100 mM). In the presence of activation, 0.4, 1.0, 1.6, and 2.1 mM concentrations were selected for microscopic evaluation; in the absence of activation, the concentrations 0.065, 0.070, and 0.075 mM were selected. Precipitation of the test substance was observed at all microscopically evaluated concentrations with metabolic activation, but was not present at any evaluated concentration without metabolic activation. The findings for Experiment II are summarized in Table 5. Toxicity was observed at concentrations of 1.0 mM (63% rel. mitotic index) and 1.6 mM (62% rel. cell density) and higher with activation and at 0.075 mM (45% rel. mitotic index; 51% rel. cell density), the highest concentration evaluated microscopically without metabolic activation. Clastogenicity remained within the historical control data range for the negative and solvent controls and all dose groups in the absence and in the presence of metabolic activation. Due to inconsistency in the results obtained for the 1.6 and 2.1 mM groups, 400 metaphases were scored for each of these concentrations. At both concentrations, one out of the four slides evaluated showed an aberration rate (5% for both) above the

N.J. Szabo / Food and Chemical Toxicology 64 (2014) 322–334

historical negative control data range. Because the remaining three slides were in the negative control range, the reported means were within the historical control data range. As in Experiment I, no dose–response relationship was observed. CAG did not exhibit clastogenicity under the conditions of the in vitro chromosome aberration assay in the V79 Chinese hamster cell line in either Experiment I or II in the absence of metabolic activation. With metabolic activation, inconsistent results, likely related to the poor solubility properties of the test substance, were observed in both experiments and led to a final increased aberration rate in one intermediate test group (1.50 mM) in Experiment I. Because the clastogenic effect was relatively moderate and in a concentration range where precipitation occurred and a dose–response relationship was not observed, the results of the in vitro chromosome aberration assay were determined to be equivocal. The assay was considered valid because the frequency of aberrant cells induced in the positive controls was biologically significant while aberrant cell rates in the negative and solvent controls remained within the historical negative control data range in the presence and absence of S9 mix. 3.5. In vivo erythrocyte micronucleus assay In response to the administration of CAG by i.p., moderate and strong toxicities were observed in mice of the low-dose (400 mg/ kg bw) and mid-dose (1000 mg/kg bw) groups, respectively. Mice in the high-dose (2000 mg/kg bw) group, the MTD, exhibited symptoms of systemic toxicity (i.e., spontaneous activity, constricted abdomen, bradykinesia, recumbency, ataxia and halfclosed/closed eyes). Although all mean rel. PCE values in peripheral blood collected from the three dose groups and negative control group at 44 h were within the historical data ranges for the negative controls (Table 6), the rel. PCE values for the low- and mid-dose females at 44 h were significantly decreased when compared with the female negative control (P < 0.05). At 68 h, the values for the high-dose groups were significantly decreased when compared with the corresponding negative controls (P < 0.05). The decreased rel. PCE values in the test groups compared with the controls confirmed that the target cells had been exposed to the test substance. The mean micronucleus frequency values for all dose groups and times (44 h and 68 h) were within the ranges of the negative control and historical negative control values (Table 6). Although no statistically significant increase of micronu-

Table 6 Mean rel. PCE values and incidence of micronucleus induction in peripheral erythrocytes. Study groups

Mean rel. PCEa

Micronucleus induction (%)a

Male

Female

Male

Female

44 h Negative control Positive control 400 mg/kg bw 1000 mg/kg bw 2000 mg/kg bw

3.18 2.30 3.06 2.68 2.56

3.77 2.12* 2.10* 2.54* 2.92

0.18 1.75* 0.20 0.19 0.19

0.22 1.34* 0.17 0.12 0.14

68 h Negative control 2000 mg/kg bw

3.28 0.96*

3.89 1.03*

0.24 0.16*

0.20 0.13

1.19–3.85 0.30–2.83

0.08–0.43 0.93–3.76

0.08–0.34 0.68–2.84

Historical (2007–2010) Negative controlb 1.43–3.97 Positive controlc 0.30–2.21

rel. PCE = relative polychromatic erythrocytes. P < 0.05, compared to the corresponding negative control. a n = 5, unless otherwise indicated. b n = 78. c n = 72.

*

331

cleus incidence occurred in the dose groups, a significant increase in micronucleus frequency was induced in the positive controls (P < 0.05). Under the conditions of this study, CAG was not clastogenic and/or aneugenic; CAG did not induce structural and/or numerical chromosomal damage in the immature erythrocytes of the mouse. The assay was considered valid because mouse weights did not exceed ±20% of the mean weights of each sex, the incidence of micronucleated cells for the negative control vehicle remained in the historical range for negative laboratory controls, and the micronucleus frequency in the CPA positive control group was induced to a statistically significant degree (P < 0.01 for both sexes).

4. Discussion In the 13-week subchronic toxicity study daily ingestion of CAG at the highest dose administered provided150 mg/kg bw/d. This exposure was well-tolerated by the rats with no resulting toxicities or other adverse effects. Although the highest dose studied in rats is equivalent to 10,500 mg/d in a 70-kg individual, a reasonable dietary level in humans would utilize a safety factor of at least ‘100’ which would decrease exposure to at most 1.5 mg/kg bw/d or 105 mg/d in a 70-kg person, a level of exposure to CAG comparable with possible, albeit not usual, exposure from Radix Astragali in TCMs. Predictably, natural concentrations of CAG and ASTs in TCM formulations vary with the roots used. Root concentrations, in turn, are dependent on source species, environmental factors during culture, and age and season at harvest (Ma et al., 2002; Shibata et al., 1996). An example of a high endogenous concentration was estimated from A. membranaceus roots (1 kg) obtained from a Chinese drugstore in Hangzhou (Yang et al., 2005). After being powdered and extracted in 70% ethanol, 10.96 g of saponin extract were yielded. When analyzed, 83.6% of the extract (equivalent to 9.16 mg/g dry wt in the root) was composed of identifiable ASTs, nearly all of which contained CAG as the aglycone. The CAG content in the root material was estimated to be in the range of 5.34–5.73 mg/g dry wt. Based on the official accepted daily dose of 9–30 g Astragalus root in China (McKenna et al., 2002), dietary exposure to CAG from this source could potentially range from 48–171 mg/d, a range within which ingestion at 105 mg/d falls. When more generally available Radix Astragali products in the US and China are considered, estimated CAG exposures lessen: 0.078–0.115 mg/g dried root product in 5 commercial A. membranaceus root products in the US (Ganzera et al., 2001) and 0.076– 1.475 mg/g dried root product in 44 Radix Astragali dried root products (22 purchased from New York City and 22 from Hong Kong) (Xiao et al., 2011) equate to CAG exposures ranging from 0.68–44.25 mg/day. In addition to preparation-related effects, factors such as species, soil conditions during growth and season and age of the plants at harvest contribute to variability. When A. membranaceus Bge. was grown in four different soil conditions in a single year, AST content ranged from 1.51 to 3.89 mg/g dry wt in the roots, giving an estimated CAG content of 0.881–2.43 mg/g dry wt in the roots of the different soil types (Shibata et al., 1996). Seasonal variations in CAG content in roots from 2 to 3 yr-old plants of two Radix Astragali species (n = 5 of each) cultured in consistent soils were estimated to range from 0.6 mg/g in May to 1.1 mg/g in September, with roots harvested in September to October having the greater content (Ma et al., 2002). Differences in relatively young plant ages were also shown to have a moderate effect on concentration; fall-harvested roots from plants known to be 1 year old and 4 years old (the oldest age tested) gave estimated CAG concentrations of 1.2 mg/g and 1.6 mg/g, respectively (Ma et al., 2002). Taken together, the examined factors equate to reasonably expected exposures ranging from 5.4 to 72.9 mg/d CAG for a person ingesting Astragalus root in accordance with established dosing

332

N.J. Szabo / Food and Chemical Toxicology 64 (2014) 322–334

limits. These ranges and the traditional safety accorded to them are in agreement with a published observational study, wherein 114 human volunteers consumed a dietary supplement containing up to 50 mg TA-65Ò, a CAG-containing preparation from Chinese Astragalus, on a daily basis for periods of up to five years or more. Ingestion of TA-65Ò was found to be well-tolerated and without adverse effect among the volunteers (Harley et al., 2011, 2013). Although CAG concentrations have been infrequently monitored, dietary exposure to CAG is not novel to the human experience; customary exposures from TCMs are estimated to range from about 0.70 mg/d up to 70 mg/d without negative effect. Cardiotonic effects from Astragalus root in TCM are primarily associated with A. membranaceus (Fisch.) Bge. var. mongholicus (Bge.) Hsiao and A. membranaceus (Fisch.) Bge., the two authentic Radix Astragali species, even though other unspecified members in the genus have also reportedly produced the same or similar effects (WHO, 1999; Ma et al., 2002; McKenna et al., 2002; Yu et al., 2007b; Sevimli-Gür et al., 2011). A. membranaceus root extracts have shown cardioprotective activities in in vitro systems such as rat heart mitochondria (Hong et al., 1994), cardiotonic effects in rats following induced congestive heart failure (Ma et al., 2002), and hypotensive activity (attributed to vasodilation) in dogs (Chang and But, 1987; WHO, 1999; McKenna et al., 2002), cats and rats (Zhang et al., 1984). Cardiovascular activity was also reported in isolated frog and toad hearts when Radix Astragali was added as an alcohol extract (enhanced contraction amplitude and contractility), in dogs after i.p. injection (prolonged S-T intervals and biphasic T waves), and in rabbits, dogs and cats after IV administration (hypotension) (Chang and But, 1987; WHO, 1999; McKenna et al., 2002). When only the saponin fraction was tested, isolated rat hearts still experienced a positive inotropic effect and the resting potentials of cultured rat myocardial cells decreased (Wang, 1992; WHO, 1999). The cardiovascular effects induced by Radix Astragali saponins have been demonstrated to be due to AST IV, which expresses activity in a fashion similar to the cardiotonic glycosides (Li and Cao, 2002; Verotta and El-Sebakhy, 2001). In the scientific literature, CAG is not recognized to exert cardiovascular effects. Indeed, daily exposure to 150 mg/kg bw CAG in the 90-day subchronic toxicity study did not induce cardiovascular effects. Neither the blood pressure measures nor the serum enzymes revealed any statistically significant differences between the dose groups and their respective controls in the main toxicity or recovery studies. As a further indication that CAG did not affect blood pressure, measured values for rats in Groups 5 and 6 in the recovery study (W eek 16) and Groups 1 and 4 in the main toxicity study (Week 12) were comparable by sex. Enzymes AST, CK and LDH were monitored because all are present in myocardial tissue in sufficient quantities that even minor localized damage would result in markedly increased circulating activities. CAG consumption exerted no identifiable effect on the serum activities of these enzymes. In additional support of the negative findings of the cardio-component, no adverse macroscopic or microscopic observations noted for the heart were related to test substance administration. Although in female rats of Groups 3 and 4, absolute and relative heart weights were statistically increased compared with those of the control group, the changes which may have been adaptive were not toxicologically relevant. The changes lacked any kind of clinical or histopathologic correlate, and the Group 4 increases were each intermediate in value to Groups 2 and 3. Further, no similar observations were made in recovery study females. In addition, findings for the FOB conducted during the in-life phase of the main toxicity study were normal and unaffected by CAG administration; neither erratic nor subdued behaviors were reported. Taken all together, under the conditions of this study, CAG was not found to induce cardiac effects.

In the course of isolating and identifying the primary bioactive saponins in Astragalus spp., CAG, AST IV, and in CAG-containing preparations such as TA-65Ò have been identified as small-molecule telomerase activators that can induce elongation of very short telomeres (Fauce et al., 2008; Harley et al., 2011;de Jesus et al., 2011; Yang et al., 2012; Yung et al., 2012; Zhou et al., 2012). As suggested by studies in mice, the increase in telomere length appears to occur without increasing global cancer incidence (de Jesus et al., 2011). Telomeres which are the protective caps consisting of repeated ‘TTAGGG’ units located at the ends of chromosomes, are necessary for proper cell division. Telomeres gradually shorten with each replication of a cell’s DNA due to incomplete replication of the telomeres (de Jesus et al., 2011; Harley et al., 2011; Mulambalah et al., 2012; Yung et al., 2012). Each year a human typically loses 15–60 bp of telomeric DNA, a modest rate of loss (Harley et al., 2011). When telomeres eventually become too short to support DNA replication, cellular division stops and cells enter replicative senescence. Shortened telomeres and cellular senescence are linked with the natural aging process and also the premature aging that accompanies certain genetic conditions (i.e., Down’s Syndrome, aplastic anemia, dyskeratosis congenital, idiopathic pulmonary fibrosis), chronic disease states (i.e., cytomegalovirus, human immunodeficiency virus), and chronic stress (de Jesus et al., 2011; Harley et al., 2011; Mulambalah et al., 2012). Regeneration of telomeres requires activation of telomerase, an enzyme present in relatively high levels in embryonic and fetal cells and certain stem cells, but in extremely low levels in most somatic adult cells. Telomerase elongates DNA by preferentially adding new ‘TTAGGG’ repeat units onto the ends of the shortest existing telomeres (de Jesus et al., 2011). Telomerase activators encourage the reactivation of telomerase which slows telomere shortening and may promote a younger cell phenotype or expression profile as suggested by studies in human lymphocytes (Fauce et al., 2008), mice (de Jesus et al., 2011) and humans (Harley et al., 2011, 2013). Daily ingestion of CAG at up to 150 mg/kg bw/day via oral gavage was well-tolerated by the rats in the 13-week subchronic toxicity study with 4-week recovery. No biologically relevant effects attributable to the administration of CAG were identified from the in-life observations, ophthalmology, urinalysis, hematology, clinical chemistry, organ weights, gross pathology, or histopathology in main toxicity or recovery group animals. Although statistical significance was shown for several parameters, each change was only of sporadic incidence; did not demonstrate a dose-dependent relationship; was also observed in the control group; was not correlated to other clinical and/or histopathologic change; lacked toxicological relevance; and/or was within the historically observed ranges in the age and strain of rat used in the study (Pettersen et al., 1996; Derelanko, 2000). In addition, no adverse cardiac-related effects as monitored by changes in blood pressure, selected cardiac enzymes, and gross and histopathologic examination were identified in either sex of any dose group in the main toxicity or recovery phases. The three genotoxicity assays provided strong overall support that CAG lacks mutagenic and/or clastogenic potential. CAG did not exhibit mutagenicity under the conditions of the bacterial reverse mutation assay in S. typhimurium or E. coli tester strains with or without metabolic activation at levels up to 5000 lg/plate. CAG did not exhibit clastogenicity under the conditions of the in vitro chromosome aberration assay in both experiments in the absence of metabolic activation; in the presence of metabolic activation, inconsistent results, likely related to the nature of the test substance and its relatively poor solubility properties, were observed in both experiments that led to a final increased aberration rate in one mid-range test group (1.50 mM) in one experiment. Because the effect was relatively moderate and in a concentration range

N.J. Szabo / Food and Chemical Toxicology 64 (2014) 322–334

where precipitation occurred and no dose-dependence was observed, the final conclusion is that the in vitro chromosome aberration assay was equivocal. CAG did not exhibit clastogenicity or aneugenicity under the conditions of the in vivo erythrocyte micronucleus assay. The mean values of aberrant cells in all 44- and 68-h negative control and test groups remained within the historical data range for the negative control. Similarly, mean mitotic index values remained in the ranges of their respective negative controls for the 44-h male and female test groups and 68-h female test group. In the 68-h male test group the mean mitotic index was significantly lower than the corresponding negative control, but this finding was determined to be due to biological variability among the animals and was therefore not judged to be biologically relevant. No increase in micronucleation frequency above historical negative control ranges was induced by treatment with CAG. In conclusion, as demonstrated by the findings of the 13-week subchronic repeated oral dose study in rats with a 4-week recovery phase and three genotoxicity assays, the daily administration of CAG was well-tolerated in the rat and did not induce any toxic or genotoxic effect. The oral no-observed-adverse-effect level NOAEL of CAG was 150 mg/kg bw/day, the highest dose tested in male and female rats. Conflict of Interest The author has a financial relationship with the sponsor of the studies and manuscript, Telomerase Activation Sciences, New York City, New York. Acknowledgments The 13-week repeated dose study with recovery phase for CAG in rats was conducted at Product Safety Labs in Dayton, New Jersey, USA. The mutagenic assay testing CAG was conducted at BioReliance in Rockville, Maryland. The clastogenic assays testing CAG were conducted at BSL Bioservice in Planegg, Germany. References Anderson, D.M.W., 1989. Evidence for the safety of gum tragacanth (Asiatic Astragalus spp.) and modern criteria for the evaluation of food additives. Food Addit. Contam. 6, 1–12. Bedir, E., Çalis, I., Aquino, R., Piacente, S., Pizza, C., 1998a. Cycloartane triterpene glycosides from the roots of Astragalus brachypterus and Astragalus microcephalus. J. Nat. Prod. 61, 1469–1472. Bedir, E., Çalis, I., Zerbe, O., Sticher, O., 1998b. Cyclocephaloside I: A novel cycloartane-type glycoside from Astragalus microcephalus. J. Nat. Prod. 61, 503–505. Bedir, E., Çalis, I., Aquino, R., Piacente, S., Pizza, C., 1999a. Secondary metabolites from the roots of Astragalus trojanus. J. Nat. Prod. 62, 563–568. Bedir, E., Çalis, I., Aquino, R., Piacente, S., Pizza, C., 1999b. Trojanoside H: A cycloartane-type glycoside from the aerial parts of Astragalus trojanus. Phytochemistry 51, 1017–1020. Bedir, E., Çalis, I., Dunbar, C., Sharan, R., Buolamwini, J.K., Khan, I.A., 2001a. Two novel cycloartane-type triterpene glycosides from the roots of Astragalus prusianus. Tetrahedron 57, 5961–5966. Bedir, E., Tatli, I.I., Çalis, I., Khan, I.A., 2001b. Trojanosides I-K: New cycloartane-type glycosides from the aerial parts of Astragalus trojanus. Chem. Pharm. Bull. 49, 1482–1486. Castillo, C., Valencia, I., Reyes, G., Hong, E., 1993. 3-Nitropropionic acid, obtained from Astragalus species, has vasodilator and antihypertensive properties. Drug Dev. Res. 28, 183–188. Chang, H.M., But, P.P.H. (Eds.), 1987. Pharmacology and Applications of Chinese Materia Medica, vol. 2. World Scientific Publishing, Singapore. de Jesus, B.B., Schneeberger, K., Vera, E., Tejera, A., Harley, C.B., Blasco, M.A., 2011. The telomerase activator TA-65 elongates short telomeres and increases health span of adult/old mice without increasing cancer incidence. Aging Cell 10, 604– 621. Derelanko, M.J., 2000. Clinical pathology. In: Toxicologist’s Pocket Handbook. CRC Press, New York, pp. 97–122. Fauce, S.R., Jamieson, B.D., Chin, A.C., Mitsuyasu, R.T., Parish, S.T., Ng, H.L., Ramirez Kitchen, C.M., Yang, O.O., Harley, C.B., Effros, R.B., 2008. Telomerase-based pharmacologic enhancement of antiviral function of human CD8+ T lymphocytes. J. Immunol. 181, 7400–7406.

333

Ganzera, M., Bedir, E., Çalis, I., Khan, I.A., 2001. Separation of Astragalus saponins by reversed phase high performance liquid chromatography and evaporative light scattering detection. Chromatographia 53, 131–134. Gregory, G., 1819. Gum. A New and Complete Dictionary of Arts and Sciences: Including the Latest Improvement and Discovery and the Present State of Every Branch of Human Knowledge, vol. 2. Collins and Co., London. Harley, C.B., Liu, W., Blasco, M., Vera, E., Andrews, W.H., Briggs, L.A., Raffaele, J.M., 2011. A natural product telomerase activator as part of a health maintenance program. Rejuvenation Res. 14, 45–56. Harley, C.B., Liu, W., Flom, P.L., Raffaele, J.M., 2013. A natural product telomerase activator as part of a health maintenance program: metabolic and cardiovascular response. Rejuvenation Res. 16, 386–395. Hong, C.Y., Lo, Y.C., Tan, F.C., Wei, Y.H., Chen, C.F., 1994. Astragalus membranaceus and Polygonum multiflorum protect rat heart mitochondria against lipid peroxidation. Am. J. Chin. Med. 22, 63–70. Kitagawa, I., Wang, H.K., Takagi, A., Fuchida, M., Miura, I., Yoshikawa, M., 1983. Saponin and Sapogenol. XXXIV. Chemical constituents of Astragali Radix, the root of Astragalus membranaceus Bunge. (1). Cycloastragenol, the 9,19cyclolanostane-type aglycone of astragalosides, and the artifact aglycone astragenol. Chem. Pharm. Bull. 31, 689–697. Li, Z.P., Cao, Q., 2002. Effects of astragaloside IV on myocardial calcium transport and cardiac function in ischemic rats. Acta Pharmacol. Sin. 23, 898–904. Ma, X.Q., Shi, Q., Duan, J.A., Dong, T.T.X., Tsim, K.W.K., 2002. Chemical analysis of Radix Astragali (Huangqi) in China: a comparison with its adulterants and seasonal variations. J. Agric. Food Chem. 50, 4861–4866. McKenna, D.J., Jones, K., Hughes, K., Humphrey, S., 2002. Astragalus. In: Botanical Medicines. The Desk Reference for Major Herbal Supplements, second ed. The Haworth Herbal Press, New York, pp. 1–17. Movafeghi, A., Djozan, Dj., Razeghi, J.A., Baheri, T., 2010. Identification of volatile organic compounds in leaves, roots and gum of Astragalus compactus Lam. using solid phase microextraction followed by GC–MS analysis. Nat. Prod. Res. 24, 703–709. Mulambalah, C.S., Siamba, D.N., Vulule, J.M., 2012. The molecular and medical aspects of the aging process. Res. J. Med. Sci. 6, 117–123. National Research Council of the National Academies, 2011. Guide for the Care and Use of Laboratory Animals. Institute of Laboratory Animal Research, Division of Earth and Life Studies. National Academy Press, Washington, DC. Pettersen, J.C., Morrissey, R.L., Saunders, D.R., Pavkov, K.L., Luempert III, L.G., Turnier, J.C., Matheson, D.W., Schwartz, D.R., 1996. A 2-year comparison study of Crl:CD BR and Hsd: Sprague–Dawley SD rats. Fund. Appl. Toxicol. 33, 196–211. Ralphs, M.H., Creamer, R., Baucom, D., Gardner, D.R., Welsh, S.L., Graham, J.D., Hart, C., Cook, D., Stegelmeier, B.L., 2008. Relationship between the endophyte Embellisia spp. and the toxic alkaloid swainsonine in major locoweed species (Astragalus and Oxytropis). J. Chem. Ecol. 34, 32–38. Rios, J.L., Waterman, P.G., 1997. A review of the pharmacology and toxicology of Astragalus. Phytother. Res. 11, 411–418. _ Atilla, P., Genç, R., Çakar, N., Delilog˘lu-Gürhan, I., _ Bedir, Sevimli-Gür, C., Onbasßilar, I., E., 2011. In vitro growth stimulatory and in vivo wound healing studies on cycloartane-type saponins of Astragalus genus. J. Ethnopharmacol. 134, 844– 850. Shibata, T., Sakai, E., Nishi, K., Anetai, M., 1996. Growth and glycoside contents of Astragalus membranaceus Bunge (Leguminosae) cultivated in different soil groups. Nat. Med. 50, 296–299. Sors, T.G., Martin, C.P., Salt, D.E., 2009. Characterization of selenocysteine methyltransferases from Astragalus species with contrasting selenium accumulation capacity. Plant. J. 59, 110–122. Verotta, L., El-Sebakhy, N.A., 2001. Cycloartane and oleanane saponins from Astragalus sp.. Stud. Nat. Prod. Chem. 25, 179–234. Vogel, H.J., Bonner, D.M., 1956. Acetylornithinase of Escherichia coli: partial purification and some properties. J. Biol. Chem. 218, 97–106. Wang, Q., 1992. Inotropic action of Astragalus membranaceus Bge. saponins and its possible mechanism. China J. Chin. Mater. Med. 17, 557–559. WHO (World Health Organization), 1999. Radix Astragali. In: WHO monographs on selected medicinal plants, vol. 1. World Health Organization, Geneva, Switzerland. pp. 50–58. Xiao, W.L., Motley, T.J., Unachukwu, U.J., Lau, C.B.S., Jiang, B., Hong, F., Leung, P.C., Wang, Q.F., Livingston, P.O., Cassileth, B.R., Kennelly, E.J., 2011. Chemical and genetic assessment of variability in commercial Radix Astragali (Astragalus spp.) by ion trap LC–MS and nuclear ribosomal DNA barcoding sequence analyses. J. Agric. Food Chem. 59, 1548–1556. Yang, Z.G., Sun, H.X., Fang, W.H., 2005. Haemolytic activities and adjuvant effect of Astragalus membranaceus saponins (AMS) on the immune responses to ovalbumin in mice. Vaccine 23, 5196–5203. Yang, W.Z., Ye, M., Huang, F.X., He, W.N., Guoa, D.A., 2012. Biocatalysis of cycloastragenol by filamentous fungi to produce unexpected triterpenes. Adv. Synth. Catal. 354, 527–539. Yu, S.Y., Yang, H.T.O., Yang, J.Y., Huang, X.L., Yang, T., Duan, J.P., Cheng, J.P., Chen, Y.X., Yang, Y.J., Qiong, P., 2007a. Subchronic toxicity studies of Radix Astragali extract in rats and dogs. J. Ethnopharmacol. 110, 352–355. Yu, Q.T., Qi, L.W., Li, P., Yi, L., Zhao, J., Bi, Z., 2007b. Determination of seventeen main flavonoids and saponins in the medicinal plant Huang-qi (Radix-Astragali) by HPLC–DAD–ELSD. J. Sep. Sci. 30, 1292–1299. Yung, L.Y., Lam, W.S., Ho, M.K.C., Hu, Y., Ip, F.C.F., Pang, H., Chin, A.C., Harley, C.B., Ip, N.Y., Wong, Y.H., 2012. Astragaloside IV and cycloastragenol stimulate the phosphorylation of estracellular signal-regulated protein kinase in multiple cell types. Planta Med. 78, 115–121.

334

N.J. Szabo / Food and Chemical Toxicology 64 (2014) 322–334

Zhang, Y.D., Wang, Y.L., Shen, J.P., Li, D.X., 1984. Effects on blood pressure and inflammation of Astragalus saponin 1, a principle isolated from Astragalus membranaceus Bge. Acta Pharm. Sin. 19, 333–337 (as cited in McKenna et al., 2002).

Zhou, R.N., Song, Y.L., Ruan, J.Q., Wang, Y.T., Yan, R., 2012. Pharmacokinetic evidence on the contribution of intestinal bacterial conversion to beneficial effects of astragaloside IV, a marker compound of Astragali Radix, in traditional oral use of the herb. Drug Metab. Pharmacokinet. 27, 586–597.