Food and Chemical Toxicology 90 (2016) 95e101

Contents lists available at ScienceDirect

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

A developmental toxicity study of 3S, 30 S-Astaxanthin in New Zealand white rabbits Steffen Schneider a, Werner Mellert a, Stefan Schulte b, Bennard van Ravenzwaay a, * a b

BASF SE, Experimental Toxicology and Ecology, 67056, Ludwigshafen, Germany BASF SE, Product Safety, 67056, Ludwigshafen, Germany

a r t i c l e i n f o

a b s t r a c t

Article history: Received 18 February 2015 Received in revised form 21 January 2016 Accepted 2 February 2016 Available online xxx

Astaxanthin, a naturally occurring pigment used to give the characteristic orange-pink colour to salmonid fish reared in aquaculture, is also marketed as a dietary supplement. Synthetic 3S, 30 S-Astaxanthin was tested for potential harmful effects on the in utero development of New Zealand white rabbits in a study according to international regulatory guidelines. There were two control groups, one being a placebo administration and three dose levels corresponding to 100, 200, and 400 mg of 3S, 30 S-Astaxanthin per kg body weight/day. The group sizes varied from 23 to 27 litters, providing approximately 200 fetuses per group for evaluation of developmental toxicity. There were no significant effects on the health of the does, nor on the size and viability of the litters. Malformations, both external and internal, were rare and occurred in all groups, including controls with no indication of a treatment relationship. Variations were much more common, being found in all litters. However, when examined by type and frequency, no pattern emerged indicating a relationship to administration of the test substance. It is concluded that administration of 3S, 30 S-Astaxanthin in a gelatin/carbohydrate powder formulation throughout pregnancy up to 400 mg/kg body weight/day is without harmful effects on reproduction or fetal development. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Developmental toxicity Astaxanthin Rabbit

1. Introduction Astaxanthins are carotenoids found naturally in algae, fish and crustaceans and are associated with the attractive pink colour of wild salmon flesh. This colour can only be obtained in aquaculture by using natural organisms containing these pigments (e.g. shrimp waste) or by mixing Astaxanthin with the diet. This is obtained either from microalgae (Haematococcus pluvialis) (Lorenz and Cysewski, 2000) or from fully synthetic sources. The three transisomers exist in nature in variable ratios whereas the synthetic material is either a racemic mixture 1:2:1 (3S,30 S:3R,30 S:3R,30 R) (Moretti et al., 2006) or a pure enantiomer. Many carotenoids have significant biological activity, in particular their antioxidant properties (Lordan et al., 2011). Focus on this aspect has led to the characterisation of carotenoids as being generally beneficial (Rao and Rao, 2007). Some carotenoids, notably beta-carotene, are precursors for the synthesis of vitamin A in herbivores and omnivores.

* Corresponding author. E-mail address: [email protected] (B. van Ravenzwaay). http://dx.doi.org/10.1016/j.fct.2016.02.001 0278-6915/© 2016 Elsevier Ltd. All rights reserved.

Excessive dosing with vitamin A leads to a well known pattern of toxicity (hypervitaminosis). However, the cleavage of provitamin A carotenoids is highly regulated and vitamin A toxicity from this source is considered to be practically impossible (Penniston and Tanumihardjo, 2006). Preformed vitamin A toxicity includes a risk of teratogenicity (Teratology Society, 1987), but, while retinoic acid teratogenicity has been demonstrated at high doses in animal experiments, the evidence of harm in humans from vitamin A supplementation appears quite limited (Azaïs-Braesco and Pascal, 2000). Of course, carotenoids are a chemical family with great structural and biological diversity (Britton, 1995) and no general assumptions can be made about potential for adverse developmental effects. In a recent EFSA-Opinion on the safety and efficacy of a synthetic, racemic Astaxanthin (EFSA, 2014) it was reported that there was no evidence of any embryotoxicity, fetotoxicity or teratogenicity in rats and rabbits in GLP-compliant developmental toxicity studies. For the pure S,S0 -Astaxathin isomer toxicological information is limited. With regard to systemic toxicity, two recent sub-chronic studies, using as test materials either a natural astaxanthin-rich carotenoid extract (containing mainly the S,S0 -

96

S. Schneider et al. / Food and Chemical Toxicology 90 (2016) 95e101

Astaxanthin form; EFSA, 2007a) or a synthetic S,S0 -Astaxanthin in a powder formulation, showed consistently no specific organ toxicity (Buesen et al., 2015; Katsumata et al., 2014). However, in the scientific literature no studies specifically on reproductive toxicity have been published so far. This is considered to be relevant information for the safety assessment of S,S0 -Astaxanthin in food applications. Accordingly, an experimental study was designed in rabbits which is a well-accepted animal model for human developmental risk (Foote and Carney, 2000). For this purpose, 3S, 30 SAstaxanthin was included into a gelatin/carbohydrate powder formulation in order to improve stability and bioavailability of the crystalline molecule. The formulation is intended for fortification of certain foods and for use in dietary supplements with an anticipated overall daily exposure of the consumer not above 10 mg Astaxanthin per day. 2. Materials and methods This study was performed in compliance with the guidelines for testing of chemicals as prescribed by United States EPA OPPTS 870.3700 1998, OECD 414 (2001) and European Commission Regulation No. 440/2008. The study was performed in an AAALAC accredited laboratory and according to OECD good laboratory practice (OECD, 1997). It was notified to the local authority under the file number 23 177-07. 3S, 30 S-Astaxanthin (syn. 3,30 -Dihydroxy-b,b-carotene-4,40 dione, CAS-no. 472-61-7) was obtained by chemical synthesis and had a purity of 95.6 g/100 g as determined by UV photometric analysis. Astaxanthin was formulated into powder beadlets due to the known low oral bioavailability of unprocessed carotenoids (Mercke Odeberg et al., 2003) and the limited stability of the pure crystalline substance during storage. The composition of the powder formulation was as follows: 3S, 30 S-Astaxanthin: 21.1%; Porcine gelatin: 23%; Sucrose: 47%; Sodium ascorbate: 2%; Mixed tocopherol (70%): 2.8%; Corn starch (native): 19%; Residual water: 6.2%. The content of Astaxanthin in the formulation was determined spectrophotometrically. In order to avoid any unspecific effects induced by formulation ingredients a placebo control group was included which was composed of all the formulation ingredients without the addition of Astaxanthin. Groups of adult female New Zealand white rabbits (Charles River) were artificially inseminated and allocated at random to the various dose groups. The day of insemination was designated as Gestational Day 0 (GD 0, i.e. beginning of the study) and the following day as GD 1. Due to the large number of animals (150) involved this was done in six batches of animals over a period of nine days. They were dosed by gavage from gestational day 6e28 with 10 ml/kg body weight of deionised water (test group 0), blank formulation (placebo; test group 1) or 3S, 30 S-Astaxanthin at concentrations calculated to give 100, 200 or 400 mg/kg body weight of the test substance (test groups 2e4). These doses were selected based on information available for racemic Astaxanthin, which did not induce maternal toxicity or developmental toxicity at the highest tested dose of 400 mg/kg body weight/day in an unpublished study with rabbits (EFSA, 2014) and based on a previous maternal toxicity study with 3S, 30 S-Astaxanthin (BASF SE, 2014). The viscosity of the powdered formulation, when mixed with deionised water, precluded administration of more than 400 mg 3S, 30 S-Astaxanthin/kg body weight of in a volume of 10 ml/kg body weight. The lower doses were prepared by further dilution of the powdered formulation to give a standard volume. The test substance preparations were prepared at the beginning of the administration period and thereafter at maximum intervals of 11 days, which took into account the period of established stability. The content, the homogeneity and the stability of 3S, 30 S-

Astaxanthin in the vehicle (deionised water) was investigated by HPLC analysis and photodiode array detection (DAD). The food used was pelleted Kliba maintenance diet rabbit and guinea pig “GLP”, supplied by Provimi Kliba SA, Kaiseraugst, Switzerland. Food and drinking water (potable tap water in water bottles) were available ad libitum throughout the study. The does were regularly examined during pregnancy and weighed on GD 0, 2, 4, 6, 9, 11, 14, 16, 19, 21, 23, 25, 28 and 29. Only does that survived and were pregnant were used in the subsequent calculations. The does were killed by injection of pentobarbital (Narcoren®, 2 ml/animal) on gestational day 29 and the standard measures were taken: uterine weight, corpora lutea, live fetuses, dead fetuses and dead implantations (early and late). Evaluations of all other parameters were performed by technicians who were unaware of the treatment group allocation in order to prevent bias. The following calculations were derived: conception rate, pre- and post-implantation loss. After opening the uteri, all fetuses were examined for external abnormalities. Furthermore, the viability of the fetuses and the condition of placentas, umbilical cords, fetal membranes, and fluids were examined. Individual placental weights were recorded. Thereafter, the fetuses were sacrificed with pentobarbital. After the fetuses had been sacrificed, the abdomen and thorax were opened in order to examine the organs in situ before they were removed. The heart and the kidneys were sectioned in order to evaluate the internal structure. The sex of the fetuses was determined by examination of the gonads in situ. Half of the fetuses at random, plus any with relevant external abnormalities, had their heads removed for fixation in Bouin's solution and Wilson section (Wilson and Warkany, 1965). After fixation in ethyl alcohol all skeletons (including those without heads), were cleared and double-stained to reveal the skeletal structure: bone and cartilage (Kimmel and Trammel, 1981). In the present study, the internationally harmonized glossary of Wise et al. (1997) and the updated version Makris et al. (2009) was essentially used to describe findings in fetal morphology. Classification of these findings was based on the terms and definitions proposed by Chahoud and Solecki (Chahoud et al., 1999; Solecki et al., 2001, 2003): A malformation is regarded as a permanent structural change that is likely to adversely affect survival or health. A variation represents a change that also occurs in the fetuses of control animals and/or is unlikely to adversely affect survival or health. This includes delays in growth or morphogenesis that have otherwise followed a normal pattern of development. The term “unclassified observation” was used for those fetal findings, which could not be classified as malformations or variations. Statistical analyses: All statistical evaluations for clinical, necropsy and fetal examinations compared treatment groups to the placebo control group. In addition, the placebo control group was compared to the vehicle control group. Parametric tests such as DUNNETT (two-sided), or non-parametric tests such as FISHER'S EXACT test (one-sided) or WILCOXON-test (one-sided) were used to test for the hypothesis either of equal means or equal proportions or medians. Historical control data of the test facility have been utilized to enhance the assessment of the findings in the present study. The historical data were collected from 10 comparable studies conducted between January 2009 and January 2013, using the same strain of New Zealand white rabbits (Crl:KBL(NZW)) as in the present study. 3. Results The stability of Astaxanthin in the application vehicle (deionized

S. Schneider et al. / Food and Chemical Toxicology 90 (2016) 95e101

water) was demonstrated over a period of 11 days. The actual recoveries as calculated against the theoretical value ranged from 79 to 107% and were considered in an acceptable range of 80e120%. No contamination of the blank or placebo solution was detected. Concentration and homogeneity analysis of the test material in the formulated dosage solutions were confirmed analytically and were considered satisfactory: the concentrations ranged between 88 and 96%of expected nominal values. The extraction efficiency for Astaxanthin was determined by analysing the neat test material (i.e. Astaxanthin powder formulation) and the found recovery rate (ca. 76% ± 1%, n ¼ 4) was used to calculate a recovery factor to compensate for extraction losses. The obtained Astaxanthin concentrations in the dosage preparations ranged between 87 and 100% for the individual analyses with a low relative standard deviation (5%)w. The main reproduction findings of the study can be found in Table 1. Adults: There were a total of 12 non-pregnant animals, which resulted in a large excess of pregnant animals at terminal sacrifice available for study (23e27 litters per group), compared to the minimum guideline requirement of 16 litters. Five does died during the study. One low-dose female died intercurrently on GD 19 showing clinical signs such as reduced defecation on GD 18. One low-dose female was sacrificed moribund on GD 24, as it was in a reduced nutritional state. The causes for these deaths remain unclear due to the unspecific clinical signs and necropsy findings. Incidental cases of mortality are found to be in the range of 3% for control animals of the same strain of the same labaratory. Therefore, none of these deaths was considered to be related to the test item. Two mid-dose females were sacrificed after abortion ahead of schedule. Spontaneous abortions in single does are not uncommon findings in the strain of rabbits used for this study as confirmed by the historical control data of the same laboratory (7 abortions in 249 pregnant control females during the last 5 years prior to the study). They were thus considered to be spontaneous incidents. A low-dose female was found dead on GD 13. The gross pathological examination of this animal revealed findings indicative of gavage error without any apparent relation to treatment (there were none in the high dose group). There were no apparent treatment effects on food consumption or body weight gain (see Table 2 and Fig. 1), all dose groups being similar to the placebo control group. The vehicle control group had

97

slightly higher food consumption than the placebo group, but body weight gain was similar. No treatment related effects were found on placental weight or on necropsy findings of does. With regard to the reproduction data, there were no test substance-related and/or biologically relevant differences between the different test groups in the conception rate, mean number of corpora lutea, implantation sites and resorptions, as well as the values of pre- and the postimplantation losses derived thereof, and in the number of viable foetuses (Table 1). All differences observed are considered to reflect the normal range of fluctuations for animals of the rabbit strain and age. Fetuses: There were no treatment related effects on sex ratio, placental weight, litter size or pup weight; all dosed groups were comparable to the vehicle and placebo control (Table 1). The total number of fetuses available for examination varied from 180 from 23 litters (200 mg/kg group) to 237 from 27 litters (400 mg/kg group) (Table 3). External malformations and variations were rare: malformations consisting of one omphalocele in the 200 mg/kg group and one in the 400 mg/kg group, while one external variation was noted, namely paw hyperflexion in a dead fetus of the lowest dose group (100 mg/kg). There was also one unclassified observation in the lowest dose group which was polyhydramnios. These observations are considered to be incidental and not treatment related as they were either not related to dose or are occasionally seen in control rabbits at this facility (historical control range up to 1% affected foetuses per study). Soft tissue malformations were also rare: 5 malformations were noted, 3 in the placebo control and 2 in the middle dose group. There were none in the low and high dose groups or in the vehicle control group. The malformations were distributed thus:  placebo: absent subclavian 2 cases, microphthalmia 1 case,  mid dose group: absent gall bladder 1 case, small spleen 1 case. Soft tissue variations were more common affecting 4e8 fetuses per group and about 20% of litters, see Table 3. There was no indication of a treatment related trend in these findings, nor was there any specific variation appearing more often in any of the groups. A similar incidence was found in both the placebo and vehicle control groups. Some unclassified soft tissue observations were recorded: a

Table 1 Effect of 3S,30 S-Astaxanthin on the developmental toxicity in New Zealand white rabbit offspring showing an overview of the effects on main observed reproductive parameters.

No. of inseminated animals No. of non-pregnant animals Maternal mortality (No.) Dams with viable fetuses Dams with all resorptions Maternal body weight change (days 0e29) (grams; mean ± SD) Maternal body weight change corrected for pregnant uterine weight(days 6e29) (grams; mean ± SD) Gravid uterine weight (grams; mean ± SD) Corpora lutea (mean ± SD) Implantation sites (mean ± SD) Preimplanation loss (mean% ± SD) Postimplantation loss (mean% ± SD) Total resorptions (mean ± SD) Live fetuses (mean ± SD) No. of male and female fetuses per litter (mean ± SD) Fetal weights (grams; mean ± SD)

Test group 0 0 mg/ kg bw/d vehicle

Test group 1 0 mg/ kg bw/d placebo

Test group 2 100 mg/kg bw/d

Test group 3 200 mg/kg bw/d

Test group 4 400 mg/kg bw/d

30 1 0 29 0 350 (207) 279 (210)

30 3 0 27 0 350 (245) 305 (242)

30 4 3 23 0 408 (255) 297 (238)

30 2 2 (sacrificed) 24 2 412 (227) 194 (290)

30 2 0 27 1 382 (297) 287 (229)

468 (109) 9.0 (2.3) 8.7 (2.62) 5.4 (11.5) 6.2 (7.84) 0.6 (0.8) 8.1 (2.39) 4.3 (2.12) 3.8 (1.95) 40.5 (5.87)

468 (104) 9.5 (2.4) 8.9 (2.83) 7.9 (16.6) 5.6 (8.40) 0.6 (0.8) 8.4 (2.60) 4.1 (1.95) 4.3 (2.12) 39.4 (6.05)

503 (96) 9.2 (1.9) 9.1 (1.93) 2.4 (3.8) 5.1 (8.34) 0.4 (0.8) 8.6 (1.97 4.4 (1.50) 4.2 (1.92) 40.2 (4.94)

397 (172) 7.3 (3.4) 7.3 (3.37) 12.6 (21.3) 12.1 (27.84) 0.3 (0.6) 7.4 (2.93) 4.0 (2.09) 3.4 (1.67) 40.5 (5.58)

475 (137) 8.9 (2.9) 8.9 (2.87) 8,6 (15.5) 7.9 (19.14) 0.5 (0.7) 8.8 (2.34) 4.9 (1.46) 3.9 (1.90) 39.1 (5.96)

98

S. Schneider et al. / Food and Chemical Toxicology 90 (2016) 95e101

Table 2 Mean food consumption of New Zealand white rabbits during gestation. Treatment with S,S0 -Astaxanthin from days 6e28. Test group 0 0 mg/kg bw/ d vehicle (N ¼ 29) Food consumption (grams; mean ± SD) Days 0 167 (3) e6 Days 6 128 (27) e28

Test group 1 0 mg/kg bw/ d placebo (N ¼ 27)

Test group 2 100 mg/kg bw/ Test group 3 200 mg/kg bw/ Test group 4 400 mg/kg bw/ d (N ¼ 23) d (N ¼ 26) d (N ¼ 28)

181 (4)

185 (3)

181 (6)

181 (3)

115 (28)

127 (31)

126 (25)

122 (29)

mg/kg bw/d ¼ milligram 3S,30 S-Astaxanthin per kilogram body weight per day. N ¼ number of pregnant animals at terminal sacrifice.

Fig. 1. Mean body weight New Zealand white rabbits during gestation. Treatment with S,S0 -Astaxanthin from days 6e28.

Table 3 Effect of 3S,30 S-Astaxanthin on total soft tissue variations in New Zealand white rabbit offspring.

Litter Fetuses Fetal incidence Litter incidence Affected fetuses/ litter

N N N (%) N (%) Mean

Test group 0 0 mg/kg bw/ d vehicle

Test group 1 0 mg/kg bw/ d placebo

Test group 2 100 mg/ kg bw/d

Test group 3 200 mg/ kg bw/d

Test group 4 400 mg/ kg bw/d

29 234 3 (1.3) 3 (10) 1.5

27 226 7 (3.1) 5 (19) 3.0

23 199 4 (2.0) 4 (17) 2.1

24 180 4 (2.2) 4 (17) 2.1

27 237 8 (3.4) 7 (26) 3.3

mg/kg bw/d ¼ milligram 3S,30 S-Astaxanthin per kilogram body weight per day. N ¼ number; % ¼ per cent.

fluid-filled abdomen and thorax in one fetus of the low dose group (100 mg/kg), atelectasis in one fetus of mid dose group (200 mg/kg) and a blood coagulum around urinary bladder in one fetus in the Placebo group and one in the low dose group (100 mg/kg). These findings did not occur in the high dose and were not considered biologically relevant. As it is common in mammals including rabbits, there were several skeletal alterations but the distribution appeared to be random (Carney and Kimmel, 2007; Daston and Seed, 2007). The incidence of skeletal malformations being two for the placebo, three for the low dose, one for the mid dose and three for the high dose (Table 4). Variations were common, implicating all the litters and around 95% of the individual fetuses. Individual types of variations were compared using a Wilcoxon onesided test for statistical significance. Any variations which were statistically different are shown in Table 5. The only variation that was statistically significantly different

from placebo and slightly outside the historical range was: unilateral ossification of sternebra with unchanged cartilage. However, this was in the mid dose group and the high dose group was comparable to the placebo and historical values. Some isolated cartilage findings without impact on the respective bony structures, which were designated as unclassified cartilage observations, occurred in all test groups. The observed unclassified cartilage findings did not show any relation to dosing (fetal incidence from 8.5% in the low dose to 11% in the placebo) and were therefore considered to be spontaneous in nature. When all malformations (external, soft tissue and skeletal) are considered together a slightly unexpected pattern emerges as shown in Table 6. There were malformations in five placebo-control, three lowdose, four mid-dose and four high-dose fetuses. The scattered occurrence of these malformations in single fetuses throughout all test groups including the placebo controls without a consistent

S. Schneider et al. / Food and Chemical Toxicology 90 (2016) 95e101

99

Table 4 Effect of 3S,30 S-Astaxanthin on individual fetal skeletal malformations in New Zealand white rabbit offspring. Test group

Doe no. e fetus no., sex

Finding

0 (0 mg/kg bw/d) Vehicle 1 (0 mg/kg bw/d) Placebo

No malformation observed 48e01 M 56e01 F 70e07 M 75e09 M 79-01 D 112e05 M 137e08 M 149e03 M 149e07 M

severely malformed vertebral column lumbar hemivertebra absent lumbar vertebra misshapen interparietal sutures fused absent interparietal short rib misshapen thoracic vertebra thoracic hemivertebra

2 (100 mg/kg bw/d)

3 (200 mg/kg bw/d) 4 (400 mg/kg bw/d)

mg/kg bw/d ¼ milligram 3S,30 S-Astaxanthin per kilogram body weight per day. No. ¼ number; M ¼ male; F ¼ female, D ¼ dead.

Table 5 Occurrence of statistically significant skeletal variations (expressed as mean percentage of affected fetuses/litter) after administration of 3S,30 S-Astaxanthin to New Zealand white rabbits from gestational days 6e28. Finding

Test group 0 0 mg/ kg bw/d vehicle

Test group 1 0 mg/ kg bw/d placebo

Test group 2 100 mg/kg bw/d

Test group 3 200 mg/kg bw/d

Test group 4 400 mg/kg bw/d

Historical mean % (range)

Extra ossification site between cervical centers Unilateral ossification of sternebra; unchanged cartilage Incomplete ossification of pubis; cartilage present

4.7

0.0

0.0

0.0

1.5*

0.5 (0.0e2.5)

1.7

1.8

2.2

7.6*

2.6

1.4 (0.0e5.8)

1.7

0.0

0.3

2.1*

0.7

1.2 (0.0e4.3)

mg/kg bw/d ¼ milligram 3S,30 S-Astaxanthin per kilogram body weight per day. * ¼ p  0.05 (Wilcoxon-test [one-sided]).

Table 6 Incidence of total fetal malformations (external þ visceral þ skeletal) after administration of 3S, 30 S-Astaxanthin to New Zealand white rabbits on gestational days 6e28. Test group 0 0 mg/ kg bw/d Litter Fetuses Fetal incidence Litter incidence Affected fetuses/ litter

N 29 N 234 N (%) 0.0 N (%) 0.0 Mean 0.0 %

Test group 1 0 mg/kg bw/ d placeboa

Test group 2 100 mg/kg bw/ Test group 3 200 mg/ db kg bw/db

Test group 400 mg/kg bw/ db

27 226 5 (2.2) 5 (19)* 2.3**

23 199 3 (1.5) 3 (13) 1.2

27 237 4 (1.7) 3 (11) 1.5

24 180 4 (2.2) 4 (17) 2.6

mg/kg bw/d ¼ milligram per kilogram body weight per day; N ¼ number; % ¼ per cent. * ¼ p  0.05 (Fisher's exact test [one-sided]). ** ¼ p  0.01 (Wilcoxon-test [one-sided]). a Comparison to test group 0. b Comparison to test group 1.

pattern, without a clear doseeresponse relationship and/or at incidences which were generally similar to historical control rates does not suggest any test substance-induced origin of these findings. In addition, there were no individual differences seen per observation type. While no malformations were seen in the vehicle control, the complete absence of any fetal malformation in that group is rather unusual. Thus, the total incidence of malformations in the placebo-control group was statistically significantly higher than the vehicle control. However, the placebo incidence is well within the historical control range, and the statistical difference with the vehicle control is considered to be incidental. If total variations are compared all litters and almost all fetuses were affected. This is usual in rabbits (Carney and Kimmel, 2007; Daston and Seed, 2007) and is mostly due to the high incidence of skeletal variations (see Table 5). However, no pattern by type or dose group was revealed by more detailed analysis of the types of variations.

4. Discussion Toxicity studies have previously been performed with Astaxanthin and have generally demonstrated the innocuous nature of this substance (EFSA, 2005 and EFSA, 2007b). However, these studies are for the most part only available as summaries in the public domain such that EFSA were obliged to rely on summaries for part of their 2005 evaluation. On the other hand, many studies have been performed investigating a range of clinical endpoints where Astaxanthin might have beneficial health effects. These studies mostly concern anti-oxidant or anti-inflammatory properties and include both in vitro, animal and human studies (Lordan et al., 2011; Yuan et al., 2011). Fully detailed acute and subchronic studies in rats have been published on a dehydrated algal product from Haematococcus pluvialis containing about 3% of Astaxanthin mixed esters (Stewart et al., 2008), a natural astaxanthin-rich carotenoid extract of Paracoccus carotinifaciens (Katsumata et al., 2014) and synthetic S,S0 Astaxanthin in a powder formulation (Buesen et al., 2015). They

100

S. Schneider et al. / Food and Chemical Toxicology 90 (2016) 95e101

consistently reported no significant toxicity up to a daily dose of at least 500 mg/kg body weight/day. Reproductive and developmental studies for synthetic Astaxanthin (presumably in the form of a racemic mixture) also exist (EFSA, 2005, 2007b, 2014; Vega et al., 2015) both in rats and rabbits, but the rabbit study is not available in the peer reviewed literature. According to these reviews, two developmental toxicity studies were performed in Fü-Albino rats with a beadlet formulation containing 6.1% Astaxanthin and Swiss Hare rabbits with a suspension of crystalline Astaxanthin in rape seed oil. In both studies, the no-observed-adverse-effect levels (NOAEL) regarding maternal toxicity and developmental toxicity were found at the highest tested dose levels (1000 mg/kg body weight/day in rats and 400 mg/kg body weight/day in rabbits) (EFSA, 2014). In the present developmental toxicity study, the pure 3S, 30 SAstaxanthin enantiomer was investigated in a formulation intended for dietary use. The study followed the most up to date regulatory guidelines and the conditions of Good Laboratory Practice. The highest dose administered was 400 mg/kg body weight which was the maximum that could be achieved while respecting the requirement to administer a maximum amount of 10 ml/kg body weight of dosing solution. This dose level did not produce any toxicity to the dams, but for practical and ethical reasons it was considered to be as high a dose as could reasonably be administered. In a human being weighing 60 kg, this dose level would have been equivalent to 24 g of Astaxanthin per day. Comparing this dose with the anticipated human exposure level of not more than 10 mg Astaxanthin per day, a large Margin of Safety of 2400 can be calculated (¼ 24 g/10 mg) which does not indicate a risk for the consumer. To our knowledge toxicokinetic studies with Astaxanthin in rabbits have not been published so far. To assess the potential bioavailability of the compound, we therefore have to rely on information from other species. In a subchronic study (Stewart et al., 2008) investigating an Astaxanthin rich biomass did not observe an accumulation of the substance in the plasma of rats over treatment periods of 90 days when compared with the plasma levels at the beginning of treatment. It is known that presence of dietary lipids can increase the absorption rate of lipophilic carotentoids in humans (Kistler et al., 2002; Mercke Odeberg et al., 2003; Osterlie et al., 2000) which indicates that Astaxanthin needs to be administered in an appropriate matrix in order to be taken up in the gastrointestinal tract. Therefore, the test substance was formulated into a gelatin/carbohydrate matrix which is from our experience known to be a suitable vehicle for the administration of carotenoids in food and feed applications. Therefore, if there would have been an effect of the vehicle, it would have been to increase bioavailability. In our study, the availability of dietary lipids could have been a limiting factor for the gastrointestinal absorption of Astaxanthin using a standard laboratory diet for rabbits. Taken these facts together, it cannot be excluded that the low bioavailability could have contributed to the absence of toxicity. In an EFSA opinion (EFSA, 2014) it was previously reported that Astaxanthin administered in rape seed oil to rabbits did not induce any signs of maternal or developmental toxic effects at the highest tested dose level of 400 mg/kg body weight/day which is in line with our results for the pure S,S0 -Astaxanthin isomer in a powder formulation. Regarding developmental toxic effects, the incidence of malformations in this study was quite low, an average incidence of 3 per group of about 200 fetuses (about 25 litters). Although the vehicle control group had no malformations, which is unusual, the highest incidence was in the placebo group (5 fetuses) while the highest dose group had 3 fetuses with malformations. Overall, there was no indication that 3S, 30 S-Astaxanthin increased the frequency of malformations.

As far as variations are concerned the litter incidence was 100% which is due to the well-known and documented high incidence of skeletal variations in New Zealand white rabbits (Ema et al., 2012). However, looking at this on a more detailed level of specific variations did not suggest a treatment effect either. Only one specific variation lay outside the historical range: and this concerned the mid-dose group. The other dose groups and placebo controls were in the normal range. Given that this is only one incidence out of many evaluated, it can be considered a chance event. Overall, it can be concluded that daily administration of up to 400 mg/kg body weight of 3S, 30 S-Astaxanthin throughout the pregnancy of New Zealand white rabbits in a gelatin/carbohydrate powder formulation intended for dietary use did not reveal any potential for developmental toxicity following oral administration. Conflicts of interest The study was sponsored by BASF SE, 67056 Ludwigshafen, Germany. Acknowledgements The authors would like to thank the staff of BASF's Reproductive Toxicology Laboratory for their excellent technical assistance. Furthermore, we wish to thank U. Blum for her help with the revision and submission of this manuscript. Transparency document Transparency document related to this article can be found online at http://dx.doi.org/10.1016/j.fct.2016.02.001. References Azaïs-Braesco, V., Pascal, G., 2000. Vitamin A in pregnancy: requirements and safety limits. Am. J. Clin. Nutr. 71, 1325se1333s. BASF SE, 2014. 3S, 30 S-Astaxanthin, Maternal Toxicity Study in New Zealand White Rabbits (unpublished data). Britton, G., 1995. Structure and properties of carotenoids in relation to function. FASEB J. 9, 1551e1558. €ters, S., Carvalho, S., Buesen, R., Schulte, S., Strauss, V., Treumann, S., Becker, M., Gro van Ravenzwaay, B., 2015. Safety assessment of [3S,3S0 ]-astaxanthin e subchronic toxicity study in rats. Food Chem. Toxicol. 81, 129e136. Carney, E.W., Kimmel, C.A., 2007. Interpretation of skeletal variations for human risk assessment: delayed ossification and wavy ribs. Birth Defects Res. (Part B) 80, 473e496. Chahoud, I., Buschmann, J., Clark, R., Druga, A., Falke, H., Faqi, A., et al., 1999. Classification terms in developmental toxicology: need for harmonization. Reprod. Toxicol. 13, 77e82. Daston, G.P., Seed, J., 2007. Skeletal malformations and variations in developmental toxicity studies: interpretation issues for human risk assessment. Birth Defects Res. (Part B) 80, 421e424. EFSA, 2005. Opinion of the scientific panel on additives and products or substances used in animal feed on the request from the European Commission on the safety of use of colouring agents in animal nutrition part I. General principles and astaxanthin. EFSA J. 291, 1e40. Available online: http://www.efsa.europa. eu/de/efsajournal/pub/291.htm. EFSA, 2007a. Safety and efficacy of Panaferd-AX (red carotenoid-rich bacterium Paracoccus carotinifaciens) as feed additive for salmon and trout1. Scientific opinion of the panel on additives and products or substances used in animal feed. EFSA J. 546, 1e30. Available online: http://www.efsa.europa.eu/de/ efsajournal/pub/546.htm. EFSA, 2007b. Safety and efficacy of CAROPHYLL® stay-pink (astaxanthin dimethyldisuccinate) as feed additive for salmon and trout. Scientific opinion of the panel on additives and products or substances used in animal feed. EFSA J. 574, 1e25. Available online: http://www.efsa.europa.eu/de/efsajournal/pub/574. htm. EFSA, 2014. Scientific Opinion on the safety and efficacy of astaxanthin (CAROPHYLL® Pink 10% CWS) for salmonids and ornamental fish. EFSA J. 12 (6), 3725, 35 pp. Available online: http://www.efsa.europa.eu/de/efsajournal/doc/ 3724.pdf. Ema, M., Aoyama, H., Arima, A., Asano, Y., Chihara, K., Endoh, K., Fujii, S., Hara, H., Higuchi, H., Hishikawa, A., Hojo, H., Horimoto, M., Hoshino, N., Hosokawa, Y., Inada, H., Inoue, A., Itoh, K., Izumi, H., Maeda, M., Matsumoto, K., Matsuo, S.,

S. Schneider et al. / Food and Chemical Toxicology 90 (2016) 95e101 Matsuura, I., Mineshima, H., Miwa, Y., Miyata, H., Mizoguchi, Y., Nakano, N., Naya, M., Nishizawa, H., Noritake, K., Noyori, H., Ohta, T., Oku, H., Shimizu, T., Shimomura, K., Shiozawa, K., Takakura, I., Tanaka, R., Uesugi, T., Yabe, K., Yamauchi, T., Yokoi, R., 2012. Historical control data on prenatal developmental toxicity studies in rabbits. Congenit. Anom. (Kyoto) 52, 155e161. EPA, 1998. Health Effects Test Guidelines, OPPTS 870.3700: Prenatal Developmental Toxicity Study. European Commission Regulation (EC) No 440/2008 of 30 May 2008 laying down test methods pursuant to Regulation (EC) No 1907/2006 of the European Parliament and of the Council on the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), Part B: Methods for the determination of toxicity and other health effects: Prenatal Developmental Toxicity Study; Official Journal of the European Union, No. L 142 Foote, R.H., Carney, E.W., 2000. The rabbit as a model for reproductive and developmental toxicity studies. Reprod. Toxicol. 14, 477e493. Katsumata, T., Ishibashi, T., Kyle, D., 2014. A sub-chronic toxicity evaluation of a natural astaxanthin-rich carotenoid extract of paracoccus carotinifaciens in rats. Toxicol. Rep. 1, 582e588. Kimmel, C.A., Trammel, C., 1981. A rapid procedure for routine double staining of cartilage and bone in fetal and adult animals. Stain Technol. 56, 271 e 27. Kistler, A., Liechti, H., Pichard, L., Wolz, E., Oesterhelt, G., Hayes, A., Maurel, P., 2002. Metabolism and CYP-inducer properties of astaxanthin in man and primary huma hepatocytes. Arch. Toxicol. 75, 665e675. Lordan, S., Ross, R.P., Stanton, C., 2011. Marine bioactives as functional food ingredients: potential to reduce the incidence of chronic diseases. Mar. Drugs 9, 1056e1100. Lorenz, R.T., Cysewski, G.R., 2000. Commercial potential for Haematococcus microalgae as a natural source of astaxanthin. Trends Biotechnol. 18, 160e167. Makris, S.L., Solomon, H.M., Clark, R., Shiota, K., Barbellion, S., Buschmann, J., et al., 2009. Terminology of developmental abnormalities in common laboratory mammals (Version 2). Birth Defects Res. (Part B) 86, 227e327. € glund, P., 2003. Oral bioavailMercke Odeberg, J.M., Lignell, A., Pettersson, A., Ho ability of the antioxidant astaxanthin in humans is enhanced by incorporation of lipid based formulations. Eur. J. Pharm. Sci. 9, 299e304. Moretti, V.M., Mentasti, T., Bellagamba, F., Luzzana, U., Caprino, F., Turchini, G.M., Giani, I., Valfre, F., 2006. Determination of astaxanthin stereoisomers and colour attributesin flesh of rainbow trout (Oncorhynchus mykiss) as a tool to

101

distinguish the dietary pigmentation source. Food Addit. Contam. 23, 1056e1063. OECD Guideline for Testing of Chemicals, 2001. Proposal for Updating Guideline 414: Prenatal: Developmental Toxicity Study. OECD, 1997. Principles of Good Laboratory Practice revised. Osterlie, M., Bjerkeng, B., Liaaen-Jensen, S., 2000. Plasma appearance and distribution of astaxanthin E/Z and R/S isomers in plasma lipoproteins of men after single dose administration of astaxanthin. J. Nutr. Biochem. 11, 482e490. Penniston, K.L., Tanumihardjo, S.A., 2006. The acute and chronic toxic effects of vitamin A. Am. J. Clin. Nutr. 83, 191e201. Rao, A.V., Rao, L.G., 2007. Carotenoids and human health. Pharmacol. Res. 55, 207e216. Solecki, R., Bürgin, H., Buschmann, J., Clark, R., Duverger, M., Fialkowski, O., et al., 2001. Harmonisation of rat fetal skeletal terminology and classification. Report of the third workshop on terminology in developmental toxicology. Reprod. Toxicol. 15, 713e772. Solecki, R., Bergmann, B., Bürgin, H., Buschmann, J., Clark, R., Druga, A., et al., 2003. Harmonization of rat fetal external and visceral terminology and classification. Report of the fourth workshop on the terminology in developmental toxicology. Reprod. Toxicol. 17, 625e637. Stewart, J.S., Lignell, A., Pettersson, A., Elfving, E., Soni, M.G., 2008. Safety assessment of astaxanthin-rich microalgae biomass: acute and subchronic toxicity studies in rats. Food Chem. Toxicol. 46, 3030e3036. Teratology Society, 1987. Teratology society position paper: recommendations for vitamin a use during pregnancy. Teratology 35, 269e275. Vega, K., Edwards, J., Beilstein, P., 2015. Subchronic (13-week) toxicity and prenatal developmental toxicity studies of dietary astaxanthin in rats. Regul. Toxicol. Pharmacol. http://dx.doi.org/10.1016/j.yrtph.2015.10.013 ([Epub ahead of print]). Wilson, J.G., Warkany, J. (Eds.), 1965. Teratology: Principles and Techniques. Univ. of Chicago Press, Chicago (. Wise, D., Beck, S., Beltrame, D., Beyer, B., Chahoud, I., Clark, R.L., et al., 1997. Terminology of developmental abnormalities in common laboratory mammals (Version 1). Teratology 55, 249e292. Yuan, J.-P., Peng, J., Yin, K., Wang, J.-H., 2011. Potential health-promoting effects of astaxanthin: a high-value carotenoid mostly from microalgae. Mol. Nutr. Food Res. 55, 150e165.

A developmental toxicity study of 3S, 3'S-Astaxanthin in New Zealand white rabbits.

Astaxanthin, a naturally occurring pigment used to give the characteristic orange-pink colour to salmonid fish reared in aquaculture, is also marketed...
376KB Sizes 4 Downloads 16 Views