Safety assessment of Gossence™ (galactooligosaccharides): Genotoxicity and general toxicity studies in Sprague Dawley rats

Abstract

Galactooligosaccharides (GOS) used as prebiotics are one of the major constituents of the infant milk formulas. GOS (Gossence™) is produced by a patented process of biotransformation of lactose; hence toxicology studies were carried out to assess its safety. The objective of the present study was to evaluate the general and genetic toxicity of Gossence™. In 14-day and subchronic (90-day) oral toxicity studies in Sprague Dawley rats, daily administration of GOS at dose levels of 1000, 2000, or 5000 mg/kg (equivalent to 1347, 2694, and 6735 mg/kg/day of Gossence™, respectively) did not cause any mortality, or clinical signs, and changes in body weights, feed consumption, hematology, clinical chemistry, and urinalysis. In 90-day study, no changes in ophthalmological and neurological findings were observed. Significant increases in the cecum weights (with and/or without content) at dose levels of ≥2000 mg/kg were observed in both 14-day and 90-day studies. Based on the results of 90-day study, the no-observed-adverse-effect-level for GOS is 5000 mg/kg/day which is equivalent to 6735 mg Gossence™/kg/day. In the bacterial reverse mutation test, there was no significant increase in the mean numbers of revertants at the tested concentrations. Gossence™ was not mutagenic up to 5000 µg/plate. In chromosomal aberration test, there was no statistically significant increase in the number of percent aberrant metaphase for the Gossence™. Gossence™ is non-clastogenic (negative) in the in vitro chromosomal aberration test using human peripheral blood lymphocyte during short and prolonged treatment.

Keywords

Galactooligosaccharides safety oral toxicity genotoxicity

Introduction

Galactooligosaccharides (GOS) are non-digestible carbohydrate-based food ingredient that selectively increase the beneficial microflora of the intestine leading to health benefits that are extensively recognized.^1
–3 GOS, also known as oligoglalactosyllactose, oligogalactose, oligolactose, or transgalactooligosaccharide, belong to the group of prebiotics. GOS are synthesized from lactose through a transgalactosylation reaction catalyzed by an enzyme β-galactosidase.^3
–5 GOS provide their health benefits by two main mechanisms: first by selective proliferation of bifidobacteria and lactobacilli in the guts, which provide resistance against colonization of pathogens there by reducing infections and second mechanism by production of short fatty acids, which reduces the risk of cancer, increases mineral absorption, and controls serum lipid and cholesterol levels.^1,2. Due to high benefits, GOS are currently employed in various commercial commodities such as breakfast cereals, beverages, snack bars, bakery products, and so on. GOS have become the center of attention in the field of functional food because of their known health benefits chiefly in digestion. Currently, GOS are claimed to be belong to prebiotics that selectively stimulate the growth of beneficial bacteria in the lower part of human intestine.^6
–8 Human breast milk is the natural source of GOS for infant and contains approximately 100 times more GOS than cow’s milk. The infants fed with the human breast milk have bifidobacteria-predominated microflora. For this reason, GOS are most common component of infant milk formula/composition and recognized as safe and suitable alternative to human milk.^9,10 GOS wide range of health benefits include stool improvement, allergy alleviation, weight management, anti-cancer properties, anti-inflammatory action, and aid in lipid metabolism. GOS is mainly added to infant formula as it is able to stimulate growth of beneficial intestinal microflora such as bifidobacterium and reinforce microbiota similar to that of in breastfed infants.^11
–13

Gossence™ is a GOS containing product manufactured by the action of enzymes from whole microbial cells (Sporobolomyces singularis MTCC5491) in a manner like other GOS products in the market. In this report, the safety of Gossence™ (GOS) was demonstrated by repeat dose oral toxicity study for 90-days followed by 28-day recovery period in Sprague Dawley rats and by assessing for its genotoxicity potential.

Materials and methods

Test item details

Gossence™ (GOS), referred to as test item throughout this article, was supplied by Tata Chemicals Limited and stored at room temperature. The details of physicochemical properties of Gossence™ (GOS) have been summarized in Table 1.

Table 1.

Test item details.

Batch No.	GP0D17001
CAS No., chemical name	6587-31-1, galactooligosaccharides
Purity (on % dry matter)	76.28
Composition	Total galactooligosaccharides ≥ 76.28% Lactose and others ≤ 23.72% Lactose 17.10% Galactose and glucose 3.90% Glycerol 1.71% Arabitol 1.01%
Physical appearance	Fine white powder
Solubility	Readily soluble in water
Density	1.4
pH	5.54
Structure

Compliance statement

The studies were performed in compliance with Organization for Economic Co-operation and Development (OECD) Principles of Good Laboratory Practice (as revised in 1997), ENV/MC/CHEM (98) 17, OECD, Paris, 1998. (No. 1 in OECD Series on Good Laboratory Practice and Compliance Monitoring) concerning Mutual Acceptance of Data in the Assessment of Chemicals, [C (81)30(FINAL)] and Council Decision-Recommendation [C (95)8(FINAL)]. All animals were handled humanely with due regard for their welfare and complied with the recommendations of Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC) and Regulations of Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA), Government of India.

Animals

Sprague Dawley rats were quarantined for 7 days upon arrival at the facility. Animals were examined at receipt and acclimatized in experimental room for 5 days before start of the study. Rats were housed in a controlled environment with a temperature range of 19.5–24.8°C, a relative humidity range of 49–69%, a light/dark cycle of 12 h each, and at least 15 fresh air changes per hour. Rats were housed individually with provision of environmental enrichment. Autoclaved corncob (Sparconn Life Sciences, Karnataka, India) was used as bedding. Rats were fed ad libitum (except during overnight fasting prior to blood collection) with certified Rodent Pellet feed (Altromin, Maintenance Diet, Germany). Potable water autoclaved and filtered through reverse osmosis was provided ad libitum to all animals. Microbiological load and chemical contaminants analyses were performed for feed, water, and bedding material. On the last day of acclimatization, rats were randomly assigned to four main groups and two recovery groups using Pristima version 7.2.0 (Xybion Medical Systems, Lawrenceville, NJ). The body weight variation of rats did not exceed +20% of the mean body weight in each sex and group. Care of animals complied with the recommendations of AAALAC and CPCSEA, Government of India. The study was designed to use the fewest number of animals possible. The “Form B” for carrying out animal experimentation was reviewed and approved by the Institutional Animal Ethics Committee (IAEC Protocol No: SYNGENE/IAEC/794/01-2017).

Study design

14-day dose range finding study in rat

The 14-day dose range finding study was performed to select the dose levels for 90-day repeat dose study in rats. For this study, young adult healthy male and female Sprague Dawley rats were used (procured from Vivo Biotech Ltd., Hyderabad, India). The Gossence™ (GOS) was mixed in purified water and administered daily through gavage route at the dose levels of 1000, 2000, and 5000 mg/kg/day to separate groups of rats (N = 6/sex/group). Purified water was administered to control group rats. The dose volume administered was 10 mL/kg body weight. Dose formulations were analyzed for active ingredient (4′-galactosyllactose with CAS registry number 6587-31-1, a major component of GOS) by validated high performance liquid chromatography (HPLC) method on day 1 and day 12 of treatment period. The rats were observed for clinical signs or mortality, and body weights, feed consumption, and feed conversion efficiency were measured. All the rats were killed under isoflurane anesthesia and blood samples were collected by retro-orbital puncture on day 15 for clinical pathology (hematology, coagulation, and clinical chemistry) analyses. Urine was collected on day 15 from overnight fasted rats for urinalysis. Necropsy and gross pathological examination of all animals were performed on day-15. Organ collection, fixation, weighing, and microscopic examination of organs were done as summarized in Table 4. Histopathology examination was performed on rats of control and high dose groups. Based on test item-related changes, the cecum was evaluated from animals of all dose groups.

90-day sub-chronic toxicity study in rat

The 90-day study was conducted as per OECD Guideline No. 408 for Testing of Chemicals, “Repeated Dose 90-Day Oral Toxicity Study in Rodents.” For this study, young adult healthy male and female Sprague Dawley rats were used (Vivo Biotech Ltd.). The doses were selected based on the results of 14-day repeat dose toxicity study. The Gossence™ (GOS) was mixed in purified water and administered daily through gavage route at the dose levels of 1000, 2000, and 5000 mg/kg/day to separate groups of rats (N = 10 for each main group and N = 6 for each recovery group). Purified water was administered to control group rats. The dose volume administered was 10 mL/kg body weight across all the groups. The study design is summarized in Table 2.

Table 2.

Study design.

				Main/recovery group
Group No.	Treatment	Dose of GOS (mg/kg/day)	Dose concentration (mg/mL)	Males	Females
G1/G1R	Purified water	0	0	10/6	10/6
G2	Gossence™ (galactooligosaccharides)	1000	100	10	10
G3		2000	200	10	10
G4/G4R		5000	500	10/6	10/6

R: recovery group; GOS: galactooligosaccharides.

The animals in recovery groups did not receive vehicle or Gossence™ for 28 days following the 90-day dosing period. The doses were selected based on the results of 14-day repeat dose toxicity study. The dose volume was calculated for individual animals on the first day of the treatment period (day 1) and was adjusted according to the most recent body weights recorded during the treatment period.

Dose formulations were analyzed for active ingredient (4′-galactosyllactose with CAS registry number 6587-31-1, major component of GOS) by validated HPLC method at monthly interval for three times during the treatment period, that is, on day 1 and months 2 and 3.

Parameters evaluated

Observations

Morbidity/mortality check and clinical observations were performed daily. Detailed clinical examination, body weights, and food consumption were monitored at weekly interval (±2 days). Fasting body weight was recorded prior to termination. Ophthalmological examination was performed with an ophthalmoscope (Direct Ophthalmoscope; Welch Allyn, New York, USA) prior to start of treatment and at the end of the treatment period (13th week of treatment phase). Functional observational battery (neurobehavioral examinations) and motor activity were conducted during 13th week of the dosing phase.

Clinical pathology

The animals were fasted overnight (12–16 h) and terminal blood collection was done on day 91 for main groups and on day 119 (after 28 days of recovery period) for recovery groups for clinical pathology (hematology, coagulation, and clinical chemistry) analyses (Table 3). Blood was collected via retro-orbital plexus puncture under mild isoflurane anesthesia. The hematological, coagulation, and clinical chemistry parameters were determined using the ADVIA 2120i hematology system, automated coagulation analyzer (STA compact), and Dimension Xpand Plus clinical chemistry system, respectively.

Table 3.

Parameters evaluated in hematology, coagulation, clinical chemistry, and urinalysis.

Hematology	Coagulation	Clinical chemistry	Urinalysis
Red blood cell count, hematocrit, hemoglobin concentration, mean corpuscular hemoglobin, mean corpuscular hemoglobin concentration, mean corpuscular volume, platelet count, mean platelet volume, total white blood cell count, differential white blood cell (absolute count), reticulocytes (absolute count)	Activated partial thromboplastin time, prothrombin time, fibrinogen	A/G ratio—(calculated), alanine aminotransferase, albumin, alkaline phosphatase, aspartate aminotransferase, blood urea nitrogen, calcium, chloride, creatinine, gamma-glutamyl trans-peptidase, globulin (calculated), glucose, inorganic phosphorus, potassium, sodium, total bilirubin, total cholesterol, total protein, triglycerides.	Volume, color, appearance, glucose, bilirubin, ketone, blood, pH, protein, urobilinogen, nitrite, leukocyte, specific gravity

Urinalysis

Urine was collected individually from all animals at the end of treatment period (day 91 for main group) and at the end of recovery period (day 119 for recovery group). The urine samples were analyzed for the listed parameters in Table 3:

Gross necropsy, organ weights, and histopathology

All animals of main groups and recovery groups were killed in a stratified manner by exsanguination under deep isoflurane anesthesia and necropsied on day 91 of the dosing phase (terminal sacrifice) and day 119 (recovery sacrifice), respectively. All animals were examined for gross changes. The list of organs collected for organ weights and histopathology are indicated in Table 4. The organs were preserved in 10% neutral buffered formalin, except for testes and eyes with optic nerve, which were preserved in modified Davidson’s fixative and Davidson’s fixative, respectively. The paired organs were weighed together. The organ weight ratios as percentage of body and brain weight were calculated. All preserved organs from vehicle control and high dose (5000 mg/kg/day) groups and cecum from mid dose (2000 mg/kg/day) and low dose (1000 mg/kg/day) groups were subjected to histopathological evaluation.

Table 4.

List of organs collected, weighed, preserved, and examined microscopically.

S. No.	Tissues/organs	Organs weighed	Collected and preserved	Examined microscopically
1	Adrenals	X	X	X
2	Animal identification mark	–	X	–
3	Aorta	–	X	X
4	Axillary lymph nodes	–	X	X
5	Biceps femoris muscles	–	X	X
6	Bone marrow smear^a	–	X	–
7	Brain^b	X	X	X
8	Cecum^c	X	X	X
9	Colon	–	X	X
10	Diaphragm	–	X	X
11	Duodenum	–	X	X
12	Epididymides	X	X	X
13	Esophagus	–	X	X
14	Eyes	–	X	X
15	Femur bone with joint	–	X	X
16	Gastrocnemius muscle	–	X	–
17	Gross change	–	X	X
18	Harderian glands	–	X	X
19	Heart	X	X	X
20	Ileum with Peyer’s patches	–	X	X
21	Jejunum	–	X	X
22	Kidneys	X	X	X
23	Lacrimal glands	–	X	X
24	Larynx	–	X	X
25	Liver	X	X	X
26	Lungs^d	X	X	X
27	Lymph nodes (mandibular and mesenteric)	–	X	X
28	Mammary gland^e	–	X	X
29	Optic nerve	–	X	X
30	Ovaries	X	X	X
31	Oviducts	–	X	X
32	Pancreas	–	X	X
33	Pharynx	–	X	X
34	Pituitary^f	X	X	X
35	Prostate, seminal vesicles and coagulating glands^g	X	X	X
36	Quadriceps muscle	–	X	–
37	Rectum	–	X	X
38	Salivary glands (mandibular, sublingual, and parotid)	–	X	X
39	Sciatic nerve	–	X	X
40	Skin (inguinal region)	–	X	X
41	Soleus muscle	–	X	–
42	Spinal cord^h	–	X	X
43	Spleen	X	X	X
44	Sternum with marrow	–	X	X
45	Stomach	–	X	X
46	Testes	X	X	X
47	Thymus	X	X	X
48	Thyroid with parathyroid^f	X	X	X
49	Tongue	–	X	X
50	Trachea	–	X	X
51	Urinary bladder	–	X	X
52	Ureters	–	X	X
53	Uterus with cervix	X	X	X
54	Vagina	–	X	X

X: activity performed.

^aBone marrow smears were stained with Giemsa stain, however not evaluated, based on experimental results.

^bRepresentative regions including cerebrum, cerebellum, and medulla/pons.

^cTied at both ends and weighed with and without content, after the cecum was weighed with content, the content was rinsed with saline properly and again weighed without content.

^dIncluding main bronchi and bronchioles.

^eSite of mammary gland in males.

^fWeighed post fixation.

^gWeighed as a whole; subsequently prostate was separated and weighed. The derived weight was presented for the seminal vesicles and coagulating glands.

^hCervical, mid thoracic, and lumbar.

Data compilation

The Pristima version 7.2.0 (Xybion Medical Systems) for online data capturing in toxicology studies was used for recording but not limited to daily observations, body weights, food input and output, ophthalmological examination, home cage measurements, handheld measurements, open field measurements, sensory response measurements, grip strength, hind limb foot splay, rectal temperature, clinical pathology (hematology, coagulation, clinical chemistry and urinalysis), terminal body weights, organ weights, gross pathology, and histopathology data.

Statistical analysis

Pristima captured data were analyzed using Pristima built-in statistical tests. For comparative statistics, data were evaluated using the Levene’s test for homogeneity of variances with significance at 5% level. Data determined to be homogeneous were evaluated for analysis of variance (ANOVA) to verify significance at 5% level. Pairwise comparisons of each treated group with the control group were made using a Dunnett test, to identify statistical differences (at 5% level). If the data were found to be non-homogeneous, ANOVA was done using suitable transformation. Where the homogeneity tests were significant even after transformation, data were evaluated using a Kruskal–Wallis test for group factor significance. If significance (at 5% level) existed between groups, Dunn’s pairwise comparison was performed. For comparison of two groups (G1R and G4R), data were subjected to student t-test. Statistical analysis for food conversion efficiency and motor activity were performed using validated statistical software (SigmaPlot 12.3). Data with statistically significant differences were designated by superscripts of +/− symbols: significantly higher or lower than the control group/control recovery group.

Genotoxicity

Bacterial reverse mutation test

Bacterial reverse mutation test was performed according to OECD test guideline 471 (OECD, 1997) to evaluate the mutagenic potential of GOS (or its metabolites) by measuring its ability to induce reverse mutations at selected loci of Salmonella typhimurium tester strains TA98, TA100, TA1535, and TA1537 and at the tryptophan locus of Escherichia coli tester strain WP2 uvrA, in the presence and absence of Aroclor 1254-induced rat liver S9 (Lot No. 3849 Xenometrix AG, Switzerland). In the preliminary toxicity assay, the maximum dose tested was 5000 µg per plate. All dose levels of test item and vehicle controls were plated in duplicates. For the mutagenicity assay (initial and confirmatory assays), a minimum of five dose levels along with appropriate vehicle and positive controls were plated with tester strains TA98, TA100, TA1535, TA1537, and WP2 uvrA in the presence and absence of rat liver S9 activation. GOS was dissolved in distilled water and sterilized through a 0.45 mm filter. Dose levels of GOS, vehicle controls, and positive controls were plated in triplicate. For initial assay, a five dose levels of 50, 159, 501, 1582, and 5000 µg/plate were tested, in the presence and absence of S9, using the plate incorporation treatment.^14,15 The dose solution concentrations were adjusted with the potency correction factor (1.347) to compensate for the purity. In the confirmatory assay, the test system was exposed to the test item via the pre incubation methodology described by Yahagi et al.¹⁶ at dose levels of 99, 265, 699, 1869, and 5000 µg/plate, in the presence and absence of S9. Positive controls included 2-aminoanthracene (Lot No. STBD3302V; Sigma–Aldrich, Saint Louis, MO, USA), 2-nitrofluorene (Lot No. S43858V; Sigma–Aldrich), 9-aminoacridine (Lot No. BCBN7149V; Sigma–Aldrich), sodium azide (Lot No. MKBP4386V; Sigma–Aldrich), and 4-nitroquinoline N-oxide (Lot No. WXBB6231V; Sigma–Aldrich). A response was considered positive when the following criteria were met: (i) a significant and twofold increase of revertant colonies over the respective negative controls for strains TA98, TA100, and TA102 and (ii) a significant and threefold increase of revertant colonies for strains TA1535 and TA1537, with a clear dose-dependent response.

Fold increase (FI) = \frac{Mean of individual revertant colonies in the treated group}{Mean of individual revertant colonies in the vehicle control group}

In vitro chromosomal aberration test with HPBL cells

In vitro mammalian chromosomal aberration test was performed according to OECD test guideline 473 (OECD, Adopted 29 July 2016). The clastogenic potential of GOS was evaluated based upon its ability to induce chromosome aberrations in human peripheral blood lymphocytes (HPBL) cells, in the absence and presence of an Aroclor-induced S9 activation system. The human blood was collected after approval from Independent Ethics Committee (IEC) through Protocol No. CLCD-039-16. Required volume of blood samples were obtained from healthy young human donors 27 and 23 years of age for initial and confirmatory chromosome aberration assay, respectively. The volunteers were non-smoker, non-alcoholic, and not received medication for a month before donating blood. The blood was withdrawn into heparinized vacutainers. Three individual experimental conditions were maintained during the cytotoxicity and mutagenicity studies, with testing in the absence (−S9) and presence (+S9) of metabolic activation system (rat liver S9 homogenate) for short duration treatment period (3–6 h). Also, the evaluation was carried out in the absence (−S9) of metabolic activation system for prolonged duration treatment period (20–21 h).

During shipment of blood samples, the temperature was between 20–23°C and 22–26°C, and relative humidity was between 69%–72% and 61%–72%, for the samples received during cytotoxicity test and mutagenicity study, respectively. GOS was dissolved in distilled water and sterilized through a 0.45 mm filter. The initial chromosome aberration assay was conducted using standard procedures of Evans 1976 by exposing cultures of HPBL cells to five concentrations of GOS as well as solvent controls.¹⁷ The dose levels of GOS tested were 0.3125, 0.625, 1.25, 2.5, and 5.0 mg/mL. The dose solution concentrations were adjusted with the potency correction factor (1.347) to compensate for the purity. In the non-activated test system, treatment was for 3 h and 21 h; in the S9 activated test system, treatment time was for 3 h at 37°C in 5% CO₂.

The confirmatory assay was conducted by exposing HPBL cells to GOS at dose levels of 1.25, 2.5, and 5.0 mg/mL and to positive and solvent controls continuously for 21 h in the absence of S9 activation (Lot No. 3849 Xenometrix AG, Switzerland).

HPBL cells were exposed to the negative (sterile water) and positive (0.650 mg of ethyl methane sulfonate) controls for 3 and 21 h, in the absence of S9 (Lot. No. A307003; Spectro Chemicals Ltd. Mumbai, India) and to 0.055 mg of cyclophosphamide monohydrate (another positive control) for 3 h, in the presence of S9 (CPA, Lot. No. MKBX1822V; Sigma–Aldrich)/mL culture medium, respectively.

To ensure evaluation of first division metaphase cells, the dividing cells were arrested in the metaphase and were harvested for microscopic evaluation of chromosome aberrations at approximately 20 h after the initiation of treatment. The clastogenic potential of GOS was measured by its ability to increase structural aberrations in a dose–responsive manner compared to the solvent control group. GOS was also assessed for its ability to induce numerical chromosome aberrations. Colchicine (Lot. No. SLBQ8131V; Sigma–Aldrich) was added 2 h before the end of the experiment at a concentration of 0.2 µg/mL to stop cell division. Cells were washed, fixed with Carnoy’s solution (methanol:glacial acetic acid, 3:1), and stained with 5% Giemsa’s staining solution (SLBS0364V; Sigma–Aldrich). Structural aberrations (chromatid break and exchange, chromosome break and exchange), numerical aberrations (polyploids, more than 38 chromosomes), and endo-reduplication were determined under blinded conditions by the evaluation of 300 metaphase chromosomes (4 slides/group; 75 metaphase chromosomes per slide) from each group. Total aberrations were recorded to give incidence (%). Percent aberrant metaphases (excluding gaps) were subjected to statistical analysis. The analyses were performed using validated statistical software (GraphPad Prism 5.02). Data were subjected to a one-tailed Fisher Exact test. All analysis and comparisons were evaluated at the 5% level.

Results

Stability and dose confirmation analysis

The stability of Gossence™ (GOS) in the vehicle (purified water) was determined by analyzing major content of GOS “4′-Galactosyllactose” at Syngene International Limited using validated HPLC method. The formulations were stable in vehicle at concentrations of 100, 200, and 500 mg/mL for 7 days at 2–8°C. Homogeneity analysis was not performed since the formulation was clear solution. Dose confirmation analyses were within the acceptable limits (%variation ± 10% and %RSD < 10%) for 4′-galactosyllactose (major component of GOS) concentrations in the formulation (1000, 2000, and 5000 mg/kg) prepared for dosing on day 1, month 2 (day 36), and month 3 (day 85) of the study. The vehicle control showed no test item contamination.

14-day dose range finding study in rat

Mortality and clinical signs

Repeated oral administration of Gossence™ (GOS) at dose levels of 1000, 2000, and 5000 mg/kg/day consecutively for 14-days to Sprague Dawley rats was well tolerated without any incidence of morbidity or mortality. There were no clinical signs of toxicity in any of the animals.

Body weights

At high dose (5000 mg/kg/day), there was a slight (13%) decrease in the overall body weight gain (days 1–14) in males and this change was correlated to statistically significant reduction (14%) in the overall (days 1–14) food consumption.

Food consumption and food conversion efficiency

At 2000 mg/kg/day, food consumption on days 1–14 was slightly reduced (7%) as compared to control. The food conversion efficiency in the test item treatment groups was comparable to the control vehicle group.

Clinical pathology, organ weights, and histopathology

No test item-related changes were observed in hematology, coagulation, clinical chemistry, and urinalysis parameters. At necropsy, increased size of the cecum was observed at 5000 mg/kg (male—5 of 6, female—4 of 6) and 2000 mg/kg (males only—2 of 6). Test item–related statistically significant increase in mean absolute and relative weights of cecum (with and without cecal content) was observed in males at ≥2000 mg/kg and in females at 5000 mg/kg (data not shown in narrative). Increased cecal weights correlated with the microscopic finding of mucosal epithelial hypertrophy at 5000 mg/kg. No test item–related histopathology changes were observed in cecum at lower dose levels

90-day subchronic toxicity study in rat

Mortality, clinical signs, and ophthalmological examination

There was no mortality, clinical sign, and ophthalmic findings at any dose levels in this study.

Body weights

The mean body weight and body weight gain is presented in Figures 1 to 4. The body weight at 1000 and 2000 mg/kg/day in both sexes and 5000 mg/kg/day in females were not affected by treatment with GOS, when compared to the control group. A slight decrease in mean final body weight (day 90) was observed in males at high dose (5000 mg/kg/day) as compared to the control group, which was considered as test item related (Figure 1). This effect was not considered adverse because the magnitude of decreased body weight gain was not significant and it completely recovered and comparable to the control group at the end of recovery period (day 28).

Figure 1.

Growth curves of males in terms of mean body weight (g): G1, G2, G3, and G4 represent vehicle control, 1000, 2000, and 5000 mg GOS/kg/day, respectively. GOS: galactooligosaccharides.

Figure 2.

Growth curves of females in terms of mean body weight (g): G1, G2, G3, and G4 represent vehicle control, 1000, 2000, and 5000 mg GOS/kg/day, respectively. GOS: galactooligosaccharides.

Figure 3.

Mean males body weight gain during dosing period. G1, G2, G3, and G4 represent vehicle control, 1000, 2000, and 5000 mg GOS/kg/day, respectively. GOS: galactooligosaccharides.

Figure 4.

Mean females body weight gain during dosing period. G1, G2, G3, and G4 represent vehicle control, 1000, 2000, and 5000 mg GOS/kg/day, respectively. GOS: galactooligosaccharides.

Food consumption and food conversion efficiency

The overall food consumption (days 1–90) was comparable with the control group up to dose levels of 2000 mg/kg/day in males and up to dose levels of 5000 mg/kg/day in females (Figures 5 and 6). Statistically significant reduction in food consumption was observed in male rats at high dose (5000 mg/kg/day) on days 50–57 (13%), 57–64 (13%), 64–71 (15%), 71–78 (8%), and 85–90 (15%) as compared to the control group. The overall food consumption (days 1–90) was statistically significantly decreased to 14% as compared to the control group. Since, changes in food consumption were minimal and completely recovered at the end of recovery period (day 28) in animals administered 5000 mg/kg/day. The food conversion efficiency in the dosed groups was comparable to the control vehicle group (Figures 7 and 8).

Figure 5.

Food consumption curves of males in terms of mean food consumption. G1, G2, G3, and G4 represent vehicle control, 1000, 2000, and 5000 mg GOS/kg/day, respectively. GOS: galactooligosaccharides.

Figure 6.

Food consumption curves of females in terms of mean food consumption. G1, G2, G3, and G4 represent vehicle control, 1000, 2000, and 5000 mg GOS/kg/day, respectively. GOS: galactooligosaccharides.

Figure 7.

Food conversion efficiency of males in terms of (%) G1, G2, G3, and G4 represent vehicle control, 1000, 2000, and 5000 mg GOS/kg/day, respectively. GOS: galactooligosaccharides.

Figure 8.

Food conversion efficiency of females in terms of (%) G1, G2, G3, and G4 represent vehicle control, 1000, 2000, and 5000 mg GOS/kg/day, respectively. GOS: galactooligosaccharides.

Functional observation battery test and motor activity

No treatment-related changes observed in neurological/functional examination and motor activity parameters carried out at the end of dosing period for main groups.

Clinical pathology, organ weights, and histopathology

There were no test item–related changes in hematology, coagulation, clinical chemistry, and urinalysis parameters. At necropsy, increased size of the cecum was observed in 6 of 10 males and 7 of 10 females at 5000 mg/kg/day. Test item–related significantly increased absolute and relative weights (relative to body weight and brain weight) of cecum (with content) was observed in males at ≥2000 and females at 5000 mg/kg/day. Cecal weights (without content) were significantly increased only at 5000 mg/kg/day in both male and female rats (Tables 5 and 6), which correlated with the microscopic finding of minimal mucosal hypertrophy. No test item–related histopathology microscopic changes were observed in the cecum at lower dose levels. Organ weight, gross, and histopathology changes were not observed in the cecum of rats at the recovery sacrifice indicating complete recovery of these changes.

Table 5.

Gossence™ (galactooligosaccharides) changes in cecal weights in males.

	Cecum weight with content			Cecum weight without content
Dose (mg/kg/day)	1000	2000	5000	1000	2000	5000
Absolute weight	↑23	↑35^a	↑102^a	↓1	↑15	↑36^a
Relative to body weight	↑26	↑37^a	↑116^a	↑2	↑16	↑44^a
Relative to brain weight	↑22	↑34^a	↑101^a	↓2	↑13	↑35^a

^aStatistically significant at p ≤ 0.05. Absolute and relative weight values are presented as % increase (↑) or decrease (↓) compared to the concurrent control group.

Table 6.

Gossence™ (galactooligosaccharides) changes in cecal weights in females.

Dose (mg/kg/day)	1000	2000	5000	1000	2000	5000
Absolute weight	↑24	↑41	↑132^a	↑3	↑14	↑37
Relative to body weight	↑25	↑45	132^a	↑5	↑18	↑38^a
Relative to brain weight	↑30	↑37	↑131^a	↑8	↑12	↑37

^aStatistically significant at p ≤ 0.05. Absolute and relative weight values are presented as %increase (↑) compared to the concurrent control group.

Bacterial reverse mutation test

The Gossence™ (GOS) was soluble in water and formed clear solution at 50 mg/mL. Based on the solubility test, the highest concentration of 5000 µg/plate was selected for the preliminary cytotoxicity test.

In the preliminary cytotoxicity test, the test item was tested at concentrations of 78.125, 156.25, 312.5, 625, 1250, 2500, and 5000 µg/plate. Test item did not precipitate on the Vogel Bonner (VB) agar plates at all the tested concentrations, in the presence and absence of S9. In the tester strain TA 100, the growth of the bacterial background lawn and mean number revertant colonies were comparable to that of the respective vehicle control plates, in the presence or absence of S9. In the preliminary cytotoxicity study, the Gossence™ (GOS) was not cytotoxic up to 5000 µg/plate in tester strain TA 100, in the presence or absence of S9 (data not shown).

The initial assay was conducted using the test item concentrations of 50, 159, 501, 1582, and 5000 µg/plate, in the presence or absence of S9, using the plate incorporation treatment method. The test item did not precipitate on the VB agar plates at all tested concentrations, in the presence and absence of S9. In all the tester strains, the growth of the bacterial background lawn and mean number of revertant colonies were comparable to that of the respective vehicle control plates, in the presence or absence of S9. There was no twofold increase in the mean numbers of revertants in test strains TA 98, TA 100, and WP2 uvrA (pKM101) or threefold in test strains TA 1535 and TA 1537, at any of the tested concentrations. From the initial assay results, it is concluded that the test item was not cytotoxic or mutagenic up to 5000 µg/plate, in the presence and absence of S9 (Table 7).

Table 7.

Bacterial reverse mutation test on Gossence™ (galactooligosaccharides): Initial assay.

		Number of revertants (number of colonies/plate)
Without or with S9 mix	Test item concentration (µg/plate)	TA98			TA100			TA1535			TA1537			WP2 uvrA
Without or with S9 mix	Test item concentration (µg/plate)	Mean	SD	FI	Mean	SD	FI	Mean	SD	FI	Mean	SD	FI	Mean	SD	FI
S9 mix (+)	VC (water)	23	3	NA	124	5	NA	10	2	NA	6	3	NA	94	4	NA
	50	25	2	1	127	4	1	12	2	1	6	2	1	93	3	1
	159	24	3	1	123	4	1	11	2	1	6	2	1	93	5	1
	501	25	3	1	124	3	1	10	2	1	5	1	1	95	6	1
	1582	26	2	1	123	3	1	10	1	1	6	2	1	96	7	1
	5000	24	3	1	126	6	1	12	2	1	5	2	1	93	3	1
	PC	410	11	18	761	55	6	117	9	12	109	3	18	407	14	4
S9 mix (−)	VC (water)	25	4	NA	125	4	NA	15	2	NA	7	2	NA	98	5	NA
	50	24	2	1	122	2	1	14	2	1	6	2	1	93	4	1
	159	23	3	1	125	4	1	14	2	1	6	2	1	96	5	1
	501	24	4	1	123	3	1	16	2	1	6	2	1	99	6	1
	1582	23	4	1	126	3	1	14	2	1	7	3	1	95	3	1
	5000	24	2	1	124	5	1	14	2	1	6	2	1	98	7	1
	PC	402	14	16	667	36	5	233	19	16	151	14	22	710	19	7

VC: vehicle control; PC: positive control; S9 (−): absence of S9; S9 (+): presence of S9; SD: standard deviation; FI: fold increase.

The confirmatory assay was conducted to confirm the negative results of initial assay by varying the concentrations of test item and the method of treatment. During the confirmatory test, the test item was tested at concentrations of 99, 265, 699, 1869, and 5000 µg/plate, in the presence and absence of S9, using pre-incubation method. The test item did not precipitate on the VB agar plates at all the tested concentrations, in the presence or absence of S9. Results of confirmatory assay showed that the bacterial background lawn and mean number revertant colonies were similar as that of initial assay. It is concluded from confirmatory assay that the Gossence™ (GOS) was not cytotoxic or mutagenic up to 5000 µg/plate, in the presence or absence of S9 (Table 8). The stock solution of 50 mg/mL test item concentration was analyzed for the dose confirmation during initial and confirmatory assay. The mean percent recovery of Gossence™ (GOS) was 100.27% and 103.03% for initial and confirmatory assay, respectively. In both the assays, no peaks of test item were observed from the vehicle control samples (water) (Table 10). Under the test conditions, the results indicate that Gossence™ (GOS) is not mutagenic in the presence or absence of S9.

Table 8.

Bacterial reverse mutation test on Gossence™ (galactooligosaccharides): Confirmatory assay.

Without or with S9 mix	Test item concentration (µg/plate)	Number of revertants (number of colonies/plate)
		TA98			TA100			TA1535			TA1537			WP2 uvrA
		Mean	SD	FI	Mean	SD	FI	Mean	SD	FI	Mean	SD	FI	Mean	SD	FI
S9 mix (+)	VC (water)	20	4	NA	99	8	NA	11	2	NA	7	3	NA	92	3	NA
	50	22	3	1	102	8	1	13	3	1	6	3	1	91	2	1
	159	24	3	1	112	8	1	12	3	1	8	4	1	92	3	1
	501	21	1	1	108	9	1	10	2	1	7	4	1	95	4	1
	1582	22	4	1	115	9	1	12	2	1	10	3	1	92	3	1
	5000	20	3	1	106	8	1	12	2	1	9	3	1	96	5	1
	PC	405	19	20	730	47	7	112	12	10	112	10	16	421	37	5
S9 mix (−)	VC (water)	20	3	NA	109	7	NA	13	3	NA	9	3	NA	102	7	NA
	50	22	2	1	110	8	1	12	2	1	7	3	1	100	6	1
	159	22	4	1	112	6	1	11	3	1	10	2	1	103	8	1
	501	22	3	1	110	9	1	13	3	1	6	2	1	97	4	1
	1582	22	2	1	114	9	1	12	2	1	8	3	1	98	3	1
	5000	21	3	1	111	3	1	11	3	1	6	2	1	97	4	1
	PC	409	17	21	717	33	7	292	66	22	137	7	15	644	61	6

VC: vehicle control; PC: positive control; S9 (−): absence of S9; S9 (+): presence of S9; SD: standard deviation; FI: fold increase.

In vitro chromosomal aberration test with HPBL cells

The clastogenic potential of test item, Gossence™ (GOS), was tested in the in vitro mammalian chromosome aberration test. The Gossence™ (GOS) was evaluated for the potential to cause structural chromosomal aberration in cultured HPBL. The Gossence™ (GOS) was soluble in water and formed clear solution at 50 mg/mL. The cytotoxicity test was conducted with Gossence™ (GOS) at different concentrations (0.3125, 0.625, 1.25, 2.5, and 5.0 mg/mL) along with concurrent vehicle control (water). The Gossence™ did not precipitate in the treatment medium at all tested concentrations. The pH of the treatment media at all tested concentrations were comparable with concurrent vehicle control group. The cytotoxicity (% reduction) was determined using mitotic index (treated groups compared with vehicle control group). Under all experimental conditions, the preliminary cytotoxicity data obtained for the test item concentrations from 0.0312 mg/mL to 5.0 mg/mL were <8%. The preliminary cytotoxicity data obtained for all the test item concentrations were within the acceptable range of <50% (data not shown).

Based on the preliminary cytotoxicity test, the concentrations selected for the confirmatory mutagenicity study were 1.25, 2.5, and 5.0 mg/mL of treatment medium, along with concurrent vehicle and the appropriate positive controls. Ethyl methane sulphonate (short and prolonged duration, −S9) and cyclophosphamide monohydrate (short duration, +S9) were employed as clastogenic positive controls. All the groups were tested in duplicate. Under all experimental conditions, pH of treatment media treated with Gossence™ (GOS) was comparable to that of vehicle control treatment media at the start and end of the treatment. The Gossence™ (GOS) at all tested concentrations did not precipitate in the treatment media.

During the confirmatory study, cytotoxicity data at all tested concentrations ranged from 4% to 7% (short duration, −S9), 2% to 8% (short duration, +S9), and 3% to 10% (prolonged duration, −S9) compared with the vehicle control group. In the positive control group, the cytotoxicity ranged from 5% to 13% compared with the vehicle control group. Under all the experimental conditions, the cytotoxicity data obtained for all the test item concentrations were within the acceptable range of <50% in the mutagenicity study (Table 9).

Table 9.

Chromosomal aberration test of Gossence™ (galactooligosaccharides).

Treatment time (hour)	S9 mix	Test concentration (mg/mL)	Total number of metaphases read	Total number of normal metaphases	Total number of aberrant metaphases		Aberrant metaphases mean % (excluding gap)	Total number of aberrations mean (excluding gap)
Treatment time (hour)	S9 mix	Test concentration (mg/mL)	Total number of metaphases read	Total number of normal metaphases	Including gap	Excluding gap	Aberrant metaphases mean % (excluding gap)	Total number of aberrations mean (excluding gap)
Short duration (3–6 h)	−	VC (water)	300	298	2	0	0	0
		1.25	300	298	2	2	1	2
		2.5	300	298	2	1	0	1
		5.0	300	298	2	0	0	0
		PC (EMS—0.650 mg/mL)	300	275	25	21	7^a	24
Short duration (3–6 h)	+	VC (water)	300	298	2	0	0	0
		1.25	300	297	3	0	0	0
		2.5	300	299	1	0	0	0
		5.0	300	300	0	0	0	0
		Positive control (CPA—0.055 mg/mL)	300	271	29	22	7^a	28
Long duration (20–21 h)	−	VC (water)	300	296	4	1	0	1
		1.25	300	298	2	1	0	1
		2.5	300	296	4	0	0	0
		5.0	300	297	3	1	0	1
		PC (EMS—0.650 mg/mL)	300	267	33	21	7^a	28

VC: vehicle control; PC: positive control; EMS: ethyl methane sulfonate; CPA: cyclophosphamide monohydrate; S9 (−): absence of S9; S9 (+): presence of S9.

^aStatistically significant percent increase in aberrant metaphase when compared with vehicle control group (p < 0.05).

Under all the three experimental conditions, there was no statistically significant increase in the number of percent aberrant metaphase for the Gossence™ (GOS), when compared with the vehicle control group. As expected, statistically significant increase in the number of percent aberrant metaphase was observed for all the positive controls, when compared to the vehicle control group in the short (−S9 and +S9) and prolonged (−S9) duration treatment.

The mean percent recovery of Gossence™ (GOS) at 50 mg/mL stock concentration was 97.67%. No Gossence™ (GOS) was observed in the vehicle control sample group (water) (Table 10). It is concluded that the test item, Gossence™ (GOS), is non-clastogenic (negative) in the in vitro chromosomal aberration test using HPBL during short (absence and presence of S9) and prolonged (absence of S9) duration treatment.

Table 10.

Results of dose formulation analyses.

Bacterial reverse mutation test
Test item	Stock concentration (mg/mL)	Replicate	Recovery concentration (mg/mL)	Mean recovery concentration (mg/mL)	Mean recovery (%)
Gossence™ (galactooligosaccharides)	50 (initial assay)	R1	40.78	40.78	100.27
	50 (initial assay)	R2	40.78	40.78	100.27
	50 (confirmatory assay)	R1	41.75	41.92	103.03
	50 (confirmatory assay)	R2	42.08	41.92	103.03
Chromosomal aberration test
Gossence™ (galactooligosasccharides)	50	R1	40.43	39.73	97.67
Gossence™ (galactooligosasccharides)	50	R2	39.03	39.73	97.67

Discussion

Characterization of toxicities of test item during preclinical safety assessment is an important aspect of drug development process.

The 90-day repeat dose oral toxicity study was conducted to evaluate systemic toxicity and to determine no observed adverse effect level (NOAEL) of Gossence™ (GOS) following once daily administration at dose levels of 1000, 2000 or 5000 mg/kg/day by gavage to Sprague Dawley rats and to assess the reversibility, persistence, or delayed occurrence of toxic effects after a 28-day recovery period.

There was no test item (Gossence™)-related mortality, clinical signs, and opthalomological findings in any of the treated animals at any dose levels. Functional observation battery revealed no test item–related effects on home cage, handling, open-field, sensory, neuromuscular, and physiological observations up to the highest dose of 5000 mg/kg/day, which was consistent with the published literature.^11,18,19 At high dose level (5000 mg/kg/day), there was a slight (13%) decrease in the overall body weight gain from day 1 to day 90 in males, which correlated with a statistically significant reduction (14%) in the overall food consumption during this period. However, these parameters were comparable to concurrent controls during recovery period suggesting complete reversibility of these effects. Decreased food consumption with Gossence™ (GOS) administration is consistent with the published literature on GOS and considered as non-adverse effect because magnitude of these effects was minimal and did not affect the normal physiological activity of the animals.¹⁸

There were no test item–related changes in hematology, coagulation, clinical chemistry, and urinalysis.

There were no test item–related gross changes in any of the organs examined except for increased size of the cecum in high-dose (5000 mg/kg/day) males and females, which was associated with the increased cecal weight with and without cecal content. Increased cecal weight with content was also noted in mid-dose (2000 mg/kg/day) males with no gross change. These cecal changes were correlated with the microscopic finding of mucosal hypertrophy of the cecum in animals administered at 5000 mg/kg/day, which was completely reversed after 28-day recovery period. The change in cecum (increased weights associated with gross and microscopic changes) was considered as compensatory and adaptive response to the administration of GOS.² These findings in the cecum were consistent with earlier reports for compounds of similar class.^20

–23

Cecal enlargement along with mucosal hypertrophy has been observed as an adaptive response in several rodent species to other food ingredient such as modified starches, polyols, some fibers and lactose, and these ingredients share the feature of being poorly absorbed and osmotically active.²¹ Studies have demonstrated that consumption of pectin and nondigestible oligosaccharides can cause histological changes in cecum in rats.^22,24 The cecum is an area of bacterial fermentation, and cecal hypertrophy is thought to occur due to trophic effects of increased amounts of short fatty acids that are produced by bacterial fermentation after large amount of nonabsorbed carbohydrate and dietary fiber enter the cecum and colon.²⁵

Daily administration of Gossence™ (GOS) for 90 days via gavage to Sprague Dawley rats was tolerated well without any systemic toxicity up to the dose level of 5000 mg/kg/day; therefore, no-observed-adverse-effect level (NOAEL) was considered as 5000 mg GOS/kg/day, which is equivalent to 6735 mg Gossence™/kg/day, within the context of this study. In this study, Gossence™ (GOS) was found to be non-genotoxic in both Ames test and in vitro chromosomal aberration test with HPBL, which is consistent with published reports with similar class of compounds.¹¹ In vivo micronucleus test was negative with GOS at a doable maximum dose.² The results of the current study and published literature suggest that GOS do not have genotoxic potential.

Conclusion

In conclusion, based on the results of the 90-day study, GOS was found to be safe (non-toxic) at up to 5000 mg/kg/day in Sprague Dawley rats. The NOAEL for GOS is 5000 mg GOS/kg/day body weight/day which is equivalent to 6735 mg Gossence™/kg/day. In vitro genotoxicity tests showed no mutagenesis in the Ames assay or in Escherichia coli WP2 uvrA, and no chromosomal aberrations in cultured HPBL.

Footnotes

Acknowledgements

The authors are thankful to the management of Tata Chemicals Limited and Syngene International Limited.

Author contributions

MJ, DPT, SN, and AKD contributed to the conception or design of the work. BR, AJ, DV, and GC contributed to the data collection. BR, AJ, DV, GC, PB, and MK contributed to the data analysis and interpretation. BR, AJ, DV, GC, and CJ drafted the manuscript. All authors critically reviewed and approved the manuscript.

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The work reported in this publication was funded solely by Tata Chemicals Limited with mutual agreement with Syngene International Limited. All authors are (or were) employees of Tata Chemicals Limited or Syngene International Limited.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was funded solely by Tata Chemicals Limited with mutual agreement with Syngene International Limited.

ORCID iD

Ashok Kumar Dubey

References

Sangwan

Tomar

Singh

. Galactooligosaccharides: novel components of designer foods. J Food Sci 2011; 76: R103–R111.

Zhou

Kruger

Ravi

. Safety evaluation of galacto-oligosaccharides: subchronic oral toxicity study in Sprague-Dawley rats. Toxicol Res Appl 2017; 1: 1–12.

Hingu

Shah

. Review: role of galactooligosaccharides as prebiotic. Microbes 2013; 5: 15–29.

Lamsal

. Production, health aspects and potential food uses of dairy prebiotic galactooligosaccharides. J Sci Food Agr 2012; 92: 2020–2028.

Hsu

Lee

Chou

. Enzymatic production of galactooligosaccharides by β-galactosidase from Bifidobacterium longum BCRC 15708. J Agr Food Chem 2007; 55: 2225–2230.

Macfarlane

Cummings

. Review article: prebiotics in the gastrointestinal tract. Aliment Pharm Ther 2006; 24: 701–714.

Schaafsma

. Lactose and lactose derivatives as bioactive ingredients in human nutrition. Int Dairy J 2008; 18: 458–465.

Macfarlane

Steed

Macfarlane

. Bacterial metabolism and health-related effects of galacto-oligosaccharides and other prebiotics. J Appl Microbiol 2008; 104: 305–344.

Fanaro

Boehm

Garssen

. Galacto-oligosaccharides and long-chain fructo-oligosaccharides as prebiotics in infant formulas: a review. Acta Paediatr Suppl 2005; 94: 22–26.

10.

Boehm

Knol

Ben

. Intestinal microbiology in early life: specific prebiotics can have similar functionalities as human-milk oligosaccharides. Am J Clin Nutr 2013; 98: 561S–571S.

11.

Kobayashi

Yasutake

Uchida

. Safety of a novel galacto-oligosaccharide: genotoxicity and repeated oral dose studies. Hum Exp Toxicol 2009; 28: 19–630.

12.

Oku

. Oligosaccharides with beneficial health effects: a Japanese perspective. Nutr Rev 1996; 54: S59–S66.

13.

Salminen

Bouley

Boutron

. Functional food science and gastrointestinal physiology and function. Br J Nutr 1998; 80: S147–S171.

14.

Maron

Ames

. Revised methods for the Salmonella mutagenicity test. Mutat Res 1983; 113: 173–215.

15.

Ames

. Methods for detecting carcinogens and mutagens with the Salmonella/mammalian-microsome mutagenicity test. Mutat Res 1975; 31: 347–364.

16.

Yahagi

Nagao

Seino

. Mutagenicities of N-nitrosamines on Salmonella. Mutat Res 1977; 48: 121–129.

17.

Evans

. Cytological methods for detecting chemical mutagens. Chem mutagens, princ methods detect 1976; 4: 1–29.

18.

Anthony

Merriman

Heimbach

. 90-day oral (gavage) study in rats with galactooligosaccharides syrup. Food Chem Toxicol 2006; 44: 819–826.

19.

Kobayash

Ishida

Kaneko

. A 6-week oral gavage toxicity study of a novel galacto-oligosaccharide in juvenile rats. Hum Exp Toxicol 2014; 33: 722–728.

20.

Greaves

. In: Greaves

(ed.) Digestive System, Histopathology of preclinical toxicity studies. 4th ed. Boston: Academic Press, 2012, pp. 325–431.

21.

Haschek

Rousseaux

Wallig

. In: Haschek

Rousseaux

Wallig

(eds) Fundamentals of toxicologic pathology. 2nd ed. San Diego: Academic Press, 2010, pp. 163–196.

22.

Genta

Cabrera

Grau

. Subchronic 4-month oral toxicity study of dried Smallanthus sonchifolius (yacon) roots as a diet supplement in rats. Food Chem Toxicol 2005; 43: 1657–1665.

23.

Jain

Gote

Dubey

. Safety evaluation of fructooligosaccharide (FOSSENCE^TM): Acute, 14-day, and subchronic oral toxicity study in Wistar rats. Toxicol Res Appl 2018; 2: 1–20.

24.

Demigne

Remesy

. Stimulation of absorption of volatile fatty acids and minerals in the cecum of rats adapted to a very high fiber diet. J Nutr 1985; 115: 53–60.

25.

Levine

. Regulation of intestinal mucosal growth. In: Morisset

Solomon

(eds) Growth of the gastrointestinal tract: gastrointestinal hormones and growth factors. Boca Raton: CRC Press, 1991, pp. 176–188.