The Ramazzini Institute 13-week pilot study: aims and experimental design

The protocol of the pilot study commences with exposure from gestation day (GD) 6 (implantation) continuously through pregnancy and lactation. To satisfy the need to consider multiple effects across multiple life stages, at weaning the offspring were assigned to two testing cohorts at random, so as to have minimal differences in body weight among groups (standard deviation < 10% of the average). The first cohort (6-week cohort) was continuously dosed until full sexual maturity (Post Natal Day-PND 73 ± 2), then sacrificed. The second cohort (13-week cohort) was continuously dosed until adulthood (PND 125 ± 2), then sacrificed. Both cohorts were analyzed for post-natal developmental landmarks, microbiome, target organs toxicity and clinical pathology.

The design of the pilot study has been developed by the Ramazzini Institute in collaboration with all Institutions taking part in the overall Glyphosate Study. All of the in vivo experimental phases of the study were performed at the Ramazzini Institute, while the other collaborating Institutions have independently assessed different outcomes and endpoints of interest. In this paper, we present the study design, the first evaluation of in vivo parameters and the determination of glyphosate and its major metabolite AMPA in urine.

Experimental model

The study was conducted following the rules established by the Italian law regulating the use and humane treatment of animals for scientific purposes [Decreto Legislativo (D.Lgs.) N. 26, 2014. Attuazione della direttiva n. 2010/63/UE in materia di protezione degli animali utilizzati a fini scientifici. - G.U. Serie Generale, n. 61 del 14 Marzo 2014]. Before starting, the protocol was examined by the Internal Ethical Committee for approval. The protocol of the experiment was also approved and formally authorized by the ad hoc commission of the Italian Ministry of Health (ministerial approval n. 710/2015-PR). The experiment was performed on both male and female SD rats, which belong to the colony used at the Cesare Maltoni Cancer Research Center laboratories of the Ramazzini Institute (CMCRC/RI) for over 40 years. An animal disease screening program enforced by the Italian Health Authority and Research Organization for Animal Health is in place and ongoing on sentinel animals belonging to the RI colony.

Female breeders SD rats were placed individually in Polycarbonate cage (42x26x18cm; Tecniplast Buguggiate, Varese, Italy) with a single unrelated male until evidence of copulation was observed. After mating, matched females were housed separately during gestation and delivery. Newborns were housed with their mothers until weaning. Weaned offspring were housed, by sex and treatment group, not more than 3 per each cage. Cages were identified by a card indicating: study protocol code, experimental and pedigree numbers, dosage group. A shallow layer of white fir wood shavings served as bedding (supplier: Giuseppe Bordignon, Treviso, Italy). Analysis of chemical characteristics (pH, ashes, dry weight, specific weight) and possible contamination (metals, aflatoxin, polychlorobiphenyls, organophosphorus and organochlorine pesticides) of the bedding was performed by CONSULAB Laboratories (Treviso, Italy). The cages were placed on racks, inside a single room prepared for the experiment at 22 °C ± 3 °C temperature and 50 ± 20% relative humidity. Daily checks on temperature and humidity were performed. The light was artificial and a light/dark cycle of 12 h was maintained.

During the experiment SD rats received ad libitum the standard “Corticella” pellet feed supplied by Laboratorio Dottori Piccioni Srl (Piccioni Laboratory, Milan, Italy). The constituents of the diet are: ground corn (23%), barley milled (15%), soybean meal extract (20.6%), wheat middling (24%), wheat bran (2%), spray dried whey (2.5%), di-calcium phosphate (2%), calcium carbonate (1.1%), chicken meal (6%), carob bean gum (3%), sodium chloride (0.5%), mixed vitamins (0.3%). Every day, the animals drank fresh municipal tap water from glass bottles ad libitum. Both feed and water were periodically analyzed to identify possible chemical or microbiological contaminants or impurities; the analyses are included in the documentation of the experiment. The pelleted feed was tested for possible glyphosate contamination in compliance with Commission Regulation (EU) No 293/2013 [maximum residue levels (MRLs) < 1 mg/kg]. Tap drinking water was tested for possible glyphosate contamination in compliance with Directive 2008/105/EC, D.Lgs. 152/2006, Directive2006/118/EC (active substances in pesticides, including their relevant metabolites, degradation and reaction products < 0.1 μg/l).

Active ingredient glyphosate (Pestanal™ analytical standard, CAS number 1071–83-6, purity > 99,5%) was supplied from Sigma-Aldrich (Milan, Italy). The commercial formulation Roundup Bioflow (containing 360 g/L of glyphosate acid in the form of 480 g/l isopropylamine salts of glyphosate (41.5%), water (42.5%) and surfactant (16%; chemical name, CAS number and/or exact percentage have been withheld as a trade secret) was supplied from a local agricultural consortium (Consorzio Agrario dell’Emilia, Bologna, Italy). The original containers/bottles of glyphosate and Roundup were stored in its original container and kept in a ventilated storage cabinet at room temperature (22 °C ± 3 °C) throughout the study. Purity data for each batch of glyphosate and Roundup were provided by the supplier. The opening and the use date of the different batches of test substances were recorded in the raw data. An aliquot of each lot of the test article is maintained in the ventilated storage cabinet, until 5 years from the end of the main experiment. The solutions of glyphosate and Roundup were prepared by the addition of appropriate volume of tap drinking water.

Experimental plan

Each of twenty-four virgin female SD rats (17 weeks old, 270-315 g) was cohabited outbred with one breeder male rat of the same age and strain. Every day, the females were examined for presence of sperm. Gestational day (GD) 0 was defined as the one in which the sperm was found in vaginal smears. The day on which parturition was completed was designated as lactating day (LD) 0 for the dam and PND 0 for the offspring. Each dam and delivered litter was co-housed in common nesting box during the postpartum period. Following the NTP MOG design, on PND 28, thus 28 days after the last litter was delivered, the offspring were weaned and identified by ear punch according to the Jackson Laboratory system. Sequentially, they were allocated in the same treatment group of their mother in order to have 18 males (8 for the 6-week cohort and 10 for the 13-week cohort) and 18 females (8 for the 6-week cohort and 10 for the 13-week cohort) for each dose group. No more than 2 males and 2 females from the same litter were included in the same cohort/treatment group. Altogether, 108 SD rats (54 males and 54 females) were enrolled in the post-weaning treatment phase. The experimental plan of the pilot study is outlined in Table 1. A summary of the endpoints and relative monitoring time points evaluated in the pilot study, both in dams and in the offspring (6-week and 13-week cohorts) is presented in Table 2.

Two groups of SD rats were treated with either glyphosate or Roundup diluted in tap water administered ad libitum and one group received only tap water as control. Roundup was diluted in tap water in order to obtain an equivalent dose of glyphosate of 1.75 mg/kg bw/day. During gestational and lactational periods, embryos and newborns (F1) received the test compounds mainly through their dams (F0). Glyphosate and Roundup water formulations during these periods were freshly prepared on a daily base depending on individual body weight and water consumption of dams as measured at each scheduled time point (see below). After weaning, until the end of the experiment (PND 73 ± 2 or 125 ± 2), the test substances were administered in tap water to F1 animals on the basis of the average body weight and average water consumption per sex and per experimental group, as measured at each scheduled time point (see below). Males and females were considered separately because of their difference in weight gain, body weight and water consumption.

At least every week, the exposure doses were recalculated and registered. The actual levels of test compounds that reached the fetus during gestation or that were ingested postnatally by the offspring during the period of lactation were not estimated in the present study.

Animals were monitored during the entire experimental period. The following procedures were performed:

Health status control: from the start of the experiment, animals were checked three times daily, except on Sundays and non-working days, when they were only checked twice. All observed variations from normal status were recorded.

Clinical control: status, behavior and clinical observation on the experimental animals were checked before the start of the treatment, and at least every two days until the end of the experiment. Any findings listed below were then recorded: alterations of skin, hair, eyes and mucosa; modification in production of secretions or excretions and in autonomic activity; respiratory symptoms; postural changes or changes in walk; presence of tonic or clonic contractions; unusual stereotypes and behavior.

Dams’ body weights were recorded on GD 0, 3, 6 and then daily during gestation until parturition. During lactation, dams’ body weights were recorded at LD 1, 4, 7, 10, 13, 16, 19, 21 and 25 (last measurement before weaning). Pups’ body weight by sex and litter was determined on PND 1, 4, 7, 10, 13, 16, 19, 21 and 25. After weaning, the body weight was measured twice a week, until PND 73 ± 2, then weekly until PND 125 ± 2 and before terminal sacrifices; the means of individual body weights were calculated for each group and sex.

Dams’ feed and water consumption were recorded twice weekly during gestation (GD 0, 3, 6, 9, 12, 15, 18, 21), whereas during lactation were measured at LD 1, 4, 7, 10, 13, 16, 19, 21, 25 and 28.

After weaning the daily water and feed consumption per cage were measured twice a week, until PND 73 ± 2, then weekly until PND 125 ± 2; the means of individual consumptions were calculated for each group and sex.

The day before the terminal sacrifices, all the animals were located individually in metabolic cages and starved for around 16 h. During this time, the animals had free access to water alone or to the programmed test compound solutions. The day after, in the morning, samples of at least 5 ml of spontaneous urine from each animal were collected and put in separate labelled tubes. Urine samples for analysis of glyphosate and AMPA excretion were obtained from 3 dams/group and from 10 (5 males + 5 females) rats/group belonging to the 6-week and 13-week cohorts.

Glyphosate and aminomethylphosphonic acid (AMPA) detection

Analyses of glyphosate and its metabolite AMPA in drinking water, feed and urine were performed by Neotron Laboratories (Modena, Italy), an officially accredited laboratory by Accredia (Lab. N. 0026) according to European regulation UNI CEI EN ISO/IEC 17025:2005. The specification and results are maintained in the experimental documentation. The analytical method is based on liquid chromatography tandem mass spectrometry (LC-MS/MS) [42–45]. The limit of quantification (LQ) for glyphosate and AMPA corresponded to 0.10 μg/l in water, 50 μg/kg in feed, and 1 μg/kg in urine.

Statistical analysis

Summary statistics, means ± standard deviations (sd), were calculated for continuous variables. For body weight, water and feed consumption over time further analyses were performed using multilevel mixed-effect linear regression models, to control for within subject correlation across time; moreover we have considered also the litter effect during the lactation period. Analysis of variance and Dunnett’s tests (when applicable) were also performed to compare body weight gain in different periods and consumption of food and water as mean consumption in several periods.

All tests were two tailed, with alpha set at 0.05. Statistical analyses were perfomed by using STATA version10 (Stata Corporation, College StationTexas, USA).