EPA REGISTRATION DIVISION COMPANY NOTICE OF FILING FOR PESTICIDE PETITIONS PUBLISHED IN THE FEDERAL REGISTER  

EPA Registration Division contact: Marion Johnson (703) 305-6788

Pesticide Petition #6E8511

	EPA has received a pesticide petition (6E8511) from Interregional Research Project Number 4 (IR-4), 500 College Road East, Suite 201 W, Princeton, NJ 08540 proposing, pursuant to section 408(d) of the Federal Food, Drug, and Cosmetic Act (FFDCA), 21 U.S.C. 346a(d), to amend 40 CFR part 180 by establishing a tolerance for residues of sulfur dioxide, including its metabolites and degradates, in or on fig at 25 ppm.  Compliance with the tolerance level is to be determined by measuring only those sulfite residues convertible to sulfur dioxide, expressed as sulfur dioxide in or on figs. EPA has determined that the petition contains data or information regarding the elements set forth in section 408 (d)(2) of  FDDCA; however, EPA has not fully evaluated the sufficiency of the submitted data at this time or whether the data supports granting of the petition. Additional data may be needed before EPA rules on the petition.

A.	Residue Chemistry                                        

 Plant metabolism   

The nature of the residue in grapes is adequately understood.  Grapes were treated with radiolabeled sulfur dioxide once a week for 16 weeks.  The main residues in or on grapes treated with sulfur dioxide were sulfite and sulfate.  Sulfate is a natural component in the human body required for biosynthesis of sulfur-containing compounds.  Sulfate is found in food (as a food additive) and in drinking water at levels that may exceed 100 ppm.  Sulfate is not considered a residue of concern since it is ubiquitous in nature, is an important nutrient, and found in low levels in grapes.  Although sulfites readily biodegrade, are quickly oxidized, are rapidly excreted from the body, sulfite is considered a residue of concern.    

 Analytical method
   
An analytical enforcement method, the Monier-Williams Procedure for Sulfites (21 CFR Part 101 Appendix A), is available for enforcement of tolerances for sulfites in food.  Minor method modifications were made to improve the performance of the method.  This method involves extraction of fig samples with boiling hydrochloric acid.  This extraction converts sulfite residues to sulfur dioxide.  The sulfur dioxide is collected in a hydrogen peroxide solution which oxidizes the sulfur dioxide to sulfuric acid.  The sulfuric acid residues are quantified by titration.  The residues are expressed as sulfur dioxide.  Method suitability was evaluated both prior to fig sample analysis and also concurrently with sample analysis.  The lowest level of method validation (LLMV) was 25 ppm.  Recoveries at this level were in the ranges 38 to 121% (average recovery =  67+16%) for sulfur dioxide.  The LOD for sulfur dioxide was 10.3 ppm and the LOQ was 30.9 ppm. 

 Magnitude of residues
   
Four post-harvest trials were conducted with mature figs using either sulfur dioxide gas, sulfur dioxide slow-release pads, or a combination of sulfur dioxide gas followed by slow-release pads.  

Commercially mature figs were collected from the field and transported to the post-harvest treatment site.  The figs for fumigation were either stored in a cold storage room or in coolers with ice or held at ambient temperature until treatment the following day.  The figs for pad treatment were placed in cold storage and treated within 1 day.  

Prior to fumigation, the figs were removed from storage and placed in open fig trays that fit inside the fumigation chamber.  In two trials, the figs were treated with sulfur dioxide gas to provide a concentration over time (CT) of 25 ppm-hour in one trial and 250 ppm-hour (10X exaggerated rate) in a second trial.  The concentration of sulfur dioxide was measured each minute, and when an overall CT near 25 ppm-hour (or 250 ppm-hour) was reached, the chamber was exhausted and the figs were removed.

Prior to pad treatment, the figs were placed in trays inside typical fresh fig boxes.  A sulfur dioxide slow release pad was placed on top of the figs, ensuring the pad covered the entire upper surface of the fig box. After pad placement, the boxes were put in cold storage.  In a second trial, figs were fumigated using sulfur dioxide gas and then the figs were put in fig boxes and a slow release pad was placed shiny side up on top of the figs. 

No residues above the LLMV of 25 ppm were observed in figs from any of the treatments.

B.	Toxicological Profile 

Evaluations performed by the World Health Organization (WHO), the International Agency for Research on Cancer (IARC), and the Agency for Toxic Substances and Disease Registry (ATSDR) were relied upon by EPA for the safety finding for sulfur dioxide made in the May 2007 RED assessment on inorganic sulfites, which includes the chemicals sulfur dioxide and sodium metabisulfite.  These assessments are based on peer-reviewed evaluations performed by the Cosmetic Ingredient Review (a program established in 1976 by the Cosmetic, Toiletry & Fragrance Association, now known as the Personal Care Products Council (PCPC)), with the support of the U.S. Food and Drug Administration (FDA) and the Consumer Federation of America (CFA); the Organization for Economic Cooperation and Development-Screening Information Data Set and from other open literature sources.  Additional information summarized below is based on public literature data.

People may be exposed to small amounts of sulfur through the food supply. However, since sulfur does not cause any relevant toxic effects, no quantitative dietary risk assessment is needed. Short-term studies show that sulfur is of very low acute oral toxicity and does not irritate the skin (it has been placed in Toxicity Category IV, the least toxic category, for these effects). 

Sulfur dioxide (21 CFR 182.3862) is listed as Generally Recognized as Safe (GRAS) by the FDA as a preservative in certain foods. The Select Committee on GRAS Substances concluded that:
``There is no evidence in the available information on sulfur dioxide that demonstrates, or suggests reasonable grounds to suspect, a hazard to the public when used at levels that are now current and in the manner now practiced.'' This conclusion was based on the knowledge that orally administered sulfite is very rapidly oxidized to sulfate in all species studied. The metabolic removal of sulfite appears to be the critical defense mechanism. The WHO has emphasized the use of appropriate labeling for alerting individuals who cannot tolerate sulfites. After receiving and reviewing reports of adverse reactions in certain individuals following ingestion of sulfiting agents used as preservatives in food products, beverages, and fresh fruits and vegetables, the FDA requires ingredient labels to list sulfite concentrations in excess of 10 ppm."  

Several regulatory endpoints and standards for ambient air concentrations of sulfur dioxide have been established at the state, Federal and international levels. The endpoint selected by the Agency for the bystander inhalation risk assessment is 0.25 ppm sulfur dioxide, with one-hour exposure duration. The 0.25 ppm concentration is based on an ambient air quality standard set by the California Air Resources Board. This endpoint is deemed most applicable to this exposure scenario, as it is based on effects of concern for bystanders (such as bronchoconstriction, shortness of breath, wheezing, and chest tightness during physical activity in persons with asthma).

Sodium metabisulfite (21CFR §182.3766) is listed as Generally Recognized as Safe (GRAS) by the FDA as a preservative in certain foods.  Sodium metabisulfite is also used up to a concentration of 1% as an antioxidant in hair care products and as a reducing agent in cosmetic formulations (CIR 2003). Sources of sulfur dioxide include the combustion of fossil fuels, smelting of sulfide ores, volcanic emissions, and other natural sources. Sulfur dioxide is also used to manufacture hydrosulfites, to bleach wood pulp and paper, to process, disinfect, and bleach food, for waste and water treatment, in metal and ore refining, and in oil refining.

1.  Acute Toxicity

Sulfur dioxide and sulfites (sodium metabisulfite) have different toxicities.  A large amount of published data exists on the toxicity of sulfur dioxide, sodium bisulfite, and sodium metabisulfite in animals and humans.  

a.  Sulfur dioxide:  No acute oral, acute dermal, eye and dermal irritation and dermal sensitization data are available for sulfur dioxide.  Groups of 8 male CD outbred rats were exposed for 4 hours via the inhalation route to sulfur dioxide at concentrations of 224, 593, 965, 1168 or 1319 ppm.  There were no deaths at 224 and 593 ppm.  In the 965 ppm group, 3 out of 8 rats died.  At 593 ppm, 5 out of 8 rats died within 1 to 48 hours after termination of exposure.  All rats in the 1319 ppm group died within 1 to 24 hours after termination of exposure.  The acute oral LC50 of sulfur dioxide was >965 and less than 1168 ppm under the conditions of this study.  Sulfur dioxide acts mainly as an irritant.  Contact with moisture in the mucous membranes converts sulfur dioxide to sulfurous acid, which is the direct irritant.  The lungs are the most sensitive organ. Lung effects can be observed after exposure of 5-10 minutes. The most sensitive endpoints are increased airway resistance (sRaw) and decreased forced expiratory volume in 1 second (FEV1). Asthmatics are the group that is most sensitive to sulfur dioxide exposure, and exercise exacerbates this sensitivity.  Effects have been seen in asthmatics at sulfur dioxide concentrations as low as 0.2 ppm.  Exposure of the general population to 20 ppm sulfur dioxide can lead to symptoms similar to an asthmatic.  Eye and skin irritation may occur at this higher concentration.    

b.  Inorganic Sulfites:  The acute oral LD50 of sodium sulfite is >3,560 mg/kg bw in rats and 820-920 mg/kg bw in mice.  Acute inhalation of sodium sulfite aerosols caused bronchoconstriction in guinea pigs (LOAEL 0.204 mg/m[3]).  Sodium sulfite was not irritating to the skin or eyes in OECD TG 404 and 405 studies.  No dermal sensitization studies in animals are available but in some humans, sensitization from topical contact has been reported.  Some symptoms observed in sulfite-sensitive humans after exposure via oral ingestion of sulfites include flushing, tingling, pruritis, dysphagia, chest pain, urticaria, angioedema, acute broncho-constriction, hypotension, anaphylactic shock, and loss of consciousness.  Broncho-constriction appears to be the most commonly reported adverse reaction to sulfites in a sub-group of asthmatic subjects. Data indicate that the current regulatory standard of 10 ppm of sulfites in/on grapes and figs is protective, even for the sensitive population. Sulfite is present in the human body as a normal metabolite and intermediate of sulfur-containing amino acids, from metabolism of sulfur dioxide inhaled via polluted air, and from ingestion of sulfite-containing agents used widely in foods and beverages.  

c.  Sodium metabisulfite:  The acute oral LD50 of sodium metabisulfite in rats is reported to be 1540 mg/kg in one study and between 1500 and 2250 mg/kg in a second study.  The acute dermal LD50 is > 2000 mg/kg.  No acute inhalation data has been reported.   Sodium metabisulfite is not a skin sensitizer, is not a skin irritant, and is irritating to the eyes.

2. Subchronic Toxicity

a.  Sulfur dioxide: Male, Sprague-Dawley rats (30/group) were exposed to sulfur dioxide at concentrations of 0, 300 and 400 ppm by whole body inhalation exposure for 3 hours/day, 5 days/week for 3, 4 or 6 weeks.  Both exposure concentrations caused a generalized increase in Goblet cells which reached at maximum after 3 weeks of treatment.  The number of tracheal-bronchial Goblet cells was doubled in the 300 ppm group compared with controls.  The greatest density was observed in the proximal intra-pulmonary airways, which according to the study authors is normally devoid of Goblet cells.  In the 400 ppm group, some tracheal epithelial stripping, decreased number of tracheal Goblet cells and only marginally affected density of the bronchial Goblet cells was observed.  In this group, maximal Goblet cell hyperplasia was seen in the distal bronchial airways which decreased with duration of exposure.  After 15 days of exposure, both groups had significant amounts of free, airway mucus which decreased with the duration of exposure. 

Male, Sprague-Dawley rats (70/group) were exposed to sulfur dioxide at concentrations of 0 (filtered air), 10 and 30 ppm by whole body inhalation exposure for 6 hours/day, 5 days/week for 21 weeks.  Mean measured concentrations were 10.1 and 29.9 ppm.  After 21 weeks of treatment, 3/14 rats in the 30 ppm group had Goblet-cell hyperplasia of the trachea and 2/14 rats each in the 10 and 30 ppm groups had tracheal epithelial hyperplasia compared to none of these lesions in the control group.  In the lung at 21 weeks, 4/14 rats had bronchial hyperplasia, 6/14 and 14/14 had mucoid degeneration in the 10 and 30 ppm groups, respectively, and 8/14 and 14/14 in the 10 and 30 ppm groups, respectively had desquamation of the bronchial epithelium compared to 0/14 of each of these lesions in the control group.  After 4 weeks of treatment, increased incidences of Goblet-cell hyperplasia of the trachea and bronchial hyperplasia, mucoid degeneration and desquamation of the bronchial epithelium in the lung were increased compared to the control group.  Relative liver, kidney and lung weights were unaffected by treatment; however, relative testes weights were statistically increased compared to the controls at 30 ppm.

Groups of Hartley albino guinea pigs (20/sex/group) were exposed to sulfur dioxide concentrations of 0, 0.13, 1.01 or 5.72 ppm by whole body inhalation exposure for approximately 22 hours/day, seven days/week for 3 months.  Mean measured concentrations of sulfur dioxide in each group during the study were measured.  There were no effects on mortality, body weights, measured hematology and clinical chemistry parameters or lung function tests.   Animals in the high dose group had a lower incidence and severity of spontaneous lung disease.  In the 5.72 ppm group, an increase in incidence and severity of hepatocytic cytoplasmic vacuolation was observed when compared to the control group.  The effect on the liver could be an adaptive effect at this level of sulfur dioxide; however, in rats sulfur dioxide has been shown to decrease CYP1A1 and 1A2 in lungs and livers of rats and to inhibit ethoxyresorufin O-demethylase (EROD) and methyoxyresorufin O-demethylase (MROD) activities at 28 and 56 mg/m3 (equivalent to 10.7 and 21.4 ppm, respectively).   The NOAEL is considered to be the high dose 5.72 ppm and the NOEL 1.01 ppm based on liver histopathology.

b. Sodium metabisulfite:  Male Wistar rats (10/group) were fed diets containing sodium metabisulfite at dietary levels of 0, 0.5, 1.0, 2.0, 6.0 and 8.0% for up to 8 weeks.  No clinical signs of toxicity were observed at dietary levels up 2.0%.  Diarrhea was observed at the 4% level.  Body weights were statistically decreased at the high dose by the end of study weeks 4 and 8.  Although not indicated as statistically significant, food consumption and food efficiency were decreased in male rats of the high dose group.  Hemoglobin, hematocrit and erythrocytes were statistically decreased in the groups fed dietary levels of 2.0 and 6.0%.  Erythrocytes of animals in the 6% sodium bisulfite group showed hypochromasia, polychromasia and anisocytosis.  Occult blood was present in the feces of animals fed diets at >1% but did not persist in the chronic portion of the study at this dose level.  Liver weights were unaffected by treatment with sodium metabisulfite; however, spleen weights were increased in males fed diets containing 4% sodium metabisulfite or higher. 

Microscopically, hyperkeratosis and acanthosis was observed in the forestomach at >1%.  Ulcers and papillomatous elevations with massive hyperkeratosis occurred in animals in the 6 and 8% dietary groups.  Hemorrhagic erosions, necrosis of surface and glandular cells and cellular inflammatory infiltrates were found in the glandular stomach of the 4, 6 and 8% groups fed metabisulfite for 28 and 56 days.  An increase in hematopoietic activity was observed in the spleen at the same dietary levels (4, 6 and 8%).  The NOAEL can be considered to be a dietary level of 1% in the diet (~500 mg/kg/day) based on the anemia observed at 2% in the diet (~1000 mg/kg/day) in the diet and higher.  The NOEL is 0.5% (~250 mg/kg/day) based on microscopic lesions of the forestomach and fecal occult blood resulting from the forestomach lesions at 1% dietary levels and above.

3.  Developmental and Reproductive Toxicity

a.  Sulfur dioxide:  In a non-guideline study, pregnant CF-1 mice or New Zealand white rabbits were exposed to 25 and 70 ppm sulfur dioxide, respectively, by whole body inhalation for 7 hours/day from days 6 through 15 of gestation for mice or from gestation days 6 through 18 for rabbits.  A filter air-exposed group was included as the control for each species.  No mortality or clinical signs of toxicity were observed and no effects on body or liver weight were noted.  In a preliminary study, a decreased in body weight gain was observed at these doses levels in mice and rabbits.  Food consumption was statistically reduced early in the treatment period for mice and rabbits, but data were not shown.  No effects were observed on the reproductive parameters.  Fetal body weights were slightly (<5%), but statistically, decreased in mice, but not in rabbits.  There was no statistical increase in any external, visceral or skeletal findings in sulfur dioxide treated mice or rabbits.  The maternal NOAEL in this study is considered to be less than 25 ppm in mice and less than 70 ppm in rabbits based on reduced food consumption and effects on body weight observed in the probe study.  The developmental NOAEL is considered to be 25 ppm in mice and 70 ppm in rabbits.  The decrease in body weight in fetal mice was not considered to be treatment related because it was so slight (1.05 versus 1.00 g for controls versus the treated group) and even conservatively can be considered very close to a NOAEL.

Reproductive effects were not observed in rats exposed to 5-30 ppm sulfur dioxide for a period from 9 days prior to mating until 12-14 days of pregnancy; or in mice exposed to 25 ppm sulfur dioxide 7 hours/day on gestation days 6-15; or in rabbits exposed to 70 ppm sulfur dioxide 7 hours/day on gestation days 6-18 (ATSDR).

b.  Sodium bisulfite:  Wistar-derived female rats were administered doses of 0, 1.0, 5.0, 24.0 or 110 mg/kg/day sodium bisulfite by intubation in water from gestation day 6 through 15.  There were 21, 23, 24, 22 and 23 pregnant rats that were evaluated in the control, 1.0, 5.0, 24.0 and 110 mg/kg/day groups, respectively.  Statistics were apparently not done on the data obtained. There were no dam deaths or abortions.  Body weight and body weight gain were unaffected by treated as were reproductive parameters and fetal body weight.  No external or visceral findings were reported.  A small increase in the incidence of wavy ribs was observed at the high dose compared to the control group on a fetal and litter basis (approximately a doubling).  There was no increase in this finding on a litter basis in the other test groups.  There was also an increase in the incidence of incomplete closure of the skull at the 1.0, 5.0 and 110 mg/kg/day doses when compared to the control group, which was not dose related.  The study author concluded that:  "[t]he administration of up to 110 mg/g of the test material to pregnant rats for 10 consecutive days had no clearly discernible effect on nidation or on maternal or fetal survival.  The number of abnormalities seen in either soft or skeletal tissues of the test groups did not differ from the number occurring spontaneously in the sham-treated controls."  Therefore, the NOAEL for maternal and developmental toxicity is considered to be 110 mg/kg/day.

Dutch-belted rabbits were administered doses of 0, 1.0, 4.64, 21.6 or 100 mg/kg/day sodium bisulfite by intubation in water from gestation day 6 through 18.  There were 14, 11, 13, 10 and 13 pregnant rabbits that were evaluated in the control, 1.0, 4.64, 21.6 and 100 mg/kg/day groups, respectively.  There were two deaths and/or abortions in the 21.6 mg/kg/day group and one in the 100 mg/kg/day group.  Statistics were apparently not done on the data obtained.  A slight decrease in body weight gain (GD 6 to 18) occurred in the 1.0 and 100 mg/kg/day groups.  Fetal body weight were also decreased in the 1.0 and 100 mg/kg/day groups.  The average number of corpora lutea/dam, number of live litters, the number of live fetuses/dam, sex ratio and fetal survival were unaffected by treatment.  The number of implantation sites/dam and fetal body weights were slightly decreased in the 1.0 and 100 mg/kg/day groups.  No treatment-related external, visceral or skeletal findings were observed.  The NOEL for maternal and developmental toxicity could conservatively be considered to be 21.6 mg/kg/day based on the minimal findings observed at 100 mg/kg/day; however, because these effects were minimal, the NOAEL is considered to be 100 mg/kg/day, the highest dose tested.  The study author concluded that:  "[t]he administration of up to 100 mg/g of the test material to pregnant rabbits for 13 consecutive days had no clearly discernible effect on nidation or on maternal or fetal survival.  The number of abnormalities seen in either soft or skeletal tissues of the test groups did not differ from the number occurring spontaneously in the sham-treated controls.

Outbred CD-1 female mice were administered doses of 0, 2, 7, 32 or 150 mg/kg/day sodium bisulfite by intubation in water from gestation day 6 through 15.  There were 20, 21, 21, 21 and 21 pregnant mice that were evaluated in the control, 2, 7, 32 and 150 mg/kg/day groups, respectively.  Statistics were apparently not done on the data obtained.  There were no deaths or abortions.  There were no effects of treatment on dam body weights or body weight gain or on fetal body weights.  Reproductive parameters were unaffected in the treated groups compared to the control group, except for a doubling of dead pups at the high dose (3 dead in the control group and 6 at the high dose).  No external or visceral findings were reported.  An increased in the incidence of sternebra and hyoid findings was reported for the treated groups on a fetal basis only.  The increase in these findings was not dose related.  Because the findings were not dose related and are common findings in this strain of mice the increase in the treated groups was not considered to be treatment related.  The NOAEL for maternal and developmental toxicity is considered to be 150 mg/kg/day, the highest dose tested.  The study author concluded that:  "[t]he administration of up to 100 mg/g of the test material to pregnant mice for 10 consecutive days had no clearly discernible effect on nidation or on maternal or fetal survival.  The number of abnormalities seen in either soft or skeletal tissues of the test groups did not differ from the number occurring spontaneously in the sham-treated controls."

c.  Sodium metabisulfite:  Developmental studies reported by the CIR 2003 indicate sodium metabisulfite produced no adverse findings, either maternal or fetal, in mice up to 160 mg/kg in a water solution, in rats up to 110 mg/kg in the diet, in hamsters up to 120 mg/kg in the diet, or in rabbits up to 123 mg/kg in the diet. These results are supported by developmental information reported by WHO (1999) in which no effects were observed on implantation, or on maternal or fetal survival in sodium metabisulfite doses of up to 150, 110 and 120 mg/kg bw in mice, rats, and hamsters, respectively (WHO Series 18). 

Male and female Wistar rats (F0) were fed diets containing 0, 0.125, 0.25, 0.5, 1.0 and 2.0% for 21 weeks prior to mating and throughout three generations.  There was no effect on fertility, birth weight of pups in the first and second generations or pup mortality.  During the study, rat were given diets that had been stored for a mean of 1 week at -18oC; diets were changed daily.  Mean pup body weights were statistically decreased at >0.5% sulfite in the diet; however, the effect was not dose-related.  Postnatal day (PND) 8, pup body weights were statistically decreased at the high dose in the F1a generation, at all dietary doses in the F2a generation and at 1.0 and 2.0% sulfite in the F2b generation. On PND 21, pup body weights were statistically decreased at the high dose in the F1a generation, at 1.0 and 2.0% sulfite in the F2a and F2b generations and at all dietary dose levels in the F3a generation.  Other sporadic statistically significant decreased in pup body weights were also observed.  No effects on endocrine or reproductive organ weights were observed nor were there any treatment related histopathology findings in these organs.  The NOAEL for reproductive toxicity in the sodium metabisulfite SIDS document was considered to be 2% in the diet which was equivalent to 942 mg/kg/day; however, this reviewer considers the NOAEL to be 0.5% in the diet or 217 mg/kg/day based on effects observed on pup body weights at PND 8 in the F1b, F2a and F2b generations.  The effects on day 21 were not considered because the pups would be eating diets at that time at a much dose than administered to the adults.  The effects on PND 8 in the F1b generation at all dose levels were not dose related, were variable and not seen at the lower dose levels in other generations.  Therefore, the lower body weights in the lower doses (<0.5%) were not considered relevant for setting a NOAEL.

d.  Potassium metabisulfite:  Virgin adult, female Wistar-derived rats were administered potassium metabisulfite by gavage in water at dose levels of 0, 1.55, 7.19, 33.4 or 155.0 mg/kg/day from gestation day 6 through 15.  There were a total of 23, 20, 20, 20 and 20 pregnant mice in the 0, 1.55, 7.19, 33.4 and 155.0 mg/kg/day groups, respectively.  No dams died on study.  Body weight gain was reduced approximately 20% at the high dose.   Reproductive parameters were unaffected in the treated groups compared to the control group.  Fetal body weight were comparable to those of the control group.  No treatment-related effects were observed upon external, visceral or skeletal fetal examination.  Based on the 20% reduction in maternal body weight gain observed at the high dose, the NOAEL for maternal toxicity is considered to be 33.4 mg/kg/day.  The NOAEL for developmental toxicity is the highest dose tested or 155 mg/kg/day.  The study authors concluded that:  "[t]he administration of up to 155 mg/g of the test material to pregnant rats for 10 consecutive days had no clearly discernible effect on nidation or on maternal or fetal survival.  The number of abnormalities seen in either soft or skeletal tissues of the test groups did not differ from the number occurring spontaneously in the sham-treated controls."

Virgin adult, female CD-1 outbred mice were administered potassium metabisulfite by gavage in water at dose levels of 0, 1.25, 5.47, 26.9 or 125.0 mg/kg/day from gestation day 6 through 15.  There were a total of 22, 22, 22, 24 and 21 pregnant mice in the 0, 1.25, 5.47, 26.9 and 125.0 mg/kg/day groups, respectively.  No dams died on study, and there were no effects on dam body weight or body weight gain.   Reproductive parameters were unaffected in the treated groups compared to the control group values.  Fetal body weight were comparable to those of the control group.  No treatment-related effects were observed upon external or visceral fetal examination.  An increase in the fetal incidence of sternebrae unossified was observed in all dose groups on a fetal basis; however, there was no dose response and the highest incidence was at the mid dose.  The incidence of this finding in the mouse study conducted on sodium bisulfite summarized above was 80 fetuses out of 19 litters.  The incidence of missing hyoid was also increased, primarily at the high dose.  The response was similar to that observed in the sodium bisulfite mouse developmental toxicity study summarized above.  The incidence of this finding in the control group of that study was 29 fetuses out of 14 litters.  As for the sodium bisulfite study, because the findings were not dose related and are common findings in this strain of mice, the increase in the treated groups is not considered to be treatment related.  The study authors concluded that:  "[t]he administration of up to 125 mg/g of the test material to pregnant mice for 10 consecutive days had no clearly discernible effect on nidation or on maternal or fetal survival.  The number of abnormalities seen in either soft or skeletal tissues of the test groups did not differ from the number occurring spontaneously in the sham-treated controls."  Therefore, the NOAEL for both maternal and developmental toxicity is considered to be 125 mg/kg/day.

4. Carcincogenicity.  

a.  Sulfur dioxide:  IARC evaluated information on the potential carcinogenicity of sulfur dioxide, sulfite, bisulfites and metabisulfites and classified these compounds as Group 3 for carcinogenicity (not classifiable as to their carcinogenicity in humans (1992). No adverse effect was reported at any dose on the musculoskeletal, hepatic, renal, or endocrine systems or body weight in humans. 

Three month old male and female LX mice were exposed to concentrations of sulfur dioxide by whole body inhalation of 0 and 500 ppm for 5 minutes/day, five days/week for two years.  Clinical signs noted were not provided in the report.  Mortality was presented on an individual animal basis.  Visual inspection of the mortality data did not reveal any obvious differences between treated and control mice.  Body weights were recorded in this study, but only presented for individual animals only and not statistically analyzed.  Visual inspection of the data does not indicated any obvious differences between controls and treated rats.  Necropsy and microscopic findings for all animals were also presented but only on an individual animal basis.  The incidence of primary carcinoma was slightly increased in female mice (4/30) in the sulfur dioxide group compared to the controls (0/30).  The incidence of lung adenomas was increased in males (15/28) and females (13/30) in the treated group compared to controls (11/35 males and 13/30 females).  The incidence of lung hyperplasia was slightly increased in male mice.  The method of examining the lung for lesions by looking at them under a light is suspect in this study.  Lesions could have been missed in both the control and treated group.

b.  Sodium metabisulfite:  Male and female Wistar rats (F0) were fed diets containing 0, 0.125, 0.25, 0.5, 1.0 and 2.0% for 21 weeks prior to mating for the F1a generation and at study week 34 for the F1b generation.  These dose levels were equivalent to approximately 0, 48, 106, 217, 454 and 942 mg/kg/day.  No clinical signs of toxicity were reported.  Survival was unaffected by treatment, except for a slight increase in mortality at the high dose in males in the F1 generation.  The report states that body weights were comparable among the dose groups in the F0 generation; however, in the F1 generation, body weights were decreased at the high dose level.  Reduced hemoglobin content, hematocrit and erythrocyte count was observed in F0 males at the high dose at study week 52, 78 and 100.  Occult blood was observed in the feces of the F0 and F1 generation rats fed the 2% diet.  No effects were observed on organ weights, clinical chemistry parameters or upon urinalysis.  The microscopic findings were observed in the stomach.  Forestomach and glandular stomach hyperplasia or inflammation due to local irritation was observed at dietary levels of 1 and 2%.  No other treatment-related microscopic findings were seen.  There was no increase in any tumor type.  The systemic NOAEL was considered to be the highest dose tested or ~942 mg/kg/day and the NOAEL due to local irritation was considered to be 0.5% or 217 mg/kg/day.  This reviewer considers the NOAEL to be 1% (~454 mg/kg/day) in the diet based on the slight anemia and effects on body weights observed at 2%.

c.  Potassium metabisulfite:  Fifty male and female ICR/JCL mice were administered potassium metabisulfite in drinking water (distilled water) at concentrations of 0, 1 or 2% for 24 months.  No increase in any tumor type was observed compared to the control group.  Potassium metabisulfite was not carcinogenic to ICR/JCL mice under the conditions of this study.

5.  Genotoxicity/Mutagenicity:

a.  Sulfur dioxide:  Four genotoxicity studies were conducted with sulfur dioxide, including a mouse micronucleus test, a Comet assay, an in vitro SCE and micronuclei assay on human lymphocytes, and an in vitro assay for sister chromatid exchanges and chromosomal aberrations in lymphocytes of workers at a sulfur dioxide plant.  All studies were positive.   

b.  Sodium Bisulfite:  A total of 10 studies were available on sodium bisulfite including two Ames Salmonella assays, two SCE assays (human lymphocytes and Chinese hamster ovary cells), in vitro chromosomal aberration assay (human lymphocytes), in vitro micronucleus test (human lymphocytes), yeast gene conversion assay, mouse micronucleus test, unscheduled DNA synthesis in isolated rat hepatocytes and a rat dominant lethal assay.  Sodium bisulfite was negative in the standard Ames assay, positive in some non-standard Salmonella strains, positive for inducing SCE in two assays in vitro, positive for inducing micronuclei in human lymphocytes, positive for inducing chromosomal aberrations in human lymphocytes, negative in the mouse micronucleus test, negative in the UDS assay and negative in the rat dominant lethal assay.  

b.  Sodium metabisulfite:  Two Ames Salmonella assays, an in vitro SCE assay in human lymphocytes, and in vitro (human lymphocytes) and in vivo (rat bone marrow) chromosomal aberration assays were conducted with sodium metabisulfite were conducted.  Sodium metabisulfite was clearly negative in both Ames Salmonella assays, positive in the human lymphocytes assays for SCE and chromosomal aberrations, questionably positive in one in vivo chromosomal assay when administered by gavage (control data not presented) and negative in the other more reliable in vivo chromosomal assay.  

6.  Neurotoxicity/Developmental Neurotoxicity

a.  Sulfur Dioxide:  The effects of continuous exposure to sulfur dioxide of male and female mice prior to mating, during mating and during gestation on various behavioral parameters were evaluated, and effects of parental exposure on passive avoidance learning and retention were evaluated in their offspring (cross-fostered to untreated lactating dams).  Male and female CD-1 mice (10/sex/group) were exposed to sulfur dioxide at concentrations of 0, 5, 12 or 30 ppm sulfur dioxide by whole body inhalation exposure 9 days prior to mating, during mating and until pregnancy day 12 to 14.  Sulfur dioxide concentrations were within +10% of target.  During the chamber activity testing, rearing, grooming, social interaction and sniffing were clearly increased upon initial exposure to sulfur dioxide which could be a reaction to the initial smell of sulfur dioxide.  During the activity testing outside of the exposure chambers, some effects of treatment were observed.  Predominately, grooming was decreased in males on exposure day 9 and in females on exposure days 6 and 9 and sniffing and wall rearing were increased in females on exposure day 9 at 30 ppm.  Body weight, food consumption and water consumption were unaffected by exposure to sulfur dioxide prior to mating.  However, body weights were clearly depressed at the 30 ppm level after mating accompanied with depressed food consumption.  Exposure to sulfur dioxide had no effect on mating, litter size, sex ratio, or neonatal mortality.  In addition, there were no effects on the day of eyelid and ear opening or on tooth eruption.  Passive avoidance memory and retention were unaffected by parental exposure to sulfur dioxide.

In another study, pregnant female CD-1 mice (14/group) were exposed to concentrations of 0, 32 or 65 ppm sulfur dioxide from gestation days 7 to 18.   The mean number of pups born/litter was unaffected by exposure of the dams to either concentration of sulfur dioxide.  Pup body weights were statistically decreased on PND1 in the 65 ppm group compared to controls.  Exposure to both concentrations of sulfur dioxide statistically increased the time required for righting reflex and negative geotaxis, but had no statistical effect on aerial righting reflex on PND 12.  

7.  Immunotoxicity.  Snowden has conducted a Weight of the Evidence evaluation on the available information on the immunotoxicity of sulfur dioxide/sulfites, and has requested a waiver for this study for the following reasons:  1.  Sulfur dioxide/sulfites cause hypersensitivity in susceptible individuals, primarily asthmatics (e.g. increased bronchoconstriction with increasing concentrations);  2.  Hypersensitivity reactions do not result from immunosuppression, but immunopotentiation;  3. The EPA required guideline immunotoxicity study (TDAR) determines the immunosuppression potential of a chemical, not its potential for immunopotentiation; and 4.  Based on a weight of the evidence of the available information on the immunotoxicity of sulfur dioxide, Snowden request a waiver for a TDAR study (OCSPP 870.7800) on sulfur dioxide because such a study is not appropriate for this chemical.  

8.  Animal Metabolism:  Sulfite and bisulfite are thought to be the major ions formed on absorption of sulfur dioxide in the mucous membranes of the nose and upper respiratory tract due to the solubility of sulfur dioxide in aqueous media.  The key reactions are:
             SO2 + H2O --> H2SO3
             H2SO3 + H2O --> HSO3[-]   + H3O[+]  (1)
             HSO3[-]  + H2O --> SO3[-2]  + H3O[+]  (2)

Under acidic conditions, metabisulfite is converted to bisulfite (1) and then to sulfite or sulfurous acid, sulfur dioxide and water (2).   Molybdenum-dependent sulfite oxidase converts the absorbed (bi)sulfite to sulfate.  This enzyme occurs in highest concentrations in the liver and kidney although lower concentrations are found in the lung.  Free sulfite in food is a mixture of sulfur dioxide, bisulfite ion, and sulfite ion in chemical equilibrium dependent on the pH of the food.  At normal physiological pH values and concentrations of greater than 1 M, the equilibrium is between approximately equal proportions of sulfite and bisulfite while at the lower pH of the stomach of fasting humans, the equilibrium is essentially between bisulfite ion and free sulfur dioxide.  According to Petering, et al. (1975), the interaction of sulfur dioxide with biological molecules in an aqueous medium is equivalent to considering the reactions of the hydrated forms of sulfur dioxide, sulfite and bisulfite.   The pH of most grapes is in the acidic range, pH 2.8 to 3.82, and figs in the range of 5 to 6.   

C.	Exposure Assessment

   1. 	Dietary Exposure

1.  Food and Feed uses:  Sulfites are the residues of concern for consumption of figs.  There are other sources of sulfites in or on food which are regulated by the Food and Drug Administration (FDA). 

Exposures to sulfites when used as an active or inert pesticide ingredient are minimal because it is known to be readily biodegradable, quickly oxidized, and rapidly excreted from the body. In addition, sulfur dioxide (21 CFR 182.3862) is listed as GRAS by the FDA, with limitations, as a food preservative. Some sulfites form in certain foods through fermentation.  Concentrations of > 100 ppm can occur from food additive uses in dried fruits (excluding dark raisins and prunes), lemon and lime juices, wine, molasses, and sauerkraut juice.  Foods such as dried potatoes, grape juice, wine vinegar, gravies, fruit topping and maraschino cherries could have levels of between 50 and 100 ppm from food additives.  Lower concentrations (10 to 50 ppm) can occur in other foods such as pectin, fresh shrimp, corn syrup, sauerkraut, pickled foods, cornstarch, hominy, frozen potatoes, maple syrup, imported jams and jellies, and fresh mushrooms (CIR 2003).  The use of sulfur dioxide or inorganic sulfites as fungicides may lead to lower sulfite levels than from the food additive uses. 

2.  Drinking water exposure. Based on environmental fate information for sulfur dioxide and the requested postharvest use pattern (in closed chambers), concentrations of concern are not expected in drinking water.

   2. Non-dietary exposure

       Occupational exposure
          
Applicators can be exposed to sulfur dioxide during and after fumigation.  Products containing sulfur dioxide as the active ingredient are formulated as liquid under pressure (99.9- 100% sulfur dioxide) which turns into gas upon release of pressure. Sulfur dioxide is registered for post-harvest fumigation of grapes held in cold storage in enclosed spaces (e.g. trailers, railcars, transportation vehicles and warehouses).  These postharvest treatments control gray mold disease which is caused by Botrytis cinerea. In addition, sulfur dioxide is used in the wine industry for the sanitation of corks and barrels, and an emergency exception is in place for the use of sulfur dioxide to control Botrytis cinerea in figs.

      b. Residential (Non-occupational) exposure and risk
      
1.  Inhalation:  There are no residential uses for sulfur dioxide or sodium metabisulfite.  Exposure is possible to bystanders to sulfur dioxide from off-gassing after sulfur dioxide fumigations inside warehouses or transportation vehicles.  The endpoint selected by the Agency for the bystander inhalation risk assessment is 0.25 ppm sulfur dioxide, with one-hour exposure duration. The 0.25 ppm concentration is based on an ambient air quality standard set by the California Air Resources Board. This endpoint is deemed most applicable to this exposure scenario, as it is based on effects of concern for bystanders (such as bronchoconstriction, shortness of breath, wheezing, and chest tightness during physical activity in persons with asthma).

There is the potential for inhalation exposure to sulfur dioxide via ambient air. These exposures can be acute as well as longer-term in nature.  While ambient air data are typically required for fumigants, sulfur dioxide in ambient air comes from other environmental, non-pesticidal sources.  Those sources include combustion of fossil fuels, smelting of sulfide ores, volcanic emissions, other natural sources, and to manufacture hydrosulfites which are used to process, disinfect and bleach food, for waste and water treatment, and in metal, ore and oil refining (ASTDR 2004).  As a criteria air pollutant, sulfur dioxide is regulated by EPA's Office of Air Quality Planning and Standards (OAQPS).

D. Cumulative Effects

In accordance with the FQPA, HED must consider and aggregate (add) pesticide exposures and risks from three major sources: food, drinking water, and residential exposures. For sulfites, allergic reactions in sensitive individuals are the reason for required limitations in sulfite ingestion rather than any systemic toxicity. In the absence of significant systemic toxicity, a quantitative dietary (food and drinking water) risk assessment was not performed.  Since similar effects would not be expected to occur through dermal or inhalation exposure, and significant incidental oral exposure is not expected, the aggregate assessment includes only the dietary exposure scenario, for which quantification of risks is not appropriate. For sulfur dioxide, exposure is primarily through inhalation; significant oral and dermal exposures are not expected. Additionally, since the inhalation endpoint for sulfur dioxide is different from that for the ingested sulfite residues, aggregation across routes is not appropriate. Bystander exposure to sulfur dioxide would not need to be aggregated with other routes of exposure.  

Section 408(b)(2)(D)(v) of FFDCA requires that, when considering whether to establish, modify, or revoke a tolerance, the Agency consider ``available information'' concerning the cumulative effects of a particular pesticide's residues and ``other substances that have a common mechanism of toxicity.''  EPA has not found inorganic sulfites to share a common mechanism of toxicity with any other substances, and sulfur dioxide does not appear to produce a toxic metabolite produced by other substances. For the purposes of this tolerance action, therefore, EPA has assumed that sulfur dioxide does not have a common mechanism of toxicity with other substances. For information regarding EPA's efforts to determine which chemicals have a common mechanism of toxicity and to evaluate the cumulative effects of such chemicals, see EPA's Web site at http://www.e

E. 	Safety Factor for Infants and Children

--------------------------------------------------------------------------------
There is sufficient toxicological information for sulfur dioxide to address risks to infants and children. The available information indicates that there is no evidence of increased quantitative or qualitative susceptibility of the offspring after in utero or postnatal exposure. Based on the lack of significant toxicity in existing toxicological testing of sulfur dioxide and FDA's classification of sulfites as GRAS, EPA has not performed a quantitative risk assessment for sulfur dioxide using safety factors. For the same reason, and given the absence of any evidence of pre- or post-natal sensitivity to sulfur dioxide, EPA concludes that there is reliable data to support not using an additional safety factor to protect infants and children.
--------------------------------------------------------------------------------

--------------------------------------------------------------------------------
F. 	Aggregate Risks and Determination of Safety
--------------------------------------------------------------------------------

--------------------------------------------------------------------------------
Oral (dietary) exposure to sulfites is the main route of exposure.  Dermal and inhalation exposure is not considered to be significant.  Only aggregate assessment needs to be included in the dietary exposure scenario.  Inhalation is the primary route of exposure to sulfur dioxide.  Oral and dermal exposures to sulfur dioxide are not significant.  The inhalation endpoint for sulfur dioxide is different from that for ingested sulfites.  Therefore, bystander exposure to sulfur dioxide would not be aggregated with other routes of exposure. 
--------------------------------------------------------------------------------

--------------------------------------------------------------------------------
	5.	Determination of safety 
--------------------------------------------------------------------------------

--------------------------------------------------------------------------------
The residue levels expected from this use on figs are relatively low when compared to concentrations of sulfites in many common foods and viewed as GRAS by the FDA. Given the low fig use rate, low expected residue levels, and relatively low consumption of figs, the safety finding made by EPA in the May 2007 RED assessment for the post-harvest use on grapes may be extended to include the proposed tolerance level of 25 ppm on figs.  There is a reasonable certainty that no harm will result to the general population, or to sulfite sensitive individuals, infants and children, from aggregate exposure to residues of sulfur dioxide, including its metabolites and degradates.
--------------------------------------------------------------------------------

--------------------------------------------------------------------------------
G.	 International Tolerances

There are currently no established CODEX or Canadian and/or Mexican MRLs for sulfur dioxide in/on figs.  Mexico has adopted a grape MRL based on the tolerance established in the US.  
