EPA/690/R-22/002F | September 2022 | FINAL

United States
Environmental Protection
Agency

xvEPA

Provisional Peer-Reviewed Toxicity Values for

Dibenzothiophene
(CASRN 132-65-0)

U.S. EPA Office of Research and Development
Center for Public Health and Environmental Assessment


-------
A mA United States

Environmental Protection
»»Agency

EPA 690 R-22 002F
September 2022
https://www.epa.gov/pprtv

Provisional Peer-Reviewed Toxicity Values for

Dibenzothiophene
(CASRN 132-65-0)

Center for Public Health and Environmental Assessment
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268


-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS

CHEMICAL MANAGERS

Lucina E. Lizarraga, PhD

Center for Public Health and Environmental Assessment, Cincinnati, OH
Jeffry L. Dean II, PhD

Center for Public Health and Environmental Assessment, Cincinnati, OH

CONTRIBUTORS

Chelsea A. Weitekamp, PhD

Center for Public Health and Environmental Assessment, Research Triangle Park, NC
James A. Weaver, PhD, DABT

Center for Public Health and Environmental Assessment, Research Triangle Park, NC

SCIENTIFIC TECHNICAL LEAD

Lucina E. Lizarraga, PhD

Center for Public Health and Environmental Assessment, Cincinnati, OH

DRAFT DOCUMENT PREPARED BY

SRC, Inc.

7502 Round Pond Road
North Syracuse, NY 13212

PRIMARY INTERNAL REVIEWERS

Daniel D. Petersen, MS, PhD, DABT, ATS, ERT

Center for Public Health and Environmental Assessment, Cincinnati, OH

Suryanarayana V. Vulimiri, BVSc, PhD, DABT

Center for Public Health and Environmental Assessment, Washington, DC

PRIMARY EXTERNAL REVIEWERS

Eastern Research Group, Inc.

110 Hartwell Avenue
Lexington, MA 02421-3136

PPRTV PROGRAM MANAGEMENT

Teresa L. Shannon

Center for Public Health and Environmental Assessment, Cincinnati, OH
J. Phillip Kaiser, PhD, DABT

Center for Public Health and Environmental Assessment, Cincinnati, OH

in

Dib enzothi ophene


-------
Questions regarding the content of this PPRTV assessment should be directed to the U.S. EPA
Office of Research and Development (ORD) Center for Public Health and Environmental
Assessment (CPHEA) website at https://ecomments cpa.gov/pprtv.

iv

Dib enzothi ophene


-------
TABLE OF CONTENTS

COMMONLY USED ABBREVIATIONS AND ACRONYMS	vi

BACKGROUND	1

QUALITY ASSURANCE	1

DISCLAIMERS	2

QUESTIONS REGARDING PPRTVs	2

1.	INTRODUCTION	3

2.	REVIEW OF POTENTIALLY RELEVANT DATA (NONCANCER AND CANCER)	7

2.1.	HUMAN STUDIES	11

2.1.1.	Oral Exposures	11

2.1.2.	Inhalation Exposures	11

2.2.	ANIMAL STUDIES	11

2.2.1.	Oral Exposures	11

2.2.2.	Inhalation Exposures	15

2.3.	OTHER DATA (SHORT-TERM TESTS, OTHER EXAMINATIONS)	15

2.3.1.	Genotoxicity	15

2.3.2.	Metabolism/Toxicokinetic and Supporting Animal Studies	19

3.	DERIVATION 01 PROVISIONAL VALUES	23

3.1. DERIVATION OF ORAL REFERENCE DOSES	23

3 .2. DERIVATION OF INHALATION REFERENCE CONCENTRATIONS	23

3.3.	SUMMARY OF NONCANCER PROVISIONAL REFERENCE VALUES	23

3.4.	CANCER WEIGHT-OF-EVIDENCE DESCRIPTOR	23

3.5.	DERIVATION OF PROVISIONAL CANCER RISK ESTIMATES	24

APPENDIX A. SCREENING PROVISIONAL VALUES	25

APPENDIX D. REFERENCES	55

v

Dib enzothi ophene


-------
COMMONLY USED ABBREVIATIONS AND ACRONYMS

a2u-g

alpha 2u-globulin

IVF

in vitro fertilization

ACGIH

American Conference of Governmental

LC50

median lethal concentration



Industrial Hygienists

LD50

median lethal dose

AIC

Akaike's information criterion

LOAEL

lowest-observed-adverse-effect level

ALD

approximate lethal dosage

MN

micronuclei

ALT

alanine aminotransferase

MNPCE

micronucleated polychromatic

AR

androgen receptor



erythrocyte

AST

aspartate aminotransferase

MOA

mode of action

atm

atmosphere

MTD

maximum tolerated dose

ATSDR

Agency for Toxic Substances and

NAG

7V-acetyl-P-D-glucosaminidase



Disease Registry

NCI

National Cancer Institute

BMC

benchmark concentration

NOAEL

no-observed-adverse-effect level

BMCL

benchmark concentration lower

NTP

National Toxicology Program



confidence limit

NZW

New Zealand White (rabbit breed)

BMD

benchmark dose

OCT

ornithine carbamoyl transferase

BMDL

benchmark dose lower confidence limit

ORD

Office of Research and Development

BMDS

Benchmark Dose Software

PBPK

physiologically based pharmacokinetic

BMR

benchmark response

PCNA

proliferating cell nuclear antigen

BUN

blood urea nitrogen

PND

postnatal day

BW

body weight

POD

point of departure

CA

chromosomal aberration

PODadj

duration-adjusted POD

CAS

Chemical Abstracts Service

QSAR

quantitative structure-activity

CASRN

Chemical Abstracts Service registry



relationship



number

RBC

red blood cell

CBI

covalent binding index

RDS

replicative DNA synthesis

CHO

Chinese hamster ovary (cell line cells)

RfC

inhalation reference concentration

CL

confidence limit

RfD

oral reference dose

CNS

central nervous system

RGDR

regional gas dose ratio

CPHEA

Center for Public Health and

RNA

ribonucleic acid



Environmental Assessment

SAR

structure-activity relationship

CPN

chronic progressive nephropathy

SCE

sister chromatid exchange

CYP450

cytochrome P450

SD

standard deviation

DAF

dosimetric adjustment factor

SDH

sorbitol dehydrogenase

DEN

diethylnitrosamine

SE

standard error

DMSO

dimethylsulfoxide

SGOT

serum glutamic oxaloacetic

DNA

deoxyribonucleic acid



transaminase, also known as AST

EPA

Environmental Protection Agency

SGPT

serum glutamic pyruvic transaminase,

ER

estrogen receptor



also known as ALT

FDA

Food and Drug Administration

SSD

systemic scleroderma

FEVi

forced expiratory volume of 1 second

TCA

trichloroacetic acid

GD

gestation day

TCE

trichloroethylene

GDH

glutamate dehydrogenase

TWA

time-weighted average

GGT

y-glutamyl transferase

UF

uncertainty factor

GSH

glutathione

UFa

interspecies uncertainty factor

GST

g 1 ii ta t h i o nc - V-1 ra ns fc ra sc

UFc

composite uncertainty factor

Hb/g-A

animal blood-gas partition coefficient

UFd

database uncertainty factor

Hb/g-H

human blood-gas partition coefficient

UFh

intraspecies uncertainty factor

HEC

human equivalent concentration

UFl

LOAEL-to-NOAEL uncertainty factor

HED

human equivalent dose

UFS

subchronic-to-chronic uncertainty factor

i.p.

intraperitoneal

U.S.

United States of America

IRIS

Integrated Risk Information System

WBC

white blood cell

Abbreviations and acronyms not listed on this page are defined upon first use in the
PPRTV document.

vi

Dib enzothi ophene


-------
EPA/690/R-22/002F

PROVISIONAL PEER-REVIEWED TOXICITY VALUES FOR
DIBENZOTHIOPHENE (CASRN 132-65-0)

BACKGROUND

A Provisional Peer-Reviewed Toxicity Value (PPRTV) is defined as a toxicity value
derived for use in the Superfund program. PPRTVs are derived after a review of the relevant
scientific literature using established U.S. Environmental Protection Agency (U.S. EPA)
guidance on human health toxicity value derivations.

The purpose of this document is to provide support for the hazard and dose-response
assessment pertaining to chronic and subchronic exposures to substances of concern, to present
the major conclusions reached in the hazard identification and derivation of the PPRTVs, and to
characterize the overall confidence in these conclusions and toxicity values. It is not intended to
be a comprehensive treatise on the chemical or toxicological nature of this substance.

Currently available PPRTV assessments can be accessed on the U.S. EPA's PPRTV
website at https://www.epa.gov/pprtv. PPRTV assessments are eligible to be updated on a 5-year
cycle and revised as appropriate to incorporate new data or methodologies that might impact the
toxicity values or affect the characterization of the chemical's potential for causing adverse
human-health effects. Questions regarding nomination of chemicals for update can be sent to the
appropriate U.S. EPA eComments Chemical Safety website at
https://ecomments.epa.gov/chemicalsafetv/.

QUALITY ASSURANCE

This work was conducted under the U.S. EPA Quality Assurance (QA) program to ensure
data are of known and acceptable quality to support their intended use. Surveillance of the work
by the assessment managers and programmatic scientific leads ensured adherence to QA
processes and criteria, as well as quick and effective resolution of any problems. The QA
manager, assessment managers, and programmatic scientific leads have determined under the
QA program that this work meets all U.S. EPA quality requirements. This PPRTV was written
with guidance from the CPHEA Program Quality Assurance Project Plan (PQAPP), the QAPP
titled Program Quality Assurance Project Plan (PQAPP) for the Provisional Peer-Reviewed
Toxicity Values (PPRTVs) and Related Assessments/Documents (L-CPAD-0032718-QP), and the
PPRTV development contractor QAPP titled Quality Assurance Project Plan—Preparation of
Provisional Toxicity Value (PTV) Documents (L-CPAD-0031971-QP). As part of the QA
system, a quality product review is done prior to management clearance. A Technical Systems
Audit may be performed at the discretion of the QA staff.

All PPRTV assessments receive internal peer review by at least two CPHEA scientists
and an independent external peer review by at least three scientific experts. The reviews focus on
whether all studies have been correctly selected, interpreted, and adequately described for the
purposes of deriving a provisional reference value. The reviews also cover quantitative and
qualitative aspects of the provisional value development and address whether uncertainties
associated with the assessment have been adequately characterized.

1

Dib enzothi ophene


-------
EPA/690/R-22/002F

DISCLAIMERS

The PPRTV document provides toxicity values and information about the adverse effects
of the chemical and the evidence on which the value is based, including the strengths and
limitations of the data. All users are advised to review the information provided in this document
to ensure that the PPRTV used is appropriate for the types of exposures and circumstances at the
site in question and the risk management decision that would be supported by the risk
assessment.

Other U.S. EPA programs or external parties who may choose to use PPRTVs are
advised that Superfund resources will not generally be used to respond to challenges, if any, of
PPRTVs used in a context outside of the Superfund program.

This document has been reviewed in accordance with U.S. EPA policy and approved for
publication. Mention of trade names or commercial products does not constitute endorsement or
recommendation for use.

QUESTIONS REGARDING PPRTVS

Questions regarding the content of this PPRTV assessment should be directed to the
U.S. EPA ORD CPHEA website at https://ecomments.epa.gov/pprtv.

2

Dib enzothi ophene


-------
EPA/690/R-22/002F

1. INTRODUCTION

Dibenzothiophene (CASRN 132-65-0) is a solid, sulfur-containing, three-ringed,
heterocyclic polycyclic aromatic hydrocarbon (PAH) derivative. It is one of the organosulfur
components of petroleum and coal and is used as a chemical intermediate and as an ingredient in
cosmetics and pharmaceuticals (NI.M. 2021b; Blunter et al.. 201 1; Deutschmann et al.. 201 1).
Dibenzothiophene is listed as an active substance in commerce on the public Toxic Substances
Control Act (TSCA) inventory (U.S. EPA, 202le) and it is registered with Europe's Registration,
Evaluation, Authorization, and Restriction of Chemicals (REACH) program (ECHA. 2021).

Production and import volumes were not reported in U.S. EPA's Chemical Data
Reporting (CDR) database (U.S. EPA. 202 la). No public information on industrial production or
synthetic processes were located.

The empirical formula for dibenzothiophene is C12H8S; its structure is shown in Figure 1.
Experimental and estimated physicochemical properties identified for dibenzothiophene from
U.S. EPA (2021c) and NLM (2021b) are presented in Table 1. When more than one
experimental value was available, an experimental average is presented. Dibenzothiophene is
moderately volatile from water and moist soil surfaces based on its calculated Henry's law
constant; the soil adsorption coefficient indicates that dibenzothiophene will strongly sorb to soil
and sediment, however, which may limit volatilization from these surfaces. Due to strong
sorption and low water solubility, the potential to leach to groundwater or undergo runoff after
precipitation is low.

Figure 1. Dibenzothiophene (CASRN 132-65-0) Structure

3

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table 1. Physicochemical Properties of Dibenzothiophene
(CASRN 132-65-0)

Property (unit)

Value3

Physical state

Solidb

Boiling point (°C)

333

Melting point (°C)

97.0

Density (g/cm3)

1.23-1.25 (estimated)

Vapor pressure (mm Hg at 25°C)

0.000205 (extrapolated)13

pH (unitless)

NA

pKa (unitless)

NA

Solubility in water (mg/L at 25°C)

1.47 (converted from PhysProp NCCT value of
7.98 x 10-6 mol/L)

Log octanol/water partition coefficient (log Kow)

4.38

Henry's law constant (atm-m3/mole at 25°C)

3.38 x 10 " (calculated from vapor pressure/water
solubility)0

Soil adsorption coefficient Koc (units not reported)

5,273-20,535b

Atmospheric OH rate constant (cm5Vmolecule-sec at 25°C)

8.10 x 10-12b

Atmospheric half-life (d)

1.3 (estimated using 12-hday; 1.5 x 106OH/cm3)d

Relative vapor density (air = 1)

Not applicable for solid

Molecular weight (g/mol)

184.26

Flash point (°C)

Not applicable for solid

'Unless otherwise noted, values are from U.S. EPA (2021c).
bNLM (2021b).

°U.S. EPA (2012): calculated by EPI Suite™ using a vapor pressure of 0.000205 mm Hg and a water solubility of
1.47 mg/L.

dU.S. EPA (2012).

EPI Suite™ = Estimation Programs Interface Suite; NA = not applicable; NCCT = National Center for
Computational Toxicology; PhysProp = Physical Properties Database.

A summary of available toxicity values for dibenzothiophene from U.S. EPA and other
agencies/organizations is provided in Table 2.

4

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table 2. Summary of Available Toxicity Values for Dibenzothiophene

(CASRN 132-65-0)

Source
(parameter)ab

Value (applicability)

Notes

Reference

Noncancer

IRIS

NV

NA

U.S. EPA (202Id)

HEAST

NV

NA

U.S. EPA (2011b)

DWSHA

NV

NA

U.S. EPA (2018)

ATSDR

NV

NA

ATSDR (2021)

WHO

NV

NA

WHO (2021)

CalFPA

NV

NA

CalEPA (2021. 2020)

OSHA

NV

NA

OSHA (2021a. 2021b. 2021c)

NIOSH

NV

NA

NIOSH (2018)

ACGIH

NV

NA

ACGIH (2020)

Cancer

IRIS

NV

NA

U.S. EPA (202Id)

HEAST

NV

NA

U.S. EPA (2011b)

DWSHA

NV

NA

U.S. EPA (2018)

NTP

NV

NA

NTP (2016)

IARC (WOE)

Group 3: not classifiable as to
its carcinogenicity to humans

Based on inadequate evidence for
carcinogenicity in experimental
animals and no available data for
carcinogenicity in humans.

IARC (2013)

CalEPA

NV

NA

CalEPA (2021. 2020)

ACGIH

NV

NA

ACGIH (2020)

aSources: ACGIH = American Conference of Governmental Industrial Hygienists; ATSDR = Agency for Toxic
Substances and Disease Registry; CalEPA = California Environmental Protection Agency; DWSHA = Drinking
Water Standards and Health Advisories; HEAST = Health Effects Assessment Summary Tables;

IARC = International Agency for Research on Cancer; IRIS = Integrated Risk Information System;

NIOSH = National Institute for Occupational Safety and Health; NTP = National Toxicology Program;

OSHA = Occupational Safety and Health Administration; WHO = World Health Organization.

Parameters: WOE = weight of evidence.

NA = not applicable; NV = not available.

Literature searches were conducted in June 2019 and updated in June 2022 for studies
relevant to the derivation of provisional toxicity values for dibenzothiophene. Searches were
conducted using U.S. EPA's Health and Environmental Research Online (HERO) database of
scientific literature. HERO searches the following databases: PubMed, TOXLINE1 (including
TSCATS1), and Web of Science. The following resources were searched outside of HERO for

'Note that this version of TOXLINE is no longer updated

(https://www.nlm.nih.gov/databases/download/toxlinesubset.html'): therefore, it was not included in the literature
search update from June 2022.

5

Dib enzothi ophene


-------
EPA/690/R-22/002F

health-related values: American Conference of Governmental Industrial Hygienists (ACGIH),
Agency for Toxic Substances and Disease Registry (ATSDR), California Environmental
Protection Agency (CalEPA), Defense Technical Information Center (DTIC), European Centre
for Ecotoxicology and Toxicology of Chemicals (ECETOC), European Chemicals Agency
(ECHA), U.S. EPA Chemical Data Access Tool (CDAT), U.S. EPA ChemView, U.S. EPA
Integrated Risk Information System (IRIS), U.S. EPA Health Effects Assessment Summary
Tables (HEAST), U.S. EPA Office of Water (OW), International Agency for Research on
Cancer (IARC), U.S. EPA TSCATS2/TSCATS8e, U.S. EPA High Production Volume (HPV),
Chemicals via International Programme on Chemical Safety (IPCS) INCHEM, Japan Existing
Chemical Data Base (JECDB), Organisation for Economic Co-operation and Development
(OECD) Screening Information Data Sets (SIDS), OECD International Uniform Chemical
Information Database (IUCLID), OECD HPV, National Institute for Occupational Safety and
Health (NIOSH), National Toxicology Program (NTP), Occupational Safety and Health
Administration (OSHA), and World Health Organization (WHO).

6

Dib enzothi ophene


-------
EPA/690/R-22/002F

2. REVIEW OF POTENTIALLY RELEVANT DATA
(NONCANCER AND CANCER)

Tables 3A and 3B provide overviews of the relevant noncancer and cancer evidence
bases, respectively, for dibenzothiophene and include all repeated-dose short-term, subchronic,
and chronic studies, as well as reproductive and developmental toxicity studies identified from
the literature search. Principal studies used in the PPRTV assessment for derivation of
provisional toxicity values are identified in bold. The phrase "statistical significance" and term
"significant," used throughout the document, indicate ap-value of < 0.05 unless otherwise
specified.

7

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table 3A. Summary of Potentially Relevant Noncancer Data for Dibenzothiophene (CASRN 132-65-0)



Number of Male/Female, Strain,















Species, Study Type, Study









Reference



Category3

Duration, Reported Doses

Dosimetryb

Critical Effects

NOAELb

LOAELb

(comments)

Notes0

Human

1. Oral (mg/kg-day)

ND

2. Inhalation (mg/m3)

ND

Animal

1. Oral (mg/kg-day)

Short-term

6-12 IV1/6—12 F, Sprague Dawley,

0,3,10,30

Increased relative liver weight in

3

10

JECDB (201D

PS,



rat, unspecified oral, 28 d



both males and females and







NPR







reduced motor activity and











Reported doses: 0,3,10,



increased prothrombin time in











30 mg/kg-d



males at >10 mg/kg-d. Other effects















occurring mostly at the highest















dose (30 mg/kg-d): hepatocyte















hypertrophy in males and females















and changes in serum markers of















liver function in males (reduced















albumin protein fraction and A/G















ratio); increased relative kidney















weights and kidney lesions in















males; increased APTT in males.









8

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table 3A. Summary of Potentially Relevant Noncancer Data for Dibenzothiophene (CASRN 132-65-0)

Category3

Number of Male/Female, Strain,
Species, Study Type, Study
Duration, Reported Doses

Dosimetryb

Critical Effects

NOAELb

LOAELb

Reference
(comments)

Notes0

Subchronic
to Chronic

Male (number not specified),
albino, rat, diet, 165 d

Reported dietary concentrations: 0
(historical), 0.025, 0.05, 0.10%

0 (historical),
13, 27, 63

Increased liver weight compared with
laboratory historical controls;
histopathological changes in liver.

NDr

NDr

Thomas et al. (1942)
(The design and
reporting limitations
[including the lack of
a concurrent control
and information on
the number of test
animals] prevent the
determination of
effect levels for this
study.)

PR

2. Inhalation (mg/m3)

ND

aDuration categories are defined as follows: acute = exposure for <24 hours; short term = repeated exposure for >24 hours to <30 days; long-term

(subchronic) = repeated exposure for >30 days or <10% life span for humans (>30 days up to approximately 90 days in typically used laboratory animal species); and

chronic = repeated exposure for >10% life span for humans (>~90 days to 2 years in typically used laboratory animal species) (U.S. EPA. 2002).

bDosimetry: doses are presented as ADDs (mg/kg-day) for oral noncancer effects and as HECs (in mg/m3) for inhalation noncancer effects.

°Notes: NPR = not peer reviewed; PR = peer reviewed; PS = principal study.

ADD = adjusted daily dose; A/G = albumin/globulin; APTT = activated partial thromboplastin time; F = female(s); HEC = human equivalent concentration;

LOAEL = lowest-observed-adverse-effect level; M = male(s); ND = no data; NDr = not determined; NOAEL = no-observed-adverse-effect level.

9

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table 3B. Summary of Potentially Relevant Cancer Data for Dibenzothiophene (CASRN 132-65-0)

Category

Number of Male/Female, Strain, Species,
Study Type, Reported Doses, Duration

Dosimetry

Critical Effects

Reference
(comments)

Notes

Human

1. Oral (mg/kg-day)

ND

2. Inhalation (mg/m3)

ND

Animal

1. Oral (mg/kg-day)

ND

2. Inhalation (mg/m3)

ND

ND = no data.

10

Dib enzothi ophene


-------
EPA/690/R-22/002F

2.1.	HUMAN STUDIES

2.1.1.	Oral Exposures

No studies were identified.

2.1.2.	Inhalation Exposures

No studies were identified.

2.2.	ANIMAL STUDIES
2.2.1. Oral Exposures

The effects of oral exposure to dibenzothiophene in animals have been evaluated in
short-term (JHCDB. 2011) and chronic (Thomas et al.. 1942) studies in rats.

Short-Term Studies

JECDB (2011)

In an OECD Test Guideline (TG) 407 study from the Japanese literature (JECDB. 2011).
groups of 12, 6, 6, and 12 Crl:CD(SD) rats/sex received dibenzothiophene at doses of 0, 3, 10, or
30 mg/kg-day, respectively, by oral administration (additional details not available in English
from the existing Japanese language report) for 28 days. At the end of exposure,
six rats/sex/group were sacrificed; six rats/sex in the control and high-dose groups were followed
for an additional 14 days untreated (recovery) prior to sacrifice. The animals were observed for
clinical signs of toxicity, and body weight and food intake were measured once each week.
Detailed clinical observations of the animals in cages, during handling, and in open field were
performed weekly. During exposure week 4 and recovery week 2, the animals were subjected to
functional observational battery (FOB), assessing reactivity (visual, touch, auditory, pain,
proprioceptive), righting reflex, grip strength, and motor activity. Blood and urine were collected
at the end of exposure and at the end of the recovery period. Hematology parameters included
erythrocytes (red blood cells [RBCs] and reticulocyte counts, hemoglobin, hematocrit, mean
corpuscular volume [MCV], mean corpuscular hemoglobin [MCH], mean corpuscular
hemoglobin concentration [MCHC]), platelet counts, prothrombin time [PT], activated partial
thromboplastin time [APTT], and white blood cells [WBCs; total and differential counts]).

Serum chemistry was evaluated including total protein, albumin, globulins, albumin/globulin
(A/G) ratio, aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline
phosphatase (ALP), y-glutamyl transferase (GGT), total bilirubin, glucose, total cholesterol,
triglycerides, blood urea nitrogen (BUN), creatinine, and electrolytes. Urinalysis parameters
included pH, protein, glucose, ketone bodies, urobilinogen, bilirubin, occult blood, color,
volume, and specific gravity. All animals received gross necropsy. The following organs were
weighed in all animals: liver, kidney, spleen, heart, brain, pituitary, thymus, thyroid, adrenal, and
reproductive organs (testis, epididymis, prostate, seminal vesicle, ovary, and uterus).
Histopathology results were reported for the following organs: lung, cecum, ileum, pancreas,
liver, kidney, testis, epididymis, prostate, and pituitary gland (other organs may have been
examined as well).

Both male and female rats at the high dose of 30 mg/kg-day consumed less food than
controls on administration day 7 (-17 and -14% relative to controls, respectively, p < 0.01) but
not at other time points (14, 21, and 28 days). Male rats receiving 30 mg/kg-day exhibited lower
body weights (-7%) on Days 7 and 14, but there were no statistically significant differences in
body weight at Day 21 or 28 or after the recovery period, or in body-weight gains over the full
treatment or recovery periods (terminal body weights changes are displayed in Table B-l).

11

Dib enzothi ophene


-------
EPA/690/R-22/002F

Females exhibited no significant differences in body weight. Salivation was observed in small
numbers of males and females (n = 1 or 2) at the high dose; no other clinical signs were noted.
Reactivity, righting reflex, and forelimb and hindlimb grip strength were not significantly
affected by exposure at any dose. At the end of exposure, statistically significant decreases in
motor activity counts were observed at doses of 10 and 30 mg/kg-day in male rats (-63 and
-53%, respectively; see Table B-l for more details); no significant differences were observed for
females. There were no differences in motor activity between control and high-dose rats after the
recovery period. No treatment-related urinalysis changes were apparent. Male rats exhibited
statistically significant longer PT (23-36% at >10 mg/kg-day) and APTT (41% at 30 mg/kg-day;
see Table B-l) relative to controls at the end of treatment, while females did not; there were no
other effects on hematology at the end of treatment and none after recovery. Statistically
significant clinical chemistry findings in males at the end of exposure were increased calcium
(+5%>), increased alpha 2u-globulin (au-g) and P globulin protein fraction percentages (+14%),
and decreased albumin protein fraction percentages (-6%) and A/G ratios (—13%) in males
receiving 30 mg/kg-day dibenzothiophene (see Table B-l). Females receiving the highest dose
had significantly higher total cholesterol than controls (+54%) (see Table B-l). After the
recovery period, the only observed changes were significant decreases in blood glucose in males
and decreases in calcium and increases in chloride and ai globulin protein fraction percent in
females (data not shown). Although there were no significant changes in most serum
hepatocellular/hepatobiliary markers (ALT, AST, ALP, GGT, and total bilirubin [data not
shown]) in male and female rats, the decreases in albumin protein fraction and A/G ratio in
males at 30 mg/kg-day are indicative of potential liver damage and are consistent with other liver
effects observed in exposed rats (see below for more details).

Dose-related, statistically significant increases in relative liver weights occurred at
>10 mg/kg-day in males (11—38%) and females (10—27%) (see Table B-l). Absolute liver weight
was statistically significantly increased at the low dose in females (18%) and at the high dose in
males and females (29-31%); however, the changes did not follow a dose-response gradient.
Female rats displayed biologically significant (>10%) increases in absolute liver weight at all
doses, while male rats achieved biologically significant increases in absolute liver weight at only
the low and high doses. Males, but not females, exhibited dose-related, significant increases in
relative kidney weights at >10 mg/kg-day (+9% at 10 mg/kg-day and +12% at 30 mg/kg-day).
Absolute kidney weights increased significantly in females at 3 mg/kg-day (+14%) but there was
no dose-response correspondence. Gross necropsy findings at the end of the exposure period
consisted of dark brown discoloration of the liver in six of six female rats at 30 mg/kg-day (no
other female groups and no males exhibited this change). Histopathology findings in the liver
(see Table B-2) consisted primarily of centrilobular hepatocyte hypertrophy (six of six high-dose
animals of both sexes and one of six females at 10 mg/kg-day; graded as slight in all cases). One
high-dose male rat had a finding of slight focal liver necrosis. These lesions were not observed in
the control group. Males (six of six in the high-dose group, two of six in the mid-dose group, and
one of six in the low-dose group) also exhibited hyaline droplets and eosinophilic bodies (all
graded as slight) in the proximal tubular epithelium of the kidney (see Table B-2). The kidney
changes, but not the liver changes, persisted until the end of the recovery period in some rats.

No-observed-adverse-effect level (NOAEL) and lowest-observed-adverse-effect level
(LOAEL) values of 3 and 10 mg/kg-day, respectively, are identified from this study based on
>10% increases in relative liver weights in both sexes considered biologically significant, as well
as statistically significant reductions in motor activity and increased PT in males. There is some

12

Dib enzothi ophene


-------
EPA/690/R-22/002F

uncertainty due to lack of a full English language report. Although >10% increases in absolute
liver weights were observed at a dose of 3 mg/kg-day in rats, the changes in absolute liver
weights did not follow a dose-response gradient. Further, the increase in relative liver weight in
the mid- and high-dose groups is supported by increased incidence of hepatocyte lesions in male
and female rats at >10 mg/kg-day (mostly hypertrophy but also possible evidence of necrosis)
and significant decreases in albumin protein fraction and A/G ratio in males at 30 mg/kg-day.
Biologically significant increases (>10%) in relative kidney weights were observed in males at
30 mg/kg-day and these animals also showed evidence of kidney lesions (100% incidence of
hyaline droplets and eosinophilic bodies). The administered doses of 0, 3, 10, and 30 mg/kg-day
correspond to human equivalent doses (HEDs) of 0, 0.75, 2.5, and 7.4 mg/kg-day for males, and
0, 0.68, 2.2, and 6.7 mg/kg-day for females, respectively.2

Subchronic/Chrottic Studies

Thomas et al. (1942)

In a published, peer-re viewed study, Thomas et al. (1942) administered dibenzothiophene
(purity not reported) in the diet of male albino rats (source and number not reported) aged 25-
28 days with an average body weight of 48 g at the beginning of the study. The animals received
0.25, 0.50, or 1.00% dibenzothiophene in the diet for the first 4 days of the dosing period
(number of animals per dose group not reported). Because of low food intakes and decreases in
body weight, the doses were decreased to 0.025, 0.050, or 0.100% dibenzothiophene for the
remainder of the 165-day dosing period. Adjusted daily doses (ADDs) are estimated to be 13, 27,
and 63 mg/kg-day, respectively, based on total dibenzothiophene consumption reported by the
study authors and time-weighted average (TWA) body weights obtained by digitizing the growth
curves provided by the study authors. Animals were housed five to a cage; other details
regarding animal husbandry were not provided. Appearance and behavior were recorded by the
study authors "throughout the duration of the study." Food and water were provided ad libitum;
animals and food cups were weighed twice a week for the duration of the study. Experimental
data for each exposure group were compared with data for age- or body-weight-matched
historical control animals; the type of historical control used for each endpoint is listed below
with the results for that endpoint.

At study termination, animals were sacrificed, and histopathological examinations were
performed. The study authors noted that they used a necropsy technique previously described by
Wilson et al. (1938); the spleen, liver, adrenal glands, kidneys, testes, ovaries, and heart were
weighed under this necropsy protocol. Histopathological sections of the liver, spleen, adrenal
gland, heart, bladder, intestine, lung, testis, and stomach were prepared from five animals in each
exposure group and stained with hematoxylin and eosin (Thomas et al.. 1942). Frozen sections of
the livers from three animals in the high-dose group and all animals in the low-dose group were
stained with Sudan IV. Blood was collected on Days 107 and 157 from the tails of five high-dose
animals and analyzed for hemoglobin and for erythrocyte, reticulocyte, and total and differential
WBC counts. Although the study authors indicated statistical significance of their findings, no
information was provided regarding their statistical methods. This study was performed prior to

2Administered doses were converted to HEDs by multiplying by dosimetric adjustment factors (DAFs) of 0.250,
0.248, and 0.247 for low-, mid-, and high-dose males and 0.226, 0.223, and 0.222 for low-, mid-, and high-dose
females calculated as follows: DAF = (BWa1/4 ^ BWhI/4), where BWa = animal body weight, and BWh = human
body weight. Study-specific TWA animal body weights of 0.272, 0.264, and 0.259 kg for low-, mid-, and high-dose
males, and 0.182, 0.174, and 0.171 kg for low-, mid-, and high-dose females were used. For humans, the reference
value of 70 kg was used for body weight, as recommended by U.S. EPA (1988).

13

Dib enzothi ophene


-------
EPA/690/R-22/002F

the adoption of Good Laboratory Practice (GLP), and little information regarding the laboratory
procedures was provided. The study authors also reported a second experiment examining the
presence of dibenzothiophene metabolites in the urine of rabbits (see Table 4B in Section 2.3.2).

No deaths or clinical signs of toxicity were reported during the study. Body weights
throughout the course of the study were presented graphically, and mean terminal body weights
were provided in numerical form for each exposure group (see Table B-3). A dose-dependent
decrease in body weight was observed; however, the study authors attributed this to reduced food
consumption and did not consider it a direct effect of dibenzothiophene. For the evaluation of
organ weights, animals dosed with dibenzothiophene were compared with laboratory historical
controls matched according to body weight. As a result, the differences from the control group
approximate a change in relative (to body weight) organ weight. The only significant effects on
organ weight observed were in the liver and spleen. Although statistical significance for weight
changes in both the liver and spleen were noted by the study authors, neither an indication of the
dose at which significance occurred nor any levels of significance were reported. Data for these
organs are presented in Table B-3. Liver weights increased (7-115%) in a dose-dependent
manner, with changes >10% occurring at >27 mg/kg-day. Spleen weights decreased (29-57%) in
a dose-dependent manner. The decreased spleen weight may be related to the decreased food
consumption, as spleen weight has been shown to decrease disproportionately to body weight
when food consumption is decreased (Peters and Boyd. 1966). Gross examination revealed that
the livers in the mid- and high-dose animals were large and presented a yellowish, fatty
appearance. Spleens appeared normal except for a reduction in their sizes upon gross
examination. Liver and kidney histopathological lesions were reported by the study authors;
however, incidence was not reported, and no control group was examined. Histopathology of
livers from the high-dose animals revealed extensive fatty metamorphosis of the hepatic cells,
abnormal fat accumulation, and irregular vacuolation of the parenchymal cells extending
throughout the lobules. Livers from high-dose animals also had some cells with indistinct
borders where it appeared that adjacent cells had fused. Other liver cells had a rim of
homogenous, deeply stained cytoplasm surrounding groups of vacuoles. Similar changes, but
less severe, were observed in the mid-dose group. The liver effects observed in the low-dose
group were described as "still less severe" than those observed at the mid-dose. There was no
evidence of fibrosis or necrosis, and the Kupffer cells were unchanged. Kidneys of all exposed
animals had slight-to-moderate, light brown, granular pigmentation in the epithelial cells of the
proximal convoluted tubules, but there was no evidence of cell destruction. Histopathological
abnormalities in other organs, including the spleen, were not observed. Hematological effects
were compared to age-matched controls. There were no hematological effects observed based on
the blood analyses of the high-dose animals when compared with age-matched laboratory
historical controls. In addition, the study authors noted that similar blood counts were seen in
previously published hematological data from untreated animals and in animals treated with the
closely related compound, diphenylene oxide.

The outstanding limitations in the study design and data reporting, primarily the lack of a
concurrent control group or reporting on the number of test animals, prevent further
interpretation of the results or the determination of NOAEL and LOAEL values. The

14

Dib enzothi ophene


-------
EPA/690/R-22/002F

administered doses of 0, 13, 27, and 63 mg/kg-day correspond to HEDs of 0, 2.9, 5.9, and
13 mg/kg-day, respectively.3

Reproductive/Developmental Studies

No studies were identified.

2.2.2. Inhalation Exposures

No studies were identified.

2.3. OTHER DATA (SHORT-TERM TESTS, OTHER EXAMINATIONS)
2.3.1. Genotoxicity

The available genotoxicity data for dibenzothiophene are limited and primarily indicate a
lack of genotoxic activity. Dibenzothiophene was negative for mutagenicity in the Ames test
involving Salmonella typhimurium strains at doses up to 5,000 (.ig/plate (JECDB. 2010a; Madill
et al.. 1999; Mcfall et al.. 1984; Pelrov et al.. 1983; Dickson and Adams. 1980). with or without
metabolic activation. A positive result was reported for mutation in the photoluminescent
bacterium, Vibrio fisheri, in the Mutatox test (without activation), although the study authors
noted that a positive response in this assay can occur without deoxyribonucleic acid (DNA)
damage (Madill et al.. 1999). Studies in mammalian cells were negative for mutation in Chinese
hamster ovary (CHO) cells at doses up to 100 (.ig/mL (Rasmussen et al.. 1991) and chromosomal
aberrations (CAs) in Chinese hamster lung fibroblast (CHL) cells at doses up to 116 jag/m L for
24 hours with S9 or up to 1,850 |ig/L for 6 hours without activation (JHCDB. 2010b). A study in
cultured rainbow trout liver RTL-W1 cells reported induction of micronucleus formation by
dibenzothiophene, with an EC25 (the concentration causing 25% of the maximum effect level of
the standard, 4-nitroquinoline oxide) of 10.8 mg/mL (3.2 mg/L after correction for estimated
losses due to volatilization, sorption, etc.) (Brinkmann et al.. 2014). This is of uncertain
relevance to mammals, however, A mat et al. (2004) observed weak DNA adduct formation at
cytotoxic concentrations in HepG2 human hepatocellular carcinoma cells exposed to >50 |iM
dibenzothiophene. These studies are further described in Table 4A.

3Administered doses were converted to HEDs by multiplying by DAFs of 0.225, 0.219, and 0.208 for low-, mid-,
and high-dose rats calculated as follows: DAF = (BWa1/4 ^ BWh1/4), where BWa = animal body weight, and
BWh= human body weight. Study-specific estimated average animal body weights of 0.179, 0.161, and 0.130 kg for
low-, mid-, and high-dose rats were used. For humans, the reference value of 70 kg was used for body weight, as
recommended by U.S. EPA (1988).

15	Dib enzothi ophene


-------
EPA/690/R-22/002F

Table 4A. Summary of Dibenzothiophene (CASRN 132-65-0) Genotoxicity

Endpoint

Test System

Doses/
Concentrations
Tested3

Results
Without
Activationb

Results
With
Activationb

Comments

Reference

Genotoxicity studies in prokaryotic organisms

Reverse
mutation

Ames assay using Salmonella
typhimurium strain TA98 treated with
10-100 |ig dibenzothiophene per
plate dissolved in DMSO and
incubated at 37°C for 48 h with
Aroclor 1254-induced rat-liver S9
homogenate activation (S9
concentrations of 4, 10, or 20%)

100 (ig/plate

ND



Not mutagenic at any dose; S9 volume did not
affect activity.

Mcfall et al.
(1984)

Ames assay using S. typhimurium
strains TA98, TA100, TA1535,
TA1537, and TA1538 treated with an
unreported quantity of
dibenzothiophene dissolved in
DMSO with Aroclor 1254-induced
rat-liver S9 homogenate activation

NR

ND



Not mutagenic; mutagenicity results presented as
revertant ratio (number of revertants per
plate/number of spontaneous revertants);
dibenzothiophene reportedly had "no mutagenic
response" with an average revertant ratio <2.0.

Dickson and

Adams (1980)

Ames assay using S. typhimurium
strains TA98, TA100, TA1535, and
TA1537 treated with 2-500 |ig
dibenzothiophene per plate dissolved
in DMSO with and without
Aroclor 1254-induced rat-liver S9
homogenate activation

500 (ig/plate





Not mutagenic at any dose.

Pel to v et al.

(1983)

16

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table 4A. Summary of Dibenzothiophene (CASRN 132-65-0) Genotoxicity

Endpoint

Test System

Doses/
Concentrations
Tested3

Results
Without
Activationb

Results
With
Activationb

Comments

Reference



Ames assay (preincubation modified
plate incorporation test) using
S. typhimurium strains TA98 and
TA100 treated with unspecified
three-log dose dilutions of
dibenzothiophene in DMSO with or
without S9 activation; control
experiments conducted with DMSO
as negative control and BaP and
4-nitroqui no 1 i nc- \ -o.xide as positive
controls

NR





Not mutagenic at any dose.

Madill et al.
(1999)



Ames assay using S. typhimurium
strains TA98, TA100, TA1535, and
TA1537 and Escherichia coli
WP2uvrA treated with
dibenzothiophene in DMSO at 78.1-
5,000 ng/plate with or without S9
activation; negative and positive
controls included

5,000 ng/plate





Not mutagenic at any dose. Precipitation and
growth inhibition were seen in all strains at the
higher dose levels tested (1,250-5,000 |ig/plate).

JECDB (2010a)

Mutation

Mutatox assay in which
photoluminescent bacterium Vibrio
fisheri were incubated with 0.01-
5 ng/tube dibenzothiophene in
methanol for 45 min with or without
S9 activation; control experiments
conducted with methanol as negative
control and BaP and phenol as
positive controls

0.38 ng/tube

+



In absence of activation, positive response
(twofold increase in light output vs. negative
control) in 5 of 10 tubes in dilution series, with
0.38 ng/tube dibenzothiophene being the lowest
effective dose. In presence of activation, negative
at all doses up to 5 ng/tube. Positive response in
this assay can occur without DNA damage
(e.g., phenol, the positive control, is
nongenotoxic and noncarcinogenic).

Madill et al.
(1999)

Genotoxicity studies in nonmammalian eukaryotic cells—in vitro

Micronucleus
formation

Micronucleus assay conducted in
rainbow trout liver RTL-W1 cells
using 0.3-41 mg/L dibenzothiophene

10.8 mg/L
(EC25)

+

ND

EC25 = 10.8 mg/L for micronucleus induction
(3.2 mg/L after correction for estimated losses
due to volatilization, sorption, etc.).

Brinktnatm et al.

(2014)

17

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table 4A. Summary of Dibenzothiophene (CASRN 132-65-0) Genotoxicity

Endpoint

Test System

Doses/
Concentrations
Tested3

Results
Without
Activationb

Results
With
Activationb

Comments

Reference

Genotoxicity studies in mammalian cells—in vitro

Mutation

Chinese hamster ovary
(CHO-K1BH4) cells treated with 1-
100 (ig/mL dibenzothiophene with
Ham's F12 medium, activated with
4% Aroclor-induced rat-liver S9
solution and incubated for 5 h;
control experiments conducted with
DMSO as the control and methyl
methane sulfonate as a positive
control

100 (ig/mL

ND



Not mutagenic at any dose.

Rasiiius.se ti et al.

(1991)

CAs

Cultured Chinese hamster lung
fibroblast CHL/IU cells incubated
with dibenzothiophene at
7.23-116 |ig/mL for 6-24 h (without
S9) or 57.8-1,850 ng/mL for 6 h
(with S9)

1,850 ng/mL





No effect on structural or numerical aberrations
at any dose. Toxicity was observed at the higher
doses in the tests without S9 (86.9-116 ng/mL)
but not in the tests with S9.

JECDB (2010b)

DNA adduct
formation

Cultured HepG2 human
hepatocellular carcinoma cells
incubated with 0.25-150 |iM
dibenzothiophene (dissolved in
methanol) for 24 or 48 h; control
experiments conducted with culture
medium alone as negative control and
BaP in DMSO as positive control.

50 \M

±

ND

Weak DNA adduct formation at cytotoxic
concentrations.

A mat et al. (2004)

aLowest effective dose for positive results, highest dose tested for negative results.
b+ = positive; ± = weakly positive; - = negative.

BaP = benzoic/1 pyre nc: CA = chromosomal aberration; DMSO = dimethylsulfoxide; DNA = deoxyribonucleic acid; EC25 = the concentration causing 25% of the
maximum effect level of the standard, 4-nitroquinoline oxide; ND = no data.

18

Dib enzothi ophene


-------
EPA/690/R-22/002F

2.3.2. Metabolism/Toxicokinetic and Supporting Animal Studies

Additional studies investigating the metabolism of dibenzothiophene in rats (Jacob et al.. 1991;
Vignier et al., 1985), elimination of dibenzothiophene in the urine of rabbits (Thomas et al., 1942), acute
toxicity of dibenzothiophene in mice (Leigfaton. 1989), and toxicity of dibenzothiophene by weekly
injection in rats (Silva et al .. 2015), as well as in vitro studies of effects of dibenzothiophene on
aggregation of platelets (Chaudhurv et al .. 1988) and viability of differentiated SK-N-SH human
neuroblastoma cells (Sarma et al.. 2017) are also available. See Table 4B for details of these studies.

19

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table 4B. Other Studies

Test

Materials and Methods

Results

Conclusions

Reference

Supporting Animal Studies

Short-term

CD-I mice treated via gavage.

Pilot ranse-findins studies: 4 mice/sex/dose.

Pilot ranee finding studies: No treatment-related

Without induction: acute LD5o
of 470 mg/kg; with prior
induction of MFO, acute LD5o
of 335 mg/kg; preinduction of
MFO potentiated the toxicity of
dibenzothiophene.

Leighton (1989)

hematological changes seen; no treatment-related
histological lesions seen in the kidney, duodenum,
spleen, or heart; liver lesions included centrilobular
or periacinar degeneration and necrosis.

LDsu experiments: All mortality occurred within
72 h of treatment and was increased in groups with
prior induction of MFO; animals were sluggish;
gross lesions in mice found dead included
pulmonary congestion and edema, mild to
moderate hydrothorax, intestinal hemorrhage, and
mottled livers; all MFO-induced mice had mild
fibrinous peritonitis; histological lesions included
severe centrilobular hepatic necrosis across doses
in both experiments, necrosis of lymphocytes in
thymic cortices at >540 mg/kg in Experiment 1 and
>265 mg/kg in Experiment 2, and degenerative
changes in the walls of small arteries in the lung in
five mice dosed with 265-492 mg/kg
(Experiment 2).



single dose of 0-3,250 mg/kg or
four consecutive daily doses of 0-325 mg/kg;
necropsy performed 24 h after last dose; blood
taken from hearts of mice and examined for
hematological effects; liver, kidney, spleen,
heart, lungs, thymus, and duodenum examined at
necropsy.

LDsu Experiment 1: 12 male mice/dose.
12 vehicle controls, 8 untreated controls; single
doses of 0, 260, 374, 540, 777, 1,118, or
1,609 mg/kg; LD5o determined at 7 d; surviving
mice sacrificed on day 14 for histology of liver,
lung, heart, and thymus.

LDsu Experiment 2: 12 mice/treatment aroiiD.
5 preinduced vehicle controls; MFO pretreatment
(one i.p. injection of 3-methylcholanthrene
[80 mg/kg] followed by daily i.p. injections of
phenobarbital [50 mg/kg] in sterile saline for 3 d)
followed 24 h later by single doses of 0, 215,
265, 325, 400, 492, 605, or 744 mg/kg.

20

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table 4B. Other Studies

Test

Materials and Methods

Results

Conclusions

Reference

Injection

Male Wistar rats were given weekly i.p.
injections of saline (n = 5), soy oil (n = 5) or
30 mg/kg dibenzothiophene in soy oil (n = 15)
for 10 wk and were examined after a latency
period of 14 wk for hematology (RBC,
hematocrit, WBC total and differential), serum
chemistry (amylase, ALT, AST), organ weights
(liver and spleen), histology (liver, spleen, lungs,
and intestines), immunohistochemistry (using
anti-CEA and anti-CD44 antibodies) and
proteomic analysis. Body weights were recorded
weekly.

There were no treatment-related effects on body
weight, hematology, serum chemistry, or organ
weights. Histopathological examination showed
effects only in the intestines, including increased
incidence of cellular atypia in the mucosa and
submucosa, mucosa inflammation, necrosis, and
hyperplasia of lymphoid nodules in both the large
and small intestines. Counts of cells positive for
antibody labelling were increased threefold (either
CEA or CD44), suggesting the presence of
preneoplasia. Proteomic analysis identified
23 proteins showing altered levels (>1.5-fold
change versus controls) in the small intestine, with
functions including heat-shock response;
cytoskeleton organization; antioxidant activity; cell
signaling; carbohydrate, lipid and nucleotide
metabolism; and protein folding.

Dibenzothiophene produced
dysplastic lesions in the large
and small intestines of male rats
treated weekly by injection of
30 mg/kg in soy oil for 10 wk
and examined at 24 wk.

Silva et al.
(2015)

Metabolism/toxico kinetic

Metabolism/
Toxicokinetic

Male Wistar rat (number not specified), treated
with daily i.p. injections of 40 mg/kg
dibenzothiophene for 3 d, 3-methylcholanthrene
for 3 d, 500 mg/kg Aroclor 1254 for 5 d, or twice
daily i.p. injections of 40 mg/kg phenobarbital
for 4 d, then starved for 24 h after final injection;
liver microsomes isolated; in vitro oxidation
assay performed using dibenzothiophene
(0.02-0.50 mM) and rat liver microsomal
suspension (10 |iL).

Dibenzothiophene metabolic pathway determined
to be S-oxidation with metabolites of
dibenzothiophene-5-oxide (primary) and
dibenzothiophene -5 -dioxide (secondary);

Aroclor 1254, 3-methylcholanthrene, and
phenobarbital increased rate of formation of
sulfoxide, but dibenzothiophene pretreatment had
no effect; carbon monoxide inhibited sulfoxidation.

Dibenzothiophene metabolite,
dibenzothiophene-5-oxide, was
further oxidized to
dibenzothiophene-5-dioxide;
CYP450 monooxygenases most
likely involved in the
metabolism.

Vienier et al.
(1985)

21

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table 4B. Other Studies

Test

Materials and Methods

Results

Conclusions

Reference



Male Wistar rat (number not specified), treated
with i.p. injections of 40 mg
5,6-benzoflavone/kg for 3 d, 200 mg Aroclor/kg
once, or 80 mg phenobarbital/kg in 0.9% NaCl
over 3 d and sacrificed 24 h after last dose;
microsomes from four animals per group were
incubated with 50 (imol/L dibenzothiophene for
20 min at 37°C and analyzed; solvent-only
controls.

Metabolic products were sulfoxide (main product)
and sulfone; no pretreatments affected sulfoxide
formation, but pretreatments with phenobarbital
and Aroclor increased sulfone formation.

Dibenzothiophene metabolites
(using rat microsomes) were
sulfoxide and sulfone,
controlled by different enzymes;
only the one responsible for
sulfone formation can be
induced by CYP450 inducers
such as phenobarbital.

Jacob et al.

(1991)

One rabbit (sex and strain not specified) given an
emulsion of 2 g dibenzothiophene in water
administered via stomach tube; urine collected
(time not specified) and analyzed.

Main excretion product was mono-hydroxy-
diphenylene sulfone.

Dibenzothiophene oxidized to
mono-hydro xy-diphenylene
sulfone in the rabbit.

Thomas et al.
(1942)

Mode of action/mechanistic

In vitro

Aggregation of platelets from blood collected
from male Sprague Dawley rats was measured
photometrically after incubation for 2 min with
1. 3. or 5 |iL of dibenzothiophene in DMSO or
DMSO alone. Tests were also conducted with
addition of thrombin or ADP to stimulate platelet
aggregation, mobilization of internal calcium
stores within platelets, and uptake of
extracellular calcium by platelets.

Platelet aggregation was significantly reduced in a
dose-related manner by dibenzothiophene relative
to controls. Dibenzothiophene also reduced
aggregation stimulated by thrombin or ADP and
reduced extracellular calcium uptake by platelets.

Dibenzothiophene may inhibit
platelet aggregation by bringing
about alterations in the platelet
plasma membrane.

Chaudhurv et al.

(1988)

Differentiated SK-N-SH human neuroblastoma
cells were cultured for 24 h with 5-100 |iIVI
dibenzothiophene and assessed for viability and
production of ROS relative to unexposed
controls.

Neuroblastoma cell viability was reduced relative
to controls at all test concentration, with
LCio = 5.07 nM and LC20 = 47.26 ^M. Significant
increases in ROS were seen at >10 ^M.

Dibenzothiophene induces
neuronal cell damage by a
mechanism that involves
generation of oxidative stress.

Sarma et al.

(2017)

ADP = adenosine diphosphate; ALT = alanine aminotransferase; AST = aspartate aminotransferase; CYP450 = cytochrome P450; DMSO = dimethyl sulfoxide;
i.p. = intraperitoneal; MFO = mixed function oxidase; NaCl = sodium chloride; RBC = red blood cell; ROS = reactive oxygen species; WBC = white blood cell.

22

Dib enzothi ophene


-------
EPA/690/R-22/002F

3. DERIVATION OF PROVISIONAL VALUES

3.1.	DERIVATION OF ORAL REFERENCE DOSES

The database of repeat-dose oral studies for dibenzothiophene is limited to a
non-peer-revievved, 28-day study in Japanese with only tables and figures in English (JHCDB.
2011) and a peer-reviewed, 165-day study from 1942 that relied on historical laboratory control
groups instead of a concurrent control (Thomas et al.. 1942). Due to the shortcomings of these
studies, reference doses (RfDs) cannot be confidently derived here. However, the studies provide
sufficient data to develop a screening subchronic provisional reference dose (p-RfD) value
(see Appendix A).

3.2.	DERIVATION OF INHALATION REFERENCE CONCENTRATIONS

No subchronic or chronic provisional reference concentration (p-RfC) can be derived
because no inhalation studies on exposure to dibenzothiophene were identified.

The feasibility of using an analogue approach was attempted for the derivation of
screening-level p-RfC values via read-across but no candidate analogues with inhalation toxicity
values were identified (see Appendix A).

3.3.	SUMMARY OF NONCANCER PROVISIONAL REFERENCE VALUES

Table 5 presents a summary of the noncancer provisional references values.

Table 5. Summary of Noncancer Risk Estimates for Dibenzothiophene

(CASRN 132-65-0)

Toxicity Type
(units)

Species/
Sex

Critical Effect

p-Reference
Value

POD

Method

POD

(HED/HEC)

UFc

Principal Study

Screening
subchronic p-RfD
(mg/kg-d)
(see Appendix A)

Rat/F

Increased

hepatocyte

hypertrophy

3 x 1(T3

BMDL io

1.04

300

JHCDB (20ID

Chronic p-RfD
(mg/kg-d)

NDr

Subchronic p-RfC
(mg/m3)

NDr

Chronic p-RfC
(mg/m3)

NDr

BMD = benchmark dose; BMDL = 95% benchmark dose lower confidence limit on the BMD (subscripts denote
BMR: i.e., 10 = dose associated with 10% extra risk); HEC = human equivalent concentration; HED = human
equivalent dose; NDr = not determined; POD = point of departure; p-RfC = provisional reference concentration;
p-RfD = provisional reference dose; UFC = composite uncertainty factor.

3.4. CANCER WEIGHT-OF-EVIDENCE DESCRIPTOR

No human or animal data were located on the carcinogenicity of dibenzothiophene by
oral or inhalation exposure. One available injection study observed dysplastic lesions in the
intestines of rats treated for 10 weeks, suggesting that dibenzothiophene may have some
carcinogenic potential (Silva et al.. 2015). Genotoxicity studies were largely negative, including

23

Dib enzothi ophene


-------
EPA/690/R-22/002F

multiple Ames tests for mutation in bacteria and assays for mutation and CAs in mammalian
cells (JHCDB. 2010a. b; Madill et al.. 1999; Rasmussen et al.. 1991; Mcfall et al.. 1984; Pelrov
et al.. 1983; Dickson and Adams, 1980). Under the U.S. EPA Cancer Guidelines (U.S. EPA,
2005), there is "Inadequate Information to Assess Carcinogenic PotentiaT of dibenzothiophene
by oral or inhalation exposure (see Table 6).

Table 6. Cancer WOE Descriptor for Dibenzothiophene (CASRN 132-65-0)

Possible WOE Descriptor

Designation

Route of Entry
(oral, inhalation,
or both)

Comments

"Carcinogenic to Humans"

NS

NA

No human carcinogenicity data are available.

"Likely to Be Carcinogenic
to Humans"

NS

NA

No adequate animal cancer bioassays or human
cancer data are available.

"Suggestive Evidence of
Carcinogenic Potential"

NS

NA

No adequate animal cancer bioassays or human
cancer data are available.

"Inadequate Information to
Assess Carcinogenic
Potential"

Selected

Both

Selected due to the lack of adequate data on
carcinogenicity. One injection study provided
limited evidence of carcinogenic potential.

"Not Likely to Be
Carcinogenic to Humans"

NS

NA

No evidence of noncarcinogenicity is available.

NA = not applicable; NS = not selected; WOE = weight-of-evidence.

3.5. DERIVATION OF PROVISIONAL CANCER RISK ESTIMATES

The absence of suitable data precludes the development of cancer risk estimates for
dibenzothiophene (see Table 7).

Table 7. Summary of Cancer Risk Estimates for Dibenzothiophene

(CASRN 132-65-0)

Toxicity Type

Species/Sex

Tumor Type

Cancer Value

Principal Study

p-OSF (mg/kg-d) 1

NDr

p-IUR (mg/m3) 1

NDr

NDr = not determined; p-IUR = provisional inhalation unit risk; p-OSF = provisional oral slope factor.

24

Dib enzothi ophene


-------
EPA/690/R-22/002F

APPENDIX A. SCREENING PROVISIONAL VALUES

Due to the lack of evidence described in the main Provisional Peer-Reviewed Toxicity
Value (PPRTV) document, it is inappropriate to derive provisional toxicity values for
dibenzothiophene. However, some information is available for this chemical, which although
insufficient to support derivation of a provisional toxicity value under current guidelines, may be
of limited use to risk assessors. In such cases, the Center for Public Health and Environmental
Assessment (CPHEA) summarizes available information in an appendix and develops a
"screening value." Appendices receive the same level of internal and external scientific peer
review as the provisional reference values to ensure their appropriateness within the limitations
detailed in the document. Users of screening toxicity values in an appendix to a PPRTV
assessment should understand that there could be more uncertainty associated with deriving an
appendix screening toxicity value than for a value presented in the body of the assessment.
Questions or concerns about the appropriate use of screening values should be directed to the
CPHEA.

A screening subchronic provisional reference doses (p-RfD) was derived for
dibenzothiophene as described in the section below. For inhalation, an alternative analogue
approach was evaluated (see APPLICATION OF AN ALTERNATIVE ANALOGUE
APPROACH (METHODS) below), but suitable analogues were not identified and a screening
value was not derived.

DERIVATION OF SCREENING PROVISIONAL REFERENCE DOSES

As discussed in the main body of the report, the available repeat-dose oral studies for
dibenzothiophene include only JECDB (2011) and Thomas et al. (19421 both of which have
limitations precluding their use in deriving provisional toxicity values. In order to account for the
uncertainty associated with basing a toxicity assessment on these studies, the assessment is
considered a screening-level assessment.

The 28-day oral exposure study by JECDB (2011) is limited by unpublished status, lack
of peer review, and use of Japanese language with only tables and figures in English. There was
enough material presented in English, however, to ascertain that the study appeared to be
adequately designed and conducted, and to provide dose-response information on a wide range
of endpoints suitable for use in quantitative toxicity assessment, including body weight, food
consumption, clinical observations, functional observational battery (FOB), hematology, serum
chemistry, urinalysis, and selected organ weight and histopathology (see study summary in
Section 2.2.1 for more details). Liver effects were a sensitive target for dibenzothiophene in the
JECDB (2011) study. Dose-related and biologically significant (>10%) increases in relative liver
weight were reported in both male and female rats at >10 mg/kg-day and were a primary basis
for the study reported no-observed-adverse-effect level (NOAEL)/lowest-observed-adverse-
effect level (LOAEL) values (NOAEL = 3 mg/kg-day and LOAEL =10 mg/kg-day). The
relative liver weight changes were accompanied by increased incidence of hepatocyte
hypertrophy in both sexes at >10 mg/kg-day (1/6 females at 10 mg/kg-day and 6/6 males and
females at 30 mg/kg-day) and possible evidence of structural degeneration (slight necrosis in
1/6 animals) and changes in serum markers of liver damage (decreased albumin protein fraction
and albumin/globulin [A/G] ratio]) in males at 30 mg/kg-day. Although increases (>10%) in
absolute liver weights were observed in rats at >3 mg/kg-day, the changes at the lowest dose

25

Dib enzothi ophene


-------
EPA/690/R-22/002F

were not supported by corroborative evidence of liver toxicity and overall pattern of effects
lacked a dose-response relationship (see Table B-l). The findings across organ weight,
histopathology, and clinical chemistry measures provide coherent evidence of liver toxicity after
short-term oral exposure to dibenzothiophene.

Other treatment-related effects observed at the JECDB (2011) study LOAEL of
10 mg/kg-day included significant reductions in motor activity, although there was no
corroborative evidence from other FOB assays evaluating reactivity (visual, touch, auditory,
pain, proprioceptive), righting reflex, or grip strength. Additionally, significant increases in
prothrombin time (PT) were reported in males at >10 mg/kg-day, accompanied by significant
increases in activated partial thromboplastin time (APTT) at 30 mg/kg-day in these animals.
Prolonged clotting times (i.e., increased PT and APTT) are consistent with findings of
dibenzothiophene-induced platelet aggregation in vitro (Chaudhurv et al.. 1988). and decreased
motor activity is consistent with findings of decreased viability of differentiated SK-N-SH
human neuroblastoma cells with in vitro dibenzothiophene exposure (Sarma et al.. 2017).
However, there is limited in vivo evidence to determine the biological significance of the
changes in motor activity and prolonged clotting times in males.

Male rats in the JECDB (2011) study exhibited biologically significant increases (>10%)
in relative kidney weights at 30 mg/kg-day. Dose-related increases in the incidence of hyaline
droplets and eosinophilic bodies in the proximal tubular epithelium of the kidney (one of six,
two of six, and six of six animals at 3, 10, and 30 mg/kg-day, respectively) also occurred in
males. Accumulation of hyaline droplets (also described as cytoplasmic eosinophilic bodies
containing protein) are commonly associated with alpha 2u-globulin (a2u-g)-mediated
nephropathy (Hard et al.. 1999). a male rat-specific nephropathy not considered relevant to
humans. According to (U.S. EPA. 1991). three criteria are required for evaluating the relevance
of kidney lesions in males based on possible involvement of a2u-g: (1) observation of an
increase in number and size of hyaline droplets only in male kidneys; (2) identification of the
protein contained in the hyaline droplets as a2u-g; and (3) observation of additional events in the
pathological sequence of lesions associated with a2u-g disease (i.e., single cell necrosis,
exfoliation of epithelial cells into tubular lumen, and granular casts). The evidence for
dibenzothiophene is limited to increases in hyaline droplets occurring only in male rats in the
28-day JECDB (2011) study. Given that the study is in Japanese, it is unclear whether the study
authors performed any specialized staining for detection of a2u-g and no additional observations
were made regarding other events in the pathological sequence of the development of a2u-g
disease. Further, there is no supporting evidence for a2u-g-, including the 165-day study by
Thomas et al. (1942). Thomas et al. (1942) reported slight-to-moderate, light brown, granular
pigmentation in the epithelial cells of the proximal convoluted tubules of male rats (with no
evidence of cell destruction) but no other details were provided; therefore, the toxicological
significance of the findings is unknown. The limitations in the database for dibenzothiophene
prevent further interpretation of the relevance of the male rat kidney lesions in the JECDB
(2011) study. Given the uncertainty and lack of information for further evaluation, these kidney
lesions in male rats (hyaline droplets and eosinophilc bodies) were not further considered for
dose-response analysis.

26

Dib enzothi ophene


-------
EPA/690/R-22/002F

The 165-day dietary exposure study in rats by Thomas et al. (1942) had outstanding
limitations such as the lack of a concurrent control group, instead making inferences based on
historical control groups. Other limitations included the use of males only, lack of reporting on
the number of test animals per group, and incomplete data reporting for histopathological
outcomes. The major study findings were increases in liver weight that reached 35% at
27 mg/kg-day and 115% at 63 mg/kg-day (over body-weight-matched laboratory historical
controls). Histopathological lesions in the liver (i.e., fat accumulation, irregular vacuolation of
the parenchymal cells [hepatocytes] throughout the lobules, and indications that adjacent cells
had fused) were also observed at all doses; however, incidence of lesions was not provided, and
severity was described as much less in the low- and mid-dose groups (13 and 27 mg/kg-day)
compared to the high-dose group (63 mg/kg-day). Although the study limitations add
considerable uncertainty to the interpretation of the findings or the determination of
NOAEL/LOAEL values, these observations are consistent with the liver effects in the JECDB
(2011) study, providing supportive evidence of dibenzothiophene-induced liver toxicity.

Overall, the increases in relative liver weight and liver lesions (primarily hypertrophy)
and decreases in serum markers of liver function (albumin protein fraction and A/G ratio)
provide coherent evidence of liver effects in rats at >10 mg/kg-day after 28-day exposure
(JECDB. 2011). Although the relevance of male kidney lesions reported in the JECDB (2011)
study is unclear, the changes in relative kidney weights in males at 30 mg/kg-day were
considered biologically significant (>10%). Therefore, both the liver effects and relative kidney
weight changes from this study were considered further for the derivation of screening p-RfDs.
Other treatment-related effects (decreased motor activity, increased PT and APTT and increased
hyaline droplets in males) in the JECDB (2011) study were not advanced for dose-response
analysis due to the limitations in the database for dibenzothiophene, which prevent further
determination of the toxicological significance of the findings.

Derivation of Screening Subchronic Provisional Reference Dose

Data for liver effects in male and female rats and increased relative kidney weights in
male rats from the JECDB (2011) study were modeled using the U.S. Environmental Protection
Agency (U.S. EPA) Benchmark Dose Software (BMDS, Version 3.2). Despite the
non-peer-reviewed status and lack of full English language report, the study used an adequate
design (28-day rat study), included multiple doses and a comprehensive array of toxicity
endpoints, and identified sensitive health effects that are suitable for the derivation of the
screening subchronic p-RfD (JECDB. 2011). For liver effects, dose-related increases in relative
liver weight in males and females at >10 mg/kg-day were modeled as continuous data using a
benchmark response (BMR) of 10% relative deviation (RD) because a 10% change in liver
weight is considered a minimally biologically significant response in adult animals. Hepatocyte
hypertrophy was modeled in females as dichotomous data, applying a standard BMR of 10%
extra risk (ER). Hepatocyte hypertrophy in males was not modeled given that the effects were
only observed in the high-dose group. Although the decreases in some serum markers of liver
function (albumin protein fraction and A/G ratio) in male rats provide supporting evidence for
dibenzothiophene-induced liver effects, these endpoints were not considered for dose-response
assessment since more sensitive and relevant markers of liver toxicity were available
(i.e., relative liver weight and hepatocyte hypertrophy). Increased relative kidney weight in males
were modeled as continuous data using a BMR of 10%, which is considered biologically
significant. Human equivalent doses (HEDs) in mg/kg-day were used as the dose metric for
BMD analysis.

27

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table A-l shows the data for liver and kidney endpoints that were considered for dose-
response assessment and Table A-2 summarizes the BMD modeling results and provides
candidate points of departure (PODs) for the derivation of the screening subchronic p-RfD.
Details of model fit for each data set are presented in Appendix C. Candidate PODs that could
not be evaluated via BMD analysis (i.e., hepatocyte hypertrophy in males) are presented as
NOAEL/LOAEL values.

Table A-l. Data for Sensitive Endpoints in Male and Female
Sprague Dawley Rats After Oral Treatment with
Dibenzothiophene for 28 Days"

Endpoint

ADD [HED] in mg/kg-db



0

3 [0.68 female,
0.75 male]

10 [2.2 female,
2.5 male]

30 [6.7 female,
7.4 male]

Increased relative liver
weight in males0

3.233 ±0.247

3.512 ±0.271
(+9%)

3.578 ±0.153*
(+11%)

4.465 ± 0.208**
(+38%)

Increased relative liver
weight in females0

3.123 ±0.170

3.355 ±0.145
(+7%)

3.450 ±0.299*
(+10%)

3.970 ±0.187**
(+27%)

Increased hepatocyte
hypertrophy in males'1

0/6 (0%)

0/6 (0%)

0/6 (0%)

6/6 (100%)

Increased hepatocyte
hypertrophy in females'1

0/6 (0%)

0/6 (0%)

1/6 (17%)

6/6 (100%)

Increased relative kidney
weight in males0

0.732 ±0.040

0.735 ±0.023
(+0%)

0.798 ±0.052*
(+9%)

0.823 ±0.031**
(+12%)

aJECD6 (2011).

bADDs were converted into HEDs (HED = ADD x DAF) using DAFs of 0.250, 0.248, and 0.247 for low-, mid-,
and high-dose males and 0.226, 0.223, and 0.222 for low-, mid-, and high-dose females calculated as follows:
DAF = (BWa1/4 + BWh1'4), where BWa = animal body weight, and BWh = human body weight. Study-specific
TWA animal body weights of 0.272, 0.264, and 0.259 kg for low-, mid-, and high-dose males, and 0.182, 0.174,
and 0.171 kg for low-, mid-, and high-dose females were used. For humans, the reference value of 70 kg was used
for body weight, as recommended by U.S. EPA (1988).

Data are means ± SD; n = 6 for all data points; value in parentheses is % change relative to control = ([treatment
mean - control mean] + control mean) x 100.

dData are number of animals showing changes/ total number of animals examined (% incidence).

* Significantly different from control (p < 0.05) by Dunnett's test as reported by the study authors.

**Significantly different from control (p < 0.01) by Dunnett's test as reported by the study authors.

ADD = adjusted daily dose; DAF = dosimetric adjustment factor; HED = human equivalent dose; SD = standard
deviation; TWA = time-weighted average.

28

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table A-2. Comparison of Candidate POD Values in Male and Female
Sprague Dawley Rats After Oral Treatment with Dibenzothiophene

for 28 Days"

Endpoint

Best-Fitting Model

BMR

BMDL (HED)

(mg/kg-d)

POD type

POD (HED)

(mg/kg-d)

Increased relative liver
weight in males

Exponential 3
(constant variance)

10% RD from
control (0.1 RD)

2.01

BMDL

2.01

Increased relative liver
weight in females

Linear (constant
variance)

10% RD from
control (0.1 RD)

2.19

BMDL

2.19

Increased hepatocyte
hypertrophy in males

Data not amenable for BMD modeling13

NOAEL

2.5

Increased hepatocyte
hypertrophy in females

Probit (constant
variance)

10% ER from
control (0.1 ER)

1.04

BMDL

1.04

Increased relative
kidney weight in males

Exponential 4
(constant variance)

10% RD from
control (0.1 RD)

1.36

BMDL

1.36

aJECD6 (2011).

bData were not considered amenable for BMD modeling given that incidence was 100% at the highest dose and 0%
at lower doses.

BMD = benchmark dose; BMDL = benchmark dose lower confidence limit; BMR = benchmark response;
ER = Extra Risk; HED = human equivalent dose; NOAEL = no-observed-adverse-effect level; POD = point of
departure; RD = relative deviation.

The 10% benchmark dose lower confidence limit (BMDLio) (HED) of 1.04 mg/kg-day
for increased hepatocyte hypertrophy in female rats in the JECDB (2011) study is the lowest
POD in the available database and is expected to be protective of other health effects associated
with dibenzothiophene oral exposure. The significance of dibenzothiophene-induced liver effects
is based on coherent evidence across organ weights (increased relative liver weight),
histopathology (primarily hypertrophy with some evidence of necrosis), and serum markers of
liver function (decreased albumin protein fraction and A/G ratio) in rats at >10 mg/kg-day after
28-day oral exposure (JECDB. 2011). Supportive evidence of potential liver toxicity was also
found after dietary exposure for 165 days in males rats (increased liver weight and fatty
accumulation in the liver at >27 mg/kg-day) (Thomas et al.. 1942) and acute gavage range-
finding and median lethal dose (LD50) experiments in mice (centrilobular degeneration and
necrosis across 260-1,609 mg/kg) (Leighton. 1989). Altogether, the weight of evidence suggests
that the liver is a primary target for dibenzothiophene via oral exposure and the BMDLio [HED]
of 1.04 mg/kg-day for increased hepatocyte hypertrophy in female rats exposed for 28 days
(JECDB. 2011) is selected as the most sensitive POD for the derivation of the subchronic p-RfD.

29

Dib enzothi ophene


-------
EPA/690/R-22/002F

The screening subchronic p-RfD of 3 x 10 3 mg/kg-day for dibenzothiophene is derived
by applying a composite uncertainty factor (UFc) of 300 (reflecting an interspecies uncertainty
factor [UFa] of 3, an interindividual variability uncertainty factor [UFh] of 10, and a database
uncertainty factor [UFd] of 10) to the selected POD of 1.04 mg/kg-day, as follows:

Screening Subchronic p-RfD = POD (HED) UFc

= 1.04 mg/kg-day -^300
= 3 x 10"3 mg/kg-day

Table A-3 summarizes the uncertainty factors for the screening subchronic p-RfD for
dib enzothi ophene.

Table A-3. Uncertainty Factors for the Screening Subchronic p-RfD for
Dibenzothiophene (CASRN 132-65-0)

UF

Value

Justification

UFa

3

A UFa of 3 (100 5) is applied to account for uncertainty in characterizing the toxicodynamic
differences between rats and humans following oral dibenzothiophene exposure. The toxicokinetic
uncertainty has been accounted for by calculation of an HED through application of a DAF as
outlined in the U.S. EPA's Recommended Use of Body Weight4 as the Default Method in Derivation
of the Oral Reference Dose (U.S. EPA, 2011c).

UFd

10

A UFd of 10 is applied to account for deficiencies and uncertainties in the database. The repeat-dose
oral database for dibenzothiophene includes a non-peer-reviewed, 28-day rat study in Japanese and a
chronic rat study with significant limitations (primarily lack of a concurrent control and reporting on
the number of test animals). No reproductive or developmental toxicity studies are available by any
route of exposure.

UFh

10

A UFh of 10 is applied for interindividual variability to account for human-to-human variability in
susceptibility in the absence of quantitative information to assess the toxicokinetics and
toxicodynamics of dibenzothiophene in humans.

UFl

1

A UFl of 1 is applied because the POD is a BMDL.

UFS

1

A UF s of 1 is applied because the POD was derived from a study of suitable duration (28 days) for a
subchronic value.

UFC

300

Composite UF = UFa x UFd x UFh x UFl x UFs.

DAF = dosimetric adjustment factor; HED = human equivalent dose; POD = point of departure;

BMDL = benchmark dose lower confidence limit; p-RfD = provisional reference dose; UF = uncertainty factor;
UFa = interspecies uncertainty factor; UFC = composite uncertainty factor; UFD = database uncertainty factor;
UFh = intraspecies uncertainty factor; UFL = LOAEL-to-NOAEL uncertainty factor; UFS = subchronic-to-chronic
uncertainty factor.

Derivation of Screening Chronic Provisional Reference Dose

The POD used for derivation of the screening subchronic p-RfD based on increased
hepatocyte hypertrophy in female rats (BMDLio [HED] of 1.04 mg/kg-day) from the 28-day
study by JECDB (2011), cannot be used directly for derivation of the screening chronic p-RfD,
due to the short duration of the critical study. The available 165-day chronic study by Thomas et
al. (1942) reported increases in liver weight and liver histopathology at >27 mg/kg-day, which
are similar to the doses associated with liver effects in the 28-day JECDB (2011) study
(>10 mg/kg-day). However, the limitations in the study design and data reporting in Thomas et

30

Dib enzothi ophene


-------
EPA/690/R-22/002F

al. (1942) raise significant concerns regarding the interpretation of the study findings. Overall,
the lack of adequate data to inform whether the liver or other health effects associated with
dibenzothiophene worsen with chronic exposure prevent the derivation of a screening chronic
p-RfD.

APPLICATION OF AN ALTERNATIVE ANALOGUE APPROACH (METHODS)

The analogue approach allows for the use of data from related compounds to calculate
screening values when data for the compound of interest are limited or unavailable. Details
regarding searches and methods for analogue analysis are presented in Wang et al. (2012). Three
types of potential analogues (structural, metabolic, and toxicity-like) are identified to facilitate
the final analogue chemical selection. The analogue approach may or may not be route-specific
or applicable to multiple routes of exposure. All information was considered together as part of
the final weight-of-evidence (WOE) approach to select the most suitable analogue both
toxicologically and chemically.

An expanded analogue identification approach was developed to collect a more
comprehensive set of candidate analogues for the compounds undergoing a U.S. EPA PPRTV
screening-level assessment. As described below, this method includes application of a variety of
tools and methods for identifying candidate analogues that are similar to the target chemical
based on chemical structure and key features; metabolic relationships; or related toxic effects and
mechanisms of action.

To identify structurally-related compounds, an initial pool of analogues is identified using
automated tools, including ChemlDplus (ChcmlDplus. 2021), CompTox Chemicals Dashboard
(U.S. HP A. 202 lb), and Organisation for Economic Co-operation and Development (OECD)
Quantitative Structure-Activity Relationship (QSAR) Toolbox (OECD, 2021). to conduct
structural similarity searches. Additional analogues identified as ChemlDplus-related substances,
parent, salts, and mixtures, and CompTox-related substances are considered. CompTox GenRA
analogues are collected using the methods available on the publicly available GenRA Beta
version, which may include Morgan fingerprints, Torsion fingerprints, ToxPrints and ToxCast,
Tox21, and ToxRef data. For compounds that have very few analogues identified by structure
similarity using a similarity threshold of 0.8 or 80%, substructure searches in the QSAR Toolbox
may be performed, or similarity searches may be rerun using a reduced similarity threshold
(e.g., 70 or 60%). The compiled list of candidate analogues is batch run through the CompTox
Chemicals Dashboard where QSAR-ready simplified molecular-input line-entry system
(SMILES) are collected and toxicity data availability is determined (e.g., from the Agency for
Toxic Substances and Disease Registry [ATSDR], Office of Environmental Health Hazard
Assessment [OEHHA), California Environmental Protection Agency [CalEPA], U.S. EPA
Integrated Risk Information System [IRIS], PPRTVs). The batch output information is then
uploaded into the Chemical Assessment Clustering Engine (ChemACE) (U.S. HP A. 201 la),
which clusters the chemicals based on chemical fragments and displays the toxicity data
availability for each candidate. The ChemACE output is reviewed by an experienced chemist,
who narrows the list of structural analogues based on known or expected structure-toxicity
relationships, reactivity, and known or expected metabolic pathways.

31

Dib enzothi ophene


-------
EPA/690/R-22/002F

Toxicokinetic studies identified from the literature searches performed for this PPRTV
assessment were used to identify metabolic analogues (metabolites and metabolic precursors).
Metabolites were also identified from the two OECD QSAR Toolbox metabolism simulators (in
vivo rat metabolism simulator and rat liver S9 metabolism simulator). Targeted PubMed
searches were conducted to identify metabolic precursors and other compounds that share any of
the observed or predicted metabolites identified for the target chemical. Metabolic analogues are
then added to the pool of candidate analogues and toxicity data availability is determined
(e.g., from AT SDR, OEHHA, CalEPA, U.S. EPA IRIS, PPRTVs).

In vivo toxicity data for the target chemical (if available from the literature searches) are
evaluated to determine whether specific or characteristic toxicity was observed
(e.g., cholinesterase inhibition, inhibition of oxidative phosphorylation). In addition, in vitro
mechanistic data identified from the literature searches or obtained from tools including GenRA,
ToxCast/Tox21, and Comparative Toxicogenomics Database (CTD) (Davis et al.. 2021) were
evaluated for this purpose. Data from CompTox Chemicals Dashboard ToxCast/Tox21 are
collected to determine bioactivity of the target chemical in in vitro assays that may indicate
potential mechanism(s) of action. The GenRA option within the Dashboard also offers an option
to search for analogues based on similarities in activity in ToxCast/Tox21 in vitro assays. Using
the ToxCast/Tox21 bioactivity data, nearest neighbors identified with similarity indices of
>0.5 may be considered potential candidate analogues. The CTD (Davis et al.. 2021) is searched
to identify compounds with gene interactions similar to interactions induced by the target
chemical; compounds with gene interactions similar to the target chemical (with a similarity
index >0.5) may be considered potential candidate analogues. These compounds are then added
to the pool of candidate analogues, and toxicity data availability is determined (e.g., from
AT SDR, OEHHA, CalEPA, U.S. EPA IRIS, PPRTVs).

The tools used for the expanded analogue searches were selected because they are
publicly available, which allows for transparency and reproducibility of the results, and because
they are supported by U.S. and OECD agencies, updated regularly, and widely used. The
application of a variety of different tools and methods to identify candidate analogues serves to
minimize the limitations of any individual tool with respect to the pool of chemicals included,
chemical fragments considered, and methods for assessing similarity. Further, the inclusion of
techniques to identify analogues based on metabolism and toxicity or bioactivity expands the
pool of candidates beyond those based exclusively on structural similarity.

Analogue Search Results for Dibenzothiophene

Candidate analogues for dibenzothiophene were identified based on structural
relationships, metabolic relationships, and toxicity/mechanisms/mode-of-action (MOA)
relationships. For candidates identified through these approaches, U.S. EPA (IRIS and PPRTV),
ATSDR, and CalEPA sources were searched for subchronic, intermediate, and chronic inhalation
toxicity values. No candidate analogues with inhalation toxicity values were identified. Details
are provided below.

32

Dib enzothi ophene


-------
EPA/690/R-22/002F

Identification of Structural Analogues with Established Toxicity Values

Dibenzothiophene is not a member of an existing OECD or New Chemical category.
Candidate structural analogues for dibenzothiophene were identified using similarity searches in
the OECD Toolbox, U.S. EPA CompTox Chemicals Dashboard, and ChemlDplus tools. A total
of 24 unique structural analogues were identified for dibenzothiophene in the Dashboard, OECD
QSAR Toolbox, and ChemlDplus (>80% similarity threshold) (NLM, 2021a; OECD, 2020).

The list of potential analogues was manually reviewed and the following criteria were
applied to select candidate analogues for further evaluation based on the structural features
expected to influence toxicokinetics and/or toxicity,:

•	Contains one thiophene ring fused with 1-3 benzene rings, and

•	Only methyl, ethyl, or propyl alkyl substituents are present.

Using these criteria, all 24 structural analogues initially identified were considered
candidate analogues for dibenzothiophene (see Table A-4). No inhalation toxicity values were
identified for any of the candidate structural analogues.

Table A-4. Candidate Structural Analogues Identified for Dibenzothiophene



c£o



Tool (Method)3

Analogue (CASRNs) Selected for Toxicity Value Searches

Structure

Dashboard
(Tanimoto),
OECD Toolbox,
and ChemlDplus
(method not
described)

Benzo|/> |naphtho[2,1 -c/|thiophcne (239-35-0)



Benzo|/) |naphtho|2.3-t/|thiophene (243-46-9)



Dashboard
(Tanimoto)
ChemlDplus
(method not
described)

2,8-Dimethyldibenzo[b,c/|thiophene (1207-15-4)



2-Methyldibenzothiophene (30995-64-3)



OECD Toolbox,
and ChemlDplus
(method not
described)

3-Methyldibenzothiophene (16587-52-3)



Dashboard
(Tanimoto)

Naphtho(2,l-6)thiophene (233-02-3)



33

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table A-4. Candidate Structural Analogues Identified for Dibenzothiophene







Tool (Method)3

Analogue (CASRNs) Selected for Toxicity Value Searches

Structure



Naphtho(1,2-6)thiophene (234-41-3)





NaplUho(2.3-/;)thiophcnc (268-77-9)

%



Anthra(2.3-/))thiophcnc (22108-55-0)

CCXI>



Anthra(2.1 -/>)thiophcnc (227-56-5)





Anthra(l,2-/>)thiophene (227-86-1)





Bcnzo(/))naphtho(2.3-t/)thiophcnc. 8-methyl- (24964-07-6)





Bcnzo(/>)naplUho(2.3-t/)thiophcnc. 9-methyl- (41895-72-1)





Bcnzo(/))naphtho(2.1 -c/)thiophcnc. 2-methyl- (4567-43-5)





Bcnzo(/))naphtho(2.1 -c/)thiophcnc. 3-methyl- (4567-45-7)





Bcnzo(/))naphtho(2.3-t/)thiophcnc. 2-methyl- (83656-84-2)

C—^



Bcnzo(/>)naphtho(2.1 -c/)thiophcnc. 8-methyl- (83821-53-8)

/ \ /~

34

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table A-4. Candidate Structural Analogues Identified for Dibenzothiophene







Tool (Method)3

Analogue (CASRNs) Selected for Toxicity Value Searches

Structure

ChemlDplus
(method not
described)

Dibenzothiophene, 2-methyl- (20928-02-3)

CHj



Dibenzothiophene, 1-methyl- (31317-07-4)

CH3

cn6



Dimethyldibenzothiophene (70021-47-5)

H;C



Dibenzothiophene, 4-methyl- (7372-88-5)

ch3



1 -Ethyldibcnzo|/),t/|thiophene (79313-22-7)

h3c



4-Ethyldibenzothiophene (89816-99-9)

HjC—



1 -Propyldibenzo |/>,c/|thiophcnc (79313 -23 -8)

Cp©

CH3

a80% similarity threshold was applied.

OECD = Organisation for Economic Co-operation and Development.

Identification of Toxicokinetic Precursors or Metabolites with Established Toxicity
Values

The main metabolite in urine from a rabbit exposed orally to dibenzothiophene was
mono-hydroxy-diphenylene sulfone (Thomas et al.. 1942). In rat liver microsomes incubated
with dibenzothiophene, the identified metabolites were dibenzothiophene-5-oxide and
dibenzothiophene-5-dioxide (dibenzothiophene sulfone) (Jacob et al.. 1991; Vignier et al.. 1985).
Predicted metabolites were collected from the OECD QSAR Toolbox. PubMed searches
(searching "dibenzothiophene" or "132-65-0" and "metabolite") were conducted to identify
metabolic precursors to dibenzothiophene. No metabolic precursors were identified. PubMed
was also searched to identify other compounds that are metabolized to any of the observed or
predicted metabolites of dibenzothiophene (searching the metabolite name or [CASRN if

35

Dib enzothi ophene


-------
EPA/690/R-22/002F

available] and "metabolite"). No compounds that share at least one metabolite with
dibenzothiophene were identified in these searches.

Table A-5 summarizes the 18 candidate metabolic analogues for dibenzothiophene
(3 observed metabolites and an additional 15 unique predicted metabolites). Searches for
relevant toxicity values for the candidate metabolic analogues of dibenzothiophene did not
identify inhalation toxicity values for any of the observed/predicted metabolites.

Table A-5. Candidate Metabolic Analogues of Dibenzothiophene

Relationship to Dibenzothiophene

Compound

Metabolic precursor

None identified

Metabolite

Mono-hydroxy-dibenzothiophene sulfone (location of hydroxy group not
specified)



Dibenzothiophene 5-oxide (CASRN 1013-23-6)



Dibenzothiophene sulfone (CASRN 1016-05-3)



2-Hydroxydibenzothiophene (CASRN 22439-65-2)



3-Hydroxydibenzothiophene (CASRN 69747-77-9)



4-Hydroxydibenzothiophene (CASRN 24444-75-5)



2,3 -Dihydroxy dibenzothiophene3



3,4 -Dihyroxy dibenzothiophene3



3,7 -Dihydroxy dibenzothiophene3



3,4,7-Trihydroxy dibenzothiophene3



2,3,7-Trihydroxy dibenzothiophene3



2-Hydroxydibenzothiophene 5-oxide3



3-Hydroxydibenzothiophene 5-oxide3



4-Hydroxydibenzothiophene 5-oxide3



2,3-Hydroxydibenzothiophene 5-oxide3



3,4-Hydroxydibenzothiophene 5-oxide3



3,7-Hydroxydibenzothiophene 5-oxide3



1-Hydroxydibenzothiophene (CASRN 69747-83-7)

Shares common metabolite(s)

None identified

aCASRN not available for this metabolite.

Identification of Analogues on the Basis of Toxicity/Mechanistic/Mode-of-Action
Information and Established Toxicity Values

Available toxicity and mechanistic data for dibenzothiophene were evaluated to
determine whether these data would suggest candidate analogues. The data were reviewed to
determine whether there were in vivo toxicity data suggesting specific, characteristic toxicity
(e.g., cholinesterase inhibition, inhibition of oxidative phosphorylation) that could be used to
identify candidate analogues. The limited available in vivo animal data on dibenzothiophene
administered orally indicate that the liver is the primary target organ and increased hepatocyte

36

Dib enzothi ophene


-------
EPA/690/R-22/002F

hypertrophy in female rats exposed for 28 days. JECDB (2011) was used as a critical effect for
the derivation of the screening subchronic p-RfD value (see "DERIVATION OF SCREENING
PROVISIONAL REFERENCE DOSES" section for more details). However, the available
information was not sufficient to suggest specific, characteristic toxicity that could be used to
identify candidate analogues.

Dibenzothiophene was active in 25 ToxCast/Tox21, 6 EDSP21, and 83 PubChem
bioactivity assays reported in the U.S. EPA CompTox Chemicals Dashboard. The GenRA option
within the Dashboard offers an option to search for analogues based on similarities in activity in
ToxCast in vitro assays. Using the ToxCast bioactivity data, none of the nearest neighbors
identified by GenRA had similarity indices >0.5 (the highest index was 0.28 for
pentachl oroani sol e).

The CTD identified several compounds with gene interactions similar to interactions
induced by dibenzothiophene (Davis et at., 2021). In the CTD, similarity is measured by the
Jaccard index, calculated as the size of the intersection of interacting genes for chemical A and
chemical B divided by the size of the union of those genes (range 0 [no similarity] to 1 [complete
similarity]). Among the compounds with gene interactions similar to dibenzothiophene, the
numbers of common gene interactions ranged from 23 to 145, and similarity indices ranged from
0.03 to 0.16; the compound with the highest similarity index (0.16) was pyrene. There were no
compounds with a similarity index over 0.5.

Summary

Searches for structural, metabolic, and toxicity/mechanistic analogues for
dibenzothiophene yielded a total of 42 unique candidate analogues: 24 structural analogues and
18 metabolites. None of the identified candidate analogues have inhalation toxicity values from
authoritative sources such as U.S. EPA, ATSDR, or CalEPA.

Because no candidate analogues with inhalation toxicity values were identified for
dibenzothiophene, the alternative analogue approach was unable to derive screening reference
inhalation concentrations for dibenzothiophene.

37

Dib enzothi ophene


-------
EPA/690/R-22/002F

APPENDIX B. DATA TABLES

Table B-l. Selected Endpoints in Male and Female Sprague Dawley Rats
After Oral Treatment with Dibenzothiophene for 28 Days"

Males: ADD [HED] in mg/kg-db

Endpointc'd

0

3 [0.75]

10 [2.5]

30 [7.4]

Motor activity (total count)

1,399.3 ±407.4

1388.3 ±480.2
(-0.8%)

516.7 ±238.8**
(-63%)

661.8 ± 322.1**
(-53%)

PT (sec)

16.62 ±0.87

17.35 ± 1.23
(+4%)

20.48 ±2.37**
(+23%)

22.62 ±4.19**
(+36%)

APTT (sec)

28.23 ±2.79

28.23 ±3.37
(+0%)

33.23 ±2.16
(+18%)

39.92 ±6.67**
(+41%)

Calcium (mg/dL)

9.55 ±0.1

9.57 ±0.28
(+0.2%)

9.82 ±0.38
(+3%)

10.07 ±0.34
(+5%)*

A/G ratio

1.208 ±0.123

1.185 ±0.065
(-2%)

1.175 ±0.049
(-3%)

1.055 ±0.07
(-13%)**

Albumin protein fraction (%)

54.65 ± 1.66

54.22 ± 1.35
(-0.8%)

53.97 ± 1.07
(-1%)

51.30 ± 1.67

(-6%)**

a2u-g protein fraction (%)

7.93 ±0.35

7.52 ±0.69
(-5%)

8.27 ±0.2
(+4%)

9.03 ±0.41
(+14%)**

Beta globulin protein fraction
(%)

14.85 ±0.48

15.02 ± 0.44 (+1%)

15.42 ±0.47
(+4%)

16.95 ±0.6
(+14%)**

Terminal body weight (g)

341.8 ± 13.7

345.8 ± 19.3
(+1%)

327.3 ±28.7
(-4%)

323.3 ±26.1
(-5%)

Absolute liver weight (g)

11.082 ± 1.213

12.170 ± 1.438
(+10%)

11.708 ± 1.117

(+6%)

14.468 ± 1.736**
(+31%)

Relative liver weight (%)

3.233 ±0.247

3.512 ±0.271
(+9%)

3.578 ±0.153*
(+11%)

4.465 ± 0.208**
(+38%)

Absolute kidney weight (g)

2.498 ± 0.204

2.535 ±0.088
(+1%)

2.615 ±0.355
(+5%)

2.653 ±0.172
(+6%)

Relative kidney weight (%)

0.732 ± 0.040

0.735 ± 0.023
(+0%)

0.798 ±0.052*
(+9%)

0.823 ±0.031**
(+12%)

38

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table B-l. Selected Endpoints in Male and Female Sprague Dawley Rats
After Oral Treatment with Dibenzothiophene for 28 Days"

Females: ADD [HED] in mg/kg-d

Endpoint

0

3 [0.68]

10 [2.2]

30 [6.7]

Terminal body weight (g)

192.7 ±21.4

211.7 ± 7.9
(+10%)

195.7 ± 12.8
(+2%)

195.7 ± 18.3
(+2%)

Total cholesterol (mg/dL)

51.8 ± 14.5

58.3 ± 13.8
(+13%)

51.5 ± 11.6
(+0.6%)

79.7 ±23.4*
(+54%)

Absolute liver weight (g)

6.008 ± 0.644

7.100 ±0.347*
(+18%)

6.733 ±0.555
(+12%)

7.767 ±0.795**
(+29%)

Relative liver weight (%)

3.123 ±0.170

3.355 ±0.145
(+7%)

3.450 ±0.299*
(+10%)

3.970 ±0.187**
(+27%)

Absolute kidney weight (g)

1.548 ±0.141

1.765 ±0.154*
(+14%)

1.613 ±0.124
(+4%)

1.593 ±0.154
(+3%)

Relative kidney weight (%)

0.808 ±0.059

0.837 ±0.074
(+4%)

0.827 ±0.058
(+2%)

0.818 ±0.058
(+1%)

aJECD6 (2011).

bADDs were converted into HEDs (HED = ADD x DAF) using DAFs of 0.250, 0.248, and 0.247 for low-, mid-,
and high-dose males and 0.226, 0.223, and 0.222 for low-, mid-, and high-dose females calculated as follows:
DAF = (BWa1/4 + BWh1'4), where BWa = animal body weight, and BWh = human body weight. Study-specific
TWA animal body weights of 0.272, 0.264, and 0.259 kg for low-, mid-, and high-dose males, and 0.182, 0.174,
and 0.171 kg for low-, mid-, and high-dose females were used. For humans, the reference value of 70 kg was used
for body weight, as recommended by U.S. EPA (1988).

Data are means ± SD; n = 6 for all data points, except n = 12 for motor activity in control and high-dose groups.
dValue in parentheses is % change relative to control = ([treatment mean - control mean] + control mean) x 100.
* Significantly different from control (p < 0.05) by Dunnett's test (motor activity, hematology, serum chemistry
and organ weights) or Mann-Whitney U-test (PT time), as reported by the study authors.

**Significantly different from control (p < 0.01) by Dunnett's test (motor activity, hematology, serum chemistry
and organ weights) or Mann-Whitney U-test (PT time), as reported by the study authors.

a2u-g = alpha 2u-globulin; ADD = adjusted daily dose; A/G = albumin/globulin; APTT = activated partial
thromboplastin time; DAF = dosimetric adjustment factor; HED = human equivalent dose; PT = prothrombin time;
SD = standard deviation; TWA = time-weighted average.

39

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table B-2. Selected Histopathological Endpoints in Male and Female
Sprague Dawley Rats After Oral Treatment with Dibenzothiophene

for 28 Days"

Males: ADD [HED] in mg/kg-db

Endpoint'

0

3 [0.75]

10 [2.5]

30 [7.4]

Liver (all lesions graded as slight):
Hepatocyte hypertrophy, centrilobular
Fatty change, periportal
Microgranuloma
Necrosis, focal

0/6 (0%)
0/6 (0%)
4/6 (67%)
0/6 (0%)

0/6 (0%)
2/6 (33%)
2/6 (33%)
0/6 (0%)

0/6 (0%)
0/6 (0%)
2/6 (33%)
0/6 (0%)

6/6 (100%)
0/6 (0%)
1/6 (17%)
1/6 (17%)

Kidney (proximal tubular epithelium; all lesions
graded as slight):

Hyaline droplet
Eosinophilic body
Regeneration

0/6 (0%)
0/6 (0%)
1/6 (17%)

1/6 (17%)
1/6 (17%)
2/6 (33%)

2/6 (33%)
2/6 (33%)
0/6 (0%)

6/6 (100%)
6/6 (100%)
0/6 (0%)

Females: ADD [HED] in mg/kg-d

Endpoint

0

3 [0.68]

10 [2.2]

30 [6.7]

Liver (all lesions graded as slight):
Hepatocyte hypertrophy, centrilobular
Fatty change, periportal
Microgranuloma

0/6 (0%)
3/6 (50%)
4/6 (67%)

0/6 (0%)
2/6 (33%)
3/6 (50%)

1/6 (17%)
0/6 (0%)
3/6 (50%)

6/6 (100%)
0/6 (0%)
2/6 (33%)

aJECD6 (20ID.

bADDs were converted into HEDs (HED = ADD x DAF) using DAFs of 0.250, 0.248, and 0.247 for low-, mid-,
and high-dose males and 0.226, 0.223, and 0.222 for low-, mid-, and high-dose females calculated as follows:
DAF = (BWa1/4 ^ BWh1'4), where BWa = animal body weight, and BWh = human body weight. Study-specific
TWA animal body weights of 0.272, 0.264, and 0.259 kg for low-, mid-, and high-dose males, and 0.182, 0.174,
and 0.171 kg for low-, mid-, and high-dose females were used. For humans, the reference value of 70 kg was used
for body weight, as recommended by U.S. EPA (1988).

Data are number of animals showing changes/ total number of animals examined (% incidence).

ADD = adjusted daily dose; DAF = dosimetric adjustment factor; HED = human equivalent dose; SD = standard
deviation; TWA = time-weighted average.

40

Dib enzothi ophene


-------
EPA/690/R-22/002F

Table B-3. Body, Liver, and Spleen Weights of Male Albino Rats After
Dietary Exposure to Dibenzothiophene for 165 Daysa'b

Endpoint'

Percent in diet
ADD [HED] in mg/kg-dd e f

0§

0.025%

0h

0.050%

0'

0.100%

0

13 [2.9]

0

27 [5.9]

0

63 [13]

Terminal body
weight (g)

310

310

273

273

212

212

Absolute liver
weight (g)

10.00 ±0.11

10.70 ±0.29
(+7%)

9.50 ±0.27

12.80 ±0.48
(+35%)

8.40 ± 0.22

18.10 ±0.74
(+115%)

Absolute spleen
weight (g)

0.97 ± 0.062

0.69 ±0.015
(-29%)

0.92 ±0.072

0.64 ± 0.067
(-30%)

0.83 ±0.041

0.36 ±0.010
(-57%)

"Thomas et al. (1942).

Statistical analysis was not reported and is not conducted because number of animals per group was not reported.
°Organ weights are expressed as mean ± probable error; value in parentheses is % change relative to matched
laboratory historical control = ([treatment mean - control mean] + control mean) x 100.

dAnimals were provided dibenzothiophene in the food at 0.25, 0.50, or 1.00% for the first 4 days. Because of low
food intakes and decreases in body weight, doses were then decreased to 0.025, 0.050, or 0.100%
dibenzothiophene for the remainder of the 165-day study period. The study authors provided the amount of
dibenzothiophene consumed. The following equation was used to convert that information to mg/kg-day:
ADD = total dibenzothiophene consumption per animal over study duration x (1 + body weight) x (1 -f- days dosed)
eADDs were converted to HEDs by multiplying by DAFs of 0.225, 0.219, and 0.208 for low-, mid-, and high-dose
rats calculated as follows: DAF = (BWa1/4 + BWh1'4), where BWa = animal body weight, and BWh = human body
weight. Study-specific estimated average animal body weights of 0.179, 0.161, and 0.130 kg for low-, mid-, and
high-dose rats were used. For humans, the reference value of 70 kg was used for body weight, as recommended by
U.S. EPA (1988).

fData for each exposure group were compared with data for laboratory historical controls. For the evaluation of

organ weights, historical controls were matched according to body weight.

gMatched laboratory historical controls for 13-mg/kg-day dose group.

hMatched laboratory historical controls for 27-mg/kg-day dose group.

'Matched laboratory historical controls for 63-mg/kg-day dose group.

ADD = adjusted daily dose; DAF = dosimetric adjustment factor; HED = human equivalent dose;
TWA = time-weighted average.

41

Dib enzothi ophene


-------
EPA/690/R-22/002F

APPENDIX C. BENCHMARK DOSE MODELING RESULTS

MODELING PROCEDURE FOR DICHOTOMOUS DATA

The benchmark dose (BMD) modeling of dichotomous data was conducted with the
U.S. Environmental Protection Agency (U.S. EPA) Benchmark Dose Software (BMDS;
version 3.2). For these data, the Gamma, Logistic, Log-Logistic, Probit, Log-Probit, Hill,
Multistage, and Weibull dichotomous models available within the software were fit using a
benchmark response (BMR) of 10% extra risk. In general, the BMR should be near the low end
of the observable range of increased risk in the study. BMRs that are too low can result in widely
disparate benchmark dose lower confidence limit (BMDL) estimates from different models (high
model dependence). Adequacy of model fit is judged based on the %2 goodness-of-fitp-value
(p > 0.1), magnitude of scaled residuals (absolute value <2.0), and visual inspection of the model
fit. Among all models providing adequate fit, the BMDL from the model with the lowest
Akaike's information criterion (AIC) is selected as a potential point of departure (POD), if the
BMDLs are sufficiently close (less than approximately threefold); if the BMDLs are not
sufficiently close (greater than approximately threefold), model dependence is indicated, and the
model with the lowest reliable BMDL is selected.

MODELING PROCEDURE CONTINUOUS DATA MODELING

The BMD modeling of continuous data was conducted with the U.S. EPA BMDS
(version 3.2). For these data, the Exponential, Linear, Polynomial, and Power continuous models
available within the software were used. The continuous Hill model was not considered for the
derivation of a POD because it has five parameters and requires a data set with a minimum of six
data points (including control). The continuous models available within the software were fit
using a BMR of 1 standard deviation (SD) or alternative BMRs where appropriate as outlined in
the Benchmark Dose Technical Guidance (U.S. EPA, 2012). A BMR 10% relative deviation
(RD) for liver and kidney weights is considered a minimally biologically significant response in
adult animals and was applied in this assessment for benchmark dose (BMD) modeling purposes.
An adequate fit was judged based on the %2 goodness-of-fit p-v alue (p> 0.1), magnitude of the
scaled residuals in the vicinity of the BMR, and visual inspection of the model fit. In addition to
these three criteria forjudging adequacy of model fit, a determination was made as to whether
the variance across dose groups was constant. If a constant variance model was deemed
appropriate based on the statistical test provided in BMDS (i.e., Test 2; p-v alue > 0.1), the final
BMD results were estimated from a constant variance model. If the test for homogeneity of
variance was rejected (p-value < 0.1), the model was run again while modeling the variance as a
power function of the mean to account for this nonconstant variance. If this nonconstant variance
model did not adequately fit the data (i.e., Test 3; p-v alue < 0.1), the data set was considered
unsuitable for BMD modeling. Among all models providing adequate fit, the lowest BMDL has
been selected if the BMDLs estimated from different models varied more than threefold;
otherwise, the BMDL from the model with the lowest AIC has been selected as a potential POD
from which to derive the proposed reference value.

42

Dib enzothi ophene


-------
EPA/690/R-22/002F

BMD MODELING TO IDENTIFY POTENTIAL PODS FOR DERIVATION OF A
SCREENING SUBCHRONIC PROVISIONAL REFERENCE DOSE

Increased Relative Liver Weight in Male Sprague Dawley Rats After Oral Treatment

with Dibenzothiophene for 28 Days (JECDB, 2011)

The procedure outlined above for continuous data was applied to the data for increased
relative liver weight in male Sprague Dawley rats orally exposed to dibenzothiophene for
28 days (JECDB. 2011). The constant variance model provided an adequate fit to the variance
data, and the Exponential models 2 and 3, and the Linear model provided adequate fit to the
means. Visual inspection of the dose-response curves suggested adequate fit, BMDLs were not
10 times lower than the lowest nonzero dose, and scaled residuals did not exceed ±2 units at the
data point closest to the predefined BMR. BMDLs for models providing adequate fit were
sufficiently close (differed by less than threefold), so the model with the lowest AIC was selected
(Exponential model 3). The estimated human equivalent benchmark dose associated with 10%
relative deviation from the control (BMDo.ird) and benchmark dose lower confidence limit
associated with 10% relative deviation from the control (BMDLo.ird) values of 2.33 and
2.01 mg/kg-day, respectively, were selected from this model. The results of the BMD modeling
are summarized in Table C-l. Figure C-l shows the fit of the Exponential model 3 model to the
data.

Table C-l. BMD Modeling Results (Constant Variance) for Relative Liver
Weight in Male Sprague Dawley Rats Orally Exposed to Dibenzothiophene

for 28 Days"

Model

Variance
/>-Valucb

Means
/>-Valucb

Scaled
Residual at
Dose Nearest
BMD

AIC

BMDo.ird

(HED,
(mg/kg-d)

BMDLo.ird

(HED,
(mg/kg-d)

Exponential (model 2)°

0.56101604

0.2269728

-0.767558354

0.935991293

2.330715

2.011481

Exponential (model 3)c d

0.56101604

0.226973

-0.767697226

0.935989576

2.330708

2.011754

Exponential (model 4)°

0.56101604

0.0691553

-1.045102633

3.273097499

2.071878

1.36167

Exponential (model 5)°

0.56101604

NA

-1.042495246

5.26907005

2.07426

1.362246

Polynomial (3-degree)6

0.56101604

0.0985958

-0.429504946

2.698252338

1.771026

4.6872224

Polynomial (2-degree)6

0.56101604

0.0890464

-0.53754105

2.861669325

1.761492

4.2908893

Power0

0.56101604

0.0737363

-0.763984146

3.167966089

1.743172

4.5119764

Linear6

0.56101604

0.1921587

-1.042667628

1.269007639

1.737413

2.5442898

aJECDB (2011).

bValues <0.10 fail to meet conventional goodness-of-fit criteria.

Tower restricted to be >1.
dSelected model.

"Coefficients restricted to be positive.

AIC = Akaike's information criterion; BMD = benchmark dose; BMDL = lower confidence limit on the BMD
(subscripts denote BMR: i.e., 0.1RD = dose associated with 10% relative deviation from the control);
BMR = benchmark response; NA = test for fit is not valid; HED = human equivalent dose.

43

Dib enzothi ophene


-------
EPA/690/R-22/002F

5
4.5
4

3.5

«, 3

u"i

c

0 ? 5
a.
in
0J

11 2
1.5
1

0.5
0

01234567

Dose

Figure C-l. Fit of Exponential Model 3 to Data for Relative Liver Weight in Male
Sprague Dawley Rats Exposed to Dibenzothiophene for 28 Days (JECDB, 2011)

BMD Model Output for Figure C-l

Data

Relative liver weight in males (JECDB 2011)

[Add user notes here]

Dose

N

Mean

Std. Dev.

HED (mg/kg-d)

[Custom]

[Custom]

[Custom]

0

6

3.233

0.247

0.75

6

3.512

0.271

2.5

6

3.578

0.153

7.4

6

4.465

0.208

Model Results

Benchmark Dose

BMD

2.33070755

BMDL

2.011754168

BMDU

2.774379473

AIC

0.935989576

Test 4 P-value

0.226972953

D.O.F.

2

Frequentist Exponential Degree 3 Model with BMR of 0.1 Rel. Dev. for the BMD and 0.95
Lower Confidence Limit for the BMDL

Estimated Probability
Response at BMD
O Data
BMD
	BMDL

44

Dib enzothi ophene


-------
EPA 690 R-22 002F

Model Parameters

# of Parameters

4

Variable

Estimate

a

3.291898107

b

0.040893213

d

Bounded

log-alpha

-3.04887724

Goodness of Fit



Dose

Size

Estimated
Median

Calc'd
Median

Observed
Mean

Estimated
SD

Calc'd
SD

Observed
SD

Scaled
Residual

0

6

3.291898107

3.233

3.233

0.21774326

0.247

0.247

-0.662570721

0.75

6

3.394424526

3.512

3.512

0.21774326

0.271

0.271

1.322658242

2.5

6

3.64624315

3.578

3.578

0.21774326

0.153

0.153

-0.767697226

7.4

6

4.455209601

4.465

4.465

0.21774326

0.208

0.208

0.110136507





Likelihoods of Interest





Model

Log Likelihood*

# of
Parameters

AIC



A1

4.014929631

5

1.97014074

A2

5.042561625

8

5.91487675

A3

4.014929631

5

1.97014074

fitted

2.532005212

3

0.93598958

R

-17.66965809

2

39.3393162

* Includes additive constant of -22.05452. This constant was not included in the LL derivation prior to BMDS 3.0.

Tests of Interest



Test

-2*Log
(Likelihood
Ratio)

Test df

p-value

1

45.42443942

6

<0.0001

2

2.055263988

3

0.56101604

3

2.055263988

3

0.56101604

4

2.965848839

2

0.22697295

45

Dib enzothi ophene


-------
EPA/690/R-22/002F

Increased Relative Liver Weight in Female Sprague Dawley Rats After Oral Treatment
with Dibenzothiophene for 28 Days (JECDB, 2011)

The procedure outlined above for continuous data was applied to the data for increased
relative liver weight in female Sprague Dawley rats orally exposed to dibenzothiophene for
28 days (JECDB. 2011). The constant variance model provided an adequate fit to the variance
data and all models provided adequate fit to the means. Visual inspection of the dose-response
curves suggested adequate fit, BMDLs were not 10 times lower than the lowest nonzero dose,
and scaled residuals did not exceed ±2 units at the data point closest to the predefined BMR.
BMDLs for models providing adequate fit were sufficiently close (differed by less than
threefold), so the model with the lowest AIC was selected (Linear). The Polynomial and Power
models converged to the Linear model. The Linear model estimated human equivalent BMDo.ird
and BMDLo.ird values of 2.73 and 2.19 mg/kg-day, respectively. The results of the BMD
modeling are summarized in Table C-2. Figure C-2 shows the fit of the Linear model to the data.

Table C-2. BMD Modeling Results (Constant Variance) for Relative Liver
Weight in Female Sprague Dawley Rats Orally Exposed to
Dibenzothiophene for 28 Days"

Model

Variance
/>-Valucb

Means
/>-Valucb

Scaled
Residual at
Dose Nearest
BMD

AIC

BMDo.ird

(HED,
(mg/kg-d)

BMDLo.ird

(HED,
(mg/kg-d)

Exponential (model 2)°

0.27880585

0.354743

0.121794767

-3.39315347

2.94462

2.417078

Exponential (model 3)°

0.27880585

0.3547431

0.121823756

-3.393153703

2.944643

2.419279

Exponential (model 4)°

0.27880585

0.2060485

-0.451460518

-1.866901424

2.210648

1.105419

Exponential (model 5)°

0.27880585

0.2060467

-0.45428488

-1.866889072

2.20753

1.105422

Polynomial (3-degree)d

0.27880585

0.3935635

-0.021131946

-3.600850872

2.733293

2.185768

Polynomial (2-degree)d

0.27880585

0.3935635

-0.021131946

-3.600850872

2.733293

2.185768

Power0

0.27880585

0.3935635

-0.02113139

-3.600850872

2.733292

2.185999

Linearde

0.27880585

0.3935635

-0.021131734

-3.600850872

2.733293

2.185768

aJECDB (2011).

bValues <0.10 fail to meet conventional goodness-of-fit criteria.

Tower restricted to be >1.

Coefficients restricted to be positive.

"Selected model.

AIC = Akaike's information criterion; BMD = benchmark dose; BMDL = lower confidence limit on the BMD
(subscripts denote BMR: i.e., 0.1RD = dose associated with 10% relative deviation from the control);
BMR = benchmark response; HED = human equivalent dose.

46

Dib enzothi ophene


-------
EPA/690/R-22/002F

4.5
4

3.5
3

g 2.5
o

Q.

w 2
cc

1.5
1
0.5
0

0	1	2	3	4	5	6

Dose

Figure C-2. Fit of Linear Model to Data for Relative Liver Weight in Female
Sprague Dawley Rats Exposed to Dibenzothiophene for 28 Days (JECDB, 2011)

BMD Model Output for Figure C-2

Data

Relative liver weight in females (JECDB 2011)

[Add user notes here



Dose

N

Mean

Std. Dev.

HED (mg/kg-day)

[Custom]

[Custom]

[Custom]

0

6

3.123

0.17

0.68

6

3.355

0.145

2.2

6

3.45

0.299

6.7

6

3.97

0.187

Model Results

Benchmark Dose

BMD

2.733293247

BMDL

2.185768319

BMDU

3.595625864

AIC

-3.600850872

Test 4 P-value

0.393563466

D.O.F.

2

Frequentist Linear Model with BMR of 0.1 Rel. Dev. for the BMD and 0.95 Lower
Confidence Limit for the BMDL

-^Estimated Probability
^—Response at BMD
o Data

	BMD

	BMDL

47

Dib enzothi ophene


-------
EPA 690 R-22 002F

Model Parameters

# of

Parameters

3

Variable

Estimate

g

3.194580413

betal

0.116876655

alpha

0.039245734

Goodness of Fit



Dose

Size

Estimated
Median

Calc'd
Median

Observed
Mean

Estimated
SD

Calc'd
SD

Observed
SD

Scaled
Residual

0

6

3.194580413

3.123

3.123

0.19810536

0.17

0.17

-0.885061798

0.68

6

3.274056538

3.355

3.355

0.19810536

0.145

0.145

1.000831977

2.2

6

3.451709054

3.45

3.45

0.19810536

0.299

0.299

-0.021131734

6.7

6

3.977654002

3.97

3.97

0.19810536

0.187

0.187

-0.094638521





Likelihoods of Interest





Model

Log Likelihood*

# of
Parameters

AIC



A1

5.732938375

5

-1.46587675

A2

7.654954357

8

0.69009129

A3

5.732938375

5

-1.46587675

fitted

4.800425436

3

-3.60085087

R

-9.782299848

2

23.5645997

* Includes additive constant of -22.05452. This constant was not included in the LL derivation prior to BMDS 3.0.

Tests of Interest



Test

-2*Log(Likelihood
Ratio)

Test df

p-value

1

34.87450841

6

<0.0001

2

3.844031964

3

0.27880585

3

3.844031964

3

0.27880585

4

1.865025877

2

0.39356347

48

Dib enzothi ophene


-------
EPA/690/R-22/002F

Increased Hepatocyte Hypertrophy in Female Sprague Dawley Rats After Oral
Treatment with Dibenzothiophene for 28 Days (JECDB, 2011)

The procedure outlined above for dichotomous data was applied to the data for increased
hepatocyte hypertrophy in female Sprague Dawley rats orally exposed to dibenzothiophene for
28 days (JECDB. 2011). All models provided adequate fit (/rvalue > 0.10). However, based on
visual inspection, the Multistage degree 1 model was not found to have an adequate fit
(estimated probabilities consistently misrepresented the observed responses by -20%). All other
models provided adequate fit upon visual inspection and scaled residuals did not exceed ±2 units
at the data point closest to the predefined BMR. BMDLs for models providing adequate fit were
sufficiently close (differed by less than threefold), so the model with the lowest AIC was selected
(Probit). The Probit model estimated a human equivalent BMDo.ier and BMDLio of 2.08 and
1.04 mg/kg-day, respectively. The results of the BMD modeling are summarized in Table C-3.
Figure C-3 shows the fit of the Probit model to the data.

Table C-3. BMD Modeling Results for Hepatocyte Hypertrophy in Female
Sprague Dawley Rats Orally Exposed to Dibenzothiophene for 28 Days"

Model

/>-Valucb

Scaled Residual at
Dose Nearest BMD

AIC

BMDo.ier

(HED,
(mg/kg-d)

BMDL io
(HED,
(mg/kg-d)

Dichotomous Hill

0.9994799

-6.91397E-08

11.40673536

2.125982

1.13544

Gamma0

0.9999729

-0.000199249

9.406843019

2.027344

0.974324

Log-Logisticd

0.9971634

-4.82841E-06

11.40675978

2.104989

1.135448

Multistage Degree 3e

0.9957091

-0.085479764

7.528643645

1.781135

0.624666

Multistage Degree 2e

0.8345974

-0.433247713

8.79031287

1.257594

0.564518

Multistage Degree le f

0.2663758

-0.991744547

13.67071077

0.471974

0.252068

Weibull0

0.9999858

0.002899612

7.409558632

1.99808

0.881564

Logistic

1

4.90622E-06

7.406763669

2.121106

1.125562

Log-Probitd

0.9999999

6.34854E-10

9.406734872

2.132271

1.119394

Probit8

1

4.20484E-06

7.406737143

2.083075

1.035344

aJECDB (20ID.

bValues <0.10 fail to meet conventional goodness-of-fit criteria.

Tower restricted to be >1.

dSlope restricted to be >1.

"Betas restricted to be >0.

fModel did not pass visual fit inspection.

gSelected model.

AIC = Akaike's information criterion; BMD = benchmark dose; BMDL = 95% benchmark dose lower confidence
limit on the BMD (subscripts denote BMR: i.e., 10 = dose associated with 10% extra risk); BMR = benchmark
response; NA = test for fit is not valid; HED = human equivalent dose.

49

Dib enzothi ophene


-------
EPA 690 R-22 002F

0.9
0.8
0.7

g °"6

I0"5

Of

* 0.4
0.3
0.2

0.1 -

00-
0

Figure C-3. Fit of Probit Model to Data for Increased Hepatocyte Hypertrophy in Female
Sprague Dawley Rats Exposed to Dibenzothiophene for 28 Days (JECDB, 2011)

HMD Model Output for Figure C-3

Data

Increased hepatocyte hypertrophy in females

[Add user notes here]

Dose

N

Incidence

HED (mg/kg-day)

[Custom]

[Custom]

0

6

0

0.68

6

0

2.2

6

1

6.7

6

6

Model Results

Benchmark Dose

BMD

2.083075

BMDL

1.035344

BMDU

3.0208465

AIC

7.406737143

P-value

1

D.O.F.

3

Chi2

1.3185E-06

Frequentist Probit Model with BMR of 10% Extra Risk for the BMD and 0.95 Lower
Confidence Limit for the BMDL

Estimated Probability
Response at BMD
o Data

	IBMD

—=BMDL

50

Dib enzothi ophene


-------
EPA 690 R-22 002F

Model Parameters

# of Parameters

2

Variable

Estimate

a

-6.877871888

b

Bounded

Goodness of Fit



Dose

Estimated Probability

Expected

Observed

Size

Scaled Residual

0

3.03766E-12

1.8226E-11

0

6

-4.269E-06

0.68

2.19744E-07

1.31847E-06

0

6

-0.0011482

2.2

0.166666027

0.999996162

1

6

4.205E-06

6.7

1

6

6

6

0

Analysis of Deviance



Model

Log Likelihood

# of Parameters

Deviance

Test d.f.

P Value

Full Model

-2.703367253

4

-

-

NA

Fitted Model

-2.703368572

1

2.637E-06

3

1

Reduced Model

-14.48729404

1

23.5678536

3

<0.0001

51

Dib enzothi ophene


-------
EPA/690/R-22/002F

Increased Relative Kidney Weight in Male Sprague Dawley Rats After Oral Treatment with
Dibenzothiophene for 28 Days (JEC I) II. 2011)

The procedure outlined above for continuous data was applied to the data for increased
relative kidney weight in male Sprague Dawley rats orally exposed to dibenzothiophene for
28 days (JHCDB. 2011). The constant variance model provided an adequate fit to the variance
data and only the Exponential degree 4 model provided adequate fit to the means. Visual
inspection of the dose-response curve suggested adequate fit and scaled residuals did not exceed
±2 units at the data point closest to the predefined BMR. Therefore, the human equivalent
BMDo iRD and BMDLo.ird values of 3.08 and 1.36 mg/kg-day, respectively, for this model were
selected. The results of the BMD modeling are summarized in Table C-4. Figure C-4 shows the
fit of the Exponential 4 model to the data.

Table C-4. BMD Modeling Results (Constant Variance) for Relative Kidney
Weight in Male Sprague Dawley Rats Orally Exposed to Dibenzothiophene

for 28 Days"

Model

Variance
/>-Valucb

Means
/>-Valucb

Scaled
Residual at
Dose Nearest
BMD

AIC

BMDo. ird

(HED,
(mg/kg-d)

BMDLo.ird

(HED,
(mg/kg-d)

Exponential (model 2)°

0.24433599

0.0810132

-0.493677602

-82.13382255

6.069332

4.448717

Exponential (model 3)°

0.24433599

0.0810136

-0.493536392

-82.13383205

6.069183

4.452747

Exponential (model 4)c d

0.24433599

0.1801994

0.64553924

-83.36413069

3.077499

1.364732

Exponential (model 5)°

0.24433599

NA

0.002299382

-83.16003746

2.701189

0.816626

Polynomial (3-degree)6

0.24433599

0.0905573

-0.519162188

-82.35656545

5.918103

4.248669

Polynomial (2-degree)6

0.24433599

0.0905573

-0.519162176

-82.35656545

5.918103

4.248669

Power0

0.24433599

0.0905573

-0.519162289

-82.35656545

5.918105

4.249527

Linear6

0.24433599

0.0905573

-0.519162316

-82.35656545

5.918103

4.248669

•'JHCDB (2011).

bValues <0.10 fail to meet conventional goodness-of-fit criteria.

Tower restricted to be >1.
dSelected model.

"Coefficients restricted to be positive.

AIC = Akaike's information criterion; BMD = benchmark dose; BMDL = lower confidence limit on the BMD
(subscripts denote BMR: i.e., 0. IRD = dose associated with 10% relative deviation from the control);
BMR = benchmark response; HED = human equivalent dose.

52

Dib enzothi ophene


-------
EPA 690 R-22 002F

0.9
0.8
0.7
0.6

01

£ 0.5
o

ZL

<2 0.4

0£

0.3
0.2
0.1
0

Frequentist Exponential Degree 4 Model with BMR of 0.1 Rel. Dev. for the BMD and 0.95
Lower Confidence Limit for the BMDL

"4

-Estimated Probability
-Response at BMD
Data
•BMD
BMDL

Figure C-4. Fit of Exponential Degree 4 Model to Data for Increased Relative Kidney
Weight in Male Sprague Dawley Rats Exposed to Dibenzothiophene for 28 Days (JECDB,

2011)

BMD Model Output for Figure C-4

Data

Increased relative kidney weight in males

[Add user notes here



Dose

N

Mean

Std. Dev.

HED (mg/kg-day)

[Custom]

[Custom]

[Custom]

0

6

0.732

0.04

0.75

6

0.735

0.023

2.5

6

0.798

0.052

7.4

6

0.823

0.031

Model Results

Benchmark Dose

BMD

3.07749939

BMDL

1.364731935

BMDU

19.93656715

AIC

-83.36413069

Test 4 P-value

0.180199393

D.O.F.

1

53

Dib enzothi ophene


-------
EPA 690 R-22 002F

Model Parameters

# of

Parameters

4

Variable

Estimate

a

0.724353256

b

0.356825599

c

1.150036123

log-alpha

-6.644715675

Goodness of Fit



Dose

Size

Estimated
Median

Calc'd
Median

Observed
Mean

Estimated
SD

Calc'd
SD

Observed
SD

Scaled
Residual

0

6

0.724353256

0.732

0.732

0.03606769

0.04

0.04

0.519318585

0.75

6

0.749871217

0.735

0.735

0.03606769

0.023

0.023

-1.009959186

2.5

6

0.788494711

0.798

0.798

0.03606769

0.052

0.052

0.64553924

7.4

6

0.825280877

0.823

0.823

0.03606769

0.031

0.031

-0.154902775





Likelihoods of Interest





Model

Log Likelihood*

# of
Parameters

AIC



A1

46.58005477

5

-83.1601095

A2

48.66183302

8

-81.323666

A3

46.58005477

5

-83.1601095

fitted

45.68206535

4

-83.3641307

R

36.61850946

2

-69.2370189

* Includes additive constant of -22.05452. This constant was not included in the LL derivation prior to BMDS 3.0.

Tests of Interest



Test

-2*Log(Likelihood
Ratio)

Test df

p-value

1

24.08664712

6

0.00050343

2

4.163556496

3

0.24433599

3

4.163556496

3

0.24433599

4

1.795978855

1

0.18019939

54

Dib enzothi ophene


-------
EPA/690/R-22/002F

APPENDIX D. REFERENCES

ACGIH (American Conference of Governmental Industrial Hygienists). (2020). 2020 TLVs and
BEIs: Based on the documentation of the threshold limit values for chemical substances
and physical agents & biological exposure indices. Cincinnati, OH.

Ant at. A; Pfohl-l.cszkowicz. A; Castegnaro. M. (2004). Genotoxic activity of thiophenes on liver
human cell line (HepG2). Polycycl Aromat Compd 24: 733-742.
http://dx.doi.org/10.1080/1040663049Q472473
ATSDR (Agency for Toxic Substances and Disease Registry). (2021). Toxic substances portal:
Toxicological profiles [Database], Atlanta, GA. Retrieved from
https://www.atsdr.cdc.gov/toxprofiledocs/index.html
Blunter. GP; Collin. G; Hoke. H. (2011). Tar and pitch. In Ullmann's encyclopedia of industrial
chemistry. Online: John Wiley & Sons.
http://dx.doi.org/10.1002/143560Q7.a26 091.pub2
Brinkmann. M; Btenkte. H; Salowskv. H; Bluhm. K; Schiwv. S; Tiehm. A; Hotter! H. (2014).
Genotoxicity of heterocyclic PAHs in the micronucleus assay with the fish liver cell line
RTL-W1. PLoS ONE 9: e85692. http://dx.doi.org/10.1371/iournat.pone.0085692
CatEPA (California Environmental Protection Agency). (2020). Consolidated table of
OEHHA/CARB approved risk assessment health values. Sacramento, California.
https://ww2.arb.ca.gov/resources/documents/consolidated-table-oehha-carb-approved-
ri sk-asscssment-heal th-values
CalEPA (California Environmental Protection Agency). (2021). OEHHA chemical database
[Database], Sacramento, CA: Office of Environmental Health Hazard Assessment.
Retrieved from https://oehha.ca.gov/chemicals
Chaudhurv. S; Khan. S; Rahimtula. AD. (1988). Comparison of the inhibitory effects of some
compounds present in crude oils on rat platelet aggregation: Role of intra- and extra-
cellular calcium. Toxicology 51: 35-46. http://dx.doi.org/10.1016/03Q0-483X(88)9Q078-9
ChcmlDplus. (2021). ChemlDplus advanced. Available online at

https://chem.nlm.nih.gov/chemidplus/

Davis. AP; Grondin. CJ; Johnson. RJ; Sciakv. D; Wiegers. J; Wiegers. TC; Mattinglv. CJ.

(2021). The comparative toxicogenomics database (CTD): update 2021 [Database]: MDI
Biological Laboratory. NC State University. Retrieved from http://ctdbase.org/
Deutschmann. O; Knozinger. H; Kochtoefl. K; Turek. T. (2011). Heterogeneous catalysis and
solid catalysts. 3. Industrial applications. In Ullmann's encyclopedia of industrial
chemistry. Online: John Wiley & Sons. http://dx.doi.org/10.1002/14356007.oQ5 o03
Dickson. JG; Adams. YD. (1980). Evaluation of mutagenicity testing of extracts from processed
oil shale (pp. 52 PP). (UWRL/Q-80/01). Logan, Utah: Utah Water Research Laboratory,
Utah State University.

ECHA (European Chemicals Agency). (2021). Substance infocard: Dibenzothiophene. Available
online at https://echa.curopa.cu/substance-intbrmation - substanceintb/100.004.613
(accessed June 22, 2021).

Hard. GC; Alden. CL; Bruner. RH; Frith. CH; Lewis. RM; Owen. RA; Krieg. K; Durchfield-

Mever. B. (1999). Non-prol iterative lesions of the kidney and lower urinary tract in rats.
In Guides for Toxicologic Pathology. Washington, D.C.: STP/ARP/AFIP.

55

Dib enzothi ophene


-------
EPA/690/R-22/002F

IARC (International Agency for Research on Cancer). (2013). Bitumens and bitumen emissions,
and some N- and S-heterocyclic polycyclic aromatic hydrocarbons [IARC Monograph],
Lyon, France. https://pubiications.iarc.fr/Book-And-Report-Series/Iarc-Monographs-On-
The-Identification-Of-Carcinogenic-Hazards-To-Humans/Bitumens-And-Bitumen-
Emissions-And-Some-Em-N-Em—And-Em-S-Em—Heterocvclic-Polvcvclic-Aromatic-
Hvdrocarbons-2013

Jacob. J; Schmotdt A; Augustin. C; Raab. G; Grimmer. G. (1991). Rat liver microsomal ring-
oxidation and s-oxidation of thiaarenes with central or peripheral thiophene rings.
Toxicology 68: 181-194. http://dx.doi.org/10.1016/0300-483X(91)90020-2
JECDB (Japan Existing Chemical Data Base). (2010a). Dibenzothiophene [CAS no. 132-65-0]:
Bacterial reverse mutation test. (SR05343). Japanese National Institute of Health
Sciences.

JECDB (Japan Existing Chemical Data Base). (2010b). Dibenzothiophene [CAS no. 132-65-0]:
In vitro mammalian chromosome aberration test. (SR05342). Japanese National Institute
of Health Sciences.

JECDB (Japan Existing Chemical Data Base). (2011). [Dibenzothiophene [CAS no. 132-65-0]:
Repeated dose 28-day oral toxicity study], (SR05343). Japanese National Institute of
Health Sciences, https://dra4.nihs.go.ip/mhlw data/home/pdf/PDF132-65-0b.pdf
Leighton. FA. (1989). Acute oral toxicity of dibenzothiophene for male CD-I mice: LD50,

lesions, and the effect of preinduction of mixed-function oxidases. Fundam Appl Toxicol
12: 787-792. http://dx.doi.org/10.1016/0272-0590(89)90010-9
Maditt. REA; Brownlee. BG; Josephv. PD; Bunce. NJ. (1999). Comparison of the Ames

Salmonella assay and Mutatox genotoxicity assay for assessing the mutagenicity of
polycyclic aromatic compounds in porewater from Athabasca oil sands mature fine
tailings. Environ Sci Technol 33: 2510-2516. http://dx.doi.org/10.1021/es981343o
Mcfatt. T; Booth. GM; Lee. ML; Tominaga. Y; Pratap. R; Castle. RN; Tediamulia. M. (1984).
Mutagenic activity of methyl-substituted tricyclic and tetracyclic aromatic sulfur
heterocycles. Mutat Res Genet Toxicol 135: 97-103. http://dx.doi.org/10.lQ16/Q165-
1218(84)90161-7

NIOSH (National Institute for Occupational Safety and Health). (2018). NIOSH pocket guide to
chemical hazards. Index of chemical abstracts service registry numbers (CAS No.).
Atlanta, GA. http://www.cdc.gov/niosh/npg/npgdcas.html
NLM (National Library of Medicine). (2021a). ChemlDplus: Dibenzothiophene. Available

online at https://chem.nlm.nih.gov/chemidplus/name/dibenzothiophene
NLM (National Library of Medicine). _(2021b). PubChem: Dibenzothiophene [CAS No. 132-65-

0], Available online at https://pubchem.ncbi.nlm.nih.gov/compound/3023
NTP (National Toxicology Program). (2016). 14th Report on carcinogens. Research Triangle

Park, NC. https://ntp.niehs.nih.gov/pubhealth/roc/index-l.html
OECD (Organisation for Economic Co-operation and Development). (2020). The OECD QSAR
toolbox [version 4.4.1]: Dibenzothiophene (CASRN: 132-65-0) [Database], Retrieved
from http://www.oecd.org/chemicalsafetv/risk-assessment/oecd-qsar-toolbox.htm
OECD (Organisation for Economic Co-operation and Development). (2021). The OECD QSAR
toolbox [version 4.5] [Database], Retrieved from

http://www.oecd.org/chemicalsafetv/risk-assessment/oecd-qsar-toolbox.htm

56

Dib enzothi ophene


-------
EPA/690/R-22/002F

OSHA (Occupational Safety & Health Administration). (2021a). Air contaminants: Occupational
safety and health standards for shipyard employment, subpart Z, toxic and hazardous
substances. (OSHA Standard 1915.1000). Washington, DC. https://www.osha.gov/iaws-
regs/regulations/ standardnumber/1915/1915.1000
OSHA (Occupational Safety & Health Administration). (2021b). Safety and health regulations
for construction: Occupational health and environmental controls: Gases, vapors, fumes,
dusts, and mists. Washington, DC. https://www.osha.gov/laws-
regs/regulations/ standardnumber/1926/1926.5 5
OSHA (Occupational Safety & Health Administration). (2021c). Table Z-l: Limits for air

contaminants. Occupational safety and health standards, subpart Z, toxic and hazardous
substances. Washington, DC. https://www.osha.gov/laws-
regs/regulations/ standardnumber/1910/1910.1000TABLEZ1
Pclrov. RA; Stewart. PL; Tominaga. Y; Iwao. M; Castle. RN; Lee. ML. (1983). Microbial

mutagenicity of 3- and 4-ring polycyclic aromatic sulfur heterocycles. Mutat Res 117:
31-40. http://dx.doi.org/10.1016/0165-1218(83)90150-7
Peters. JM; Bovd. EM. (1966). Organ weights and water levels of the rat following reduced food

intake. J Nutr 90: 354-360.

Rasmussen. K; Booth. GM; Lee. ML; Castle. RN. (1991). Mutagenicities of hydroxy-substituted
carbazoles and dibenzothiophenes using the CHO HGPRT assay. Environ Toxicol Chem
10: 1133-1137. http://dx.doi.org/10.1002/etc.56201009Q4
Sarma, SN; Blais, JM; Chan, HM. (2017). Neurotoxicity of alkylated polycyclic aromatic
compounds in human neuroblastoma cells. J Toxicol Environ Health A 80: 285-300.
http://dx.doi.org/10.1080/15287394.2017.1314840
Silva. KT; Oliveira-Castro. RA; Rodrigues. VC; de Lima. WG; Rodrigues. CV; Castro-Borges.
W; Guerra Andrade. MH. (2015). DBT- and DBT02-induced dysplasia and their
associated proteomic alterations in the small intestines of Wistar rats. J Proteome Res 14:
385-396. http://dx.doi.org/10.1021/pr5009459
Thomas. JO; Deeds. F; Cox. AJ. (1942). Chronic toxicity of diphenylene sulfide to the albino rat.

J Food Sci 7: 161-169. http://dx.doi.Org/10.llll/i.1365-2621.1942.tbl7244.x
U.S. EPA (U.S. Environmental Protection Agency). (1988). Recommendations for and
documentation of biological values for use in risk assessment [EPA Report],
(EPA600687008). Cincinnati, OH.

http://cfpub.epa. gov/ncea/cfm/recordisplay.cfm?deid=34855
U.S. EPA (U.S. Environmental Protection Agency). (1991). Alpha-2u-globulin: Association with
chemically induced renal toxicity and neoplasia in the male rat. (EPA625391019F.
PB92143668). Washington, DC: U.S. Environmental Protection Agency, National Center
for Environmental Assessment.

https://ntrl.ntis.gov/NTRL/dashboard/searchResults.xhtml?searchOuerv=PB92143668
U.S. EPA (U.S. Environmental Protection Agency). (2002). A review of the reference dose and
reference concentration processes. (EPA630P02002F). Washington, DC.
https://www.epa.gov/sites/production/files/2014-12/documents/rfd-final.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2005). Guidelines for carcinogen risk
assessment [EPA Report], (EPA630P03001F). Washington, DC.
https://www.epa.gov/sites/production/fiies/2013-
09/documents/cancer guidelines final 3-25-05.pdf

57

Dib enzothi ophene


-------
EPA/690/R-22/002F

U.S. EPA (U.S. Environmental Protection Agency). (2011a). Chemical assessment clustering

engine (ChemACE). Retrieved from https://www.epa.gov/tsca-screening-toois/chemicai-
assessment-clustering-engine-chemace
U.S. EPA (U.S. Environmental Protection Agency). (2011b). Health effects assessment summary
tables (HEAST) for superfund [EPA Report], Washington, DC. https://epa-
heast. ornl. gov/heast.php
U.S. EPA (U.S. Environmental Protection Agency). (2011c). Recommended use of body weight
3/4 as the default method in derivation of the oral reference dose. (EPA100R110001).
Washington, DC. https://www.epa.gov/sites/production/fiies/2013-
09/documents/recommended-use-of-b w3 4.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2012). Estimation Programs Interface

Suite™ (EPISuite) for Microsoft® Windows, v 4.11: Dibenzothiophene [CAS No. 132-
65-0] [Fact Sheet], Washington, DC. https://www.epa.gov/tsca-screening-tools/epi-
suitetm-estimation-program-interface
U.S. EPA (U.S. Environmental Protection Agency). (2018). 2018 Edition of the drinking water
standards and health advisories tables [EPA Report], (EPA822F18001). Washington, DC:
U.S. Environmental Protection Agency, Office of Water.
https://www.epa.gOv/svstem/files/documents/2022-01/dwtable2018.pdf
U.S. EPA (U.S. Environmental Protection Agency). (2021a). Chemical data reporting:
Dibenzothiophene [CAS No. 132-65-0], Retrieved from
https://chemview.epa.gov/chemview
U.S. EPA (U.S. Environmental Protection Agency). (2021b). CompTox chemicals dashboard.

Washington, DC. Retrieved from https://comptox.epa.gov/dashboard
U.S. EPA (U.S. Environmental Protection Agency). (2021c). CompTox chemicals dashboard:
Dibenzothiophene 132-65-0.

https://comptox.epa.gOv/dashboard/dsstoxdb/resuits7searclFOTXSID0047741#detaiis
U.S. EPA (U.S. Environmental Protection Agency). (2021d). Integrated risk information system

(IRIS) database [Database], Washington, DC. Retrieved from http://www.epa.gov/iris/
U.S. EPA (U.S. Environmental Protection Agency). (2021e). Non-confidential TSCA inventory:
Dibenzothiophene [CAS No. 132-65-0], Retrieved from https://www.epa.gov/tsca-
inventorv/how-access-tsca-inventorv#download
Vignier. V; Bcrthou. F; Dreano. Y; Flock, HH. (1985). Dibenzothiophene sulphoxidation: Anew
and fast high-performance liquid chromatographic assay of mixed-function oxidation.
Xenobiotica 15: 991-999. http://dx.doi.org/10.3109/00498258509049Q94
Wang. NC; Zhao. QJ; Wcssclkampcr. SC; Lambert JC; Petersen. D; Hess-Wilson. JK. (2012).
Application of computational toxicological approaches in human health risk assessment.
I. A tiered surrogate approach. Regul Toxicol Pharmacol 63: 10-19.
http://dx.doi.Org/10.1016/i.vrtph.2012.02.006
WHO (World Health Organization). (2021). Online catalog for the Environmental Health
Criteria (EHC) monographs [Database], Geneva, Switzerland: World Health
Organization (WHO). Retrieved from http://www.who.int/ipcs/publications/ehc/en/
Wilson. RH; McNaught. IB; Deeds. F. (1938). Chronic nicotine toxicity IV. Effect of nicotine-
containing diets on histology and weights of organs of albino rats. J Ind Hyg Toxicol 20:
468-481.

58

Dib enzothi ophene


-------