1
QSAR-based Prediction of Inhalation Toxicity
Kendall B. Wallace, Eli Petkova, Gilman D. Veith
University of Minnesota – Duluth Medical School &
International QSAR Foundation
Incorporating elements of dosimetry and reactivity to predict biological response
Human Airway
Chemical disposition
(free vapor)-
• VP
• SolH2O• Chemical Reactivity
Biological Response -
• Protein adduct -
immune surveillance
� Asthma, T-cell mediated hypersensitivity
� Irritation/inflammation/tissue necrosis
2
Factors affecting pulmonary response
noneLowLowCarbon
monoxide
Lower
terminal
airways
HighLowPhosgene
Isocyanates
Large and
intermediate
airways
HighModerateChlorine
Upper airwaysHighHighAmmonia
Upper airwaysModerateHighAcetaldehyde
Pulmonary
Toxicity
Chemical
Reactivity
Water
Solubility
Chem Name
�Although inhalation toxicity data have been compiled in selected open access databases, the entries are limited and have seldom been subjected to rigorous peer review.
�Thus, although these databases may suffice for general reference purposes, the data is frequently ambiguous and of questionable quality.
�As a result, models of inhalation toxicity derived from these databases have largely been unsuccessful and doubts have been cast regarding the validity of QSAR approaches to inhalation toxicology.
The QSAR Inhalation Toxicity Database
3
�The inhalation toxicity database (ITDB) is an effort to compile high quality inhalation data published in the open literature and government reports as well as publicly available unpublished toxicity reports using strict Q/A standards.
�ITDB has a goal of eventually becoming an international and widely distributed resource for high quality inhalation toxicity data that can be used to better characterize inhalation toxicity with minimal animal testing.
The Inhalation Toxicity Database
� We have embarked on compiling an exhaustive mammalian inhalation toxicity database using strict standards of peer review to insure only high-quality studies are included.
� Currently focus on acute (4 hr) inhalation by rats
� About 200 unique chemicals, 86 – tested for acute toxicity in rat/4h
� Limited short-term mouse data
� Expanding to include other species as well as repeat exposure and chronic inhalation data
� Preliminary analyses of the database.……….
Current Status of the ITDB
4
Modeling Assumptions
• Obstructive disorders– Low vapor pressure
– High water solubility
– High chemical reactivity
• Restrictive disorders– Low vapor pressure
– Low water solubility
– High chemical reactivity
• MoA - specific disease
• Non-specific, narcotic-like effects – Low vapor pressure
– Low water solubility
– Low chemical reactivity
LC50/rat/4h vs Vapor Pressure
Data was compiled from the literature.
•From mid 50s to present *
•All chemicals tested as vapors **
•Consistent exposure conditions ***
•Different rat strains
* Guidelines somewhat vary with time
**Specified (aimed ) in the experiment but sometimes might not be truth
***Exposure time constant, number of animals and observation periods vary-2
-1
0
1
2
3
4
5
-2 -1 0 1 2 3 4
Vapor Pressure, mmHg
LC50
,mm
ol/m
3LC50, mmol/m3
Vapor Pressure, mmHg
5
LC50 /rat/4h vs Vapor Pressure for chemicals previously classified as
NONNONNONNON----REACTIVEREACTIVEREACTIVEREACTIVE
y = 0.705x + 1.4719
R2 = 0.9277
-2
-1
0
1
2
3
4
5
-2 -1 0 1 2 3 4
Vapor pressure, mmHg
LC50, m
mol /
m3
LC50, mmol/m3
Vapor Pressure, mmHg
HYDROCARBONSHYDROCARBONSHYDROCARBONSHYDROCARBONSHYDROCARBONSHYDROCARBONSHYDROCARBONSHYDROCARBONS are a good examples for narcosis
Nonane, hexane, isoprene, butadiene, isobutylene, butane, 2-metylpentene-1, 2-metylpentene-2, styrene
-2
-1
0
1
2
3
4
5
-2 -1 0 1 2 3 4
Vapor pressure, mmHg
LC50, m
mol /
m3
LC50, mmol/m3
Vapor Pressure, mmHg
6
No similar relationship of
LC50/VP for NITRITESNITRITESNITRITESNITRITES
-2
-1
0
1
2
3
4
5
-2 -1 0 1 2 3 4
Vapor pressure, mmHg
LC50
, mm
ol/ m
3LC50, mmol/m3
Vapor Pressure, mmHg
LC50/VP relationship
for AMINES
-2
-1
0
1
2
3
4
5
-2 -1 0 1 2 3 4
Vapor pressure, mmHg
LC50, m
mol /
m3
LC50, mmol/m3
Vapor Pressure, mmHg
Allylamine, CAS 107-11-9
7
ACRYLATES & METHACRYLATES
-2
-1
0
1
2
3
4
5
-2 -1 0 1 2 3 4
Vapor pressure, mmHg
LC
50, m
mol /
m3
LC50, mmol/m3
Vapor Pressure, mmHg
For ACRYLATES & METHACRYLATES there is no relationship with Vapor Pressure but significant correlation with GSH reactivity
LC50 vs GSH reactivity
for acrylates and methacrylates
LogLC50 = 0.28logEC50 + 2.01
R2 = 0.91
0.0
0.5
1.0
1.5
2.0
2.5
3.0
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5
Log EC50,mM
LogL
C50
, mm
ol/m
3
-2
-1
0
1
2
3
4
5
-2 -1 0 1 2 3 4
Vapor pressure, mmHg
LC
50, m
mol
/ m
3LC50, mmol/m3
Vapor Pressure, mmHg
LC50, mmol/m3
EC50, mM
8
Solubility in air andLethal Concentrationvs Vapor Pressurefor narcotics (rat/4h)
-6
-5
-4
-3
-2
-1
0
-1 0 1 2 3 4
Solubility in Air / LC50
Vapor Pressure, mmHg
Solubility in air andLethal Concentrationvs Vapor Pressurefor ethers (mouse//15 min)
-3.5
-3
-2.5
-2
-1.5
-1
-0.5
0
0 0.5 1 1.5 2 2.5 3 3.5 4
Vapor Pressure, mmHG
Sol A
ir &
LC
50,
mol/l
Solubility in Air / LC50
Vapor Pressure, mmHg
9
� Fish and mammal inhalation baseline toxicity are not directly comparable because the external media are different
�However, blood thermodynamic activity for LC50(nar) is the same in fish and mammal
�At steady-state, the activity in air/water equals the activity in blood by definition :
α = С x γ
α – activity; C- concentration; γ-activity coefficient
Baseline Toxicity
� The thermodynamic activity at any concentration can be estimated by dividing by the solubility in the
medium
� activity for narcosis in fish = LC50(fish)/water solubility
activity for narcosis in rat = LC50 (rat)/air solubility
� if activity for narcosis in fish and rat were equal, the plot of LC50 versus solubility in exposure medium should be the same
Baseline Toxicity
10
Solubility in Water or Air vs LC50 in Fish or Rat
-5
-4
-3
-2
-1
0
-5 -4 -3 -2 -1 0 1
Solubility, mol/l
LC
50,
mo
l/l
LC50 fish vs WaterSolubility
LC50 rat vs AirSolubility
Solubility in Water or Air vs LC50 in Fish or Rat (combined)
LC50rat vs LC50fish*Kh
-8
-7
-6
-5
-4
-3
-2
-1
0
-5 -4 -3 -2 -1 0
log LC50 rat
Lo
g (
LC
50 f
ish
*Kh
)
LC50rat vs LC50fish*Kh
11
LogLC50 for fish or rat vs Solubility in water or air
y = 0.7847x - 1.6059
R2 = 0.889
-6
-5
-4
-3
-2
-1
0
-6 -5 -4 -3 -2 -1 0 1 2
Solubility, mol/l
LC50, mol/l
LC50fish vs LogWsol
LC50rat vs LogAirsol
LogLC50 for fish or rat vs Solubility in water or air
2.5 Endocrine active industrial chemicals:
Release and occurrence in the environment
Concentration response curves for all
mixture components
12
-11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1
-10
0
10
20
30
40
50
60
70
80
90
100
110
E2 rbtER (cyto)
SDP rbtER (cyto)
E2 hERαααα (recomb-LBD)
SDP hERαααα (recomb-LBD)
solubility limit
Binding Assays4,4'-sulfonyldiphenol
RBA %
0.0020
0.0055
Log Concentration (M)
Bin
din
g (
%)
-11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1
-10
0
10
20
30
40
50
60
70
80
90
100
110
E2 rbtER (cyto)
SDP rbtER (cyto)
E2 hERαααα (recomb-LBD)
SDP hERαααα (recomb-LBD)
solubility limit
Binding Assays4,4'-sulfonyldiphenol
RBA %
0.0020
0.0055
Log Concentration (M)
Bin
din
g (
%)
CR
TL
0
10
20
30
40
50
60
70
80
90
100
Log Concentration (M)
-10 -9 -8 -7 -6 -5 -4 -3 -2
solubility limitRBA %
0.0008
ND
Binding Assays ethylparaben
E2 rbtER (cyto)
EP rbtER (cyto)
E2 hERαααα(recomb-LBD)
EP hERαααα (recomb-LBD)
Bin
din
g (
%)
CR
TL
0
10
20
30
40
50
60
70
80
90
100
Log Concentration (M)
-10 -9 -8 -7 -6 -5 -4 -3 -2
solubility limitRBA %
0.0008
ND
Binding Assays ethylparaben
E2 rbtER (cyto)
EP rbtER (cyto)
E2 hERαααα(recomb-LBD)
EP hERαααα (recomb-LBD)
Bin
din
g (
%)
13
CR
TL
0
10
20
30
40
50
60
70
80
90
100
Log Concentration (M)
-10 -9 -8 -7 -6 -5 -4 -3 -2125
150
175
200
225
250
275
300
325
350
RBA %
0.00057
0.0098
Binding Assays resorcinol sulfide
E2 hERαααα (recomb-full)FP
RES hERαααα (recomb-full)FP
E2 rbtER (cyto)
RES rbtER (cyto)
Bin
din
g (
%)
Po
lariza
tion
(mp
)
CR
TL
0
10
20
30
40
50
60
70
80
90
100
Log Concentration (M)
-10 -9 -8 -7 -6 -5 -4 -3 -2125
150
175
200
225
250
275
300
325
350
RBA %
0.00057
0.0098
Binding Assays resorcinol sulfide
E2 hERαααα (recomb-full)FP
RES hERαααα (recomb-full)FP
E2 rbtER (cyto)
RES rbtER (cyto)
Bin
din
g (
%)
Po
lariza
tion
(mp
)
4-n-Amylaniline
CTRL
0
10
20
30
40
50
60
70
80
90
100
110
AAN rbt Vtg
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1
Control rbt Vtg
1.0×105
1.0×106
1.0×107
1.0×108
AAN rbtER (cyto)
Log Concentration (M)
[3H
]-E
2 B
ind
ing
(%
)/F
ecu
nd
ity
(%
dec
rea
se)
VT
G m
RN
A c
op
ies/4
00 n
g to
tal R
NA
14
4-n-Amylaniline
CTRL
0
10
20
30
40
50
60
70
80
90
100
110
AAN rbt Vtg
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1
Control rbt Vtg
1.0×105
1.0×106
1.0×107
1.0×108
AAN rbtER (cyto)
Log Concentration (M)
[3H
]-E
2 B
ind
ing
(%
)/F
ecu
nd
ity
(%
dec
rea
se)
VT
G m
RN
A c
op
ies/4
00 n
g to
tal R
NA
4-n-Amylaniline
CTRL
0
10
20
30
40
50
60
70
80
90
100
110
AAN rbt Vtg
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1
Control rbt Vtg
1.0×105
1.0×106
1.0×107
1.0×108
AAN rbtER (cyto)Medaka Fecundity
In vivo Water Exposure
Log Concentration (M)
[3H
]-E
2 B
ind
ing
(%
)/F
ecu
nd
ity
(%
dec
rea
se)
VT
G m
RN
A c
op
ies/4
00 n
g to
tal R
NA
15
4-n-Amylaniline
CTRL
0
10
20
30
40
50
60
70
80
90
100
110
AAN rbt Vtg
-10 -9 -8 -7 -6 -5 -4 -3 -2 -1
Control rbt Vtg
1.0×105
1.0×106
1.0×107
1.0×108
AAN rbtER (cyto)Medaka Fecundity
In vivo Water Exposure
Medaka Fecundity
Predicted in vivo LiverSteady State Concentration
Log Concentration (M)
[3H
]-E
2 B
ind
ing
(%
)/F
ecu
nd
ity
(%
dec
rea
se)
VT
G m
RN
A c
op
ies/4
00 n
g to
tal R
NA