The fish bioconcentration factor gold
standard database: practical examples
Versonnen B.1, Iaccino F.1, Jeliazkova N.2, Arijs K.1, Comber M.3, Verdonck F.1, Vangheluwe M.1
1EURAS
- member of ARCADIS, Kortrijksesteenweg 302, 9000 Gent, Belgium, [email protected]
2Ideaconsult Ltd., 4 A. Kanchev Street, 1000 Sofia, Bulgaria
3Mike Comber Consulting, Brussels, Belgium
Introduction and Aims
The classical approach to derive BCF values (OECD 305 bioconcentration test) is labour intensive, costly and requires a lot of test animals. A
database containing only high quality data was lacking. Within the LRI Eco7 project, a fish bioconcentration factor (BCF) ‘gold standard database’ has
been developed and linked with the relational database Ambit for further (Q)SAR and alternative testing development. Criteria for considering
substances as not bioaccumulative (i.e. unlikely to have a BCF > 2,000 (B) or 5000 (vB)) are mainly based on weight of evidence approaches and
expert judgement, using parameters such as molecular length, Kow and molecular diameter. The validity of the following criteria was investigated
using data from the fish BCF gold standard database in Ambit: (1) average diameter and molecular weight, (2) maximum molecular length, (3) log
Kow.
BCF principle
Database development
Gather fish BCF data
Study details,
reliability
assessment,
reliability score
Only high quality
fish BCF laboratory
data were used in
the database
Reliability summaries
Excel Database
Open source,
relational database (MySQL),
Chemical Markup Language (CML),
QSAR models
AMBIT Database
Bioaccumulation threshold criteria
Log Kow
Cut-off levels established by the TECNES working group on PBTs
6
5
5
4
BCF 5000
3
BCF 2000
2
Log BCFss
Log BCFss
Cut-off levels observed by the authors
Molecular diameter
6
1
4
BCF 5000
3
BCF 2000
BCF 5000
2
BCF 2000
Effect of molecular
size on uptake
Appropriate orientation
1
0
0
-1
-1
-2
0
2
3
4
6
8
10
12
14
0
2
Log Kow (XlogP estimate)
4
6
8
10
12
14
16
17
18
Crossectional diameter (Angstrom)
10
Molecular length
6
6
5
5
4
BCF 5000
3
BCF 2000
2
1
Log BCFss
Log BCFss
Molecular weight
cell membrane
4
BCF 5000
3
BCF 2000
Inappropriate orientation
2
1
0
0
-1
-1
0
200
400
600
800
1000
1200
Molecular weight (g/mol)
700
1100
0
5
10
15
20
25
30
35
40
45
cell membrane
Maximum length (Angstrom)
43
Conclusions
• Cut-off levels for considering substances as not bioaccumulative (i.e. unlikely to have a BCF > 2000 (B) or 5000 (vB)) are (1) log Kow < 3 or > 8, (2) molecular
weight > 600 g/mol, (3) molecular diameter > 10 Å, (4) molecular length > 15 Å.
• Only a very limited number of substances had an ‘unexpectedly’ high BCF, e.g. highly chlorinated compounds such as hexachlorobenzene
or heptachloronorbornene.
• Results of the project are made publicly available as an easily accessible relational database (more info on http://www.euras.be/bcf). This database is
embedded into the Ambit format (http://ambit.acad.bg).
This project was supported by the European Chemical Industry Council (CEFIC)
Long-Range Research Initiative