metabolism in-silico simulation laboratory of mathematical chemistry university ‘prof. as....

METABOLISM

in-silico simulation

Laboratory of Mathematical ChemistryUniversity ‘Prof. As. Zlatarov’, Bourgas, Bulgaria

Sabcho DimitrovOvanes Mekenyan

The McKim Conference on the Use QSARs and Aquatic Toxicology in Risk AssessmentJune 27-29, 2006

Research Partners from Regulation Agencies

US EPA, Athens, GA – 5 yrs coopUS EPA, Duluth, MNEnvironment CanadaNITE Japan

Research Partners from Industry

P&GExxonMobilUnileverBASF

Overview

• (Q)SARs• Metabolism of Prokaryotes• Metabolism of Eukaryotes• Simulation of metabolism

(Q)SARsBiodegradationBioaccumulationAcute ToxicityChronic ToxicityHormone ToxicitySkin sensitizationMutagenicity…

OH

O

O

O

SO

O

OH

OH

OH

O

O

OOCH3

HO

OH

OH

OH

O

Metabolism

(Q)SARs

The Five Kingdoms

Acute & Chronictoxicity

Hormone ToxicitySkin sensitizationMutagenicity

BCF

Biodegradation

Metabolism

Energy-generating component: CatabolismProduce energy (as ATP) and simple oxidized compounds

Energy-consuming component: Anabolism Build cell material

Intermediary metabolism

Common energy metabolismtricarboxylic acids cycle, glycolysis

Common biosynthetic pathwayssynthesis of amino acids, lipids, and carbohydrates

Set of reactions involving metabolites that are intermediates in the degradation or biosyntheses of biopolymers

Anabolism

Catabolism

metabolismmetabolismEnzymes often need to be highly substrate specific

IIntermediaryntermediary

Non-iNon-intermediaryntermediarymetabolismmetabolism

1. Response to environment2. Catabolism

Prokaryotes

KingdomMonera

The Kingdom Monera (Prokaryotes)

Constitutive enzymes: always produced by cellsThe enzymes that operate during glycolysis and the tricarboxylic acid cycle

Inducible enzymes: produced ("turned on") in cells in response to a particular substrate; they are produced only when needed

Operon - grouping of genes in bacteria under the control of the same regulatory system.

Bacteria tend to group genes that are functionally related together on the chromosome.

Control mechanism

Structural gene for β-galactoside permease

Structural gene for β-galactoside transacetylase

Structural gene for β-galactosidase

Lactose operon: contains genes that encode enzymes responsible for lactose metabolism

The Kingdom Monera (Prokaryotes)

Eukaryotes

Kingdoms:ProtistaPlantaFungiAnimalia

Eukaryotes

Multienzyme complex

In a number of cases, enzymes that catalyze sequential reactions in the same metabolic pathway have been found to be physically associated:

The formed enzyme complexes allow channeling of reactants between active sites – the product of one reaction can be transferred to the next active site.

• Speed up the reactions of the pathway• Protect unstable intermediates

• Active sites are found on a single, multifunctional peptide chain• Several individual enzymes are non-covalently associated• Attachment to membranes

Separate enzymes with freely diffusing pathway intermediates

(glycolysis)

Cytosolic multienzyme complex with intermediates channelled between

enzymes(pyruvate dehydrogenase complex)

Membrane bound multienzyme complex

(electron transport system)

Summary

1. The application of metabolic transformations is strongly organized.

2. This is a premise for development of metabolic simulators.

Simulation of molecular transformations

R1 C

O

O

R2

R1 C

OH

O

HO R2++ HO CC C

OH

O

C C

O

O

C

H3C C

O

O

CH2CH3

H3C C

OH

O

HO CH2CH3+

H3C C

O

O

CH CH2

H3C C

OH

O

+ HO CH CH2

HCl2C C

O

O

C

+HCl2C C

OH

O

HO C

Half-lives,250C, pH =7

2 years

7 days

4 minutes

Simulation of molecular transformations

R1 C

O

O

R2

R1 C

OH

O

HO R2+

# Transformation Rate

1

High

2

Moderate

3

Low

+X2C C

O

O

C{sp }2

X2C C

OH

O

HO C{sp }2

+C C

O

O

C{sp }2

C C

OH

OHO C{sp }2

not C{sp2} and CX2; X: F, Cl, Br

+ HO CC C

OH

O

C C

O

O

C

X: F, Cl, Br

not CX2; X: F, Cl, Br

BESS (P&G and Michigan State University)

META (MultiCASE Inc)

METEOR (Lhasa Ltd)

CATABOL (P&G and LMC, Bourgas As. Zlatarov University)

TIMES (LMC, Bourgas As. Zlatarov University)

PPS (UM-BBD, http://umbbd.msi.umn.edu/)MEPPS (under development, Lhasa Ltd)

Simulators of metabolismSimulators of metabolism

Rule based systems

Predictions: single best or multiple alternative metabolic pathways

METEOR (Lhasa Ltd)

PPS (UM-BBD)


Rule based systems


CATABOL (P&G and LMC, Bourgas As. Zlatarov University)

TIMES (LMC, Bourgas As. Zlatarov University)


Rule based systems


Probabilitykt1/2

COH

OHC O

Geminal diol decomposition

OH

O

OH

O

-oxidation

Cyclohexanone oxidation

O

OO

O

OH

+

O

Ester hydrolysis

C

CNH2

C

CO

Amine decomposition

CH3OH

-Oxidation

C NN C

C NH2 H2N C+

Azo-bond cleavage

Substrate Principal transformations Metabolites

O O

OH

CATABOLCATABOL

Simulation of catabolismSimulation of catabolism

COH

OHC O

P = 1.00


OH

O

OH

O

P = 0.99 -oxidation


O

OO

O

OH

+

O

P = 0.90

Ester hydrolysis

C

CNH2

C

CO

P = 0.75

Amine decomposition

CH3OH

P = 0.40 -Oxidation

C NN C

C NH2 H2N C+ P = 0.001

Azo-bond cleavage


O O

OH

P = 0.95

COH

OHC O

P = 1.00


OH

O

OH

O

P = 0.99-oxidation

O O

OH

P = 0.95


O

OO

O

OH

+

O

P = 0.90

Ester hydrolysis

C

CNH2

C

CO

P = 0.75

Amine decomposition

CH3OH

P = 0.40

-Oxidation

C NN C

C NH2 H2N C+P = 0.001

Azo-bond cleavage

O



OH

O

OH

O

P = 0.99-oxidation

O O

OH

P = 0.95


O

OO

O

OH

+

O

P = 0.90

Ester hydrolysis

C

CNH2

C

CO

P = 0.75

Amine decomposition

CH3OH

P = 0.40

-Oxidation

C NN C

C NH2 H2N C+P = 0.001

Azo-bond cleavage

O


Match? COH

OHC O

P = 1.00

- No!

COH

OHC O

P = 1.00


-oxidation

O O

OH

P = 0.95


O

OO

O

OH

+

O

P = 0.90

Ester hydrolysis

C

CNH2

C

CO

P = 0.75

Amine decomposition

CH3OH

P = 0.40

-Oxidation

C NN C

C NH2 H2N C+P = 0.001

Azo-bond cleavage

O


OH

O

OH

O

P = 0.99

Match? - No!

COH

OHC O

P = 1.00


OH

O

OH

O

P = 0.99-oxidation


O

OO

O

OH

+

O

P = 0.90

Ester hydrolysis

C

CNH2

C

CO

P = 0.75

Amine decomposition

CH3OH

P = 0.40

-Oxidation

C NN C

C NH2 H2N C+P = 0.001

Azo-bond cleavage


O O O

OH

P = 0.95RESULT

O

OHMatch? - Yes!

The most plausible biotransformation pathway

k0

k1

k2

k

k

CO2

P2

P

1

O2

1

CO2P1

CO2

2

O2

2

CO2

O2

CO2P

O2

n

n

n

+1n

+1n

n

+1n

-1n

- Parent chemical or metabolite ki

- Transformation and its probabilityPi

- BOD or CO2 production for a single transformation

22 , COi

Oi

Comparison of metabolic pathways

Pathway A Pathway B

IntersectionA∩B

Complement of A in BB\A or B-A

Complement of B in AA\B or A-B

Accounting (or not) for pathway structures

IntersectionA∩B



Measures of similarity/dissimilarity:

Tanimoto,Dice,etc.

Trivial measures of similarity/dissimilaritypathway structure is not accounted for

IntersectionA∩B



A in B = Card(A∩B)/Card(A)B in A = Card(A∩B)/Card(B)A out of B = Card(A\B)/Card(A)B out of A = Card(B\A)/Card(B)

Non-trivial measures of similarity/dissimilaritypathway structure is accounted for

Observed and predicted catabolism

- Observed and predicted metabolites, PredObs SS

- Observed and not predicted metabolites, PredObsPredObs \or SSSS

- Predicted and not observed metabolites, ObsPredObsPred \or SSSS

=+

Observed versus simulated pathwaysObserved versus simulated pathways

Union of pathways

CATABOL, mathematical formalismCATABOL, mathematical formalism BOD or CO2 production

%57BOD%63CalcBOD

Lyons, C. D., S. Katz, R. Bartha, Appl Environ Microbiol, 1984, 48, N 3, pp. 491-496

Simulated catabolism

CATABOL, mathematical formalismCATABOL, mathematical formalism

Quantities of metabolites

nm

mnCalcn PPQ

11 parent mol/mol ,1

k0

k1

k2

k

k

CO2

P2

P

1

O2

1

CO2P1

CO2

2

O2

2

CO2

O2

CO2P

O2

n

n

n

+1n

+1n

n

+1n

-1n

CATABOL, mathematical formalismCATABOL, mathematical formalism

Ultimate half-life

ktBOD exp1100

k2lnt 2/1

28/100/1ln

2ln

282/1 Calc

dBODt

First order kinetics k0

k1

k2

k

k

CO2

P2

P

1

O2

1

CO2P1

CO2

2

O2

2

CO2

O2

CO2P

O2

n

n

n

+1n

+1n

n

+1n

-1n

28/100/1ln 28Calc

dBODk

days 202/1 t

-1day 036.0k

%63CalcBOD

Training data

CATABOLCATABOLTDTD – time dependent model – time dependent model

Probabilistic approach First order kinetics

(7)

(8)

(9)

(10)

[S]0 – initial quantity of S

[S] – quantity of S at time t[M] – quantity of M at time tP – probability of transformation

(7’)

(8’)

(9’)

(10’)

[S]0 – initial quantity of S

[S] – quantity of S at time t[M] – quantity of M at time tk – first order kinetic constant

MSP

P 1SS 0

P 1S

S

0

P0SM

P

0S

M

MSk

kt expSS 0

ktexpS

S

0

kt exp1SM 0

kt exp1S

M

0

ktPt exp1

tPk t 1ln

Probabilistic approach & First order kinetics


1. Primary half-life – half-life of parent chemical2. Ultimate half-life – half-life by BOD3. Biodegradation as a function of time4. Metabolites quantity as a function of time5. Biodegradation within 10 days window


The model is able to predict:

Metabolite

Parent: CAS 31570-04-4

Phenol 2,4-bis(1,1-dimethylethyl), phosphite (3;1)BOD=0.038LC50=145835248 mg/l

Metabolite:BOD=0.016LC50=0.34 mg/l

Biodegradation

MetaboliteBOD=0.39LC50=1.2.106 mg/l

Parent: CAS 124-28-7N-n-Octadecyl-N, N-dimethyl amineBOD=0.83LC50=0.94 mg/l

Biodegradation

Fish liver simulator Bioccumulation

Parent

Metabolism

Phase II

Phase II

Reactivespecies

Reactivespecies

Reactivespecies

S-PrW sensitization

S-PrW sensitization

S-Pr

S sensitization

S-PrS sensitization

No sensitization

QSAR

* Dimitrov et al, 2005. Skin sensitization: Modeling based on skin metabolism simulation and formation of protein conjugates. International Journal of Toxicology 24, 189-204.

TIMES Skin Sensitization Model*

Predicted metabolism of isoeugenol

OH

O

OH

OH

O

O

.

O

O

Pr

O

O

SO

O

OH

O

O

Pr S

OH

OH

+ CH3OH

CH3 O

S

OHO

O

OOCH3

HO

OH

OH

OH

O

Glucoronidation

Sulphate conjugation

Dealkylation

Formation of semiquinone free radicals

Michael addition on quinones Free radical reaction on proteins

Skin sensitization

O

O

OHHO

HO

OOH

O

O

OHO

O

NH

SO

OHO

N

CH3

N O

OH

N

CH3

NO

O

NH O

OH

OHO

H3C N

N OH

C+

O

O

OHHO

HO

OOH

O

O

OHO

O

NH

SO

OHO

N

CH3

N O

OH

N

CH3

NO

O

NH O

OH

OHO

H3C N

N OH

C+

O

O

OHHO

HO

OOH

O

O

OHO

O

NH

SO

OHO

N

CH3

N O

OH

N

CH3

NO

O

NH O

OH

OHO

H3C N

N OH

C+

O

O

OHHO

HO

OOH

O

O

OHO

O

NH

SO

OHO

N

CH3

N O

OH

N

CH3

NO

O

NH O

OH

OHO

H3C N

N OH

C+

O

O

OHHO

HO

OOH

O

O

OHO

O

NH

SO

OHO

N

CH3

N O

OH

N

CH3

NO

O

NH O

OH

OHO

H3C N

N OH

C+

O

O

OHHO

HO

OOH

O

O

OHO

O

NH

SO

OHO

N

CH3

N O

OH

N

CH3

NO

O

NH O

OH

OHO

H3C N

N OH

C+

N-Nitrosoamine Aliphatic C-Oxidation

N-Nitrosoamine Oxidative N-DealkylationN-Nitrosoamine Oxidative N-Dealkylation

Electrophilic Species GenerationAliphatic C-Oxidation

O-Glucuronidation

Amino Acid Conjugation

Reacting with DNA

Mutagenicity

Laboratoryconditions

cDNA-expressedIndividual CYPs

Rat liver homogenate(S9 fraction)

Primary hepatocytes

MicrosomesCytosol

Liver slices In vivo

IntactOrganism

Intact Cells, Tissue, OrganismSub-cellular fractions

Next Major Activity

Research Partners from Regulation Agencies

US EPA, Athens, GAUS EPA, Duluth, MNEnvironment CanadaNITE Japan

Research Partners from Industry

P&GExxonMobilUnileverBASF

metabolism in-silico simulation laboratory of mathematical chemistry university ‘prof. as....

Documents