Professional Documents
Culture Documents
Relationship(QSAR)models
CeydaOksel
Xue Z. Wang
University ofLeeds
27.03.2013
A pan-European Infrastructure for Quality in Nanomaterials Safety Testing
NUID UCD | NHM | IOM | JRC | BFR | KIT | FUNDP | IST | UNIVLEEDS | NILU | HMGU | LMU | CIC | UU | ICN
DLO/RIKILT | WU | DGUV | TAU | VITO | SMU | TCD | FIOH | UOE | HW | CNRS | INERIS | UBRM | RIVM
Purpose ofToday
Theory ofQSAR
NewDataMining Methods
NanomaterialCharacterization
Nanotoxicity
Application ofQSARmodelinnanotoxicity
modeling
QualityNanoTrainingSchool
Part I
Theory ofQSAR
QualityNanoTrainingSchool
QSARHistory
1865
1893
Crum Brownand
Fraser[1]
expressed the
ideathat there
was a relationship
betweenactivity
andchemical
structure.
Richet[2]
correlated
toxicities of
simple organic
molecules
withtheir
solubilityin
water.
1900
Meyer[3] and
Overton[4]found
linearrelationships
betweenthe
toxicityoforganic
compoundsand
their lipophilicity.
1969
Hansch[5] published
afreeenergyrelated
modelto correlate
biologicalactivities
with
physicochemical
Properties.
Thefatheroftheconceptof
quantitativestructureactivity relationship (QSAR),
thequantitativecorrelationofthephysicochemical
propertiesofmoleculeswiththeirbiological
activities[5]
CorwinHermanHansch
(1918 2011)
What isQSAR?
QSAR(QuantitativeStructureActivity
Relationship)
TheVarianceIn
Molecular
Structure
Physicochemical
or Biological Property
A QSAR is
that relates
descriptors
compound
activity.
a statistical model
a set of structural
of a chemical
to its biological
MathematicalDependencies
Toxicity/Ecotoxicity Prediction
BulkFormNanoparticles
(widely used)(notadopted)
The presenceofparticular characteristics increase
the propability that the compund istoxic
QualityNanoTrainingSchool
[1]T.Puzyn etal.(eds.),RecentAdvancesinQSARStudies,103125. DOI10.1007/9781402097836_4,CSpringerScience+Business MediaB.V.2010
nanoQSAR
QSAR(QuantitativeStructureActivity
Relationship)
QSAR has been widely used to predict the toxicity of substances in bulk
form (especially druglike compunds) but, up to date, QSAR studies for the
prediction of nanoparticle toxicity have been rarely reported.
The European system for new chemical management (REACH) promotes
QSAR methods as an alternative way of toxicity testing.
QSARmodels are very useful incase ofthe classic chemicals butthe
concept ofnanoQSARisstill under development.
QualityNanoTrainingSchool
predict
validate
INSILICO
(computational)
Alternative,fast
developingapplications
IN VITROINVIVO
(incontainer)
(onlivingorganism)
Themostimportantsources
ofinformation
[1]A..Gajewicz,etal.,Advancingriskassessmentofengineerednanomaterials:Applicationofcomputationalapproaches,
QualityNanoTrainingSchool
Adv.DrugDeliv.Rev.(2012),doi:10.1016/j.addr.2012.05.014[1]
Purpose ofQSAR
To predict biologicalactivityand physicochemical
propertiesbyrationalmeans
To comprehendandrationalizethemechanismsofaction
withinaseriesof chemicals
Savingsinthecostofproductdevelopment
Predictionscouldreducetherequirementforlengthyand
expensiveanimaltests.
Reductionofanimaltests,thusreducing animaluseand
obviouslypainanddiscomforttoanimals
Otherareasofpromotinggreenandgreenerchemistryto
increaseefficiencyand eliminatewaste
QualityNanoTrainingSchool
2. Setofmolecular
descriptors
3.Measured biological
activity or property
DESCRIPTORS
QSARs
Physcochemical,
biological, structural
asmany as
possible
Toprovidesomeguide,itis widelyacceptedthat
betweenfiveandtencompoundsarerequired
foreverydescriptor inaQSAR[1]
4.Statistical
methods
Toxicity
Endpoints
Daphnia magna
EC50
Cancinogenicity
Mutagenicity
Rat oral LD50
Mouse inhalation
LC50
Skin sensitisation
Eye irritancy
QualityNanoTrainingSchool
[1]Aptula AO,RobertsDW(2006)Mechanisticapplicabilitydomainsfornonanimalbasedprediction
oftoxicologicalendpoints:Generalprinciples
andapplicationtoreactivetoxicity.Chem ResTox.19:10971105
Applications ofQSAR
There are a large number of applications of these models within
industry, academia and governmental (regulatory) agencies.
Therationaldesignofnumerousotherproductssuch
assurfaceactiveagents, perfumes,dyes,andfine
chemicals.
Thepredictionofavarietyofphysicochemical
propertiesofmolecules.
Theidentificationofhazardouscompoundsatearly
stagesofproductdevelopment,thepredictionof
QualityNanoTrainingSchool
toxicitytohumans
and environment.
10
2
3
4
5
adefinedendpoint;
anunambiguousalgorithm;
adefineddomainofapplicability;
appropriatemeasuresofgoodnessoffit,
robustnessandpredictivity;
amechanisticinterpretationifpossible.
Common QSARModelling
Errors
Uninformativedescriptors
Poordescriptorselectionand
chancecorrelations
Modelling complex,nonlinear
structureproperty
relationshipswithlinearmodels
IncorrectlyvalidatingQSPR
models
Notunderstandingthedomain
ofapplicabilityofmodels
Methods
QSARModelling process consists of5main steps.
Main Steps
Molecular Structure
Representation
Molecular Descriptors
Predictive Model
ModelValidation
Applicability Domainof
QSAR
QualityNanoTrainingSchool
12
Molecular Descriptors[1]
Molecular
Descriptors
ca
calculated by
Mathematical
formulas
Experimental
measurements
Theoretical
Indexes
Computational
algortihms
Different Theories
(e.g.Quantum
chemistry,
information theory,
organic chemistry)
derived
from
Molecular
Descriptors
processed
by
statistics
chemometrics
chemoinformatics
applied
in
QualityNanoTrainingSchool
[1]T.Puzyn etal.(eds.),RecentAdvancesinQSARStudies,103125. DOI10.1007/9781402097836_4,CSpringerScience+Business
MediaB.V.2010
[2] Todeschini R,Consonni V(2000)Handbookofmoleculardescriptors.WileyVCH,Weinheim
13
DataAnalysis Methods[1]
Different Statistical Methods have been used inQSARfor the extraction ofuseful information
from the data.
RegressionProblems
DataCollection
Multiplelinearregression
Partialleastsquares
Feedforwardbackpropagationneuralnetwork
Generalregressionneuralnetwork
ClassificationProblems
Molecular
Descriptor
DataAnalysis
ModelValidation
LinearDiscriminant analysis
Logisticregression
Decisiontree
Knearestneighbor
ProbabilisticNeuralNetwork
Supportvectormachine
RecentlydevelopedMethods
Kernelpartialleastsquare
Robustcontinuumregression
Locallazyregression
Fuzzyintervalnumberknearestneighbor
Fastprojectionplaneclassifier
QualityNanoTrainingSchool
[1] Ekins, S. (2007). Computational Toxicology: Risk Assessment for Pharmaceutical and Environmental Chemicals. New Jersey: Wiley-Interscience.
14
DataMiningMethods
arelativelynewfieldwhichsharesometechniqueswithtraditional
dataanalysisbutalsobringsalotofnewvisionsandtechniques.
Value
Decision
Knowledge
Information
Data:recordsofnumerical
data,symbols,images,
documents
Knowledge:
Rules:IF..THEN..
Causeeffectrelationships
Decisiontrees
Patterns:abnormal,normal
operation
Predictiveequations
QualityNanoTrainingSchool
15
DataMiningMethods
Data mining shares many techniques and tools with traditional data
analysis (such as the ones listed below). But it emphasises the analysis
of very large databases, and can deal with complicated cases which
cannot be handled by traditional methods.
Clustering
Summarisation
Classification
Regression
Conceptual Clustering
Case-based Learning
Inductive learning
Data mining research has also generated some new techniques that
were not seen in traditional data analysis, such as Link Analysis
QualityNanoTrainingSchool
16
NewDMTechniqueExample1:LinkAnalysis
Example :Tendatacases,threevariables,x1,x2,andx3
How canweautomaticallyuncoverthedependencyrelationshipbetweenthem?
WHICHRELATIONSHIP?
DATASET
x1
x2
x3
1
1
0
1
0
0
1
0
1
0
0
1
0
1
0
1
1
0
1
0
0
1
1
1
0
1
1
0
1
0
x1
x2
x3
x2
x1
x3
Thisrelationship
means,x2 is
dependent onx1,and
x3isdependent onx2,
butx1isindependent
ofx2andx3
Thisrelationship
means,x2is
dependent onx1,x3
isalso dependent on
x1,butx2andx3are
independentofeach
other
17
QualityNanoTrainingSchool
x1
x2
x3
x4
x1
x2
x3
x4
55.33
1.72
54
1.66219
92.19
71.31
3.44
55
1.60325
90.51
59.13
1.2
53
1.58399
92.74
72.3
4.02
55
1.66783
90.24
57.39
1.42
55
1.61731
91.88
68.81
6.88
55
1.69836
91.01
56.43
1.78
55
1.66228
92.8
66.61
2.31
52
1.77967
91.9
55.98
1.58
54
1.63195
92.56
63.66
2.99
52
1.81271
91.92
56.16
2.12
56
1.68034
92.61
63.85
0.24
50
1.81485
92.16
54.85
1.17
54
1.58206
92.33
67.25
53
1.72526
91.36
52.83
1.5
58
1.54998
92.22
67.19
52
1.86782
92.16
x1,x2,x3aredifferentmeasuresoffeedcomposition
x4isthelogofacombinationofoperatingconditions
y productqualitymeasure
y=good if>92,
=low ifbetween9192,
=verylow if<91
QualityNanoTrainingSchool
Aim:tofindwhichfeedcomposition
oroperationalconditionvariables
havethemostimportantimpacton
productquality
18
Automaticallygenerateddecisiontree
Productqualityismainlydeterminedbyx1andx4,almosthasnothingtodowithx2andx3
Ifx1<55.4,theprobabilitythatproductqualityisgood(y>92)ishigh
Ifx1>55.4butx4>1.8,theprobabilitythatproductqualityisgoodisalsohigh
Othertworegions,product qualityiseitherloworverylow
x1
Verylow
>55.4
<55.4
Good
2.0
>1.8
<1.8
x1
<70
91<y<92
Low
Good
x4
x4
y>92
Low
Y>92
>70
1.5
Good
Y<91
Verylow
50
60
70
Ifx1=[44.6,55.4],thenconditionalprobabilityofy=3is0.9
QualityNanoTrainingSchool
Ifx4=[1.8,2.3],thentheconditionalprobabilityofy
=3is1.0
x1
19
(1)Generationofapopulationofsolutions
<s >s
<s >s
<s >s
<s >s
Automatic
generationof
decisiontreefrom
data
This isanew
.. .... .. .. .... .. .... .. .. .... ..
approoach
(2) Repeatsteps(i) and(ii) untilthestopcriteriaare based onagenetic
programming
method
satisfied:
(i)calculatethefitnessfunctionvaluesforeach
solutioncandidate
(ii)performcrossoverandmutationtogenerate
thenextgeneration
<s >s
<s >s
<s >s
<s >s
.... ..
(3) thebestsolutioninallgenerations:thefinalsolution
QualityNanoTrainingSchool
20
Data Set 2:
Concentration effecting 50% of the population,
EC50 of algae chlorella vulgaris, by causing fluorescein
diacetate to disappear
80 compounds
1150 descriptors
QualityNanoTrainingSchool
21
YAdaptforQSARToxicityPrediction:DataSet1
Atreeforthebacteria
datausingthree
classes.
SampletreefromYAdapt
Forbacteriadata:
of 1069 descriptors,
only5descriptorsare
important! Usingthe5
descriptors,thetree
modelcanpredict with
the accuracy>96.7%
Polarsurfacearea40.368
No
Yes
Gearyautocorrelationlag4
weightedbyatomicmass
0.007
Yes
Class1
(21/21)
Gearyautocorrelationlag4
weightedbypolarizability
1.181
No
Hautocorrelationlag0weighted
bySandersonelectro
negativities2.111
Yes
Class1
(8/8)
No
Yes
Class3
(9/10)
No
Class2
(9/9)
QualityNanoTrainingSchool
BrotoMoreauautocorrelation
lag2weightedbypolarizability
4.085
Yes
Class2
(5/6)
No
Class3
(6/6)
22
YAdaptforQSARToxicityPrediction :DataSet1
Class1:
Log(1/EC50)4.0829cases
Class2: 4.08<Log(1/EC50)4.5116cases
Class3: 4.51<Log(1/EC50)4.999cases
Class4:
otherwise6cases
A decision tree from generation 46 for the bacteria data, finding four
classes from the continuous endpoint itself during induction.
Training accuracy: 93.3%
Test accuracy: 80.0%
Polarsurfacearea40.368
Yes
No
3DMoRSEsignal15weighted18weightedby
Sandersonelectronegativites0.644
Hautocorrelationlag0weightedby
Sandersonelectronegativities2.111
Yes
Class1
(16/16)
Gearyautocorrelationlag4weightedby
atomicmass0.007
Yes
Class1
(13/13)
No
Yes
No
Class2
(5/7)
No
Class2
(9/9)
QualityNanoTrainingSchool
Hautocorrelationlag0weightedby
Sandersonelectronegativities2.394
Yes
Class3
(7/9)
No
Class4
(6/6)
23
YAdaptforQSARToxicityPrediction :DataSet2
Solvationconnectivityindex 2.949
No
Yes
Selfreturningwalkcount
order83.798
Yes
Class1
(16/16)
No
Class2
(14/15)
Molecularmultiplepathcount
order392.813
Yes
No
Class3
(6/8)
Hautocorrelationoflag2weighted
bySandersonelectronegativities
0.401
Yes
Class4
(6/7)
No
Training:92.2%
2ndcomponentsymmetry
directionalWHIMindex
weightedbyvanderWaals
volume0.367
Test:81.3%
Yes
Class3
(9/10)
QualityNanoTrainingSchool
No
Class4
(8/8)
24
IntegratedProcessDataMiningEnvironment
Descriptor
calculation
DataPreprocessing
Scaling
Missingvalues
Outlieridentification
Featureextraction
DataMiningToolbox
Regression
PCAandICA
ART2networks
Kohonennetworks
Knearestneighbour
Fuzzycmeans
Decisiontreesandrules
Feedforwardneural
networks(FFNN)
Summarystatistics
Visualisation
MixtureQSARmodels
DiscoveryValidation
Statisticalsignificance
Resultsfortrainingand
testsets
Dataimport
Excel
ASCIIFiles
Database
XML
Results
Presentation
Graphs
Tables
ASCIIfiles
QualityNanoTrainingSchool
Different datamining
techniques canbe
combined intoasingle
25
system
QSARModelValidation
Crossvalidation (Q2)
Model
Validation
Model
Applicability
Model
Interpretation
Modelvalidationisrequiredtoensure
modelreliability (quality ofthe model)
Ensures thatthemodelisnotdueto
chancefactors!
26
QSARWorkflow[1]
QSARModelApplicability Domain
Descriptor 1
.
...
..
.
. . TRAININGSET
.
.
. . Descriptor 2
......
. .
...
Inside Applicability
Domain
Outside Applicability
Domain
will bepredicted
will notbepredicted
Modelapplicabilityallowsusto
decidewhetherthemodelwill be
usefulfornewdata (correct use of
the model).
Finding the boundaries ofthe
modelto see how well itwill work
for other compounds.
Notevenasignificant and
validatedQSARmodelcanbe
used for the entireuniverse of
chemicals[1]!
QualityNanoTrainingSchool
28
TOPKAT(byKomputer AssistedTechnology):
ItusesQSARstogivenumericpredictions,
basedonseveral descriptors
2. HazardExpert (by CompuDrug ):
HazardExpert,producedbyCompuDrug,isa
rulebasedsystem.Therulesuse
toxicophores,orstructuralalertsderivedfrom
QSARs
3. CASE: It canderive QSARmodelfrom users
6.DEREK:DeductiveEstimation
data.
ofRiskfromExistingKnowledge
4. ECOSAR:It usesQSARstopredicttheaquatic
(DEREK)ismarketedbyLHASA
toxicityofchemicalsforavarietyoforganisms
Ltd.It isanexpertknowledge
5. OASIS: Ituseshierarchicalrulesbasedon
basesystemthatpredicts
QSARs,whichcanbeuserderived.
whetherachemicalistoxic
Each ofthese QSARtools have their own advantages and weaknesses for toxicity
predictions
There are also several available softwarepackages for descriptor calculations (e.g.
DRAGON)
QualityNanoTrainingSchool
29
Ifthetrainingdatasetisnotlargeenough,thedatacollectedmaynotreflect
thecompletepropertyspace.Consequently,manyQSARresultscannotbe
usedtoconfidentlypredictthemostlikelycompoundsofthebestactivity.
Therearemanysuccessfulapplicationsofthismethodbut onecannot
expecttheQSARapproachtoworkwellallthetime. QSAR Modelsareeasy
tobuildbutalsoveryeasytogetwrong.
QualityNanoTrainingSchool
30
[1]Puzyn,T.,Leszczynski,J.(2012) TowardsEfficientDesigningofSafeNanomaterials:InnovativeMergeofComputational
ApproachesandExperimentalTechniques,The
RoyalSocietyofChemistry.
Part II
Nanotoxicity and
NanomaterialCharacterization
QualityNanoTrainingSchool
31
Macroconcernswithnanoworld
!
Macroconcernswithnanoworld !
Uncertainities atthe early stages ofanew technology bring new concerns!
NewTechnology
NewBenefits
NewConcerns
[1]
Safetytohumanhealthand environment
Suitability ofriskmanagementstrategies
Nostandardizedor
validatedmethods
fornanotoxicity
testing
Available toxicity
datafor NPs are
inconsistent and
confusing
Blocks the
developmentof
ENPsriskreduction
strategies[2]
[1] Chaudhry, Q. et al. (2010). Knowns, Unknowns, and Unknown Unknowns . RSC Nanoscience & Nanotechnology (p. 201-218).
Royal Society of Chemistry.
[2]Tran, L., & Chaudhry, Q. (2010). Engineered Nanoparticles and Food: An Assessment of Exposure and Hazard. RSC Nanoscience &
QualityNanoTrainingSchool
32
Nanotechnology (p 120-134). Royal Society of Chemistry.
[1]
Riskassesment
for
NMS
Riskassesment for NMS
[1]
4.Risk
Whatisthenature
andmagnitude ofthe assessment
riskandhow
accuratelycanitbe
estimated?
3.Exposure
assessment
Towhatextentis
therelevant
populationgroup
exposed tothe
harmful agent?
QualityNanoTrainingSchool
OECD (2003): Environment Directorate. Description of Selected
Key Generic Terms Used in Chemical / Hazard Assessment. OECD 33
Series on Testing and Assessment Number 44. ENV/JM/MONO(2003)15. Paris, 30-Oct-2003.
Doseresponse relationship
Theevaluationoftoxicityincludes twosteps:
Hazard Identification
Evaluation oftoxicity
Doseresponse evaluation
involves determiningtheconditionsorlevelsatwhichthepresenceof
acontaminant maytriggerabiologicalresponsefromthebody
Different amount of
CNTdust
At level of0.1mg/m3
Noeffects were
observed
Athigher level of
0.5mg/m3
Sideeffects were
detected in the
lung 34
QualityNanoTrainingSchool
Naidu, B. David (2009). Biotechnology and Nanotechnology - Regulation under Environmental, Health, and Safety Laws.. Oxford University Press.
Particle Characterization
Theonlyreasonable waytoobtain toxicity information for thelarge
numberofnanoparticleswithout testing everysingleone istorelate
thephysicochemicalcharacteristicsofNPs withtheirtoxicityinaSAR
(StructureActivityRelationship)orQSARmodel.
Nanomaterialcharacterization isone ofthe first steps intoxicity test.
However,itisavery challenging issue!!!
NanoparticleStructure
Properties
ParticleSize
SurfaceArea
Morphology
StateofDispersion,
Toxicity
QualityNanoTrainingSchool
35
Fullerenes
CarbonNTs
Graphene
Carbonbased
Nanomaterials
Different and
Exceptional
Properties
Nocertain information!!!
Recent studies showed that toxicity ofNPs canberelated to:
Sizeandshape
Sizedistribution
Agglomeration state
Porosity
Structuredependentelect.
Electronicproperties
Characterisation:
Inwhatmedium?
Surfacearea
Surfacechemistry
Surfacecharge
Crystalstructure
Composition
Configuration
Coating
Aggregationstate
Metalcontent
POSSIBLE factors[1]
inducingtoxicityof
chemicals
moreresearchisrequired
toreachaclearunderstanding
QualityNanoTrainingSchool
[1]
Puzyn, T., Leszczynska, D. & Leszczynski, J. Toward the development of Nano-QSARs: advances and challenges. Small 5, 24942509 (2009).
36
NanomaterialCharacterization
Several different methods and instruments are used to determine the properties onNMs
Shapeandaggregation
state
HighResolutionMicroscopy
Composition,purity
andsurfacechemistry
Spectroscopyand
Chromatography Methods
Surfacecharge
ZetaPotentialAnalysis
Crystalstructure
XRayDiffraction
Particlenumberand
sizedistribution
SeveralTechniques(e.g.
SEM,TEM,DLS)
SurfaceArea
BrunauerEmmettTeller
Adsorption
There isnosingle instrument that isright tool for every test.There are infact more than 400
QualityNanoTrainingSchool
37
different techniques
for particle counting and characterization
Whathaven'tweconsideredyet ?
SomeNPs
Have atendency toformclusters (e.g.aluminium oxide)
inanagglomeration state,someNPsbehavelikelargerparticles
inthemedia
Toxicitymayalsodependsonthesizeofagglomerate
NOTonoriginalparticlesizeitself
!!!
If the NPs aggregate
Howtocharacterizeaggregation?
itisimportant toidentifyhow
Textureanalysis?
thesechangesaffecttoxicity
38
Hussain, S. e. (2008). Can silver nanoparticles be useful as potential biological labels? Nanotechnology 19 .
Challenges
for nanoQSAR?
ChallengesforQSAR
Table shows the results of3different toxicity studies for silver NMs:
NP Size
type (nm)
Doses
Cell
(mg/ml)
Ag
720
0.850
Human
fibrosorcoma and
skincarcinoma
Ag
<15
0.7
Human hepatoma
Ag
15
and
100
10250
Differencesin
NMsused
(size)
[1]
Differencesin
experimental
conditions
Method
Result
XTTassay
Nontoxic
MTTAlamarblue Oxidativestres
andLDHassay
mediatedtoxicity
Kim etal.
(2009)[2]
MTT,LDH,GSH
level,ROSand
MMP
Hussain etal.
(2005)[3]
Toxic
Araro etal.
(2008)[1]
strongly affects
the results
Arora S, Jain J, Rajwade JM, Paknikar KM (2008). Cellular responses induced by silver nanoparticles: in vitro studies. Toxicol Lett 179:93100
Kim S, Choi JE, Choi J, Chung KH, Park K, Yi J, Ryu DY (2009). Oxidative stress-dependent toxicity of silver nanoparticles in human
hepatoma cells. Toxicol In Vitro 23:10761084
[3] Hussain SM, Javorina AK, Schrand AM, Duhart HM, Ali SF,
QualityNanoTrainingSchool
39
Schlager JJ (2006). The interaction of manganese nanoparticles with
PC-12 cells induces dopamine depletion. Toxicol Sci 92:456463
[2]
Lackofhighqualityexperimentaldata
Lackofrationalmodelingprocedurestoscreenlargenumbersof
structurallydiversifiednanoparticles.
40
Part III
Case Study
despite all limitations
QualityNanoTrainingSchool
41
Nanotxicology 2012,47September,Beijing
QSARToxicity(invitro)AnalysisforaPanelofEighteen
NPs
The aim of this study was to test the hypothesis that nanoparticle
toxicity is a function of some measurable physical characteristics and
that Structure Activity Relationship (SAR) might ultimately be a useful
tool for nanotoxicology. A panel of 18 nanoparticles was chosen to
include carbonbased materials and metal oxides. Comparative toxicity
of the nanoparticles in the panel was assessed, using a number of
different measures of apoptosis and necrosis; haemolytic effects and
the impact, on cell morphology were also assessed. Structural and
composition properties were measured including size distribution,
surface area, morphological parameters, metal concentration,
reactivity, free radical generation and zeta potential. Toxicity end
points were processed using principal component analysis and
clustering algorithms.
QualityNanoTrainingSchool
42
Nanooxicology 2012,47September,Beijing
Particles
Qualifier
Carbonblack
Dieselexhaustparticles
Nanotubes
Fullerene
Printex 90
EPA
Japanese
Number
N1
N2
N3
N4
Unmodified
N5
Polystyrenelatex
Amine
N6
Carboxylated
N7
7nm
N8
Aluminiumoxide
50nm
N9
300nm
N10
Ceriumoxide
N11
Nickeloxide
N12
Siliconoxide
N13
Zincoxide
N14
Rutile
N15
Titaniumdioxide
Anatase
N16
N17
QualityNanoTrainingSchool
Silver
*AHT:AlliedHighTech,**NAM:NanostructureandAmorphousMaterialInc
N18
Supplier
Degussa
EPA
VickiStone
Sigma
Polysciences
Sigma
Polysciences
KrahnChemie
AHT*
AHT*
NAM**
NAM**
NAM**
NAM**
NAM**
NAM**
NAM**
NAM**
Form
Dry
Dry
Dry
Dry
Sus.
Sus.
Sus.
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
Dry
43
Sus.
CharacterisationofPhysicochemicalProperties
Nanotoxiclogy 2012,47September,Beijing
Particle shape was analysed using LEO 1530 Scanning Electron Microscope (SEM) or Philips
CM20 Transmission Electron Microscope (TEM) aspect ratio and mean size
Surface area and porosity were measured using TriStar 3000 BET. Thirteen measurements,
including five surface areas based on different definitions, three pore volumes, three pore
sizes as well as the mean size and particle density.
Particle size and size distribution were analysed using a Malvern MasterSizer 2000. Size
distribution, and seven other size properties (mass diameter, uniformity, specific surface
area, surface area mean diameter and three mass diameters)
ThefreeradicalactivitiesweremeasuredbyEPR(Electronparamagneticresonance)
using the spin traps, DMPO and Tempone H separately. Tempone H has partial selectivity for
trapping superoxide anions, whilst DMPO traps mainly hydroxyl radicals
Particle reactivity in solution, the dithiothreitol (DTT) consumption test as a reducing species
was carried out. Since the DTT consumption test can only be conducted on dry powders, only
fourteen of the panel of nanoparticles were assessed using this assay.
Metal Concentration water soluble concentration of ten heavy metals (Ti, V, Cr, Mn, Fe, Co,
Ni, Cu, Zn and Cd) was measured
Charge:zpotentialwasmeasuredusingMalvernInstrumentsZetasizer Nano instrument
QualityNanoTrainingSchool
44
TotalNumber ofMeasurements
13 BETmeasurements(for14drysamples)
2 SEM/TEMmeasurements (forallsamples)
7 laserdiffractionsizestatisticalmeasurements(MasterSizer 2000)forallsamples
84 laserdiffractionsizedistribution measurements
2 EPRmeasurements(forallsamples)
1 reactivitymeasurement (fordrysamples)
10 metalcontentmeasurements
Total:119Measurements
Particle
name
BET Measurements
M2000
Measurements
SEM Measurements
Zeta measurements
Other
Measurements
N1
A(1,1),A(1,2),,A(1,12)
B(1,1),B(1,2)
C(1,1),C(1,2)
D(1,1),D(1,2)
E(1,1),E(1,2)
N2
A(2,1),A(2,2),,A(2,12
B(2,1),B(2,2)
C(2,1),C(2,2)
D(2,1),D(2,2)
E(2,1),E(2,2)
N18
A(18,1),A(17,2),
QualityNanoTrainingSchool
B(18,1),B(18,2)
C(18,1),C(18,2)
D(18,1),D(18,2)
E(18,1),E(18,2)
45
SEMandTEMimagesofthe18nanoparticlesamples
Nanotoxicolgy 2012,47September,Beijing
QualityNanoTrainingSchool
46
Table2.Thewatersolubleconcentrationsofvariousmetalsforthe18nanoparticlesamples
Particlenames
MetalConcentration(g/g)
Ti
Cr
Mn
Fe
Co
Ni
Cu
Zn
Cd
CarbonBlackN1
0.000
0.057
0.000
0.305
0.000
0.002
0.263
1.161
9.687
0.008
DieselExhaustN2
2.320
0.312
16.47
20.81
208.4
0.508
4.940
2.235
3181
0.387
JapaneseNanotubesN3
0.000
0.100
0.000
0.161
13.10
0.000
0.299
0.123
0.460
0.001
FullereneN4
0.000
0.084
0.000
0.000
0.000
0.000
0.534
0.142
4.837
0.000
PolystyreneLatexBeadsN5
0.000
1.172
0.000
0.000
0.000
0.000
0.000
0.625
19.56
0.000
PolystyreneLatexBeadsN6
4.368
0.779
0.343
0.199
0.000
0.006
0.440
18.23
58.71
0.010
PolystyreneLatexBeadsN7
0.000
1.018
0.000
0.000
0.000
0.000
0.000
2.588
22.323
0.000
Aluminuim OxideN8
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.714
0.000
0.000
AluminuimOxideN9
0.000
0.031
0.000
0.000
0.000
0.000
0.056
0.099
4.987
0.000
AluminuimOxideN10
0.000
0.062
0.000
0.000
0.000
0.000
0.023
0.088
0.000
0.000
CeriumOxideN11
0.000
0.004
1.058
0.128
0.000
0.000
0.616
0.191
112.9
0.007
NickelOxideN12
0.000
0.056
0.000
0.000
0.000
0.042
16963
0.286
0.000
0.038
SiliconOxideN13
0.000
0.088
0.000
0.000
0.000
0.000
14.163
0.106
0.000
0.000
ZincOxideN14
0.000
0.096
0.000
0.000
0.000
0.000
1.336
0.047
8185
9.711
TitaniumDioxideRutile N15
0.000
0.083
0.000
0.000
0.000
0.000
0.286
0.276
39.623
0.000
TitaniumDioxideAnatase N16
0.000
0.104
0.000
0.160
0.000
0.006
0.692
0.303
402.3
0.000
SilverN17
0.000
0.111
0.000
0.503
0.000
0.000
0.475
0.597
17.867
0.019
SilverN18
0.000
0.096
0.000
0.256
0.000
0.059
0.561
162.8
194.8
0.037
QualityNanoTrainingSchool
47
CharacterisationofParticleSizeand Shape
Nanotoxicology 2012,47September,Beijing
CarbonBlackN1
LEO1530GeminiFEGSEM
ScanningElectronMicroscopy
TheinformationaboutSize,
Shape canbeobtainedfrom
SEMimages
MeansizeobtainedbymeasuringtheparticlesontheSEM/TEMimage:Mean_Size_SEM
AspectratioobtainedfromSEM/TEMimages:Aspect_Ratio
QualityNanoTrainingSchool
48
Nanotoxicology 2012,47September,Beijing
Micromeritics TriStar3000
SurfaceAreaTechnique:GasAdsorption
MinimumSurfaceArea:0.01m2/g
SupportGases:NitrogenandHelium
MeasurementParameters:BETMeasurement
AdsorptionIsotherms
DesorptionIsotherms
tplots
PoreVolumeAnalysis
PoreSizeAnalysis
diameter
:
BJH_AC_SAOP
diameter:
BJH_DC_SAOP
Pore Sizes ()
3000.000 diameter:
BJH_AC_VOP
BJH Desorption cumulative volume
QualityNanoTrainingSchool
**14drypowdersamples,noavailableforthreesuspensions(N5,N6,andN7)andthenewSilversample.
49
CharacterisationofNPsMastersizer
Nanotoxicology 2012,47September,Beijing
10
9
P r o b a b ilit y ( % )
8
7
6
5
4
3
2
1
0
0.01
0.1
10
100
1000
10000
Size (micron)
MalvernMastersizer 2000
TypeofAnalyser:LaserDiffraction
SampleVolume:1000ml,150ml(HydroGandS)
MeasurementRange:20nm 2000m
QualityNanoTrainingSchool
Theinformationaboutmean
SizeandSizedistribution,can
beobtainedfromthissizer.
50
CharacterisationofNPsMastersizer
Nanotoxicology 2012,47September,Beijing
Mastersizer 2000Measurements:91**
D[4,3] theVolumeWeightedMeanorMassMomentMeanDiameter,represent
byD[4,3]
Uniformity:ameasureoftheabsolutedeviationfromthemedian,representedby
Uniformity
Specificsurfacearea:thetotalareaoftheparticlesdividedbythetotalweight,
SSA_2000
D[3,2] theSurfaceWeightedMeanorSurfaceAreaMomentMeanDiameter,
representedbyD[3,2]
D(v,0.5) thesizeinmicronsatwhich50%ofthesampleissmallerand50%is
larger.ThisvalueisalsoknownastheMassMedianDiameterorthemedianofthe
volumedistribution,representedbyD(0.5)
D(v,0.1) thesizeofparticlebelowwhich10%ofthesamplelies,representedby
D(0.1)
D(v,0.9) thesizeofparticlebelowwhich90%ofthesamplelies,representedby
D(0.9)
Sizedistribution:thereare100values.Amongthemthereare16valuesequalto
zeroforallparticles.So84 sizedistributionattributesareusedforfurtheranalysis
** Note:availableforallsamples
QualityNanoTrainingSchool
51
CharacterisationofNPsZeta Potential
Nanotoxicology 2012,47September,Beijing
MalvernZetaSizerNano
instrument
Sizeranges:0.6nm 6m
ZetaPotential
Suitable:suspensions
QualityNanoTrainingSchool
52
Toxicity Results
Cytotoxicity byLDHRelease
Viability
25.00
N6
80.00
40.00
100.00
20.00
Apoptosis
60.00
120.00
N12
N14
Viable
100.00
LDHRelease
Apoptosis
80.00
15.00
60.00
10.00
N6
40.00
N14
20.00
20.00
5.00
N6
0.00
31.25
62.50
125
250
Dose(g/ml)
0.00
0.00
31.25
62.50
125
Proinflammation effects
80
62.50
125
250
Dose(g/ml)
Dose(g/ml)
Necrosis
100
Haemolysis
90
Haemolysosatdose500g/ml
70
N6
60
Necrosis
31.25
250
50
40
N14
30
20
N3
10
80
70
60
50
40
30
20
10
0
31.25
62.50
125
Dose (g/ml)
250
53
Viability
Apoptosis
MTT
TotalToxicity
Necrosis
Haemolysis
QualityNanoTrainingSchool
54
Zincoxide
Aminated beads(N6),zincoxide(N14),Japanese
Nanotubes (N3)andnickeloxide(N12)areseparate
fromthemajorcluster.
Aminated beads(N6)hasthehighesttoxicity
valuesinnearlyallassayresults(LDH,apoptosis,
Japanesenanotubes
necrosis,haemolytic,MTTandcellmorphology
assays).
Polystrene larex (Amine)
Zincoxide(N14)hashightoxicityvalueinLDH,
apoptosis,necrosisandinflammationassays.
Nickeloxide
Japanesenanotubes (N3)havehightoxicityvalues
inviabilityandMTTassays.
Figure: Plotofthefirstandsecond PCs based
onPCAanalysisoftoxicitydata
Nickeloxide(N12)hasshownhighlytoxicinLDH
andhaemolyticassaysresults..
Thedistinctionofthefourparticles,N6,N14,N3andN12,fromother
particlesisclear.
QualityNanoTrainingSchool
55
N6 hasthehighesttoxicityvaluesinnearlyallassay
results (LDH,apoptosis,necrosis, haemolytic,MTT
andcellmorphologyassays).
N3 havehightoxicityvaluesinviabilityandMTT
assays.
56
57
N5
N6
PC1PC2PC3threedimensionalplotshows
N7
N3
thatthe18nanoparticleswereclearlygrouped
N2
N1 2
intoclusters.
N1 4
Thelargestclustercontainsparticlesthat
showedlowtoxicityvalues.
JapaneseNanotubes (N3),nickeloxide(N12)
andzincoxide(N14)anddieselexhaust
particles(N2)areoutsidethislowtoxicity
samplecluster.
There mustbeanotherreasonthatleadsto
hightoxicityvalueforN6
QualityNanoTrainingSchool
58
Results
Toexaminetheclusteringresultsmoreclearly,PC1
PC2, PC1PC3andPC2PC3twodimensionalplotsare
alsoexamined .
ItisPC2thatseparates N3(Japanesenanotubes),N12
(nickeloxide)andN14(zincoxide) fromtherest
samples,ascanbeseenfromPC1PC2andPC2PC3
plots.
QualityNanoTrainingSchool
59
Results
colour),
since
it
is
PC2
that
QualityNanoTrainingSchool
60
Results
N3
Inordertodetermineexactlywhich
physicochemicaldescriptors leadto
hightoxicityvaluesofN3,N12,N14
N14
performed.
N 12
Forthe14dryparticles,PCanalysis
withBETandDTTmeasurements gave
similarresultwith the first PCanalysis
inwhich BETand DTTdatawas
Figure: Principalcomponentanalysisofthestructural
excluded.
andcompositionaldescriptors(allsuchdescriptors
includingBETandDTTmeasurements)forthe14dry
nanoparticles.
QualityNanoTrainingSchool
61
Results
Figure: PC1andPC2contributionplotsforPCA
analysisof thecompositionaldescriptors
62
Table2.Thewatersolubleconcentrationsofvariousmetalsforthe18nanoparticlesamples
Particlenames
MetalConcentration(g/g)
Ti
Cr
Mn
Fe
Co
Ni
Cu
Zn
Cd
CarbonBlackN1
0.000
0.057
0.000
0.305
0.000
0.002
0.263
1.161
9.687
0.008
DieselExhaustN2
2.320
0.312
16.47
20.81
208.4
0.508
4.940
2.235
3181
0.387
JapaneseNanotubesN3
0.000
0.100
0.000
0.161
13.10
0.000
0.299
0.123
0.460
0.001
FullereneN4
0.000
0.084
0.000
0.000
0.000
0.000
0.534
0.142
4.837
0.000
PolystyreneLatexBeadsN5
0.000
1.172
0.000
0.000
0.000
0.000
0.000
0.625
19.56
0.000
PolystyreneLatexBeadsN6
4.368
0.779
0.343
0.199
0.000
0.006
0.440
18.23
58.71
0.010
PolystyreneLatexBeadsN7
0.000
1.018
0.000
0.000
0.000
0.000
0.000
2.588
22.323
0.000
Aluminuim OxideN8
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.714
0.000
0.000
AluminuimOxideN9
0.000
0.031
0.000
0.000
0.000
0.000
0.056
0.099
4.987
0.000
AluminuimOxideN10
0.000
0.062
0.000
0.000
0.000
0.000
0.023
0.088
0.000
0.000
CeriumOxideN11
0.000
0.004
1.058
0.128
0.000
0.000
0.616
0.191
112.9
0.007
NickelOxideN12
0.000
0.056
0.000
0.000
0.000
0.042
16963
0.286
0.000
0.038
SiliconOxideN13
0.000
0.088
0.000
0.000
0.000
0.000
14.163
0.106
0.000
0.000
ZincOxideN14
0.000
0.096
0.000
0.000
0.000
0.000
1.336
0.047
8185
9.711
TitaniumDioxideRutile N15
0.000
0.083
0.000
0.000
0.000
0.000
0.286
0.276
39.623
0.000
TitaniumDioxideAnatase N16
0.000
0.104
0.000
0.160
0.000
0.006
0.692
0.303
402.3
0.000
SilverN17
0.000
0.111
0.000
0.503
0.000
0.000
0.475
0.597
17.867
0.019
SilverN18
0.000
0.096
0.000
0.256
0.000
0.059
0.561
162.8
194.8
0.037
QualityNanoTrainingSchool
63
Conclusions
IdentifyingthefactorthatdiscriminatesN6(polystrene latexamine)fromN5
(unmodified)andN7(carboxylated)
PCAanalysis:PCAanalysisofmetalcontentsonly,orPCAanalysisofstructuraldescriptorsonly,orPCA
analysisofbothstructuralandcompositionaldescriptorstogether,couldnotdiscriminateN5,N6andN7
fromthelargestclustersofnontoxicparticlesamples.
PCAConclusion:neitherstructuraldescriptorsnormetalcontentsmadeN6differentfroN5
andN7intoxicity
Itwasatthisstageofanalysisthatwesuspectedthatthereareotherpossiblereasonsfortoxicityspecific
toN6:eitheritwasduetothechargedifference,orduetoN6beingamine,orboth.
zetapotentialforallthreepolystyrenelatexsamples.
N6(amine)
37.8,37.5,and40.3(mV),
positive
N5(unmodified)
36.2,38.8and36.8,
negative
N7(carboxylated)
54.9,55.3,and58.6.,
negative
AccordingtoVerma etal.anyparticlewithpositivechargeislikelytointeractelectrostatically sincemostcell
surfacesandotherbiologicalmembranesarenegativelychargedunderphysiologicalconditions(Verma etal.
2008).
Therefore,itismostlikelythatthelargepositivechargeofN6contributedits
observedhightoxicity,despiteitstructurallysimilartoN5andN7.
QualityNanoTrainingSchool
64
Conclusions
TheSARanalysisgeneratesthefollowingconclusions:
1. Themostlikelyreason ofhightoxicityforJapanesenanotube(N3)isitsshape(high
aspectratio).
2. Themostlikelycauseofhightoxicityforzincoxide(N14)andnickeloxide(N12)is
their highcontentsofzincandnickel.
3. Although,Aminated beads(N6)hasthehighesttoxicityvaluesinnearlyallassay
results,itwas always groupedcloselywiththeothertwopolystyrenelatexbeads,
unmodifiedbeads(N5)andcarboxylated beads(N7),whichshowednotoxicity.
Therefore,thezetapotentialofthethreepolystrene latexwas measured.The
resultsuggeststhatthelargepositivechargeofN6contributed to itsobservedhigh
toxicity.
65