Transcript Ole_Lund_June_4_2010..

Immunological Bioinformatics
Ole Lund
Challenges of the immune system
Outside
Infection
with
microbe A
Vaccine Infection Allergen ->
with
allergy
microbe B
Peptide Transplant
drugs ations
Time
Creation Creation
of self
of an
immune
system/
Tolerance
to self
Autoimmunity
(break of
tolerance to
self)
Inside
Cancer
Effect of vaccines
The Immune System
• The innate immune system
• The adaptive immune system
Adaptive immune response
• Signal induced
– Pathogens
• Antigens
– Epitopes
B Cell
T Cell
Presentation of peptides on
MHC I
Response
to1:(5x20x20
0) = 1:2000
peptides
1:5
1:200
Figure by Eric A.J. Reits
1:2
Peptide (epitope) bound to MHC
Figure by Anne Mølgaard, peptide (KVDDTFYYV) used as vaccine by Snyder et al. J Virol 78, 7052-60 (2004).
Data driven predictions
List of peptides that
have a given biological
feature
YMNGTMSQV
GILGFVFTL
ALWGFFPVV
ILKEPVHGV
ILGFVFTLT
LLFGYPVYV
GLSPTVWLS
WLSLLVPFV
FLPSDFFPS
CVGGLLTMV
FIAGNSAYE
Mathematical model (neural
network, hidden Markov model)
Search databases for other
biological sequences with the
same feature/property
>polymerase“
MERIKELRDLMSQSRTREILTKTTVDHMAIIKKYTSGRQEKNPALRMKWMMAMKYPITAD
KRIMEMIPERNEQGQTLWSKTNDAGSDRVMVSPLAVTWWNRNGPTTSTVHYPKVYKTYFE
KVERLKHGTFGPVHFRNQVKIRRRVDINPGHADLSAKEAQDVIMEVVFPNEVGARILTSE
SQLTITKEKKEELQDCKIAPLMVAYMLERELVRKTRFLPVAGGTSSVYIEVLHLTQGTCW
EQMYTPGGEVRNDDVDQSLIIAARNIVRRATVSADPLASLLEMCHSTQIGGIRMVDILRQ
NPTEEQAVDICKAAMGLRISSSFSFGGFTFKRTNGSSVKKEEEVLTGNLQTLKIKVHEGY
EEFTMVGRRATAILRKATRRLIQLIVSGRDEQSIAEAIIVAMVFSQEDCMIKAVRGDLNF
...
Prediction algorithms
MHC binding
data
Prediction
algorithms
Genome
scans
Antigen Discovery
Lauemøller et al., 2000
Influenza A virus
(A/Goose/Guangdong/1/96(H5N1))
>Segment 1
Genome
agcaaaagcaggtcaattatattcaatatggaaagaataaaagaactaagagatctaatg
tcgcagtcccgcactcgcgagatactaacaaaaaccactgtggatcatatggccataatc
aagaaatacacatcaggaagacaagagaagaaccctgctctcagaatgaaatggatgatg
gcaatgaaatatccaatcacagcagacaagagaataatggagatgattcctgaaaggaat
and 13350 other nucleotides on 8 segments
Proteins
9mer
peptides
>polymerase“
MERIKELRD
MERIKELRDLMSQSRTREILTKTTVDHMAIIKKYTSGRQEKNPALRMKWMMAMKYPITAD
ERIKELRDL
KRIMEMIPERNEQGQTLWSKTNDAGSDRVMVSPLAVTWWNRNGPTTSTVHYPKVYKTYFE
RIKELRDLM
KVERLKHGTFGPVHFRNQVKIRRRVDINPGHADLSAKEAQDVIMEVVFPNEVGARILTSE
IKELRDLMS
SQLTITKEKKEELQDCKIAPLMVAYMLERELVRKTRFLPVAGGTSSVYIEVLHLTQGTCW
KELRDLMSQ
EQMYTPGGEVRNDDVDQSLIIAARNIVRRATVSADPLASLLEMCHSTQIGGIRMVDILRQ
ELRDLMSQS
NPTEEQAVDICKAAMGLRISSSFSFGGFTFKRTNGSSVKKEEEVLTGNLQTLKIKVHEGY
LRDLMSQSR
EEFTMVGRRATAILRKATRRLIQLIVSGRDEQSIAEAIIVAMVFSQEDCMIKAVRGDLNF
RDLMSQSRT
...
DLMSQSRTR
LMSQSRTRE
and 9 other proteins
and 4376 other 9mers
Experimental validation of
Bioinformatics predictions
Peptide
Sequence
PB1591-599 VSDGGPNLY
HLA
Elispot assay1
Restriction KD (nM) + peptide - peptide
HLA-A1
6
18 ± 2
3±3
Elispot assay2
+ peptide - peptide
12 ± 4
1±1
NP44-52
HLA-A1
7
34 ± 5
4±1
13 ± 4
0±0
PB1166-174 FLKDVMESM
HLA-A2
51
74 ± 10
11 ± 6
140 ± 36
20 ± 7
PB141-49
HLA-A26
6
40 ± 3
20 ± 7
38 ± 5
24 ± 3
PB1540-548 GPATAQMAL
HLA-B7
6
7±2
2±1
13 ± 2
6±1
NP225-233
ILKGKFQTA
HLA-B8
664
9±4
1±1
19 ± 7
2±2
PA601-609
SVKEKDMTK
HLA-B8
NB
23 ± 6
1±1
119 ± 8
2±1
PB1349-357 ARLGKGYMF HLA-B27
246
10 ± 6
1±1
14 ± 4
1±1
NP383-391
SRYWAIRTR
HLA-B27
38
39 ± 6
1±1
40 ± 6
2±1
M1173-181
IRHENRMVL
HLA-B39
13
14 ± 3
3±1
84 ± 11
3±1
NP199-207
RGINDRNFW
HLA-B58
42
28 ± 5
1±1
15 ± 6
2±2
178
77 ± 20
3±2
91 ± 8
10 ± 3
88
15 ± 5
2±2
21 ± 2
2±0
CTELKLSDY
DTVNRTHQY
PB1347-355 KMARLGKGY HLA-B62
PB1566-574 TQIQTRRSF HLA-B62
Wang et al., 2006
Results are deposited in public databases
Peters B, et al. Immunogenetics. 2005 57:326-36, PLoS Biol. 2005 3:e91.
Human MHC:
~1000 variants
distributed over
12 types
Virus epitope:
up to 209
variants
Figure by Anne Mølgaard, peptide (KVDDTFYYV) used as vaccine by Snyder et al. J Virol 78, 7052-60 (2004).
Coverage of HLA alleles
Supertype Selected allele
A1
A*0101
A2
A*0201
A3
A*1101
A24
A*2401
A26 (new*)
A*2601
B7
B*0702
B8 (new*)
B*0801
B27
B*2705
B39(new*)
B*3901
B44
B*4001
B58
B*5801
B62
B*1501
Clustering in: O Lund et al., Immunogenetics. 2004 55:797-810
Polyvalent vaccines
• The equivalent of this in epitope based
vaccines is to select epitopes in a way that
that they together cover all strains.
Uneven coverage, Average coverage = 2
Epitope
Strain 1
Strain 2
Even coverage, Average coverage = 2
Strain 1
Strain 2
Processing
Proteasome specificity
• Low polymorphism
– Constitutive & Immunoproteasome
• Evolutionary conserved
• Stochastic and low
specificity
– Only 70-80% of the cleavage
sites are reproduced in
repeated experiments
Proteasome evolution (b1 unit)
Human (Hs) - Human
Drosophila (Dm) - Fly
Bos Taurus (Bota) - Cow
Oncorhynchus mykiss (Om) - Fish
…
Constitutive
Immuno- and Constitutive proteasome
specificity
Immuno
Constitutive
P1
P1’
...LVGPTPVNIIGRNMLTQL..
Predicting proteasomal
cleavage
• NetChop
– Neural network based method
• PaProc
– Weight matrix based method
• FragPredict
– Based on a statistical analysis of cleavagedetermining amino acid motifs present around
the scissile bond
• i.e. also weight matrix like
NetChop 3.0 Cterm (MHC
ligands) LDFVRFMGVMSSCNNPA
NetChop-3.0 C-term
– Trained on class I epitopes
– Most epitopes are generated by
the immuno proteasome
– Predicts the immuno
proteasome specificity
LVQEKYLEYRQVPDSDP
RTQDENPVVHFFKNIVT
TPLIPLTIFVGENTGVP
LVPVEPDKVEEATEGEN
YMLDLQPETTDLYCYEQ
PVESMETTMRSPVFTDN
ISEYRHYCYSLYGTTLE
AAVDAGMAMAGQSPVLR
QPKKVKRRLFETRELTD
LGEFYNQMMVKAGLNDD
GYGGRASDYKSAHKGLK
KTKDIVNGLRSVQTFAD
LVGFLLLKYRAREPVTK
SVDPKNYPKKKMEKRFV
SSSSTPLLYPSLALPAP
FLYGALLLAEGFYTTGA
NetChop20S-3.0
In vitro digest data from the constitutive proteasome
Toes et al., J.exp.med. 2001
Prediction performance
FP
TP
AP
AN
Sens
Aroc=0.8
Aroc=0.5
1 - spec
Predicting proteasomal
cleavage
NetChop20S-3.0
NetChop-3.0
• Relative poor predictive performance
–For MHC prediction CC~0.92 and AUC~0.95
Proteasome specificity
• NetChop is one of the best available cleavage method
– www.cbs.dtu.dk/services/NetChop-3.0
Transporter associated with
antigen processing (TAP)
What does TAP do?
TAP affinity prediction
• Transporter Associated with
antigen Processing
• Binds peptides 9-18 long
• Binding determined mostly
by N1-3 and C terminal amino
acids
Integration?
Integrating all three steps
(protesaomal cleavage, TAP
transport and MHC binding)
should lead to improved
identification of peptides
capable of eliciting CTL
responses
Identifying CTL epitopes
HLA affinity
1
2
3
4
5
6
7
8
9
...
EBN3_EBV
EBN3_EBV
EBN3_EBV
EBN3_EBV
EBN3_EBV
EBN3_EBV
EBN3_EBV
EBN3_EBV
EBN3_EBV
YQAYSSWMY
QSDETATSH
PVSPAVNQY
AYSSWMYSY
LAAGWPMGY
IVQSCNPRY
FLQRTDLSY
YTDHQTTPT
GTDVVQHQL
2.56
2.22
1.55
1.31
1.02
0.99
0.94
1.15
0.96
1.00
0.01
0.01
0.34
1.00
0.10
0.46
1.00
0.01
Proteasomal cleavage
0.03
0.28
0.97
0.99
0.97
0.97
0.99
0.01
0.02
0.34
0.88
0.01
0.02
0.22
0.50
0.02
0.42
0.03
0.99
0.04
0.22
0.01
0.01
0.05
0.82
0.02
0.99
0.02
0.83
0.21
0.75
0.18
0.01
0.07
0.04
1.00
0.01
0.51
1.00
0.94
0.01
0.01
0.01
0.01
0.02
0.75
0.30
0.02
0.92
0.06
0.01
0.63
0.02
0.46
TAP affinity
0.94
0.11
0.04
0.09
0.01
0.02
0.01
0.54
0.30
0.92 2.97 0 2.80
0.99 -0.80 0 2.28
1.00 2.63 0 1.78
1.00 3.28 1 1.58
1.00 3.01 0 1.27
0.93 3.19 0 1.24
0.96 2.79 0 1.18
0.14 -0.87 0 1.12
1.00 0.53 0 1.09
Large scale method validation
HIV A3 epitope predictions
Class II MHC binding
Human MHC II:
~1000 variants
• MHC class II binds
peptides in the class II
antigen presentation
pathway
• Binds peptides of length
9-18 (even whole
proteins can bind!)
• Binding cleft is open
• Binding core is 9 aa
Peptide:
up to 209
variants
Humoral
immunity
Cartoon by Eric Reits
Antibody - Antigen
interaction
Antigen
The antibody recognizes
structural properties of the
surface of the antigen
Fab
Epitope
Paratope
Antibody
Discontinuous B-cell epitopes
An example: An epitope
of the Outer Surface
Protein A from Borrelia
Burgdorferi (1OSP)
SLDEKNSVSVDLPGEM
KVLVSKEKNKDGKYDLI
ATVDKLELKGTSDKNN
GSGVLEGVKADKCKVK
LTISDDLGQTTLEVFKE
DGKTLVSKKVTSKDKS
STEEKFNEKGEVSEKIIT
RADGTRLEYTGIKSDGS
GKAKEVLKG
1OSP, Li et al. 1997
The DiscoTope web server
www.cbs.dtu.dk/services/DiscoTope
http://tools.immuneepitope.org/stools/discotope/discotope.do
Prediction servers at CBS for
Immunological features
• BepiPred » Linear B-cell epitopesDiscoTope »
Discontinuous B-cell epitopes
• NetChop » Proteasomal cleavages (MHC
ligands)NetCTL » Integrated class I antigen
presentation
• NetCTLpan Pan-specific integrated class I antigen
presentation
• NetMHC » Binding of peptides to MHC class I
allelesNetMHCII » Binding of peptides to MHC class II
alleles
• NetMHCIIpan » Pan-specific binding of peptides to
MHC class II HLA-DR alleles of known sequence
• NetMHCpan » Pan-specific binding of peptides to
MHC class I alleles of known sequence
• VDJsolver » Analysis of human immunoglobulin VDJ