PI Meeting University of Maryland April 2011 000 p 1 001 101 011 p 3 p 2 010 100 111 110 A model for T cell differentiation Natasa Miskov-Zivanov University of Pittsburgh
Acknowledgements 2 Faeder Lab: Department of Computational and Systems Biology, School of Medicine, University of Pittsburgh John Sekar, James Faeder Morel Lab: Department of Immunology, School of Medicine, University of Pittsburgh Michael Turner, Penelope Morel Clarke Lab: Computer Science Department, Carnegie Mellon University Paolo Zuliani, Haijun Gong, Qinsi Wang, Edmund Clarke PI Meeting, April 2011
Timeline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tatistical model checker 1$!*/ New )?A K#? -L "0#8 % mode 1&!&J "&34 l Model !67 Trace statistics Simulations :&;)#- )?O $!002 &1#28&1#- New 1,02 )%39 $30- $30N experiments More 4(8!& 4(8 $ 9 51, trace 3)0 =40 :&;)#2 s? 1$!*/ 1&!&J (CA =I@ 4(8!& 4(8 $ 9 !"82 (CM5/ %&'() 4(8!& !"82 1&!&J 4(8!& 1$!*/ 1&!&J .<8-) # 4(8!& !"82 4(8 $ 9 (CM5/ 1&!&J !"#$* ! + + .<8- 4(8!& !"82 4(8 $ 9 !"#$ Kickoff NSF Meeting PI Review PI Meeting October 2009 March 2010 October 2010 April 2011 PI Meeting, April 2011
Today’s talk 4 System Methodology Antigen presenting Model design cell (APC) Model elements Influence sets Experiments Naïve (Interaction map) Expert T cell knowledge Set of discrete values Literature for each element Influence table Circuit Model design Helper T Regulatory T analysis methods (Th) cells (Treg) cells Model rules Model simulations PI Meeting, April 2011
Origins of 5 Regulatory T cells (Treg) Treg cells mediate antigen- specific suppression of T cell activation Play a key role in maintaining tolerance Naïve T cells can be converted into Treg cells in the periphery High therapeutic potential PI Meeting, April 2011
Role in cancer 6 Antigen presenting cell (APC) Tumor cell Naïve T cell Tumor secreted cytokines ( e.g ., TGF β ) Helper T Regulatory T (Th) cells (Treg) cells Release Release cytokines that cytokines stimulate the that inhibit immune the immune response response PI Meeting, April 2011
Determinants of 7 differentiation Determine whether known mechanisms are sufficient to explain experimental observations Foxp3 transcription factor is critical for Treg function Huehn et al. Nat. Rev. Immunol. 9 , 83-89 (2009) PI Meeting, April 2011
Challenges for Modeling 8 Large number of components and interactions Rapidly evolving list of important components and interactions structural uncertainty in the model Involvement of multiple processes signaling gene regulation protein expression (cell division) PI Meeting, April 2011
Network model 9 ?D7EF?7EC@ Receptors: !"# E@GEHE7EC@ 7>?@ABCD?7EC@ $%# #*+, T cell receptor (TCR) .<8-K 3P &'( ! Co-stimulation through CD28 # ! " #*-+ &'( ! ) .<8-) &#) IL-2 receptor (IL-2R) Model elements TGF β receptor =!0/ "."/ "*02 "./0 (TGF β R) 1$!*/ )?A K#? -L "0#8 % 1&!&J "&34 Transcription Influence sets !67 (Interaction map) factors: :&;)#- &1#28&1#- )?O $!002 AP-1, NFAT, NF κ B, 1,02 )%39 SMAD3, STAT5 $30- $30N 4(8!& 4(8 $ 9 51, 3)0 =40 :&;)#2 Genes: IL-2, CD25, Foxp3 1$!*/ 1&!&J (CA =I@ 4(8!& 4(8 $ 9 !"82 (CM5/ Other important 4(8!& 1$!*/ %&'() 4(8!& !"82 1&!&J 1&!&J elements: .<8-) # 4(8!& !"82 4(8 $ 9 !"#$* ! + + PTEN, PI3K, PIP3, PDK1, (CM5/ 1&!&J Akt, mTORC1, mTORC2, .<8- 4(8!& !"82 4(8 $ 9 !"#$ TSC1-TSC2, Rheb, S6K1, pS6 PI Meeting, April 2011
Influence sets 10 Element Influence set Element Influence set PI3K TCR, CD28, IL-2, IL-2R AP-1 Fos, Jun Akt PDK1, mTORC2 ERK Ras Model elements mTORC1 Rheb, PKC- θ JNK Ras mTORC2 PI3K, S6K1 Fos ERK Influence sets (Interaction map) Foxp3 NFAT, AP-1, STAT5, Smad3 Jun JNK IL-2 NFAT, AP-1, NF κ B, Foxp3 NFAT Ca CD25 NFAT, AP-1, NF κ B, STAT5, Ca TCR Foxp3 STAT5 IL-2, IL-2R PDK1 PIP3 NF κ B PKC- θ , Akt TSC1-TSC2 Akt Smad3 TGF β , Akt, mTORC1 Rheb TSC1-TSC2 PIP3 PI3K, PTEN S6K1 mTORC1 Ras TCR, CD28, IL-2, IL-2R pS6 S6K1 PI Meeting, April 2011
Circuit design: Variables 11 Number of values for variables Example: three levels for modeling TCR necessary No antigen Low antigen dose Model elements High antigen dose Influence sets (Interaction map) Set of discrete values for each element PI Meeting, April 2011
Circuit design: Variables 12 Number of values for variables Example: three levels for modeling TCR necessary No antigen (TCR_LOW = 0, TCR_HIGH = 0) Low antigen dose (TCR_LOW = 1, TCR_HIGH = 0) Model elements High antigen dose (TCR_LOW = 0, TCR_HIGH = 1) encoded with two Boolean variables Influence sets (Interaction map) Set of discrete values for each element PI Meeting, April 2011
Circuit design: Variables 13 Number of values for variables Example: three levels for modeling TCR necessary No antigen (TCR_LOW = 0, TCR_HIGH = 0) Low antigen dose (TCR_LOW = 1, TCR_HIGH = 0) Model elements High antigen dose (TCR_LOW = 0, TCR_HIGH = 1) encoded with two Boolean variables Influence sets (Interaction map) Example: three levels for modeling PI3K necessary Low and high level of PI3K have different impact Set of discrete values for each on mTORC2 element PI Meeting, April 2011
Low Antigen Dose Trajectory 14 ;4K8KGEPGKBQFQB ?D7EF?7EC@ !"# ;4K8KBCRKBQFQB E@GEHE7EC@ ;(( 7>?@ABCD?7EC@ $%# #*+, .<8-K 3S &'( ! Trajectory Summary # ! " #*-+ &'( ! ) .<8-) &#) TCR =!0/ "."/ "*02 "./0 PI3K 1$!*/ )?A PTEN K#? -L "0#8 % "&34 1&!&J PIP3 !67 AKT )?O $!002 :&;)#- &1#28&1#- MTORC1 1,02 $30- $30N S6K1 )%39 4(8!& 4(8 $ 9 MTORC2 51, 3)0 =40 :&;)#2 STAT5 IL-2 (CA =I@ CD25 4(8 $ 9 1&!&J 4(8!& 1$!*/ FOXP3 !"82 (CM5/ 4(8!& 1$!*/ %&'() 4(8!& !"82 1&!&J 1&!&J value = ON_HIGH value = ON_ LOW 4(8!& !"82 4(8 $ 9 !"#$* ! + + (CM5/ 1&!&J value = OFF 4(8!& !"82 4(8 $ 9 !"#$ PI Meeting, April 2011
High Antigen Dose Trajectory 15 >8101IGQI1DRHRD AF;GHA;GEB !"# >8101DES1DRHRD GBIGJG;GEB >(( ;@ABCDEFA;GEB $%# #*+, ./0-1 23 &'( ! Trajectory Summary # ! " #*-+ &'( ! ) ./0-) &#) TCR ?!54 "."4 "*57 ".45 PI3K 6$!*4 PTEN )AC 1#A -M "5#0 % "&28 6&!&L PIP3 !:; AKT )AP $!557 =&>)#- &6#70&6#- MTORC1 S6K1 6,57 $25- $25O )%2< 8(0!& 8(0 $ < MTORC2 96, STAT5 2)5 ?85 =&>)#7 IL-2 CD25 (EC ?KB 8(0 $ < 6&!&L 8(0!& 6$!*4 FOXP3 !"07 (EN94 8(0!& 6$!*4 %&'() 8(0!& !"07 6&!&L 6&!&L value = ON_HIGH value = ON_ LOW 8(0!& !"07 8(0 $ < !"#$* ! + + (EN94 6&!&L value = OFF 8(0!& !"07 8(0 $ < !"#$ PI Meeting, April 2011
Circuit design: Influence tables 16 Example 1: Example 2: 2-level mTORC1 3-level PI3K, 2-level mTORC2 Rheb PI3K 0 1 0 1 2 PKC- Θ S6K1 0 0 1 0 0 1 1 1 0 1 1 0 0 1 Example 3: 3-level Foxp3 STAT5,mTOR 00 01 02 10 11 12 20 21 22 NFAT, Smad3 00 0 0 0 0 1 2 0 1 2 01 0 0 0 0 0 1 0 1 or 0 1 02 0 0 0 0 0 0 0 0 0 10 0 1 2 1 2 2 1 or 2 2 2 11 0 0 1 0 1 1 0 or 1 1 1 12 0 0 0 0 0 0 0 1 or 0 1 20 1 2 2 2 2 2 2 2 2 21 0 1 1 1 1 1 1 1 2 22 0 0 0 0 0 0 0 1 1 * : Rule 1, * : Rule 2 PI Meeting, April 2011
Example 1: 2-level mTORC1 17 Rheb 0 1 Model elements PKC- Θ mTORC1’ = Rheb and PKC- θ 0 0 0 1 0 1 ‘and’ rule means both are Influence sets necessary for activation (Interaction map) Rheb 0 1 Set of discrete PKC- Θ mTORC1’ = Rheb values for each 0 0 1 element 1 0 1 Influence tables Rheb 0 1 PKC- Θ mTORC1’ = Rheb or PKC- θ 0 0 1 ‘or’ rule means either one Model rules 1 1 1 is sufficient for activation PI Meeting, April 2011
Recommend
More recommend