(627g) Modeling Ire1p Regulation and Activation in the Yeast Upr

Saccharomyces cervisiae, or bakers' yeast, are
utilized in the biotechnology industry to express, fold, and assemble foreign
protein therapeutics [1].  Simply modifying yeast to express high levels
of heterologous protein does not necessarily maximize production and secretion,
as this action triggers the Unfolded Protein Response (UPR), which eliminates
quantities of desired protein through promoted ER-Associated Degradation
(ERAD).  The yeast UPR may even retard transcription and/or translation of
the desired protein, thus decreasing yields.  Consequently, a combined
approach of mathematical modeling and experiments has been employed in an attempt
to obtain a thorough understanding of the yeast UPR, which will be utilized to
forward engineer the system to maximize foreign protein therapeutic production.

The yeast UPR has been traditionally modeled as a negative
feedback loop in which low levels of the chaperone BiP (Binding Protein)
relative to unfolded protein (UP) in the ER signal the need for increased
production of UPR component proteins--including BiP--that help cells cope with
these high ER UP levels.  The UPR signal is transduced from the ER to the
nucleus and cytoplasm by the ER membrane-spanning endonuclease Ire1p, which
enhances translation of Hac1p, a transcription factor that upregulates BiP and
other UPR-related gene expression.  Traditional modeling identifies BiP as
the primary regulator of Ire1p:  BiP typically binds Ire1p but is
sequestered exclusively by UP when UP is in excess;  unbound Ire1p is then
free to dimerize, trans-autophosphorylate, and transduce the UPR signal
across the nuclear ER membrane [2,3].  However, recent experimental
evidence suggests BiP may simply serve as an Ire1p adjustor, and that UP
directly regulates Ire1p activation [4,5].

This work sought to identify and define fundamental
differences, using mathematical modeling and systems analysis tools, between
the traditional and newer Ire1p activation models where BiP serves as the
primary activation regulator, UP does alone, and UP does combined with
modulation by BiP.  These three activation models are the BiP-Ire1p (BI),
UP-Ire1p (UPI), and BiP-modulated (BM) models, respectively.  With the
differences identified, they could be compared against existing experimental
data, or new experiments could be designed to select the candidate(s) that best
represent the biological system.

A mechanistic, deterministic mathematical model of the yeast
UPR has been developed and implemented in the Matlab Simulink environment using
the ode15s solver.  This model contained 32 states that described the
interactions between the major UPR components--UP, Hac1p, and BiP--as well as
two critical UPR inducers--heterologous scFv and DTT.  The three Ire1p
activation regulation models were substituted into this larger UPR framework,
and their ability to reproduce currently available UPR data was evaluated.

All three models were similarly capable of reproducing this
data, so the systems analysis tool, sensitivity analysis, was employed to
comprehensively scan the models for less-apparent discrimination criteria. 
This analysis was performed using the BioSens software package for BioSpice, which
runs .bsn versions of the model in DASPK, and found that the models responded
significantly different to changes in BiP-scFv binding and UP-ER entry rates,
which are practically alterable experimentally.  The BI UPR was positively and
UPI/BM UPRs were negatively correlated to changes in the BiP-scFv binding rate,
which could be manipulated in vivo by mutating BiP-binding sites on the scFv. 
The UPI/BM UPRs had a stronger correlation to the UP-ER entry rate than the BI
UPR.  Further discrimination between the UPI and BM models comes from the
mechanistic knowledge that BiP binds Ire1p when the UPR is inactive and
releases it when the UPR is activated [4].

References

[1]        A. S. Robinson
and D. A. Lauffenburger.  Model for ER Chaperone Dynamics and Secretory
Protein Interactions.  AIChE Journal, 42 (5):  1443-1453, 1996.

[2]       
A. A. Welihinda and R. J. Kaufman.  The unfolded protein response
pathway in Saccharomyces cerevisiae. Oligomerization and trans-phosphorylation
of Ire1p (Ern1p) are required for kinase activation.  J.
Biol. Chem.  271 (30):  18181-18187, 1996.

[3]       
K. Okamura, Y. Kimata, H. Higashio, A. Tsuru, and K. Kohno.  Dissociation
of Kar2p/BiP from an ER sensory molecule, Ire1p, triggers the unfolded protein
response in yeast.  Biochem. Biophys. Res. Commun.  279 (2): 
445-450, 2000.

[4]       
Y. Kimata, D. Oikawa, Y. Shimizu, Y. Ishiwata-Kimata, and K. Kohno.  A
role for BiP as an adjustor for the endoplasmic reticulum stress-sensing
protein Ire1.  J. Biol. Chem., 167 (3):  445-456, 2004.

[5]        J. J. Credle,
J. S. Finer-Moore, F. R. Papa, R. M. Stroud, and P. Walter.  On the mechanism of sensing unfolded protein in the
endoplasmic reticulum.  PNAS 102 (52):  18773-18784, 2005.