Applications of structural
biology and bioinformatics
Outline
2Applications of structural biology and bioinformatics
 Structural biology paradigm
 Applications of structural biology and bioinformatics
 Biological research
 Drug design
 Protein engineering
 Summary
 Final remarks on the course
A structural biology paradigm…
3Biological research
 Sequence-Structure-Function
 Challenges:
 Determine structure from sequence
 Determine function from sequence/3D structure
 Modify function (by modifying sequence or external molecules)
Sequence-Structure-Function
4A structural paradigm…
Pathway prediction
Sequence to
structure Protein-protein interactions
Docking of ligands
Active site descriptors
Structure to function
Applications of structural biology and bioinformatics
5Applications of structural biology and bioinformatics
 Biological research
 Drug design
 Protein engineering
Applications of structural biology and bioinformatics
6Applications of structural biology and bioinformatics
 Biological research
 Drug design
 Protein engineering
For example?
Biological research
7Biological research
 Drug resistance of HIV protease
Drug resistance of HIV protease
8Biological research
 HIV-1 protease
 Plays critical role in viral maturation for producing viral particles
 Aspartic protease with characteristic triad Asp-Thr-Gly
 Symmetric homodimer, 99 amino acids per monomer
 3 functionally important regions in the protease structure
 Active site cavity
 Flexible flaps
 Dimer interface
Drug resistance of HIV protease
9Biological research
 HIV-1 protease
 Plays critical role in viral maturation for producing viral particles
 Aspartic protease with characteristic triad Asp-Thr-Gly
 Symmetric homodimer, 99 amino acids per monomer
 3 functionally important regions in the protease structure
 Active site cavity
 Flexible flaps
 Dimer interface
 Flap opening/closing
is crucial for catalysis
By comparing 2 crystal structures
(PDBs: 1HXW and 1TW7)
Drug resistance of HIV protease
10Biological research
 HIV-1 protease
 Plays critical role in viral maturation for producing viral particles
 Aspartic protease with characteristic triad Asp-Thr-Gly
 Symmetric homodimer, 99 amino acids per monomer
 3 functionally important regions in the protease structure
 Active site cavity
 Flexible flaps
 Dimer interface
 Flap opening/closing
is crucial for catalysis
By molecular dynamics
Drug resistance of HIV protease
11Biological research
 Protease inhibitors (PIs)
 Introduced into clinical practice in 1995 known as antiretrovirals
 Competitive inhibitors, designed to mimic the transition state
of the substrate-enzyme complex
 Binding affinity in nanomolar to picomolar range (very high)
 Currently ~ 10 different inhibitors available
 Darunavir, indinavir
 Fosamprenavir, saquinavir
 Tipranavir, ritonavir
 Amprenavir, lopinavir
 Nelfinavir, atazanavir
Drug resistance of HIV protease
12Biological research
 Drug resistance to PIs
 Drug resistance emerged against all clinically available PIs
 Resistant mutations in HIV-1 protease reduced susceptibility to
inhibitors while maintaining protease function
 Important factors in development of drug resistance
 Rapid mutation
 High rate of viral replication (108-109 virions/day)
 High error rate of HIV reverse transcriptase (≈1 in 10,000 bases)
 Long term exposure to drugs
Drug resistance of HIV protease
13Biological research
 Molecular mechanisms of drug resistance
 Deduced from comparison of structures and activities of
native and mutant proteases
Drug resistance of HIV protease
14Biological research
Major resistance
Minor resistance
Drug resistance of HIV protease
15Biological research
 Molecular mechanisms of drug resistance
 Deduced from comparison of structures and activities of
native and mutant proteases
 Several distinct mechanisms
 Active site mutations
 Mutations at dimer interface
 Mutations at distal positions
Drug resistance of HIV protease
16Biological research
 Active site mutations
 Mutation of single residue in the active site cavity eliminating direct
interactions with inhibitor
 Mutations are very conservative – ex: substitutions of hydrophobic
amino acids
Ile84Val Ile50Val
Drug resistance of HIV protease
17Biological research
 Mutations at dimer interface
 For example: Phe53Leu
 Wider separation of the two flaps
 Reduced stabilization of bound inhibitor
Phe53Leu
Drug resistance of HIV protease
18Biological research
 Mutations at distal positions
 For example: Leu90Met
 Promoted contacts with catalytic Asp25
 Reduced interaction with inhibitor
Leu90Met
Drug resistance of HIV protease
19Biological research
 Novel PIs for resistant HIV-1 protease
 Inhibitors fitting within envelope formed by bound substrate
 Inhibitors binding flaps or the dimer interface
 Inhibitors targeting main chain and conserved regions of active site
 Inhibitors targeting the gating mechanism
Drug resistance of HIV protease
20Biological research
 Novel PIs for resistant HIV-1 protease
 Inhibitors targeting the gating mechanism
o Stabilize the closed state
o Stabilize the open state
o Mixed interactions (AS and gating elements)
Closed conformation
(PDB ID: 1HVR)
Open conformation
(PDB ID: 3BC4)
Drug design
21Drug design
 Virtual screening of inhibitors of endonuclease MUS81
 Selective inhibitor of LTA4H
Drug design
22Drug design
 Methods of drug discovery
 Ligand-based
 Knowledge of active ligands
 Search for similar ones
Drug design
23Drug design
 Methods of drug discovery
 Ligand-based
 Knowledge of active ligands
 Search for similar ones
 Structure-based
 Knowledge of receptor
 Search for strong binders
 Molecular docking
Drug design
24Drug design
 Methods of drug discovery
 Ligand-based
 Knowledge of active ligands
 Search for similar ones
 Structure-based
 Knowledge of receptor
 Search for strong binders
 Molecular docking
 High-throughput screening (HTS)
 Large library of compounds
 Experimental or in silico screening
Virtual screening
25Drug design
 Structure-based VS
 Receptor-ligand docking
 Often combined with HTS
 Followed by hit optimization
 Many success stories
 Speed-up drug discovery
 Lower the costs
Virtual screening
26Drug design
Inhibitors of endonuclease MUS81
27Drug design
 DNA structure-specific endonuclease MUS81
 Endonucleases are involved in DNA reparation
 Help maintaining genomic stability
 Cancer cells often have higher replication rates
 MUS81 is a target for anti-cancer
drug development
Inhibitors of endonuclease MUS81
28Drug design
 High-throughput screening (HTS)
 Robotic platform at Center of Chemical Genetics, ASCR, Prague
 About 23,000 compounds experimentally tested
 Identified 1 effective inhibitor: IC50 = 50 μM
Inhibitors of endonuclease MUS81
29Drug design
 Structure-based VS
 Molecular docking + rescoring of binding interaction
 Binding of more than 140,000 compounds predicted
 Experimental verification on 19 potential inhibitors
 Identified 6 effective inhibitors with IC50 ≤ 50 μM
 Best inhibitor: IC50 = 5 μM
Inhibitors of endonuclease MUS81
30Drug design
 Comparison
HTS VS
Equipment (Kč) 50,000,000 500,000
Testing
Computational - 140,000
Experimental 23,000 19
Costs (Kč) 2,000,000 40,000
Time Weeks Days
Results
# of inhibitors 1 6
Best: IC50 (μM) 50 5
Selective inhibitor of LTA4H
31Drug design
 Leukotriene A4 hydrolase/aminopeptidase (LTA4H)
 Involved in chronic inflammatory and immunological diseases
 Bifunctional metalloenzyme
 Catalyzes hydrolysis of the leukotriene A4 (LTA4) into the
pro-inflammatory mediator LTB4
 Also hydrolyses the pro-inflammatory Pro-Gly-Pro
 Distinct but overlapping
binding sites and 2 tunnels
Selective inhibitor of LTA4H
32Drug design
 Leukotriene A4 hydrolase/aminopeptidase (LTA4H)
 Structural studies (crystallography) with a tripeptide analogue
revealed the aminopeptidase mechanism
Selective inhibitor of LTA4H
33Drug design
 Leukotriene A4 hydrolase/aminopeptidase (LTA4H)
 Structural studies (crystallography) with a tripeptide analogue
revealed the aminopeptidase mechanism
 This knowledge allowed designing a selective inhibitor that blocks
the hydrolysis of LTA4 but NOT the hydrolysis of Pro-Gly-Pro
 New promising lead compound
against chronic inflammation
Protein engineering
34Protein engineering
 Stabilization of dehalogenase
 Dehalogenase activity
 Lipase enantioselectivity
 De novo design of a Diels-Alderase
35Protein engineering
Enzymes: practical applications?
What are the
requirements?
36Protein engineering
Enzymes: practical applications?
 Ability to catalyse a desirable reaction
 Stable under operating conditions
 Soluble expression
 Ability to catalyse a desirable reaction
 Stable under operating conditions
 Soluble expression
 Improvement of activity or selectivity
 Robust stabilization of proteins
 Design of more soluble proteins
37Protein engineering
Enzymes: practical applications?
Protein
engineering
process
Different approaches
38Protein engineering
Stabilization of dehalogenase
39Protein engineering
 Dehalogenase DhaA
 Bacterial origin
 Hydrolytic cleavage of C-X bond
 Multiple biotechnological applications
+ Cl-
Stabilization of dehalogenase
40Protein engineering
 Dehalogenase DhaA
 Melting temperature Tm = 49 °C
 Unstable at high temperatures
 Activity half live at 60 °C τ1/2 ~ 5 min
Stabilization of dehalogenase
41Protein engineering
 Gene Site Saturation Mutagenesis
 Joint project of Diversa and DOW Chemical
 All 19 possible mutations at 315 positions tested experimentally
 → 120,000 measurements
 10 single-point mutants more stable
 Cumulative mutant:
Tm = 67 °C (18 °C ↑)
τ1/2 = 36 h (ca. 36 h ↑)
Stabilization of dehalogenase
42Protein engineering
 Rational design
 FIREPROT method
 Structure and sequence analyses
 ~5,500 possible mutants
Stabilization of dehalogenase
43Protein engineering
 Rational design
 FIREPROT method
Stabilization of dehalogenase
44Protein engineering
 Rational design
 FIREPROT method
Stabilization of dehalogenase
45Protein engineering
 Rational design
 FIREPROT method
Stabilization of dehalogenase
46Protein engineering
 Rational design
 FIREPROT method
Stabilization of dehalogenase
47Protein engineering
 Rational design
 FIREPROT method
M1L N10D .. .. .. .. M28L
Stabilization of dehalogenase
48Protein engineering
 Rational design
 FIREPROT method
Stabilization of dehalogenase
49Protein engineering
 Rational design
 FIREPROT method
1Combined mutant
Stabilization of dehalogenase
50Protein engineering
 Rational design
 FIREPROT method
 Structure and sequence analyses
 ~5,500 mutants predicted
 Experimental verification on
5 multiple-point mutants
 3 mutants more stable
 Best mutant (combined):
Tm = 74 °C (25 °C ↑)
τ1/2 = 72 h (ca. 72 h↑)
1Combined mutant
Stabilization of dehalogenase
51Protein engineering
 Comparison
GSSM Rational design
Equipment (Kč) 20,000,000 500,000
Testing
Computational - 5,500
Experimental 120,000 5
Costs (Kč) 1,000,000 80,000
Time Months Weeks
Results
# of stable mutants 11 3
Best: ΔTm (°C) 18 25
τ1/2 (h) 36 72
Dehalogenase activity
52Protein engineering
 TCP: toxic persistent pollutant
from industrial sources
 DhaA dehalogenase (poor catalyst)
 DhaA31: 5 mutations narrowed the access tunnels
 32-fold higher catalytic rate (kcat); release of product became
limiting step
DhaA31DhaAWT
+ ClTCP
DCP
DhaA
Dehalogenase activity
53Protein engineering
 Catalytic cycle: enzymes with buried active site
Dehalogenase activity
54Protein engineering
 Substrate binding: MD simulations
Distance[Å]
Time [ns]
- Tunnels need to open for binding
- Fast process for both DhaA31 and DhaAWT
- Potential substrate inhibition
Dehalogenase activity
55Protein engineering
 Reactive binding: MD simulations
Chemical step:
what do we
need?
Dehalogenase activity
56Protein engineering
 Reactive binding: MD simulations
- DhaA31 with 5.5-fold higher NAC rates
- C176Y, V245F increased interactions
- C176Y induced correct orientation of TCP
Asp 106
Asn 41
Trp 107
Near-attack
conformation (NAC)
TCP
Dehalogenase activity
57Protein engineering
 Chemical step: QM/MM calculations
- Limiting step in DhaAWT
Reaction coordinate
(dO-C)
Dehalogenase activity
58Protein engineering
 Chemical step: QM/MM calculations
- Limiting step in DhaAWT
- DhaA31 with lower G‡ by 1.6 kcal/mol
 ~14-fold higher reactivity
E
G‡
Reaction coordinate
(dO-C)
TS
Dehalogenase activity
59Protein engineering
 Product release: MD simulations
Tyr 176
Phe 149
Phe 152 Phe 168
- Limiting step in DhaA31
- DhaA31 with slower release rates
- F149, F152, F168, Y176 prevent release
Gbind[kcal/mol]
Residue
Dehalogenase activity
60Protein engineering
 Conclusions
- Catalytic improvements explained
- Key mutations identified
- New hot-spots for mutagenesis
 Lipase (EC 3.1.1.3, bacterial enzyme)
 Triacylglycerol + H2O  diacylglycerol + carboxylic acid
 Versatile biocatalysts: catalyze hydrolysis of carboxylic esters,
esterification, transesterification, etc.
 Many industrial applications
 Food, detergent, pharmaceutical, leather, textile, cosmetic,
paper industries
Lipase enantioselectivity
61Protein engineering
Triglycerides:
main constituent
of body fat
 Lipase (EC 3.1.1.3, bacterial enzyme)
 Triacylglycerol + H2O  diacylglycerol + carboxylic acid
 Versatile biocatalysts: catalyze hydrolysis of carboxylic esters,
esterification, transesterification, etc.
 Many industrial applications
 Food, detergent, pharmaceutical, leather, textile, cosmetic,
paper industries
 Used to resolve racemic mixtures
Lipase enantioselectivity
62Protein engineering
What?
 Lipase (EC 3.1.1.3, bacterial enzyme)
 Triacylglycerol + H2O  diacylglycerol + carboxylic acid
 Versatile biocatalysts: catalyze hydrolysis of carboxylic esters,
esterification, transesterification, etc.
 Many industrial applications
 Food, detergent, pharmaceutical, leather, textile, cosmetic,
paper industries
 Used to resolve racemic mixtures
Lipase enantioselectivity
63Protein engineering
(R-enantiomer)
Lipase enantioselectivity
64Protein engineering
 Lipase (EC 3.1.1.3, bacterial enzyme)
 Molecular modeling suggested residues in the tunnel controlling
substrate access are key to enantioselectivity
 Saturated mutagenesis at 3 positions
 Mutants with higher E-value and conversion %
Lipase enantioselectivity
65Protein engineering
 Lipase (EC 3.1.1.3, bacterial enzyme)
 Molecular modeling suggested residues in the tunnel controlling
substrate access are key to enantioselectivity
 Saturated mutagenesis at 3 positions
 Mutants with higher E-value and conversion %
Lipase enantioselectivity
66Protein engineering
 Lipase (EC 3.1.1.3, bacterial enzyme)
 Molecular dynamics with substrates
Time
Lipase enantioselectivity
67Protein engineering
 Lipase (EC 3.1.1.3, bacterial enzyme)
 Molecular dynamics with substrates
 Steric changes on either side
of the active site favor reactive
binding of one enantiomer
 Combined mutations favor one
enantiomer and disfavor the other
Time
De novo design of a Diels-Alderase
68Protein engineering
 Non-existing Diels-Alderase
 Goal: design biocatalyst for intermolecular Diels-Alder reaction
 very specific geometric and electronic requirements  theozymes
Diels–Alder cycloaddition
Transition state
prediction
De novo design of a Diels-Alderase
69Protein engineering
 Non-existing Diels-Alderase
 Goal: design biocatalyst for intermolecular Diels-Alder reaction
 very specific geometric and electronic requirements  theozymes
 Design: computational match with protein scaffolds and refinement
 Mutagenesis: site-directed to design active site
 Evaluated library: < 100
 Results: creation of functional & stereoselective Diels-Alderase
Transition state
prediction
Theozymes
ensemble
Scaffold
fitting
Design
optimization
Summary
 Structural biology methods are important tools to:
 Explain biological phenomena
 Increase efficiency of drug discovery
 Successfully engineer proteins for biotechnological applications
 Often produce better results than experimental brute-force
 Can reduce costs and save time
Summary 70
Summary
 Structural biology methods are important tools to:
 Explain biological phenomena
 Increase efficiency of drug discovery
 Successfully engineer proteins for biotechnological applications
 Often produce better results than experimental brute-force
 Can reduce costs and save time
 This lesson will not be on the exam!
Summary 71
References
 Congreve, M. et al. (2005) Structural biology and drug discovery. Drug Discov
Today 10: 895-907.
 Lee, D. et al. (2007) Predicting protein function from sequence and structure.
Nat Rev Mol Cell Biol. 8: 995-1005
 Lutz, S. (2010) Beyond directed evolution — semi-rational protein engineering
and design. Curr Opinion Biotechnol 21: 734–743.
 Weber, I. T. & Agniswamy, J. (2009) HIV-1 protease: structural perspectives on
drug resistance. Viruses 1: 1110-1136.
 Marques, S.M. et al. (2017) Catalytic cycle of haloalkane dehalogenases toward
unnatural substrates. J Chem Inf Model. 57: 1970-1989.
 Jessop, T.C. et al. (2009) Lead optimization and structure-based design of potent
and bioavailable deoxycytidine kinase inhibitors. Bioorganic & Medicinal
Chemistry Letters 19 6784–6787
References 72
Final remarks: evaluation
 Teachers’ evaluation
 Evaluation Survey – PLEASE respond!
Final remarks 73
Final remarks: evaluation
 Exam 1 h, 3 dates
 17 Dec. 2024, 10:00 (location: B11/333)
 7 Jan 2025, 10:00 (location: B11/333)
 28 Jan. 2025, 10:00 (location: B11/333)
 Multiple-choice exam
 25 questions
 10 points out of 25 needed to pass
 Multiple correct answers possible
 Only topics with the sign on the slides will be asked
 Teachers are available for questions. Contact me!
Final remarks 74
Final remarks: evaluation
 Questions – example 1
Final remarks 75
Choose the true statements about van der Waals interactions.
1. These are long-range interactions
2. Interaction occurs between any types of atoms
3. These interactions play a role only with charged amino acid
residues
4. These are short-range interactions
5. These interactions are entropic in nature
Final remarks: evaluation
 Questions – example 1
Final remarks 76
Choose the true statements about van der Waals interactions.
1. These are long-range interactions
2. Interaction occurs between any types of atoms
3. These interactions play a role only with charged amino acid
residues
4. These are short-range interactions
5. These interactions are entropic in nature
Points:
-1/3
+1/2
-1/3
+1/2
-1/3
Final remarks: evaluation
 Questions – example 2
Final remarks 77
Choose the true statements about homology modeling.
A) It is based on the principle that sequences are much more
conserved than 3D structures during evolution
B) The structural model of the target protein is predicted based
on the known experimental 3D structure of a related protein
C) A necessary condition for homology modeling is the
existence of a suitable template
D)Homology modeling generally provides less accurate results
than ab initio predictions
Final remarks: evaluation
 Questions – example 2
Final remarks 78
Choose the true statements about homology modeling.
A) It is based on the principle that sequences are much more
conserved than 3D structures during evolution
B) The structural model of the target protein is predicted based
on the known experimental 3D structure of a related protein
C) A necessary condition for homology modeling is the
existence of a suitable template
D)Homology modeling generally provides less accurate results
than ab initio predictions
Points:
-1/2
+1/2
+1/2
-1/2
Final remarks: evaluation
 Bring only one pen and ID card
Final remarks 79