The Nitrilases

1
The Nitrilases

• Versatile helix formers

2
Why study nitrilases?

• The glu,cys,lys catalytic triad offers versatile
  chemistry
• Use of enzymes in the manufacture of fine
  chemicals is attractive because of
  enantioselectivity.
• Use of enzymes in environmental remediation is
  attractive because organisms may not survive the
  toxins.
• Their metabolic role is not well characterised

3
Chemistry
Nitrilase - cyanide dihydratase - CynD - B.
pumilus, P.stutzeri
Cyanide hydratase - G. sorghi
Amidase G. pallidus
Carbamylase 1ERZ Agrobacterium sp.
4
The Superfamily
5
Relationship of solved structures to the
bacterial and plant nitrilases, CynD, and fungal
cyanide hydratrases
Plant nitrilases
Solved structures
CynD
Bacterial nitrilases
Fungal cyanide hydratases
6
Current uses

• Environmental remediation of cyanide
• Manufacture of acrylic acid
• Manufacture of specific enantiomers R-mandelic
  acid, R-4-cyano-3-hydroxybutyric acid,
  S-phenyllactic acid, (S)-ibuprofen,
  (S)-a-phenylglycine and (S)-naproxen.

7
Oligomerization

• All known nitrilase homologues are at least
  dimers
• There are four crystal structures of homologues
  two are dimers and two are tetramers with 222
  symmetry. Only one has known enzyme activity a
  carbamylase
• Our published work has shown three forms built
  around helical symmetry two are terminating
  spirals with 14 and 18 subunits respectively. The
  18mer changes to an open helix at pH5.4
• Most determinations of oligomeric state are based
  on gel-filtration and are unreliable but the
  numbers range from 2-18 subunits for all known
  nitrilases

8
Link between activity and oligomerization

• Nagasawa et al and two previous authors have
  shown that dimers are inactive until they
  oligomerize (in Rhodococci )
• We have shown that in case of B. pumilus an
  activity increase accompanies the pH dependent
  increase in size
• We have shown that destructive mutation of
  residues in the interfaces which associate to
  form helices abolishes activity
• Mueller et al 2006 archaeal nitrilase active as
  a dimer

9
What is it useful to know?(biotechnology
perspective)

• The configuration of the active site
• How is specificity / activity controlled
• The oligomeric configuration
• How is the oligomeric configuration controlled
• How the oligomeric configuration affects the
  active site configuration

10
What is it useful to know?(biotechnology
perspective)

• The configuration of the active site
• How is specificity / activity controlled
• The oligomeric configuration
• How is the oligomeric configuration controlled
• How the oligomeric configuration affects the
  active site configuration

11
Our Approach

• 3DEM mostly in negative stain
• Homology modelling
• Docking
• Mutation
• Species variation
• Determining crystal structures of carefully
  chosen mutants

12
The nitrilase fold
1erz D-amino acid carbamylase
1ems Nit domain of NitFhit
13
Sequence Alignment
The location of the active site
Insertions relative to the crystal homologues
14
Sequence Alignment
The components of the A surface
15
Our Structures
sorghi
pumilus
Bacillus pallidus amidase
Bacillus pumilus CynD pH 6
B. pumilus CynD pH 5.4
stutzeri
J1
Gloeocercospora sorghi cyanide hydratase
Pseudomonas stutzeri CynD pH 8
Rhodococcus rhodochrous J1 nitrilase pH 8
16
The spiral elongates by associating across the A
and C surfaces In the 14 subunit spiral
termination results from the E surface interaction
17
Homology model of a dimeric module of P.
stutzeri CynD
D
E1
C
A
E2
E2
C
E1
D
18
The dimers closest to the two fold axis have
nearly perfect 51 symmetry. The symmetry is
distorted towards the terminal dimers which move
closer to the spiral axis.
-7
-5
b
-3
d
-1
c
-6
2
a
-4
4
b
-2
6
d
1
c
a
3
5
7
19
Contacts a and b result in the terminal dimer
having an inwards tilt of 12 thus preventing the
addition of a further dimer.
-2
6
E1
a
E2
20
Contacts c and d are between helices NH2. The
contact area has a local pseudo-dyad axis.
-6
d
d
-4
glu 82
4
c
D surface
c
6
lys 86
21
Structural transition in B. pumilus nitrilase

• The transitions between pH 6 and pH 5.4 may
  involve the titration of a histidine.
• At pH 5.4 the nitrilase is a regular helix having
  9.4 subunits per turn ( for dimer model ?f
  -76.7, ?z1.58 nm )

pH 6
pH 8
Helical reconstructions were made with IHRSR
according to Egelman
pH 5.4
22
Unidirectional shadowing with tungsten enables
the upper surface to be visualised
23
Left handed 1-start helices and right handed
3-start helices seen by shadowing
Indexing based on 14 dimers in 3 turns i.e.
?f-77
B. pumilus pH 5.4
Margot Scheffer
Image made using the CM12, MIDILAB instrument at
ETH, Zurich
24
D surface
SPHVQRLLDAARDHN SLAIQKISEAAKRNE SEAVQKISAAARKNK SS
EMRRIRAAARDNQ
R. rhodochrous J1
B. pumilus
P. stutzeri
G. sorghi
Potential for two pairs of salt bridges in B.
pumilus, R. rhodochrous and G. sorghi Repulsion
in P. stutzeri - no long fibres
P. stutzeri
B. pumilus
25
The only histidines in B. pumilus that are not in
P. stutzeri.
The ATCC B. pumilus strain has no histidines in
the tail it does not form long helices
26
G. sorghi cyanide hydratase looks very similar to
B. pumilus CynD

Indexing based on 14 dimers in 3 turns i.e.
?f-77
Margot Scheffer
Image made using the CM12, MIDILAB instrument at
ETH, Zurich
27
Solving the structure with this symmetry gives a
solution in which the known handedness of the
dimer and the spiral cannot be reconciled!
28
Indexing based on 11 dimers in two turns i.e.
?f-65.45
Shadowing
Model
Cryo
29
Three dimensional reconstruction from shadowed
material using Iterative Helical Real Space
Reconstruction (IHRSR)
Inside of metal cast
Contained space
10.9 subunits per turn ( for dimer model ?f
-66 , ?z1.35 nm )
30
Cryo G. sorghi
Neg stain G. sorghi
Neg stain N. crassa
31
New interaction?
32
D surface mutations of G. sorghi cyanide hydratase
81
93
SEMRRIRAAARDN
82 84 85 87 91 92
33
The transformation of the nitrilase from R.
rhodochrous J1
We can make a construct which forms helices by
deleting C-terminal residues
Robert Ndoria Thuku
34
Fitting of dimer models of J1 nitrilase into the
3D map

• Rotation per dimer, ?? -73.6o
• Rise per dimer, ?z 15.8Å
• Pitch 77Å
• Dimers/turn 4.9
• Diameter 13 nm

D surface
97NHDRAADLLRQV86 86VQRLLDAARDHN97
C surface
90o
Robert Ndoria Thuku
35
Is it possible to explain the transition from
spirals to helices which occurs in B. pumilus
CynD at pH 5.4?
The pH 6 structure is possibly a hybrid between
a terminating 14mer and a terminating 18mer
E
90o
pH 6.0
pH 5.4
Recall that the histidines are located in the
C-terminal extension on the inside of the spiral
Hypothesis The charged histidines repel one
another disrupting the E surface interaction and
create space for an additional dimer which is
located by the D surface interactions.
36
Conclusion

• We probably know enough about the D surface and
  the C-terminal extension to control the
  oligomerisation of these enzymes
• We have a testable hypothesis for the pH
  dependent oligomerizarion of B. pumilus CynD
• It would be nice to use this knowledge to get
  atomic structures of the parts that remain
  unresolved i.e. the C surface insertions and
  the C-terminal domain

37
The amidase from G. pallidus is related to the
crystal structures
38
The Amidase has the extended C-terminal but not
the insertions
Nitrilase
Amidase
Solved structures
Extended C-terminal
39
Class which shows clear 3-fold symmetry
40
Amidase crystals from 1.2M Na citrate, 400mM
NaCl, 100mM Na acetate, pH 5.6
P4232 a130.4Å Resolution 1.72Å 1 polypeptide
chain per asymmetric unit Only 22370
unique reflections!
Laue group m-3m Systematic absences l g 2n along
0, 0, l
41
Some Eigenimages from 12698 amidase images
2
3
4?
6
2,4 and 6 fold eigenimages suggest D3 and not C3
symmetry!
42
(No Transcript)
43
(No Transcript)
44
(No Transcript)
45
(No Transcript)
46
(No Transcript)
47
(No Transcript)
48
(No Transcript)
49
(No Transcript)
50
(No Transcript)
51
(No Transcript)
52
What went in 22 identical subunit, with non
identical sidechains replaced by
alanine P4232 Restriction of resolution from 20
to 3.5 Ångstroms PHASER by Randy Read What came
out The conserved dimer interface Space to
accommodate the 65 residues per subunit at the
C-terminal Interpretable density for many of the
the sidechains
53
Michael Benedik Texas AM
Mohamed Jaffer
Muhammed Sayed
Margot Scheffer
Sean Karriem
Jeremy Woodward
Arvind Varsani
Vinod Agarkar
Robert Ndoria Thuku
Kyle Dent
Brandon Weber
Serah Kimani
54
The Sponsors
Past contributors Paul Meyers, Mark Berman,
Brendon Price, Dakshina Jandhyala, Paul Chang,
and Xing Zhang
Our thanks to
Helen Saibil, Dan Clare, Alan Roseman, Ed Egelman
and Andy Hoenger AND

• Carnegie Corporation of New York
• National Research Foundation
• Wellcome Trust
• Ford Foundation
• Bio/Chemtek
• University of Cape Town

55
Next Attraction

• First African Structural Biology Conference
• Macromolecular Structure, Health and
  Biotechnology in the Developing World
• 24 27 October, 2006
• The Wilderness
• 20 Leading international speakers
• http//sbio.uct.ac.za/conference
• Sponsored by ICGEB, DST, NRF, Carnegie
  Coroporation, Bruker and others

56
The Effect of Surface Mutations on Activity
57
P.stutzeri CynD is a 14 subunit homo-oligomeric
spiral
pH 8
pH 6
B. pumilis CynD apparently has 16 subunits but
we will return to this later