We are interested in the fundamental question: How
does a protein's primary sequence specify its three dimensional
structure? In addition, we are investigating the mechanisms
by which proteins achieve the exquisite specificity and efficiency
that are characteristic of protein-ligand interactions and
enzymatic catalysis. Our research focuses upon small proteins
that are amenable to study by a variety of biophysical, biochemical
and molecular biological techniques.
Projects
 To
view PDFs, you must install
Adobe Acrobat Reader.
Scanned PDFs are 300 dpi text/line art with color or gray-scale
images attached as necessary. We recommend Acrobat or Acrobat
Reader 5.0 or above and printing with the box next to "Expand
small pages to paper size" unchecked.
Designing
protein metal binding sites and allosteric control of conformation
We have introduced novel metal-binding sites into several
proteins, including a designed four-helix bundle protein,
α4, the B1 domain of IgG-binding protein G, and FK506
Binding Protein (FKBP). The metal-site designs are for both
structural and catalytic tetrahedral Zn(II) sites. The structural
sites enhance the stability of the proteins, whereas the catalytic
sites aim to exploit the powerful nucleophilic activity of
Zn(II)-bound water and to mimic natural enzymes such as carbonic
anhydrase and carboxypeptidase. These studies allow us to
delineate the features that are important for protein-Zn(II)
interactions and have the potential to generate proteins with
novel catalytic activities.
|
An example of a designed structural site: a
tetrahedral His3Cys site in the B1 domain of IgG binding
protein G. |
 |
Leading References
- Regan L, Clarke N. 1990. A tetrahedral
Zn(II)-binding site introduced into a designed protein.
Biochemistry 29: 10878. PDF
- Klemba M, Gardner KH, Marino S, Clarke
ND, Regan L. 1995. Novel metal-binding proteins by design.
Nat Struct Biol 2: 368. PDF
(scanned)
- Klemba M, Regan L. 1995. Characterization
of metal binding by a designed protein: single ligand substitutions
at a tetrahedral Cys2His2 site. Biochemistry 34:
10094. PDF
- Regan L. 1995. Protein design: novel metal-binding
sites. Trends Biochem Sci 20: 280. PDF
- Farinas E, Regan L. 1998. The de novo
design of a rubredoxin-like Fe site. Protein Sci
7: 1939. PDF
- Marino SF, Regan L. 1999. Secondary ligands
enhance affinity at a designed metal-binding site. Chem
Biol 6: 649.
PDF
- Marino SF, Shechner D, Regan L. 2001.
'Morphs' (MRFs): metal-reversible folding domains for differential
IgG binding. Chem Biol 8: 1221. PDF
Recent Work
- Stephen Marino, Ph.D., Andreas Plückthun
Group, Institute for Biochemistry, University of Zurich,
Switzerland
- Hiroshi Takashima, Ph.D, Chemistry, Nara
Women's University, Japan
- Kristen Lurie, Oceanside High School,
Ocenside, NY
- Aitziber López Cortajarena, Ph.D.
De novo design of allosteric control of conformation
By mutating fewer than half the residues in the predominantly
β-sheet B1 domain of protein G, we have demonstrated
that proteins with exceptionally high sequence homology can
fold into entirely different conformations. The resultant
proteins adopt the α-helical conformation of the dimeric
four-helix bundle protein Rop, directly challenging the paradigm
that proteins with highly similar sequences must fold similarly,
as expressed by Rose et al. in the Paracelsus Challenge (Proteins,
19:1-3, 1994). Two important requirements for the conformational
"switch" are the establishment of the target conformation's
hydrophobic core and adequate stability of the target conformation.
(A) The B1 domain of protein G and (B) the
Rop homodimer
We are extending this work by designing a means of allosteric
control into both the B1 and Rop target conformations. A number
of proteins, such as aspartate transcarbamoylase, require
ligation of a metal away from the active site for proper folding
and stability. We have designed tetrahedrally-coordinated
structural zinc binding sites into Rop, and we are currently
testing these designs. For B1, we have engineered a previously-designed
B1-domain tetrahedral metal binding site (see papers with
Klemba and Marino above) into the Rop sequence, along with
the residues to specify the hydrophobic core of B1. We are
currently evaluating whether metal ligation by these sites,
which can only assemble in the target conformations, can be
used as a thermodynamic driving force for proper folding into
the target conformation.
A model of a tetrahedral metal binding site
designed into the Rop monomer.
Leading References
- Dalal S, Balasubramanian S, Regan L. 1997.
Protein alchemy: changing beta-sheet into alpha-helix. Nat
Struct Biol 4: 548. PDF
(scanned)
- Dalal S, Balasubramanian S, Regan L. 1997.
Transmuting alpha helices and beta sheets. Fold Des
2: R71. PDF
- Dalal S, Regan L. 2000. Understanding
the sequence determinants of conformational switching using
protein design. Protein Sci 9: 1651. PDF
Current Contacts
Back to Project Index
Structure,
stability and folding of a four-helix bundle protein
Rop is a four-helix bundle protein whose role in vivo is
to bind to a complex of two RNA molecules, in a key step in
the regulation of replication of ColE1 plasmids. The crystal
structure of Rop has been solved at 1.7 Å resolution
and its NMR spectrum is completely assigned. These results
facilitate a detailed structural characterization of the protein
variants we create. We are using Rop as a model four-helix
bundle protein in which to study helix-helix interactions
by a systematic re-design of its hydrophobic core. Rop also
provides a useful system in which to investigate the contribution
of connecting loops to protein stability and folding. Finally,
in conjunction with the Crothers's laboratory, we are investigating
the mechanism by which Rop recognizes its RNA substrate and
the energetic contributions of the specific interactions involved.
Structures of the Rop homodimer (left) and
a mutant with a repacked
core that adopts an inverted quaternary structure (Willis
et al.).
Leading References
- Munson M, O'Brien R, Sturtevant JM, Regan
L. 1994. Redesigning the hydrophobic core of a four-helix-bundle
protein. Protein Sci 3: 2015. PDF
- Predki PF, Nayak LM, Gottlieb MB, Regan
L. 1995. Dissecting RNA-protein interactions: RNA-RNA recognition
by Rop. Cell 80: 41. PDF
(scanned)
- Predki PF, Regan L. 1995. Redesigning
the topology of a four-helix-bundle protein: monomeric Rop.
Biochemistry 34: 9834. PDF
- Munson M, Balasubramanian S, Fleming KG,
Nagi AD, O'Brien R, et al. 1996. What makes a protein a
protein? Hydrophobic core designs that specify stability
and structural properties. Protein Sci 5: 1584.
PDF
- Predki PF, Agrawal V, Brunger AT, Regan
L. 1996. Amino-acid substitutions in a surface turn modulate
protein stability. Nat Struct Biol 3: 54. PDF
(scanned)
- Munson M, Anderson KS, Regan L. 1997.
Speeding up protein folding: mutations that increase the
rate at which Rop folds and unfolds by over four orders
of magnitude. Fold Des 2: 77. PDF
- Nagi AD, Regan L. 1997. An inverse correlation
between loop length and stability in a four-helix-bundle
protein. Fold Des 2: 67. PDF
- Nagi AD, Anderson KS, Regan L. 1999. Using
loop length variants to dissect the folding pathway of a
four-helix-bundle protein. J Mol Biol 286: 257.
PDF
- Willis MA, Bishop B, Regan L, Brunger
AT. 2000. Dramatic structural and thermodynamic consequences
of repacking a protein's hydrophobic core. Structure
Fold Des 8: 1319. PDF
- Bishop B, Koay DC, Sartorelli AC, Regan
L. 2001. Reengineering granulocyte colony-stimulating factor
for enhanced stability. J Biol Chem 276: 33465.
PDF
Combinatorial biophysics: screening for Rop function
and structure
We have developed a screen for Rop function, based on the
fact that Rop reduces the copy number of ColE1 plasmids. By
expressing green fluorescent protein from a ColE1 plasmid,
and expressing Rop from a compatible vector, we can deduce
a Rop variant's ability to bind RNA by examining cellular
fluorescence. By altering the promoter for GFP expression,
we have been able to engineer both positive and negative screens
for Rop function.
|
Active Rop reduces
the copy number of a ColE1 plasmid expressing
green fluorescent protein from the lac promoter,
decreasing cellular fluoresence. |
Because of the exquisite level of understanding of how the
various positions in Rop contribute to such roles as core
formation and RNA binding, we can make libraries to address
specific hypotheses about the role of certain positions in
protein stability. Several different libraries are currently
being examined.
Current Contacts
Back to Project Index
Understanding β-sheet
formation and amyloidogenicity with model proteins
The factors that are important for α-helix formation
are much better understood than those for β-sheet formation.
This is largely because tractable model systems in which to
study β-sheet formation have been lacking. We are using
the B1 domain of Ig-binding protein G as an ideal model system
in which to study β-sheet formation. We have determined
both the intrinsic β-sheet forming propensities of the
amino acids and the energetics of pair-wise interactions across
two strands of a β-sheet. The results of these studies
allow us to formulate the first guidelines for rational β-sheet
design. We have also extended these studies to green fluorescent
protein, since the integrity of GFP's β-sheet structure
is reported at a cellular level by fluorescence.
A diagram of the host-guest site in the B1
domain of protein G used to examine β-sheet propensities,
cross-strand pairing and "aromatic rescue" of β-sheet
stability opposite glycines.
 |
An example of "aromatic
rescue" of glycine in a β sheet. Antiparallel
β sheet of a hydrogen bonded pair of cross-strand
residues from 1plc (PDB accession code) residues 82 and
94. |
Leading References
- Smith CK, Withka JM, Regan L. 1994. A
thermodynamic scale for the beta-sheet forming tendencies
of the amino acids. Biochemistry 33: 5510. PDF
- Smith CK, Regan L. 1995. Guidelines for
protein design: the energetics of beta sheet side chain
interactions. Science 270: 980. PDF
(scanned)
- Smith CK, Regan L. 1997. Construction
and design of beta-sheets. Acc Chem Res 30: 153.
PDF
- Merkel JS, Regan L. 1998. Aromatic rescue
of glycine in beta sheets. Fold Des 3: 449. PDF
- Merkel JS, Sturtevant JM, Regan L. 1999.
Sidechain interactions in parallel beta sheets: the energetics
of cross-strand pairings. Structure Fold Des 7:
1333. PDF
- Merkel JS, Regan L. 2000. Modulating protein
folding rates in vivo and in vitro by side-chain interactions
between the parallel beta strands of green fluorescent protein.
J Biol Chem 275: 29200. PDF
Mutations in the B1 domain of protein G that delay
the onset of amyloid fibril formation in vitro
We have previously reported that under certain experimental
conditions, many variants of the B1 domain of IgG-binding
protein G from Streptococcus form fibrils reproducibly.
The variant I6T53 is the focus of this study because the lag
phase in the kinetics of fibril formation by this variant
is significantly longer than that of other variants. This
lag phase is distinguished by changes in both intrinsic fluorescence
intensity and in light scattering of the protein. NMR diffusion
measurements suggest that the soluble protein during the lag
phase is monomeric. The kinetic profiles of fibril formation
are found to depend on experimental conditions. The first
kinetic phase diminishes almost completely when the reaction
is seeded with preformed amyloid fibrils. A number of studies
have shown that mutations can promote amyloid fibril formation;
here we show that protein engineering can also delay the onset
of fibril formation.
Recent Work
- Marina Ramirez-Alvarado, Ph.D., Biochemistry
& Molecular Biology, Mayo Clinic, Rochester, MN
- Melanie Cocco, Ph.D., Molecular Biology
and Biochemistry, University of California, Irvine
X-ray diffraction studies of amyloid fibrils
Variants of the B1 domain can be reproducibly induced to
form amyloid fibrils. Low resolution fiber diffraction studies
of a variety of natural fibrils reveal a "cross-β"
structure with approximately 4.7-4.8 Å strand-strand
separation and 10-11 Å sheet-sheet separation. We are
using a variety of alignment methods to obtain more ordered
fibril preparations from which to obtain high resolution structural
information.
Recent Work
- Catharine Bradford, Yale University (Undergraduate)
- Marina Ramirez-Alvarado, Ph.D., Biochemistry
& Molecular Biology, Mayo Clinic, Rochester, MN
- Jason Rahal, Yale University (Undergraduate)
Current Contacts
Collaborators
 |
(a) Amyloid
fibril formation upon "auto-seeding" of an A6A53
mutant of the B1 domain of protein G, as followed by CD
(mean residue ellipticity ar 217 nm). An electron micrograph
of the fibrils is shown below (b). |
Leading References
- Ramirez-Alvarado M, Merkel JS, Regan L.
2000. A systematic exploration of the influence of the protein
stability on amyloid fibril formation in vitro. Proc
Natl Acad Sci USA 97: 8979.
PDF
- Ramirez-Alvarado M, Regan L. 2002. Does
the location of a mutation determine the ability to form
amyloid fibrils? J Mol Biol 323: 17. PDF
- Ramirez-Alvarado M, Cocco MJ, Regan L.
2003. Mutations in the B1 domain of protein G that delay
the onset of amyloid fibril formation in vitro. Protein
Sci 12: 567. PDF
Back to Project Index
Design, structure
and binding of helical repeat domains (TPRs)
The tetratricopeptide repeat (TPR) is a 34 amino acid motif
that encodes a pair of antiparallel α-helices. These
adjacent repeats stack consecutively, in parallel, to form
remarkable and sometimes vast superhelical arrays of 3 to
22 motifs. This common structural property underpins a common
function and mechanism: the mediation of critical protein-protein
interactions in the cell. TPRs have been identified in over
300 proteins and are involved in a variety of biological processes
such as cell cycle regulation, transcriptional control, protein
transport, neurogenesis and protein folding. Our studies focus
on (i) the design and characterization of idealized TPR repeat
proteins with novel binding functions (ii) sequence and structure
analysis of natural TPR proteins to predict and understand
evolution of TPR domain functions and (iii) the characterization
of properties of long multiple repeat TPRs, both natural and
designed.
From the X-ray crystal structure of an idealized
TPR domain.
Design and structure of an idealized TPR motif
We have used a statistical, consensus-based approach to design
idealized TPR proteins. Natural TPR proteins most commonly
contain three repeats, which is thought also to be the minimal
functional unit for binding. Therefore, we first designed
a three repeat protein. The repeat amino acid sequence was
chosen from a global propensity analysis of a TPR profile,
constructed from the sequences available in the databanks.
The designed three repeat consensus protein (CTPR) was extremely
stable (Tm = 83 ºC). We further
constructed also two and one repeat TPRs, and showed these
to be stable.
We also crystallized and solved the high resolution structures
of three and two repeat CTPRs at 1.55 and 1.6 A resolution
(Main et al., 2003). A detailed analysis of these structures
provides an understanding of the TPR motif, how it is repeated
to give helical arrays with different superhelical twists,
and how a very stable framework may be constructed for future
functional designs.
Crystal structures of idealized
TPR proteins with two (left) and three (right) TPR motifs.
Engineering binding in TPRs
TPR domains diverge considerably in sequence and represent
a highly versatile class of protein-protein interaction modules
that have evolved to have a broad range of binding specificities:
The binding specificity of each TPR protein is different,
although the underlying structural motif is the same. Our
studies focus on the design and characterization of novel
ligand binding specificity using the designed, idealized TPR
motif, CTPR3 (Main et al., 2003). This protein is an ideal
scaffold on which to incorporate a variety of different binding
specificities.
Leading References
- Prodromou C, Siligardi G, O'Brien R, Woolfson
DN, Regan L, Panaretou B, Ladbury JE, Piper PW, Pearl LH.
1999. Regulation of Hsp90 ATPase activity by tetratricopeptiderepeat(TPR)-domain
co-chaperones. EMBO J 18: 754. PDF
- Main ERG, Xiong Y, Cocco MJ, D'Andrea
L, Regan L. 2003. Design of stable α-helical arrays
from an idealized TPR motif. Structure 11: 497.
PDF
Recent Work
- Luca D'Andrea, Ph.D., Institute of Biocrystallography
& Bioimaging, CNR, Naples, Italy
- Ewain Main, Ph.D., Sophie Jackson Group,
Chemistry, Cambridge University, England
- Melanie Cocco, Ph.D., Molecular Biology
& Biochemistry, University of California, Irvine
Current Contacts
Collaborators
Back to Project Index
GFP-based interaction
detection
We have designed a green gluorescent protein (GFP) reassambly
assay by which to monitor protein-protein interactions. Both
in vivo and in vitro, the dissected halves
of GFP are only reassembled to form a fluorescent molcule
if they are fused to two interacting proteins. We are developing
this as a general system by which to study the specificity
and affinity of protein-protein interactions.
A dissected GFP molecule can
be reassembled, here directed
by the binding of a leucine zipper, resulting in fluorescence.
Leading References
- Ghosh I, Hamilton AD, Regan L. 2000. Antiparallel
leucine zipper-directed protein reassembly: application
to the green fluorescent protein. J Am Chem Soc 122:
5658. PDF
Recent Work
- Indraneel Ghosh, Ph.D., Chemistry, University
of Arizona, Tucson, AZ
- Dennis Mishler, MB&B, Yale University
(rotation student)
Current Contacts
Collaborators
Back to Project Index
Genome-wide
investigation of natively-unfolded proteins
It is becoming increasingly apparent that not all proteins
are folded monomers within the cell. There are a number of
examples of proteins that are unfolded in the absence of a
“partner” molecule (alpha synuclein, for example).
The “partner” molecule can be another protein,
nucleic acid, or a small molecule. As part of the NESGC structural
genomics initiative, many proteins of unknown structure are
being produced. A significant proportion of these proteins
are unfolded. We are using a combination of computational
and experimental approaches to identify the partner molecules
that are required for these proteins to fold. The importance
of these studies is two-fold: (1) they allow the identification
of novel folds in non-monomeric proteins; and (2) they allow
the study of the thermodynamics and folding of this unusual
and biomedically relevant class of proteins.
Experimental identification of partner molecules.
A natively unfolded C. elegans protein is immobilized
on a SPR (Biacore) chip. Molecules that bind are captured
for identification from a C. elegans extract. (Red
line, background binding to chip alone; blue and black lines,
binding to chip with immobilized protein)
Leading References
- Balasubramanian S, Schneider T, Gerstein
M, Regan L. 2000. Proteomics of Mycoplasma genitalium: identification
and characterization of unannotated and atypical proteins
in a small model genome. Nucleic Acids Res 28:
3075. PDF
Recent Work
- Suganthi Balasubramanian, Ph.D., Mark
Gerstein Group, MB&B, Yale University
Current Contacts
Collaborators
Back to Project Index
Fragile X
Mental Retardation Protein (FMRP)
Fragile X syndrome is the most common form of inherited mental
retardation in the human population. The most devastating
aspects of this syndrome are mental retardation and behavioral
abnormalities, although there are also pleiotropic effects
that include testicular enlargement and a distinctive appearance
with elongated face and protruding ears. The most common cause
of the syndrome is an expansion of CGG repeats at a specific
location on the X chromosome, upstream of the gene that encodes
the Fragile X Mental Retardation Protein (FMRP). In addition
to causing chromosomal fragility, the CGG repeats are methylated
and transcription of FMRP is silenced. A key role for FMRP
in Fragile X syndrome is supported by the effect of an Ile304Asn
point mutation within the coding region. An individual with
this mutation displays a particularly severe Fragile X phenotype
in the absence of CGG repeat expansion.
FMRP belongs to the FXR family of RNA binding proteins which
contain two K-homology (KH) domains. Recent work with Drosophilia,
which possesses a single
representative of this family (dFXR), suggests that studies
on dFXR can be used to understand the functions of the FXR
family. We aim to characterize the structure and folding of
the KH domains from FMRP, its autosomal homolog FXR1 and dFXR.
In addition, we aim to identify the specific RNA targets of
these proteins and to determine the nature of the KH-RNA interaction.
We are also examining the role of the I304N mutation (present
in FMRP_KH2) in the structure and RNA binding function of
these KH domains.
Top: The domain structure
of FMRP.
Middle: NMR structure of the KH1 domain of FMRP.
Bottom: Homology map of FMRP's KH domains, with secondary
structure noted. KH2 has a large loop inserted between the
β2 and β3 strands.
Recent Work
- Shao-Min Yuan, Ph.D., Informax, Inc.,
Bethesda, MD
- Lili Aramli, Ph.D.
Current Contacts
Back to Project Index
Small-angle
X-ray diffraction of protein folding and amyloid formation
In collaboration with Prof. Lois Pollack of Cornell University,
we are using a microfabricated rapid mixing device and high
brightness synchrotron X-rays to monitor in real time the
large conformational changes that occur during macromolecular
folding. These stuides permit a time resolution that is enormously
faster (sub-millisecond) than that accessible by stopped flow
methods. To date we have begun to study early events in the
folding of several proteins and RNAs. For example, we have
identified pairs of residues between the two parallel β-strands
of Green Fluorescent Protein (GFP) which not only modulate
the stability of the native protein, but also have dramatic
effects on the rates of protein folding both in vivo and in
vitro. We have found that in wild-type and mutant GFP, secondary
structure is acquired rapidly, whereas tertiary structure
is acquired signficantly more slowly and at dramatically different
rates for the different β-strand pair variants. We are
using small angle X-ray scattering to investigate at what
stage in the folding macroscopic collapse occurs.
 |
Kratky plot of the scattering
from folded (dashed line) and unfolded (solid line) GFP |
Recent Work
- Donna Luisi, Ph.D., Wyeth BioPharma, Andover,
MA
Current Contacts
Collaborators
Back to Project Index |
