Journal of Structural Biology xxx (2013) xxxxxx

Contents lists available at ScienceDirect

Journal of Structural Biology

journal homepage: www.elsevier.com/locate/yjsbi

Fifty years of brous protein research: A personal retrospective

David A.D. Parry
Institute of Fundamental Sciences, Massey University, Private Bag 11-222, Palmerston North 4442, New Zealand

a r t i c l e

i n f o

Article history:
Available online xxxx
Keywords:
a-, b- and feather keratin
Synthetic polypeptides
Muscle
Tropomyosin
Collagen
Connective tissue
Coiled-coil proteins
Plakins
Heptads

a b s t r a c t
As a result of X-ray ber diffraction studies on brous proteins and crystallographic data on fragments
derived from them, new experimental techniques across the biophysical and biochemical spectra, sophisticated computer modeling and renement procedures, widespread use of bioinformatics and improved
specimen preparative procedures the structures of many brous proteins have now been determined to
at least low resolution. In so doing these structures have yielded insight into the relationship that exists
between sequence and conformation and this, in turn, has led to improved methodologies for predicting
structure from sequence data alone. In this personal retrospective a selection of progress made during the
past 50 years is discussed in terms of events to which the author has made some contribution.
2013 Elsevier Inc. All rights reserved.

1. Introduction
This paper charts a selection of those research developments in
the eld of brous proteins in which I have been involved and
which have both excited me and stimulated me to pursue a career
in structural biophysics. This is necessarily a very personal perspective and readers should view it in that light. Whether or not
these contributions have been of signicance is left to the reader
to decide. Although the story starts in 1963 (the commencement
of my PhD studies) it remains, some 50 years later, a work in progress, albeit now at a somewhat lesser degree of productivity than
in my hey-day. The chance to contribute a retrospective of this
type is rarely offered and I feel very privileged to have the opportunity here. By its very nature, however, I will not be able to provide as many of the details or all of the references that I would
have liked to do. The reader is therefore referred to the original literature for a more complete picture as well as appropriate attribution of the work of others. I freely acknowledge that friends and
colleagues from all over the world have made major contributions
to the work reported here, and without their assistance and input a
great deal less would have been achieved. My main aspiration in

Abbreviations: PBLG, poly-c-benzyl-L-glutamate; IF, intermediate lament; DST,

disulfosuccinimidyl-tartrate; RTT, rat tail tendon; CFDD, collagen bril diameter
distribution; PRD, plakin-repeat domain; CSIRO, Commonwealth Scientic and
Industrial Research Organization; DSIR, Department of Scientic and Industrial
Research; NIH, National Institutes of Health.
Fax: +64 6 350 5798.
E-mail address: d.parry@massey.ac.nz

writing this paper is that readers will gain an impression of how,

in my eyes, the eld of brous proteins has been such an exciting
one to be involved in over the past 50 years and, indeed, is equally
likely to be in the years ahead.

2. Synthetic polypeptides and a polymer at Kings College,

London (19631966)
In 1963 I had just completed my BSc (General) in Mathematics
and Physics at Kings College in London and was looking for a job. I
applied for a position with the British Scientic Civil Service and
was successful in getting a position that would have involved me
in designing ship hulls. Before I started, however, I was approached
by Seweryn Chomet, a lecturer in Physics at Kings College, to see
whether I would like to undertake a PhD in biophysics. I had absolutely no idea what was involved and, indeed, had never even contemplated the possibility of doing postgraduate research. After
some enquiries I took the plunge and signed up to work under
Arthur Elliott and Maurice Wilkins. I was fortunate even at that
early stage of my career in having such Supervisors. Arthur Elliott
(Fig. 1a) had worked for many years on synthetic polypeptides at
Courtaulds in the UK and was widely recognized for his expertise
in this area. Shortly beforehand, however, Courtaulds had axed
their research team in a cost-cutting measure and he was temporarily unemployed. Fortunately, Kings College Biophysics Department recognized an opportunity to gain his expertise and he
accepted a Readership after a 6 months sabbatical with Bruce Fraser and Tom MacRae at the CSIRO Division of Protein Chemistry in

1047-8477/$ - see front matter 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jsb.2013.10.010

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

Fig.1. (a) Arthur Elliott, PhD supervisor, (b) ofcial opening of the Kings College Biophysics Laboratory in Drury Lane, early in 1964. The Chancellor of the University of
London, Queen Elizabeth, The Queen Mother, was escorted by Sir John Randall, Head of the Laboratory. David Parry was the model-making postgraduate introduced to the
Queen Mother. The latter is seen here waving to the girls in the Sainsbury Store hanging out of a rst oor window next door, (c) toroidal focusing X-ray camera with the
conical camera (at right). The orientation cell with one of its beryllium windows driven by a small motor is shown at center, (d) the same camera with a at lm holder, (e) Xray diffraction pattern of an oriented solution of PBLG in m-cresol (38% concentration) showing features characteristic of an a-helix (and not a 310 helix as had been proposed
earlier by others (Luzzati et al., 1961), (f) and (g) X-ray patterns of oriented solutions of PBLG in dimethylformamide (45% and 70% concentrations, respectively) showing a
weak near-equatorial layer line (f) and then a well-dened near-equatorial layer line (g) akin to that seen in the k-m-e-f group of coiled-coil a-brous proteins.

Melbourne. Maurice Wilkins (a New Zealander, it transpired, an

attribute that would become more relevant to me later in my career than I might have expected at that time) was my other PhD
supervisor. He had just been awarded the Nobel Prize with Francis
Crick and James Watson for the structure of DNA. Interestingly,
Bruce Fraser had worked briey with Maurice Wilkins at Kings
College in the early 1950s on the structure of DNA and had produced a model that had many of the features present in the WatsonCrick structure (Fraser, 2004). The close relationship
between many of the key players in the eld of brous structures
was to become a recurring theme of my career.
The Biophysics Department was in the process of packing up its
equipment from its existing (underground) Wheatstone Laboratory in the Strand when I joined them. New premises had been
found in Drury Lane (not far from the famous theatre) and early

in 1964 Queen Elizabeth, the Queen Mother as Chancellor of the

University of London, ofcially opened the laboratory. I was chosen
to meet her as a typical postgraduate. I was supposed to be constructing molecular models as she strolled into the laboratory.
While chatting with me she spotted a group of girls from the local
Sainsburys supermarket hanging out of a rst oor room across a
narrow alleyway. She waved (Fig. 1b) and the photographer
snapped a picture that appeared in The Times newspaper and
elsewhere. Suddenly I was famous! Reality soon set in, of course,
and I spent the next month or two helping to establish the X-ray
suite, painting bits and pieces and doing odd jobs round the place.
Proper research commenced shortly thereafter.
My rst venture involved helping to solve the structure of a
polymer polycaproamide or the c-form of nylon 6 (Bradbury
et al., 1965). Arthur Elliott had started on this problem at

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

Courtaulds and I was able to help nish it off using the rst computer the Department had acquired an Elliott 806. This was a
real computer. It had ashing colored lights, a noisy tape-drive,
a tape-reader prone to chewing up my paper tapes and it made
squeaks that allowed me to work out what part of the routine
was being actioned at any particular instant. Altogether, the cform of nylon 6 proved a great introduction to the methodology
that I was going to use subsequently in solving X-ray diffraction
patterns of other brous materials.
The reason for the delay in tackling what was to be the goal of
my thesis the structure of synthetic polypeptides in solution
was that Arthur Elliott was still developing his toroidal reecting
mirror to focus X-rays. It involved him in producing a glass former
with an elliptical surface of very high axial ratio. This was then
gold-coated, placed in resin and allowed to set. The former was
then removed (hopefully without breaking it) and the gold remained attached to the resin (Elliott, 1965). The subsequent elliptical mirror proved successful in focusing an intense X-ray spot
suitable for investigating the structures of various orientated brous materials and it proved to be a key piece of equipment for
me in my research (Fig. 1c and d).
My main project was to investigate the structures of synthetic
polypeptides in solution. To do this I devised a cell (nick-named
PAM) with X-ray translucent beryllium windows. The back window was rotated relative to the front one to shear the solution that
lay between them and thereby induce some orientation. The X-ray
beam was then shone through the windows off-axis and orientated
X-ray patterns were obtained. Using PAM, an acronym for the Parry Alignation Machine (I still shudder from the pretentiousness of
that phrase), Arthur Elliott and I were able to show that poly-cbenzl-L-glutamate (PBLG) in m-cresol adopted an a-helical structure (Fig. 1e; Parry and Elliott, 1965) and not the 310 helix conformation proposed by Luzzati et al. (1961). We were also able to
produce a diffraction pattern from PBLG in dimethylformamide
that was characteristic of a coiled-coil structure in that it displayed
strong diffraction on a near-equatorial layer line (Fig. 1f and g; Parry and Elliott, 1967). This feature is a characteristic of a-brous
proteins in muscle and keratin, the structures of which were to
prove of consuming interest to me throughout my research career.
Bruce Fraser visited the Kings laboratory in 1965. He was in
Europe for the ve-yearly Wool Research Conference that was to
be held on this occasion in Paris. This was the De Gaulle era in
France when all oral contributions at conferences were to be in
French! One suspects that this proved a challenge for many overseas delegates. Bruce Fraser, however, was looking for a possible
Post-doctoral Fellow to replace Andrew Miller who had just returned to the UK after 3 years in Melbourne. The opportunity
was an exciting one for me as I had always wanted to go to Australia. I subsequently applied for and got the job. In September 1966,
with a newly minted PhD, I vacated my desk to Arthurs next PhD
student John Squire and boarded the Achille Lauro at Southampton bound for Melbourne via Italy, Malta, Egypt and Aden.
4 weeks on board ship (CSIRO paid rst class for their postdoctoral
fellows) proved to be a delightful interlude before starting my rst
real job.

3. Keratins at CSIRO, Melbourne (19661969)

The CSIRO Division of Protein Chemistry (DPC) was, at that time,
one of three research laboratories in Australia that were devoted to
wool research. As such, it reected the importance of the wool
industry to the Australian economy. Undoubtedly, the DPC was
the pre-eminent research institution of its type in the world and
was devoted to investigating the physical and chemical structure
of wool in particular, but also of other keratins. Bruce Fraser, a

brilliant theoretician and experimentalist in the eld of brous

proteins and one who was to prove a special friend and mentor
over my entire research career, led the structure group that also included Tom MacRae, Eikichi Suzuki and, later on, Peter Tulloch
(Fig. 2). The DPC was an extraordinarily stimulating place to work
in and it was well endowed nancially too. On top of that I met my
wife to be (Jenny Wilson) within a couple of hours of getting off the
boat. Everything was set for a productive 3 years and this proved to
be the case. In addition to my position at CSIRO I worked in the
evenings as a Resident Tutor at Ormond College in the University
of Melbourne, thereby gaining some teaching experience.
One of the projects undertaken in Melbourne concerned calculating the energy involved in deforming an a-helical strand into a
coiled-coil conformation (Parry and Suzuki, 1969a). The calculations in the mid-1960s were performed using a card-based computer program and input data deck rather than by paper tapes.
The decks were collected daily by a CSIRO driver who ensured that
it, as well as those from other CSIRO staff in Melbourne, was own
up overnight to the mainframe CSIRO computer in Canberra for
processing. The results were own back for delivery to the Division
on the following day (subject to not having fog at Canberra Airport). The results of the calculations showed that the energy involved in deforming a straight a-helical strand into a coiled coil
was, perhaps not surprisingly, very small indeed (Parry and Suzuki,
1969a). However, this fact was not as obvious then as it is now for
comparable structures. Additional calculations were performed for
multi-stranded a-helical ropes and these showed that coiled-coil
ropes were energetically favored over equivalent assemblies of
straight a-helices (Parry and Suzuki, 1969b). The relevance of this
work for later research in the eld proved important when the
heptad substructure characteristic of chains forming coiled-coil
structures was recognized in a wide and diverse selection of proteins, both brous and globular.
Much of my work was directed at helping to investigate the
structures of a-, b- and feather keratins. Two projects tackled by
the team at the DPC proved to be particular highlights. The rst
was aimed at solving at medium resolution the structure of feather
keratin. Some beautiful and highly detailed X-ray diffraction patterns from seagull feather rachis had been obtained by Bruce Fraser
and Tom MacRae (Fig. 3a). The 3.4 nm diameter laments, which
George Rogers had observed in the transmission electron microscope, had previously been shown to have a fourfold screw axis
with a 9.6 nm axial repeat (Fig. 3b). The high angle X-ray diffraction was characteristic of a b-structure but one with a shorter axial
rise per residue than normal (0.315 rather than 0.334 nm). After
some effort it became clear that the b-chains formed a twisted
sheet with a handedness that was opposite to that of the basic helix (Fraser et al., 1971). The opposite handedness of the twist and
the helix symmetry was, at that time, somewhat counter-intuitive
but it proved to be fundamental to the success of the project. The
fact that twisted b-sheets represent a very common feature in protein structures was not realized in the late 1960s and early 1970s
as very few crystal structures had been completed by that time.
Thus, the concept of a twisted b-sheet in proteins was rst proposed by us for feather keratin, and did not originate from the limited globular protein crystal structure data then available. Some
40 years later, when sequence data had became available for a variety of keratin b-proteins (akin to feather keratin), Bruce Fraser and
I were able to show that the feather keratin structure epitomized
an entire class of keratin b-forming laments present in the epidermal appendages of both birds and reptiles.
The second piece of research was that concerning the structure
of b-keratin. This is not a naturally occurring form of keratin but is
produced by stretching specimens of porcupine quill in steam.
Using energy calculations, infrared and X-ray diffraction data it
was possible for us to produce a model that was consistent with

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

Fig.2. A group photograph of (from left to right) of Andrew Miller, David Parry, Barbara (Doyle) Brodsky, Bruce Fraser, Tom MacRae and Eikichi Suzuki. Three of us (A.M., B.B.
and D.A.D.P.) were visiting the Division of Protein Chemistry in Melbourne at the same time in 1984.

a regular antiparallel chain arrangement in the b-sheet (Fraser

et al., 1969). Several new and important features that emerged
from these studies, however, were that there was disorder in the
packing of neighboring sheets and that the physical dimensions
of the b-crystallite were necessarily restricted to only about ten
chains per sheet and two to three sheets. This, in turn, strongly
suggested that there would be about 25 chains in cross-section
in the a-keratin intermediate lament and that there would be
equal numbers of up and down pointing chains. This result
was to prove relevant in subsequent studies involving the keratin
intermediate lament (IF) structure.

4. Tropomyosin in Boston (19691971)

After three wonderful years in Melbourne, during which time
Jenny and I got married, I was fortunate to gain a post-doctoral position with Carolyn Cohen and Don Caspar at the Childrens Cancer
Research Foundation in the Jimmy Fund Building, Boston (USA). At
that time their interest was centered on the crystal structure of
tropomyosin, a thin lament a-brous protein involved in the regulation of vertebrate skeletal muscle. I was a contributor to a couple of papers over the 2 years I spent in Boston (Cohen et al., 1971,
1972) on what were challenging crystals and other polymorphic
forms. The rst of the papers starts off with the sentence There
are few protein crystals seriously studied which give poorer Xray diffraction patterns than tropomyosin. This was, unfortunately, very true and stemmed in large part from their extremely
high water content (95%). Using the X-ray crystal data, low-resolution models could nonetheless be constructed of the lattice formed
by the head-to-tail assemblies of tropomyosin molecules (Fig. 4a
d). Computing these models to about 20 resolution rst involved
catching the shuttle bus from the Harvard Biophysics Department
(near the Jimmy Fund Building) to the Computing Center on the
main campus in Cambridge. Card-based programs were handed
over to the front desk and then you waited for them to be run
and the results put into your box. The system worked well but
was clearly time-consuming by todays standards. Subsequently,
Carolyn Cohen (Fig. 4e) and colleagues were to solve the structure
of tropomyosin at higher and higher resolutions (15 Phillips
et al., 1979; 9 Whitby et al., 1992; 7 Whitby and Phillips,
2000; 2 Brown et al., 2001) but our early low-resolution work

was nonetheless able to provide new insights into the mechanism

of muscle regulation.
The X-ray crystal patterns provided several crucial pieces of
information including the effective repeat length of the molecules
in the head-to-tail assemblies, the invariant head-to-tail connection and the bending of the molecules in a manner closely akin
to that present in the thin laments. The repeat length of the
tropomyosin assembly was equal to that of seven actin monomers
in the thin laments, thereby laying one of the foundations for a
model to explain the regulatory mechanism. Indeed, this was to
reach fruition just a year later when John Squire and I (by that time
both in Oxford) were able to develop a steric blocking model of
regulation for vertebrate skeletal muscle, a scheme that has remained essentially unchanged over the past 40 years (see later).

5. Collagen and muscle regulation in Oxford (19711973)

After a very rough crossing of the Atlantic from Boston to Southampton in the QE2 I spent two exciting years in Oxford in Professor
Sir David Phillips Laboratory of Molecular Biophysics in the
Department of Zoology. Whilst there I worked closely with Andrew
Miller on collagen structure (also with Karl Piez, Barbara Doyle
(Brodsky) and Bruce Fraser who were all visiting the laboratory
at the time), and with John Squire on the role of tropomyosin in
vertebrate muscle regulation (Fig. 6). John Squire, having completed his PhD with Arthur Elliott at Kings College, London had recently arrived in Oxford after a post-doctoral fellowship in Aarhus,
Denmark. Two particular projects stand out, both of which have
been among the most highly cited of my research contributions.
In 1973 no amino acid sequence was available for a collagen
chain from any single species. However, Karl Piez, then on sabbatical in the laboratory in Oxford, cobbled together those portions of
the sequence that were individually available for rat and calf skin
and which together formed the triple-helical region of the tropocollagen molecule. (It subsequently transpired that one tripeptide
was omitted, but the research we undertook was, fortunately, not
compromised.) Our interest centered on whether the sequence
contained any recognizable information that specied the socalled quarter stagger (or D-stagger) model of molecular aggregation proposed many years earlier on the basis of the 67 nm axial
period observed in X-ray diffraction patterns of tendon, and from

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

Fig.3. (a) X-ray diffraction pattern of sea-gull feather rachis taken by Bruce Fraser and Tom MacRae. The ber axis is vertical, (b) the lamentous structure deduced from the
X-ray data indicated subunits composed of a pair of molecules related to one another by a perpendicular diad (Fraser et al., 1971). In turn, the subunits are related to one
another by a left-handed fourfold screw axis with an axial rise of 2.4 nm and a unit twist of 90. The antiparallel twisted b-pleated sheets pack together closely to form the
core of the laments, (c) (from left to right) an undistorted sheet viewed down the hydrogen bond axis, the same sheet distorted to give a right-handed twist, and a pair of
sheets related by a perpendicular diad showing that rows of side chains in the interior interleave along the lament axis (Fraser and Parry, 2008).

the electron microscope observations of D-periodic banded collagen brils (Fig. 5a and b). With a little over 1000 amino acids in
the sequence of an a-chain I remember looking by hand to see if
there were any likely interacting groups separated by about 230
residues (if D was equal to 230 residues and the observed axial rise
per residue is 0.29 nm then the axial repeat would be close to
67 nm). The aromatic residues were indeed found in clumps with
these separations so we (and David Hulmes, then a PhD student,
in particular) decided to undertake a more complete computer
analysis where we looked for possible apolar and charged residue
interactions as a function of chain stagger. The chains were considered to be linear rather than taking into account their known triple-helical structure. The results were amazing (Fig. 5c). The
interactions peaked at 0D, 1D, 2D, 3D and 4D, thereby showing that
the mode of aggregation of molecules was, in fact, determined directly by the amino acid sequences of the proteins involved (Hulmes et al., 1973). Those in the eld had never doubted that this
was indeed very likely but now we had, for the rst time, denitive
proof that this was indeed the case, albeit for a simple linear brous structure.
In the second of the two most signicant pieces of research that
I helped carry out in Oxford the personal Kings link between John

Squire and myself was to prove very successful research-wise too.

We both came to Oxford with complementary information ripe for
analysis. I had returned from Boston with a clear view that tropomyosin must move across the surface of the actin monomers in the
thin laments during the regulatory process. Indeed, I had discussed this thought with Hugh Huxley at the Symposium on Quantitative Biology on Long Island, New York before I left Boston. Hugh
Huxley and his colleague John Haselgrove had also come to a similar conclusion. John Squire, in turn, had been in Jack Lowys team
in Aarhus that had studied the X-ray diffraction patterns of relaxed
and actively contracting muscle, and they had recognized clear differences in intensities on the second and third layer lines of the ber diffraction pattern between the two states (Vibert et al., 1972).
We decided that we should build models of the thin lament and
calculate the Fourier transforms corresponding to a range of different positions of the tropomyosin molecules. The results were
incredibly exciting (Parry and Squire, 1973). The relaxed X-ray
diffraction pattern was easily accounted for by strands of tropomyosin molecules lying on the actin surface where they would block
the attachment sites of the heads of the myosin molecules, thereby
preventing contraction. In turn, the contracting X-ray diffraction
pattern was consistent with the strands of tropomyosin molecules

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

steric blocking mechanism was thus born and remains, with

some subtle variations, the widely-accepted mode of regulation.
Indeed, there is now a great deal of other supporting evidence
available (see, for example, Lehman et al., 1994, 1995; Squire and
Morris, 1998; Brown et al., 2005; Geeves and Holmes, 2005; Poole
et al., 2006). The original model, constructed of colored rubber
balls for actin monomers and a rubber pipe for the tropomosin
strands, remains in John Squires possession (Fig. 6f).

6. A diversity of brous proteins in Palmerston North (1973-)

Jenny and I decided that we would like to return to the antipodes and consequently I sought positions in both Australia and New
Zealand. A lectureship at Massey University in Palmerston North
came up and I was thrilled to gain a permanent academic position.
It was an excellent move for me and one that I never had cause to
regret, thanks to the overwhelming support I was to receive at the
university. I could now plan my own areas of research, a part of
which was to become theoretically-based using computer modeling techniques. The computing facilities, though, were now beginning to change. Initially, however, the computer system was still
card-based. I used to carry box loads of cards to the Computing
Center that was located on the opposite side of a deep gully from
where I was housed (The ring road had not then been completed).
I could have walked the long way round but chose instead to take
the short cut down the side of the steep gully, clamber through the
bush, jump over the stream at the bottom and scale the far bank,
hopefully not dropping the card decks en route. This went on for
a number of years before desk-top computing took over. The ease
of modern day computing compared to the experiences of early
practitioners was as different as night is to day.

6.1. Tropomyosin

Fig.4. (a) and (b) X-ray diffraction patterns of two slightly different crystal forms of
tropomyosin, 2.5 precession photographs, [1 0 0] projections, (c) and (d) the
corresponding electron density maps (Cohen et al., 1971). (e) Carolyn Cohen, friend
and collaborator over many years. We worked primarily on muscle proteins and
heptad recognition and structure/function in protein sequences.

lying well away from the myosin head attachment site, thereby
allowing contraction to proceed (Fig. 6ae). This was mediated
by the levels of Ca2+ released from the surrounding muscle membrane and, subsequently, bound by troponin. The structural change
of troponin that ensued caused the strands of tropomyosin molecules to move across the surface of the thin lament. The so-called

While waiting for equipment to arrive I continued the development of the regulatory mechanism of vertebrate skeletal muscle
established in Oxford. Larry Smillie and his colleagues in Canada
had just sequenced a-tropomyosin, the rst a-brous protein to
have its primary structure completed (Stone et al., 1975). It displayed a continuous heptad substructure characterized by the
presence of apolar residues alternately three and four residues
apart, a feature rst predicted by Francis Crick 20 years earlier
(Crick, 1953). As Andrew Miller (unpublished) had already seen a
pseudo sequence repeat in tropomyosin I undertook a detailed
Fourier transform of the sequence to quantitate which repeats, if
any, were statistically signicant. Coincidentally, and concurrently,
Andrew McLachlan and Murray Stewart at Cambridge (UK) were
thinking along the same lines. Our results were fascinating and virtually identical (Fig. 7ac; Parry, 1975; McLachlan and Stewart,
1976). We showed that there was a fourteen-fold quasi-periodicity
(about 19.6 resides long) in the linear dispositions of both the
charged residues and the apolar residues. Alternatively, it could
be considered that tropomyosin had a period of length 39.2 residues (2  19.6) and that this was closely halved. The 39.2 residues
of a coiled coil conformation had the same length as that between
consecutive actin monomers in the thin lament. This meant that
each actin could be regulated by a quasi-identical portion of the
tropomyosin sequence. The two periods were subsequently
thought of as seven a-sites and seven b-sites. Tropomyosin was
thus envisaged as being able to rotate or roll over the surface of
the actin monomer with the end points being stabilized by the
a- and b-sites, respectively. The fact that the regulation could
now be understood in terms of how tropomyosin and actin interacted with one another was certainly a great thrill (Fig. 7).

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

Fig.5. (a) An electron micrograph of a negatively-stained whole-mount preparation of a collagen bril showing the D-periodic banding pattern. Each repeat contains a lightstaining molecular overlap region and a dark-staining gap region, (b) the quarter-staggered arrangement of collagen molecules (each about 4.47D long) in which molecules
are staggered axially with respect to one another by a multiple of D. This results in collagen brils with a D-periodic repeat, and (c) intermolecular apolar and ionic
interactions calculated between two collagen molecules as a function of relative axial stagger. The peaks occur at 0D, 1D, 2D, 3D and 4D, where D is about 67 nm (234
residues).

6.2. Connective tissue

When I started working on connective tissues I was fortunate to
establish a friendship and close working relationship with Alan
Craig at the Department of Scientic and Industrial Research
(DSIR). Alan was a talented electron microscopist and our complementary skills were to prove a productive combination over the

next decade. The rst experiment we undertook was to measure

the bril diameters in rat-tail tendon (RTT). This was a wellworked tissue but, strangely, there were no quantitative data on
the bril diameters. We printed all the micrographs at a standard
magnication related to a grating replica of known dimensions.
Surprisingly few electron microscopists were quantitative in those
days. Our results, collected laboriously using a ruler, were totally

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

Fig.6. The steric blocking mechanism proposed for vertebrate skeletal muscle in
1973 indicated that the binding site of HMM S-1 on actin was blocked by
tropomyosin in the off position but that the site became free as tropomosin
moved across the surface of actin. (a) The thin lament is modeled with actin
molecules (white spheres in general but with every seventh colored grey) arranged
with 13/6 symmetry, (b) the tropomyosin is shown tightly bound in the center of
the two long-period grooves in the thin lament. Troponin molecules are
represented as dark grey spheres, (c) the tropomosin is shown attached only at
38.5 nm axial intervals illustrating that attachment would be very loose unless
some supercoiling occurred or that the strands were located at some distance from
the center of the long-period grooves, (d) and (e) positions of tropomyosin
determined from model building using the X-ray diffraction data for relaxed and
contracting muscle, respectively. It can be seen that in contracting muscle that
the tropomyosin strands lie much closer to the center of the long-period grooves
than they do when they block the attachment site of HMM S-1 to actin. (f) The
original model illustrating the steric blocking mechanism was built in 1972/1973,
and was composed of rubber balls (for actin) and rubber tubing (tropomyosin)
(Parry and Squire, 1973). It now resides in John Squires study, and this picture of it
with John Squire (left) and DADP (right) was taken some 35 years later in 2008.

unexpected the collagen bril diameter distribution was clearly

bi-modal (Parry and Craig, 1977). This prompted us to look at the

same tissue over a full age span. At the earliest (foetal) stages the
brils were all very small and about the same size. Why was a bimodal distribution subsequently necessary for the RTT to function
properly as regards its mechanical role? Was RTT unique or did
other tendon/ligaments/skins display the same feature? In order
to answer these questions Alan Craig and I looked at a very diverse
range of different tissues from different species, each over a full age
span. It transpired that some tendons were bimodal, and some
were unimodal (see, for example, Parry et al., 1978a; Fig. 8). In order to understand these results we related the collagen bril diameter distributions (CFDD) to their corresponding mechanical roles
and came up with a theory that explained the observations (Parry
et al., 1978b). In its simplest terms the theory proposed that the
larger collagen brils were stronger than the smaller ones as the
former had a higher crosslink density stabilizing them. Conversely,
however, the smaller brils have a larger surface area per unit
mass, and the interactions that the smaller brils could make with
the proteoglycanwater matrix in which the brils exist would be
considerably more numerous than those made by the larger brils.
It followed that that the small brils would endow the tissue with
an ability to resist creep (i.e., plastic deformation) whilst the larger
brils would provide the extra strength required for the tissue to
withstand the stresses to which it was normally subject. This theory remains the basis for understanding the mechanical attributes
of all connective tissues. Incidentally, some years later it also transpired that some skins too had both bimodal and unimodal CFDD
(Craig et al., 1987) and the theory proved equally valid for these
cases too.
A particular connective tissue the cornea was of especial
interest. It had long been recognized that the brils in cornea all
had the same diameter. However, the diameters recorded in the literature for one species differed greatly from those reported for another species. Was this real or simply a reection of the different
experimental procedures employed? Alan Craig and I contacted
Auckland Zoo and when any animal died (from natural causes) they
would excise the eyes, put them in a bottle and post them to me.
Opening the post in those days was a strange experience for you
never knew what might be peering back out at you. Using an identical set of protocols for every specimen it soon became evident the
collagen brils in mammals, birds, reptiles and cartilaginous sh
were all identical (25 nm) but that those from the bony sh, whilst
still identical to one another within the group, were smaller (17 nm)
(Craig and Parry, 1981). This was a simple experiment but one
which conrmed that the mechanism for corneal transparency
was not species-dependent but was common to all animals. Some
time later we repeated the experiments for the bony sh using a
low temperature cryo-method of specimen preparation. The diameters were about 4050% larger than we had measured previously
using conventionally prepared specimens. Hence, in vivo the best
diameter measurement we now have for the collagen brils in
mammals, birds, reptiles and cartilaginous sh is 36 nm, and in
bony sh is 25.5 nm (Craig et al., 1986; Craig and Parry, 1989).
From measurements of the diameters of small collagen brils
prepared for electron microscopy using conventional preparative
procedures it became apparent that not all diameters were present.
The implication of the measurements was that the brils were
built by the accretion of subunits about 4 nm in diameter rather
than by single collagen molecules (Parry and Craig, 1979). This
was consistent with the concept of a collagen microbril formed
from a grouping of ve collagen molecules in section (Smith,
1968), with each molecule quarter-staggered in the manner that
gives rise to D-periodic brils seen in the electron microscope
and visualized indirectly by X-ray diffraction.
Another experiment that we undertook must rank among the
most tedious experiment ever carried out. The question we attempted to address, however, was simply stated and inherently

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

Fig.7. (a) Head-to-tail assembly of two-stranded coiled-coil tropomyosin molecules (strands are colored blue and yellow) lie in the each of the two long-period right-handed
grooves in the thin lament. Each molecule has 14 regions, each with an acidic and apolar part. These are represented by horizontal black lines. For a supercoil pitch length of
13.7 nm tropomyosin makes seven half-turns relative to the seven actin monomers with which it makes contact, thereby allowing allowing each actin to be regulated by
tropomyosin in a quasi-identical manner. The active state is thus stabilized by one set of tropomyosinactin interactions, and the relaxed state by the second set, (b) and (c)
represent the thin lament structures seen in section for the relaxed and active states, respectively. Troponin is represented as a black circle. In (b) the position of
tropomyosin blocks the site of attachment of HMM S-1 to actin, thereby preventing contraction, whereas in (c) tropomosin has moved across the surface of actin, thereby
vacating the site on actin to which HMM-S1 binds during active contraction.

interesting how long is a collagen bril? For a variety of reasons

this was not a trivial problem. As the diameters of collagen brils
are typically about 50100 nm they are too small to be seen using
optical microscopy. Using a scanning electron microscope the brils weave in and out of the bers and it is impossible to follow
any single bril for any great distance. Our approach was horrendous we took serial sections (each about 80 nm thick) and followed the course of 1368 brils over the course of 17
consecutive slices (1.4 lm). The idea was that if we found a number of ends we could, from the statistics, determine an average bril length. Sufce to say we found no ends thereby indicating the
brils in RTT were very long indeed. Also, from a knowledge of the
percentage of the specimens occupied by cells we were able to
estimate the maximum mean free path between cells and thereby
the maximum length of a bril. These approaches together implied
that the brils in mature RTT were as long as 10 mm and those in
other mature tendons were at least several mms in length (Craig

et al., 1989). Fibril lengths were thus many times the critical length
required to give the bers their structural integrity (Parry, 1988).
6.3. a-Keratins
Very frequently over the (southern) summer periods Bruce Fraser
and Tom MacRae would kindly invite me across to Melbourne for
periods of three to 6 weeks to work with them on various projects
involving a-keratin. It was always a great delight to do so. One such
project involved studying the sequences of two fractions of wool keratin, known as the Type I and Type II fragments. The conclusion
reached was that each fragment contained a regular linear disposition in both their acidic and their basic residues (Parry et al., 1977).
Furthermore, it was shown from a study of the number of potential
interchain ionic interactions that the two fragments would lie in axial register and would be parallel to one another. This was the rst
strong indication that a-keratin molecules were parallel-chain

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

10

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

Fig.8. Electron micrographs of collagen brils in transverse section from a ve year-old horse showing (a) a suspensory ligament and (b) a common digital extensor tendon,
(c) co-author Alan Craig, (d) and (e) the collagen bril diameter and mass distributions for the suspensory ligament and extensor tendon, respectively determined as a
function of age. The unimodal and bimodal forms displayed in the two mature tissues are evident, as are the similarities between the two tissues at the foetal stage and at
senescence.

heterodimers and that molecular aggregation was facilitated by

intermolecular ionic interactions. The terminology Type I and Type
II chains was subsequently used in a description of all a-keratins,
both epidermal and trichocyte. Another key result obtained on one
of these summer pilgrimages to Melbourne was the realization from
fragments of sequence data then becoming available that trichocyte
keratin was indeed one of the members of the intermediate lament
(IF) family (Dowling et al., 1983). Suddenly, hair keratin research
(largely ignored by the international community until then) became
part of the mainstream in cell biology, and papers that had rarely
been cited outside the keratin eld became highly recognized.
Another of the summer visits to Melbourne coincided with a visit
made by Peter Steinert to the DPC. Peter, an Australian by birth but

by then working at the National Institutes of Health (NIH) in Bethesda, had just shown how he could reconstitute epidermal keratin laments from individual chains (Fig. 9d). This was an extraordinarily
impressive piece of research, especially to those of us struggling
with the myriad of highly cross-linked proteins that constituted
wool. After his talk I escorted Peter to the Melbourne bus station
and we struck up a friendship that was to prove very fruitful from
both a personal and research perspective. Indeed, from 1983 to
2004 Peter and I went onto publish more than 30 papers together,
as well as a book on intermediate lament structure and a major review (Parry and Steinert, 1995, 1999). As in most successful collaborations we brought together complementary skills. Peter was an
experimentalist without peer whereas my attributes were primarily

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

in structure analysis. Frequently, I would tell Peter that some key

pieces of data were missing and I would then ask him if he could
work out how to obtain them. He would throw up his hands in
the air and say that there was absolutely no way that these could
be obtained experimentally. However, a few months later (sometimes, however, as long as two to 3 years later) he would come back
to me and say that he had now found a way to do what I had proposed. Peter was stunningly innovative, a great friend and a wonderful host during the many short research periods I was to spend
with him at the NIH. Our collaboration included describing the fundamental secondary structure of IF molecules (Fig. 9ac; Crewther
et al., 1983), distinguishing the key differences between Type I
and Type II chains (Steinert et al., 1984), delineating the subdomain
structure in the head and tail domains of epidermal keratins (Steinert et al., 1985), conrming experimentally that the two chains in
keratin IF molecules were indeed parallel and in register (Parry
et al., 1985), showing that the H1 subdomain in the head region
was important in specifying aggregation at the two- to four-molecule level (Steinert and Parry, 1993), characterizing the structure
and function of human trichohyalin (Lee et al., 1993; Steinert
et al., 2003), determining the crystal structure of the C-terminal

11

fragment of segment 2B in vimentin (Herrmann et al., 2000) and

measuring experimentally the effect of characterized trigger motifs
(Wu et al., 2000).
The most far-reaching papers of all, however, were those that
involved the characterization of the DST crosslinks induced in keratins (Fig. 9e; Steinert et al., 1993a,b), vimentin (Steinert et al.,
1993c) and various copolymers of IF chains (Steinert et al., 1999)
(Fig. 9e). This, in turn, led to the most exciting development of
all in which we showed that there was an axial structural transition between the reduced and oxidized forms of trichocyte keratin (Fig. 10; Wang et al., 2000). This work resolved many longstanding difculties in the eld where a single model had been unable to account for all of the data available. This piece of research
represented a major career highlight for both Peter and me.
Peters tragic and untimely death in 2003 was an enormous loss
both for the intermediate lament eld and for his many friends
and colleagues. One can only imagine what additional advances
might have been made had Peter still been alive.
In recent years I have also been fortunate to work closely with
Tammy (Smith) Lynch on sequence features of importance in intermediate lament chains, including those with mutations (Smith

Fig.9. (a) Model of the tri-partite structure of the trichocyte keratin molecule. The rod, as originally proposed, contained four coiled-coil segments (1A, 1B, 2A and 2B) and
three linkers (L1, L12 and L2) (Crewther et al., 1983). Modeling and subsequent crystal data show that segment 2A + L2 + a short part of segment 2B forms a continuous
bundle of near-parallel a-helices. The heads and tails refer to the N- and C-terminal domains respectively, (b) the heads are shown folded back over segment 1A and this
is believed to stabilise the two-stranded coiled-coil structure, (c) the heads are shown displaced from the rod domain and in positions where they may interact with other
cellular entities. The effect is also likely to destabilize segment 1A and allow the two a-helical strands to separate, (d) long-time friend and collaborator Peter Steinert with
whom many keratin or keratin-related projects were tackled, (e) characterization of DST and oxidative crosslinks which illustrate the principal modes of molecular assembly
(A11, A22, A12) in the IF, and the differences (in A11) that occur between the reduced and oxidative states (Wang et al., 2000).

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

12

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

et al., 2002, 2004; Parry et al., 2002; Smith and Parry, 2007, 2008),
with Juergen Schweizer, Mike Rogers and Lutz Langbein on structural aspects of medulla and keratin-associated proteins (Parry
et al., 2006; Langbein et al., 2010) and with Bruce Fraser on various
aspects of trichocyte keratin structure involved in the transition
from the reduced to oxidized forms (Fig. 10; see, for example, Fraser and Parry, 2007, 2012).
6.4. Feather and scale b-keratins
A few years after I moved from Melbourne Bruce Fraser and
Tom MacRae identied ve contiguous segments of b-strands connected by b-turns in emu feather keratin, the rst sequence of its

type to be completed (Fraser and MacRae, 1976). This region, they

suggested, was the b-rich domain that formed the core of feather
keratin laments. As Bruce Fraser and I had always harbored a
great affection for feather (and scale) keratin it was, perhaps, no
great surprise that we decided to tackle this structure again some
30 years later. The impetus to do so was largely provided by the
wealth of new sequence data that were becoming available, largely
(but not exclusively) through the efforts of Lorenzo Alibardi and his
colleagues in Italy. Among other things we were able to show that
a b-rich domain in reptilian keratin had great similarities to that
seen in feather keratin (Fraser and Parry, 1996) thereby indicating
the likelihood of there being an entire class of b-laments, and not
just a unique structure found in feather rachis. If correct, it followed that another class of laments existed in nature akin to
those dened by the well-established k-m-e-f group of a-brous
proteins dened through the pioneering work of Astbury in the
1930s.
Fitting the emu feather sequence to a low resolution model
illustrated that the inner surface of the b-sandwich formed from
an antiparallel arrangement of two b-domains was densely populated with apolar residues and that the charged and cysteine residues lay on the outer surface. In addition, the inner residues in the
repeating unit meshed neatly together in layers oriented perpendicular to the lament axis (Fig. 3c; Fraser and Parry, 2008). Further work showed that it was possible to devise a range of
structural probes based on conformational properties, hydropathy
proles, homology comparisons and Fourier analyses that could
identify the b-framework of a wide range of laments in avian
and reptilian keratins. A common pattern 34-residues in length
was thus identied (Fraser and Parry, 2011a). The lengths of the
rst and last b-strands are predicted to be shorter than those of
the central three strands. Also, detailed analyses of the N- and Cterminal domains that form the matrix in which the laments
exist revealed very signicant differences between the Archosaurs
(birds, crocodiles and turtles) and the Squamates (snakes and lizards). This work therefore conrmed the existence of a new class
of lamentous b-structures rst suggested by our work many years
earlier. The analogy with the a-brous proteins characterized more
than 50 years previously is noteworthy. An additional advance was
to explain the low-resolution net pattern seen in the X-ray diffraction patterns of feather keratin in terms of the orientations of the
laments in the short-range sheets observed by electron microscopy (Fraser and Parry, 2011b).
6.5. Heptads

Fig.10. Models illustrating the structural transition between (a) the reduced and (b)
the oxidized states in trichocyte keratin intermediate laments. The yellow
cylinders represent the helical (rod) domains of the protolaments. Each of the
eight protolaments consists of a pair of oppositely-directed molecular strands of
trichocyte keratin molecules. Note that the term protolament is a convenient one
to describe symmetry-related elements and does not imply that the protolament
has a role in IF assembly. The colored spheres represent the N-terminal domains
(green for the Up and blue for the Down strands) and C-terminal domains (red for
the Up and orange for the Down strands). Left: representation of the reduced form
of the trichocyte keratin IF. It is characterized by eight protolaments arranged on a
ring of radius 3.5 nm. Each protolament is related to its predecessor by a rotation of
45 and an axial translation of 16.85 nm. The terminal domains lie on a two-start
left-handed helix, and would lead to the expectation of diagonal banding with a
spacing of about 22 nm. Right: representation of the oxidized form of the
trichocyte keratin IF. One of the protolaments is centrally located and the other
seven surround it on a ring of radius 3.0 nm. Each of the protolaments on the ring is
related to its predecessor by a rotation of 24.3 and an axial translation of 19.82 nm.
The terminal domains are now in much closer proximity and give the impression of
being located on a one-start helix of pitch 23.5 nm (Fraser and Parry, 2005).

Crick (1953) was undoubtedly the initiator of the heptad concept and its implication for coiled-coil structure. However, almost
40 years later it became clear to me and long-time friend and colleague Carolyn Cohen that heptads were not just a feature of abrous proteins. They were present too in globular proteins, often
as oligomerization motifs. We pointed this out in several papers
(Cohen and Parry, 1986, 1990, 1994) and each subsequently became highly cited. Using the heptad concept we applied this directly in a study of the three a-helix motif that characterizes
the structure of members of the spectrin superfamily (Parry
et al., 1992). The model thus deduced proved closely similar to
that subsequently determined at atomic resolution (Yan et al.,
1993).
Studies of the limited sequences available for coiled-coil proteins revealed that two- and three-stranded coiled coils showed
subtle differences in the distribution of residue types within specic positions of the heptad (Parry, 1982; Conway and Parry,
1990, 1991). This allowed predictions to be made about the numbers of chains involved in the molecular structure from a study of
sequence alone. Several methods to identify the heptad-containing

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

sequences and the likely numbers of chains that assemble were

subsequently devised and these became increasingly sophisticated
(and successful) with time (see, for example, Lupas et al., 1991;
Berger et al., 1995; Walshaw and Woolfson, 2001; Delorenzi and
Speed, 2002; Lupas and Gruber, 2005; Woolfson, 2005).
Another feature of great interest in this eld was that very few
sequences contained an unbroken run of heptad repeats as evident
in the very rst complete sequence of an a-brous protein (atropomyosin). I collected as many heptad-containing sequences
as I could nd and was able to show that breaks in heptad phasing
corresponding to deletions of three residues (a stutter), four residues (a stammer), and six residues (a skip which is equivalent to
a pair of stutters or, alternatively, to the insertion of a single residue) were all quite common, in contrast to other possible combinations. In conjunction with Jerry Brown and Carolyn Cohen we were
able to model the stutter and stammer, respectively, in terms of a
local unwinding and overwinding of the coiled coil (Brown et al.,
1996). Subsequent X-ray crystal studies of proteins containing a
stutter (Strelkov and Burkhard, 2002; Alvarez et al., 2010) or a
stammer (Hartmann et al., 2009) have conrmed that the modeling was indeed correct.
A more recent review of coiled coil and a-helical bundle structures written in conjunction with John Squire and Bruce Fraser
(Parry et al., 2008) illustrates the massive increase in the observation of heptad (seven residues), and closely related hendecad (11
residues) and pentadecad repeats (15 residues) in diverse proteins.
Hendecad and pentadecad repeats have been studied crystallographically and shown to correspond to a near parallel bundle of
a-helices (Parry, 2006; Nicolet et al., 2010) and a right-handed
coiled coil (Khnel et al., 2004; Koretke et al., 2006; Alvarez
et al., 2010), respectively. Coiled coils with up to seven strands
have now been identied, as have coiled coils with parallel and
antiparallel strands (Parry et al., 2008). De novo design based on
the sequence features in known coiled coils has correspondingly
blossomed.
6.6. Plakins
Over the years I was fortunate enough to become involved in a
wide diversity of protein structures. My contribution generally
that of analyzing new amino acid sequences to determine regions
of coiled coil or a-helical bundles and hence assessing potential
function was often relatively minor when compared to that of
the lead authors. It was, nonetheless, very rewarding from a personal perspective. Among the numerous analyses undertaken one
family of proteins stands out. These are the plakins and they include bullous pemphigoid antigen (BPA) with John Stanley (Tanaka
et al., 1991), periplakin with Fiona Watt (Ruhrberg et al., 1997),
epiplakin with Sakuhei Fujiwara (Fujiwara et al., 2001), and plectin
(Green et al., 1992). Foremost of the plakins, however, in terms of
points of interest to emerge was undoubtedly desmoplakin (Green
et al., 1990). This involved a close collaboration with Kathy Green
over several years. We were able to recognize the extent of the
(discontinuous) rod domain from the heptad substructure, and
were able to show that it was a two-stranded structure with both
chains parallel to one another and in axial register. The C-terminal
domain contained three 176 residue long plakin-repeat domains
(PRD) termed A, B and C, each with a 38 residue sub-repeat. Comparable PRDs were also a feature of BPA (one B and one C repeat),
epiplakin (13 B repeats), and plectin (ve B and one C repeat) (see
Leung et al., 2001 for a summary of the early work). The N-terminal
domain was shown to consist largely of many a-helical bundles
with antiparallel strands (so-called plakin domains). Subsequently,
others showed these to be spectrin repeats but at the time such sequences were not available for comparison. We were able to postulate that the N-terminal domain was embedded in the plaque

13

and that the C-terminal domain was involved in interactions with

the keratin intermediate laments. All of these predictions were
subsequently veried. Structures at the atomic level have now
been completed for a large fragment of the N-terminal plakin domain (Choi and Weis, 2011) and for the B and C PRDs (Choi et al.,
2002).
7. Summary
There are many other very important topics in structural biophysics, of course, that I have not discussed here but which have
contributed enormously to the success of the eld as a whole. This
personal retrospective has, by denition, necessarily made mention of just a selection of those areas in which I have been personally involved.
Acknowledgments
I have been fortunate that my research adventure has been so
full of interest (to me at least) and has had so many twists and
turns. Life has certainly not been dull. If there is any take home
message from this paper it is that so much more can be achieved
by pooling the talents of researchers across international boundaries than could possibly be achieved by any individual, irrespective of how talented he or she might be. Throughout my career I
have had the greatest good fortune in being able to collaborate
with so many friends and colleagues. Each of them has taught
me a great deal and I am delighted to acknowledge their contributions in whatever successes my research has had. Particular mention needs to be made not only of Bruce Fraser, Peter Steinert,
Alan Craig, John Squire and Carolyn Cohen, but also of Arthur Elliott, Tom MacRae, Eikichi Suzuki, George Rogers, Don Caspar, Kathy
Green, Andrew Miller, David Hulmes, Barbara (Brodsky) Doyle, Jrgen Schweizer, Lutz Langbein, Tammy (Smith) Lynch, James Conway, Elwyn Firth, Alasdair Steven, Ueli Aebi, Sergei Strelkov and
Harald Herrmann. Together, we have been able to gain insights
into the structure and function of a diverse range of brous proteins. Although space has dictated that my research with some of
you has not been described here in detail that should not be interpreted as indicating that your contributions have been of any less
importance to me. This paper is dedicated to my wife Jenny, our
children and grandchildren, and all of you, my co-authors, which
now number close to 300!.
References
Alvarez, B.H., Gruber, M., Ursinus, A., Dunin-Horkawicz, S., Lupas, A.N., Zeth, K.,
2010. A transition from strong right-handed to canonical left-handed
supercoiling in a conserved coiled-coil segment of trimeric autotransporter
adhesins. J. Struct. Biol. 170, 236245.
Berger, B., Wilson, D.B., Wolf, E., Tonchev, T., Milla, M., Kim, P.S., 1995. Predicting
coiled coils by use of pairwise residue correlations. Proc. Natl. Acad. Sci. USA 92,
82598263.
Bradbury, E.M., Brown, L., Elliott, A., Parry, D.A.D., 1965. The structure of the c-form
of polycaproamide (nylon 6). Polymer 6, 465477.
Brown, J.H., Cohen, C., Parry, D.A.D., 1996. Heptad breaks in a-helical coiled coils:
stutters and stammers. Proteins Struct. Funct. Genet. 26, 134145.
Brown, J.H., Kim, K.-H., Jun, G., Greeneld, N.J., Dominguez, R., Volkmann, N.,
Hitchcock-DeGregori, S.E., Cohen, C., 2001. Deciphering the design of the
tropomyosin molecule. Proc. Natl. Acad. Sci. USA 98, 84968501.
Brown, J.H., Zhou, Z., Reshetnikova, L., Robinson, H., Yammani, R.D., Tobacman, L.S.,
Cohen, C., 2005. Structure of the mid-region of tropomyosin: bending and
binding sites for actin. Proc. Natl. Acad. Sci. USA 102, 1887818883.
Choi, H.J., Weis, W.I., 2011. Crystal structure of a rigid four-spectrin-repeat fragment
of the human desmoplakin plakin domain. J. Mol. Biol. 409, 800812.
Choi, H.J., Park-Snyder, S., Pascoe, L.T., Green, K.J., Weis, W.I., 2002. Structures of two
intermediate lament-binding fragments of desmoplakin reveal a unique
repeat motif structure. Nat. Struct. Biol. 9, 612620.
Cohen, C., Parry, D.A.D., 1986. a-helical coiled-coils a widespread motif in
proteins. Trends Biochem. Sci. 11, 245248.
Cohen, C., Parry, D.A.D., 1990. a-helical coiled-coils and bundles: how to design an
a-helical protein. Proteins Struct. Funct. Genet. 7, 115.

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

14

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

Cohen, C., Parry, D.A.D., 1994. a-helical coiled coils: more facts and better
predictions. Science 263, 488489.
Cohen, C., Caspar, D.L.D., Parry, D.A.D., Lucas, R.M., 1971. Tropomyosin crystal
dynamics. Cold Spring Harb. Symp. Quant. Biol. 36, 205216.
Cohen, C., Caspar, D.L.D., Johnson, J.P., Nauss, K.M., Margossian, S.S., Parry, D.A.D.,
1972. Tropomyosintroponin assembly. Cold Spring Harb. Symp. Quant. Biol.
37, 287297.
Conway, J.F., Parry, D.A.D., 1990. Structural features in the heptad substructure and
longer range repeats of two-stranded a-brous proteins. Int. J. Biol. Macromol.
12, 328334.
Conway, J.F., Parry, D.A.D., 1991. Three-stranded a-brous proteins: the heptad
repeat and its implications for structure. Int. J. Biol. Macromol. 13, 1416.
Craig, A.S., Parry, D.A.D., 1981. Collagen brils of the vertebrate corneal stroma. J.
Ultrastruct. Res. 74, 232239.
Craig, A.S., Parry, D.A.D., 1989. Quantitation of corneal bril diameters from bony
sh using low temperature preparative methods. J. Ultrastruct. Mol. Struct. Res.
102, 276278.
Craig, A.S., Robertson, J.G., Parry, D.A.D., 1986. Preservation of corneal collagen bril
structure using low temperature procedures for electron microscopy. J.
Ultrastruct. Mol. Struct. Res. 96, 172175.
Craig, A.S., Eikenberry, E.F., Parry, D.A.D., 1987. Ultrastructural organization of
skin: classication on the basis of mechanical role. Connect. Tissue Res. 16,
213223.
Craig, A.S., Birtles, M.J., Conway, J.F., Parry, D.A.D., 1989. An estimate of the mean
length of collagen brils in rat tail-tendon as a function of age. Connect. Tissue
Res. 19, 5162.
Crewther, W.G., Dowling, L.M., Steinert, P.M., Parry, D.A.D., 1983. Structure of
intermediate laments. Int. J. Biol. Macromol. 5, 267274.
Crick, F.H.C., 1953. The packing of alpha helices: simple coiled coils. Acta
Crystallogr. 6, 689697.
DeLorenzi, M., Speed, T., 2002. An HMM model for coiled-coil domains and a
comparison with PSSM-based predictions. Bioinformatics 18, 617625.
Dowling, L.M., Parry, D.A.D., Sparrow, L.G., 1983. Structural homology between hard
a-keratin and the intermediate lament proteins, desmin and vimentin. Biosci.
Rep. 3, 7378.
Elliott, A., 1965. Use of toroidal reecting surfaces in X-ray diffraction cameras. J.
Sci. Instrum. 42, 312316.
Fraser, R.D.B., 2004. The structure of deoxyribose nucleic acid. J. Struct. Biol. 145,
184186.
Fraser, R.D.B., MacRae, T.P., 1976. The molecular structure of feather keratin. In:
Proc. 16th Int. Ornith. Congress, Australian Academy of Science, Canberra, pp.
443451.
Fraser, R.D.B., Parry, D.A.D., 1996. The molecular structure of reptilian keratin. Int. J.
Biol. Macromol. 19, 207211.
Fraser, R.D.B., Parry, D.A.D., 2005. The three-dimensional structure of trichocyte
(hard a-) keratin intermediate laments: features of the molecular packing
deduced from the sites of induced crosslinks. J. Struct. Biol. 151, 171181.
Fraser, R.D.B., Parry, D.A.D., 2007. Structural changes in the trichocyte intermediate
laments accompanying the transition from the reduced to the oxidised form. J.
Struct. Biol. 159, 3645.
Fraser, R.D.B., Parry, D.A.D., 2008. Molecular packing in the feather keratin lament.
J. Struct. Biol. 162, 113.
Fraser, R.D.B., Parry, D.A.D., 2011a. The structural basis of the lament-matrix
texture in the avian/reptilian group of hard b-keratins. J. Struct. Biol. 180, 391
405.
Fraser, R.D.B., Parry, D.A.D., 2011b. The structural basis of the two-dimensional net
pattern observed in the X-ray diffraction pattern of avian keratin. J. Struct. Biol.
176, 340349.
Fraser, R.D.B., Parry, D.A.D., 2012. The role of disulde bond formation in the
structural transition observed in the intermediate laments of developing hair.
J. Struct. Biol. 173, 117124.
Fraser, R.D.B., MacRae, T.P., Parry, D.A.D., Suzuki, E., 1969. The structure of b-keratin;
Appendix I inter sheet disorder; Appendix II intra sheet disorder. Polymer
10, 810826.
Fraser, R.D.B., MacRae, T.P., Parry, D.A.D., Suzuki, E., 1971. The structure of feather
keratin; Appendix I correction of observed intensities for disorientation and
nite layer line thickness; Appendix II transformation to curvilinear
coordinates on a left-handed helical ruled surface. Polymer 12, 3556.
Fujiwara, S., Takeo, N., Otani, Y., Parry, D.A.D., Kunimatsu, M., Lu, R., Sasaki, M.,
Matsuo, N., Khaleduzzaman, M., Yoshioka, H., 2001. Epiplakin, a novel member
of the plakin family originally identied as a 450-kDa human
epidermal autoantigen: structure and tissue localization. J. Biol. Chem. 276,
1334013347.
Geeves, M.A., Holmes, K.C., 2005. The molecular mechanism of muscle contraction.
Adv. Protein Chem. 71, 161193.
Green, K.J., Parry, D.A.D., Steinert, P.M., Virata, M.L.A., Wagner, R.M., Angst, B.D.,
Nilles, L.A., 1990. Structure of the human desmoplakins: implications for
function in the desmosomal plaque. J. Biol. Chem. 265, 26032612.
Green, K.J., Virata, M.L.A., Elgart, G.W., Stanley, J.R., Parry, D.A.D., 1992. A
comparative structural analysis of desmoplakin, bullous pemphigoid antigen
and plectin: members of a new gene family involved in organization of
intermediate laments. Int. J. Biol. Macromol. 14, 145153.
Hartmann, M.D., Ridderbusch, O., Zeth, K., Albrecht, R., Testa, O., Woolfson, D.N.,
Sauer, G., Dunin-Horkawicz, S., Lupas, A.N., Alvarez, B.H., 2009. A coiled-coil
motif that sequesters ions to the hydrophobic core. Proc. Natl. Acad. Sci. USA
106, 1695016955.

Herrmann, H., Strelkov, S.V., Feja, B., Rogers, K.R., Brettel, M., Lustig, A., Hner, M.,
Parry, D.A.D., Steinert, P.M., Burkhard, P., Aebi, U., 2000. The intermediate
lament protein consensus motif of helix 2B: its atomic structure and
contribution to assembly. J. Mol. Biol. 298, 817832.
Hulmes, D.J.S., Miller, A., Parry, D.A.D., Piez, K.A., Woodhead-Galloway, J., 1973.
Analysis of the primary structure of collagen for the origins of molecular
packing. J. Mol. Biol. 79, 137148.
Koretke, K.K., Szczesny, P., Gruber, M., Lupas, A.N., 2006. Model structure of the
prototypical non-mbrial adhesin YadA of Yersinia enterocolitica. J. Struct. Biol.
155, 154161.
Khnel, K., Jarchau, T., Wolf, E., Schlichting, I., Walter, U., Wittinghofer, A., Strelkov,
S.V., 2004. The VASP tetramerization domain is a right-handed coiled-coil based
on a 15-residue repeat. Proc. Natl. Acad. Sci. USA 101, 1702717032.
Langbein, L., Yoshida, H., Praetzel-Wunder, S., Parry, D.A.D., Schweizer, J., 2010.
The keratins of the human beard hair medulla. J. Invest. Dermatol. 130,
5573.
Lee, S.-C., Kim, I.-G., Marekov, L.N., OKeefe, E.J., Parry, D.A.D., Steinert, P.M., 1993.
The structure of human trichohyalin: potential multiple roles as a functional EFhand-like calcium-binding protein, a cornied cell envelope precursor, and an
intermediate lament-associated (cross-linking) protein. J. Biol. Chem. 268,
1216412176.
Lehman, W., Craig, R., Vibert, P., 1994. Ca2+-induced tropomyosin movement in
Limulus thin laments revealed by three-dimensional reconstruction. Nature
368, 6567.
Lehman, W., Vibert, P., Ulman, P., Craig, R., 1995. Steric-blocking by tropomyosin
visualized in relaxed vertebrate muscle thin laments. J. Mol. Biol. 251, 191
196.
Leung, C.L., Liem, R.K.H., Parry, D.A.D., Green, K.J., 2001. The plakin family. J. Cell Sci.
114, 34093410.
Lupas, A.N., Gruber, M., 2005. The structure of a-helical coiled coils. Adv. Protein
Chem. 70, 3778.
Lupas, A.N., Vandyke, M., Stock, J., 1991. Predicting coiled coils from protein
sequences. Science 252, 11621164.
Luzzati, V., Cesari, M., Spach, G., Vincent, J.M., 1961. La structure du poly-L-cglutamate de benzyle en solution. Conguration en hlice diffrente de lhlice
a et transitions entre forms hlicoidales. J. Mol. Biol. 3, 566584.
McLachlan, A.D., Stewart, M., 1976. The 14-fold periodicity in alpha tropomyosin
and the interaction with actin. J. Mol. Biol. 103, 271298.
Nicolet, S., Herrmann, H., Aebi, U., Strelkov, S.V., 2010. Atomic structure of vimentin
coil 2. J. Struct. Biol. 170, 369376.
Parry, D.A.D., 1975. Analysis of the primary sequence of a-tropomyosin from rabbit
skeletal muscle. J. Mol. Biol. 98, 519535.
Parry, D.A.D., 1982. Coiled-coils in a-helix-containing proteins: analysis of
the residue types within the heptad repeat and the use of these data in
the prediction of coiled-coils in other proteins. Biosci. Rep. 2,
10171024.
Parry, D.A.D., 1988. The molecular and brillar structure of collagen and its
relationship to the mechanical properties of connective tissue. Biophys. Chem.
29, 195209.
Parry, D.A.D., 2006. Hendecad repeat in segment 2A and linker L2 of intermediate
lament chains implies the possibility of a right-handed coiled coil structure. J.
Struct. Biol. 155, 370374.
Parry, D.A.D., Craig, A.S., 1977. Quantitative electron microscope observations of the
collagen brils in rat tail tendon. Biopolymers 16, 10151031.
Parry, D.A.D., Craig, A.S., 1979. Electron microscope evidence for an 80 unit in
collagen brils. Nature 282, 213215.
Parry, D.A.D., Elliott, A., 1965. X-ray diffraction patterns of liquid crystalline
solutions of poly-c-benzyl-L-glutamate. Nature 206, 616617.
Parry, D.A.D., Elliott, A., 1967. The structure of a paracrystalline phase of poly-cbenzyl-L-glutamate in dimethylformamide. J. Mol. Biol. 25, 113.
Parry, D.A.D., Squire, J.M., 1973. The structural role of tropomyosin in muscle
regulation: analysis of the X-ray diffraction patterns from relaxed and
contracting muscles. J. Mol. Biol. 75, 3355.
Parry, D.A.D., Steinert, P.M., 1995. Intermediate Filament Structure. Springer-Verlag,
Heidelberg, pp. 1183.
Parry, D.A.D., Steinert, P.M., 1999. Intermediate laments: molecular architecture,
assembly, dynamics and polymorphism. Q. Rev. Biophys. 32, 99187.
Parry, D.A.D., Suzuki, E., 1969a. Intrachain potential energy of the a-helix and of a
coiled-coil strand. Biopolymers 7, 189197.
Parry, D.A.D., Suzuki, E., 1969b. Interchain packing energies of multistranded ahelical ropes. Biopolymers 7, 199206.
Parry, D.A.D., Crewther, W.G., Fraser, R.D.B., MacRae, T.P., 1977. Structure of akeratin: structural implication of the amino acid sequences of the type I and
type II chain segments. J. Mol. Biol. 113, 449454.
Parry, D.A.D., Craig, A.S., Barnes, G.R.G., 1978a. Tendon and ligament from the horse:
an ultrastructural study of collagen brils and elastic bres as a function of age.
Proc. R. Soc. Lond. B 203, 293303.
Parry, D.A.D., Barnes, G.R.G., Craig, A.S., 1978b. A comparison of the size distribution
of collagen brils in connective tissues as a function of age and a possible
relationship between bril size distribution and mechanical properties. Proc. R.
Soc. Lond. B 203, 305321.
Parry, D.A.D., Steven, A.C., Steinert, P.M., 1985. The coiled-coil molecules of
intermediate laments consist of two parallel chains in exact axial register.
Biochem. Biophys. Res. Commun. 127, 10121018.
Parry, D.A.D., Dixon, T.W., Cohen, C., 1992. Analysis of the three-a-helix motif in the
spectrin superfamily of proteins. Biophys. J. 61, 858867.

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010

D.A.D. Parry / Journal of Structural Biology xxx (2013) xxxxxx

Parry, D.A.D., Marekov, L.N., Steinert, P.M., Smith, T.A., 2002. A role for the 1A and L1
rod domain segments in head domain organization and function of
intermediate laments: structural analysis of trichocyte keratin. J. Struct. Biol.
137, 97108.
Parry, D.A.D., Smith, T.A., Rogers, M.A., Schweizer, J., 2006. Human hair keratinassociated proteins: sequence regularities and structural implications. J. Struct.
Biol. 155, 361369.
Parry, D.A.D., Fraser, R.D.B., Squire, J.M., 2008. Fifty years of coiled-coils and ahelical bundles: a close relationship between sequence and structure. J. Struct.
Biol. 163, 258269.
Phillips, G.N., Lattman, E.E., Cummins, P., Lee, K.Y., Cohen, C., 1979. Crystal structure
and molecular interactions of tropomyosin. Nature 278, 413417.
Poole, K.J.V., Lorenz, M., Evans, G., Rosenbaum, G., Pirani, A., Craig, R., Tobacman, L.S.,
Lehman, W., Holmes, K.C., 2006. A comparison of muscle thin lament models
obtained from electron microscopy reconstructions and low-angle X-ray bre
diagrams from non-overlap muscles. J. Struct. Biol. 155, 273284.
Ruhrberg, C., Hajibagheri, M.A.N., Parry, D.A.D., Watt, F.M., 1997. Periplakin, a novel
component of cornied envelopes and desmosomes that belongs to the plakin
family and forms complexes with envoplakin. J. Cell Biol. 139, 18351849.
Smith, J.W., 1968. Molecular pattern in native collagen. Nature 256, 157158.
Smith, T.A., Parry, D.A.D., 2007. Sequence analyses of type I and type II chains in
human hair and epithelial keratin intermediate laments: promiscuous
obligate heterodimers, type II template for molecule formation and a
rationale for heterodimer formation. J. Struct. Biol. 158, 344357.
Smith, T.A., Parry, D.A.D., 2008. Three-dimensional modelling of interchain
sequence similarities and differences in the coiled-coil segments of keratin
intermediate lament heterodimers highlight features important in assembly. J.
Struct. Biol. 162, 139151.
Smith, T.A., Strelkov, S.V., Burkhard, P., Aebi, U., Parry, D.A.D., 2002. Sequence
comparisons of intermediate lament chains: evidence of a unique functional/
structural role for coiled-coil segment 1A and linker L1. J. Struct. Biol. 137, 128
145.
Smith, T.A., Steinert, P.M., Parry, D.A.D., 2004. Modeling effects of mutations in
coiled-coil structures: case study using epidermolysis bullosa simplex
mutations in segment 1A of K5/K14 intermediate laments. Proteins Struct.
Funct. Bioinf. 55, 10431052.
Squire, J.M., Morris, E., 1998. A new look at thin lament regulation in vertebrate
skeletal muscle. FASEB J. 12, 761771.
Steinert, P.M., Parry, D.A.D., 1993. The conserved H1 domain of the type II keratin 1
chain plays an essential role in the alignment of nearest neighbor molecules in
mouse and human keratin 1/keratin 10 intermediate laments at the two- to
four- molecule level of structure. J. Biol. Chem. 268, 28782887.
Steinert, P.M., Parry, D.A.D., Racoosin, E.L., Idler, W.W., Steven, A.C., Trus, B.L., Roop,
D.R., 1984. The complete cDNA and deduced amino acid sequence of a type II
mouse epidermal keratin of 60,000 Da: analysis of sequence differences
between type I and type II keratins. Proc. Natl. Acad. Sci. USA 81,
57095713.
Steinert, P.M., Parry, D.A.D., Idler, W.W., Johnson, L.D., Steven, A.C., Roop, D.R., 1985.
Amino acid sequences of mouse and human epidermal type II keratins of Mr
67,000 provide a systematic basis for the structural and functional diversity of

15

the end domains of keratin intermediate lament subunits. J. Biol. Chem. 260,
71427149.
Steinert, P.M., Marekov, L.N., Fraser, R.D.B., Parry, D.A.D., 1993a. Keratin
intermediate lament structure: crosslinking studies yield quantitative
information on molecular dimensions and mechanism of assembly. J. Mol.
Biol. 230, 436452.
Steinert, P.M., Marekov, L.N., Parry, D.A.D., 1993b. Conservation of the structure of
keratin intermediate laments: molecular mechanism by which different
keratin molecules integrate into pre-existing keratin intermediate laments
during differentiation. Biochemistry 32, 1004610056.
Steinert, P.M., Marekov, L.N., Parry, D.A.D., 1993c. Diversity of intermediate lament
structure: evidence that the alignment of coiled-coil molecules in vimentin is
different from that in keratin intermediate laments. J. Biol. Chem. 268, 24916
24925.
Steinert, P.M., Marekov, L.N., Parry, D.A.D., 1999. Molecular parameters of type IV ainternexin and type IVtype III a-internexinvimentin copolymer intermediate
laments. J. Biol. Chem. 274, 16571666.
Steinert, P.M., Parry, D.A.D., Marekov, L.N., 2003. Trichohyalin mechanically
strengthens the hair follicle: multiple cross-bridging roles in the inner root
sheath. J. Biol. Chem. 278, 4140941419.
Stone, D., Sodek, J., Johnson, P., Smillie, L.B., 1975. Tropomyosin: correlation of
amino acid sequence and structure. Proc. 9th FEBS Meeting (Budapest), vol. 31.
North Holland Publishing Company, Amsterdam, pp. 125136.
Strelkov, S.V., Burkhard, P., 2002. Analysis of a-helical coiled coils with the program
TWISTER reveals a structural mechanism for stutter compensation. J. Struct.
Biol. 137, 5464.
Tanaka, T., Parry, D.A.D., Klaus-Kovtun, V., Steinert, P.M., Stanley, J.R., 1991.
Comparison of molecularly cloned bullous pemphigoid antigen to
desmoplakin I conrms that they dene a new family of cell adhesion
junction plaque proteins. J. Biol. Chem. 266, 1255512559.
Vibert, P.J., Haselgrove, J.C., Lowy, J., Poulsen, F.R., 1972. Structural changes in actincontaining laments of muscle. Nat. New Biol. 236, 182183.
Walshaw, J., Woolfson, D.N., 2001. SOCKET: a program for identifying and analyzing
coiled-coil motifs within protein structures. J. Mol. Biol. 307, 14271450.
Wang, H., Parry, D.A.D., Jones, L.N., Idler, W.W., Marekov, L.N., Steinert, P.M., 2000. In
vitro assembly and structure of trichocyte keratin intermediate laments: a
novel role for stabilization by disulde bonding. J. Cell Biol. 151, 14591468.
Whitby, F.G., Phillips, G.N., 2000. Crystal structure of tropomyosin at 7 resolution.
Proteins Struct. Funct. Genet. 38, 4959.
Whitby, F.G., Kent, H., Stewart, F., Stewart, M., Xie, X., Hatch, V., Cohen, C., Phillips,
G.N., 1992. Structure of tropomyosin at 9 resolution. J. Mol. Biol. 227, 441
452.
Woolfson, D.N., 2005. The design of coiled-coil structures and assemblies. Adv.
Protein Chem. 70, 79112.
Wu, K.C., Bryan, J.T., Morasso, M.I., Jang, S.I., Lee, J.H., Yang, J.M., Marekov, L.N., Parry,
D.A.D., Steinert, P.M., 2000. Coiled-coil trigger motifs in the 1B and 2B rod
domain segments are required for the stability of keratin intermediate
laments. Mol. Biol. Cell 11, 35393558.
Yan, Y., Winograd, E., Viel, A., Cronin, T., Harrison, S.C., Branton, D., 1993. Crystal
structure of the repetitive segments of spectrin. Science 262, 20272030.

Please cite this article in press as: Parry, D.A.D. Fifty years of brous protein research: A personal retrospective. J. Struct. Biol. (2013), http://dx.doi.org/
10.1016/j.jsb.2013.10.010