The chiral structure of porous chitin within the wing-scales of Callophrys rubi.

Abstract

The structure of the porous three-dimensional reticulated pattern in the wing scales of the butterfly Callophrys rubi (the Green Hairstreak) is explored in detail, via scanning and transmission electron microscopy. A full 3D tomographic reconstruction of a section of this material reveals that the predominantly chitin material is assembled in the wing scale to form a structure whose geometry bears a remarkable correspondence to the srs net, well-known in solid state chemistry and soft materials science. The porous solid is bounded to an excellent approximation by a parallel surface to the Gyroid, a three-periodic minimal surface with cubic crystallographic symmetry I4₁32, as foreshadowed by Stavenga and Michielson. The scale of the structure is commensurate with the wavelength of visible light, with an edge of the conventional cubic unit cell of the parallel-Gyroid of approximately 310 nm. The genesis of this structure is discussed, and we suggest it affords a remarkable example of templating of a chiral material via soft matter, analogous to the formation of mesoporous silica via surfactant assemblies in solution. In the butterfly, the templating is achieved by the lipid-protein membranes within the smooth endoplasmic reticulum (while it remains in the chrysalis), that likely form cubic membranes, folded according to the form of the Gyroid. The subsequent formation of the chiral hard chitin framework is suggested to be driven by the gradual polymerisation of the chitin precursors, whose inherent chiral assembly in solution (during growth) promotes the formation of a single enantiomer.

Full text

Page 1

1
2
The chiral structure of porous chitin within the wing-scales of Callophrys rubi
3
G.E. Schröder-Turka,b,⇑, S. Wickhamc, H. Averdunka, F. Brinke, J.D. Fitz Geraldd, L. Poladianc, M.C.J. Largec,
S.T. Hydea
4
5
6
7
8
9
aDept. Applied Mathematics, Research School of Physics and Engineering, The Australian National University, Canberra 0200, Australia
bTheoretische Physik I, Friedrich-Alexander-Universität Erlangen-Nürnberg, Staudtstr. 7, 91058 Erlangen, Germany
cSchool of Physics, University of Sydney, NSW 2006, Australia
dResearch School of Earth Sciences, The Australian National University, Canberra 0200, Australia
eCenter for Advanced Microscopy, The Australian National University, Canberra 0200, Australia
10
1 2
13
14
15
16
17
a r t i c l ei n f o
Article history:
Received 21 October 2010
Received in revised form 20 December 2010
Accepted 12 January 2011
Available online xxxx
18
19
20
21
22
23
24
25
26
27
Keywords:
Cubic membranes
Chitin
Chirality
Biophotonics
Circular polarisation
Self-assembly
Gyroid
Green Hairstreak
2 8
a b s t r a c t
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
The structure of the porous three-dimensional reticulated pattern in the wing scales of the butterfly Call-
ophrys rubi (the Green Hairstreak) is explored in detail, via scanning and transmission electron micros-
copy. A full 3D tomographic reconstruction of a fragment of this material reveals that the
predominantly chitin material is assembled in the wing scale to form a structure whose geometry bears
a remarkable correspondence to the srs net, well-known in solid state chemistry and soft materials sci-
ence. The porous solid is bounded to an excellent approximation by a parallel/cmc surface to the Gyroid, a
three-periodic minimal surface with cubic crystallographic symmetry I4132, as foreshadowed by Sta-
venga and Michielson. The scale of the structure is commensurate with the wavelength of visible light,
with an edge of the conventional cubic unit cell of the cmc-Gyroid of approximately 310 nm. The genesis
of this structure is discussed, and we suggest it affords a remarkable example of templating of a chiral
material via soft matter, analogous to the formation of mesoporous silica via surfactant assemblies in
solution. In the butterfly, the templating is achieved by the lipid–protein membranes within the smooth
endoplasmic reticulum (while it remains in the chrysalis), that likely form cubic membranes, folded
according to the form of the Gyroid. The subsequent formation of the chiral hard chitin framework is sug-
gested to be driven by the gradual polymerisation of the chitin precursors, whose inherent chiral assem-
bly in solution (during growth) promotes the formation of a single enantiomer.
? 2011 Published by Elsevier Inc.
46
47
48
1. Introduction
49
50
51
52
53
54
55
56
57
58
59
60
61
The basis of structural colour is the interaction of light with a
periodic structure in one, two or three dimensions whose periodic-
ity is of a comparable size to the wavelength. These structures are
known as photonic crystals, and can result in strong reflection for a
range of wavelengths. The reflections are due to constructive inter-
ference, which creates a photonic band gap – a range of wave-
lengths that cannot propagate through the crystal. The central
wavelength and the width of the photonic band gap both depend
on the direction of propagation through the crystal. Complete pho-
tonic bandgaps imply that there is a range of wavelengths that can-
not propagate through the crystal from any direction. In principle
this property allows the structure to appear the same colour when
viewed from any angle – however, a requirement of such a
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
bandgap is that the microstructure has a refractive index contrast
exceeding two (Joannopoulos et al., 2008).
The technological applications of photonic crystals however go
far beyond colouration. The exquisite control they offer over the
emission and transmission of light has led them to be used for such
diverse applications as optical security devices (van Renesse,
1997), solar cells (Bermel et al., 2007), low threshold lasers
(Akahane et al., 2003; Gong et al., 2010) and for displays (Ha
et al., 2008) and as an enabling technology for photonic chips
(Joannopoulos et al., 2008; Ha et al., 2008). The fact that such struc-
tures also exist in nature (such as in the wing-scales of certain but-
terfly species) is not only intriguing, but also they offer important
fabrication and design insights. Three-dimensional photonic crys-
tals are challenging to make, and the refractive index contrasts in
organic materials is much smaller than is possible for inorganic
systems, making it impossible to achieve complete photonic
bandgaps. The vast database of natural optical microstructures
found in biology has emerged through many generations of evolu-
tionary optimisation and the microstructures offer ingenious
implementation of polarisation effects, compound structures and
randomisation to achieve optical functionality, despite the limited
1047-8477/$ - see front matter ? 2011 Published by Elsevier Inc.
doi:10.1016/j.jsb.2011.01.004
⇑Corresponding author at: Friedrich-Alexander-Universität Erlangen-Nürnberg,
Theoretische Physik I, Staudtstr. 7B, 91058 Erlangen, Germany. Fax: +49 9131
8528444.
E-mail address: Gerd.Schroeder-Turk@physik.uni-erlangen.de (G.E. Schröder-
Turk).
Journal of Structural Biology xxx (2011) xxx–xxx
Contents lists available at ScienceDirect
Journal of Structural Biology
journal homepage: www.elsevier.com/locate/yjsbi
YJSBI 5942No. of Pages 6, Model 5G
2 February 2011
Please cite this article in press as: Schröder-Turk, G.E., et al. The chiral structure of porous chitin within the wing-scales of Callophrys rubi. J. Struct. Biol.
(2011), doi:10.1016/j.jsb.2011.01.004

Page 2

83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
contrast in refractive indices available to biological materials.
Such solutions may be particularly important for low cost poly-
mer-based photonic devices, where the index contrast is also
restricted.
Butterflies exhibit a variety of optical microstructures (Ghirad-
ella, 1984; Ingram and Parker, 2008), and three-dimensional pho-
tonic crystals have now been identified in the wing-scales of
several butterfly species, including the papilionids Parides sesostris
and Teinopalpus imperialis and the lycaenids Mitoura gryneus, Mito-
ura siva, Callophrys dunetorum and Callophrys rubi. Photonic activity
is induced by polymerised chitin material (with lesser fractions of
unidentified biomolecular species) that is structured at optical
wavelengths. However the size of these structures, their complex
topology and natural variation within a single wing-scale and be-
tween distinct specimens has made conclusive structural assigna-
tion difficult, if not uncertain. The earliest proposed structure for
C. rubi was a simple cubic array of polymeric chitin spheres
(Morris, 1975), while later studies suggested face-centred cubic
packings (Ghiradella and Radigan, 2005) and most recently, a
three-dimensional connected network related to the Gyroid
structure (Michielsen and Stavenga, 2008; Michielsen et al.,
2010; Saranathan et al., 2010). Face-centred cubic structures have
been proposed for a number of species (Vukusic and Sambles,
2003; Prum et al., 2006; Kertész et al., 2006), while a triclinic struc-
ture has been proposed for T. imperialis (Argyros et al., 2002). To
date, structural studies have relied on indirect methods, from
analysis of earlier electron micrographs of two-dimensional
sections (Michielsen and Stavenga, 2008; Michielsen et al., 2010)
to small-angle scattering X-ray (Saranathan et al., 2010).
Here we give the first direct three-dimensional structural data
for the organised chitin network found in wing-scales of C. rubi.
The excellent resolution of the data allows us to quantitatively
compare the structure to the Gyroid, resolving definitively any
doubts regarding the occurrence of this intriguing structure in
the wing-scales of C. rubi. We have performed electron tomography
on a single sample of the wing scales of C. rubi. The conclusion,
based on skeletonisation of the chitin phase and on explicit com-
parison of the imaged interface to a mathematical model surface,
is that the spatial structure of the investigated probe is commensu-
rate with the channel structure to one side of the Gyroid surface (a)
single Gyroid structure, based on the srs net (O’Keeffe et al., 2008)
withcubic symmetry group
a = (311 ± 5) nm. The structure is illustrated in Fig. 1.
I4132,and latticeparameter
126
2. Structure determination from 3D electron tomography
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
Dried specimen of C. rubi were purchased from a commercial in-
sect supplier (www.insectcompany.com). Small samples of the
green wing areas were prepared for TEM tomography data collec-
tion in the standard manner (Ghiradella, 1985). Wing pieces of
approximately 1 cm2were treated with a primary fixative (2.5%
glutaraldehyde in 0.1 M phosphate butter at pH 7.2), rinsed with
0.1 M buffer and then treated with a secondary fixative (1% Os-
mium tetraoxide in 0.1 M phosphate buffer). They were then dehy-
drated in a graded series of ethanol (first 50% ethanol in water,
then 70%, 90%, 99%, and finally 100% ethanol), infiltrated in Spurr’s
Resin under mild vacuum (approx. 400 torr) and left to polymerise
at 60 ?C. The embedded blocks were sectioned using an ultramicro-
tome and the sections picked up on copper grids and stained with
uranyl acetate. A slice of thickness 500 nm was chosen, as speci-
mens of this thickness had been successfully imaged by TEM pre-
viously (Argyros et al., 2002). Colloidal gold particles (diameter
10 nm) were embedded in the sample on both sides of the sections
to act as fiducial markers for the tomographic reconstruction. Dual-
axis tilt series were obtained on a Tecnai TF30 300 kV Transmission
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
Electron Microscope, with a Gatan 650 tilt-rotate holder, using the
SerialEM program for automated data collection. Tilt images were
collected in 1? increments for the range [?60?, 60?], at a magnifica-
tion of 20,000?. Separate samples were prepared for SEM and the
TEM (Figs. 4,5) by mounting wing scales on copper grids then
exposing the scales to a 3–5 kV Ar ion beam for 1–10 minutes.
The scanning electron samples were milled in a beam set at 15?
to the exposed flat scale sample, which was glued to the grid then
imaged in a Hitachi 4300S/N FESEM operating at 15 keV. Transmis-
sion electron micrographs samples were suspended within grid
gaps and milled from both sides for a similar time. Those samples
were observed in a Philips CM300, operated at 300 kV.
Tomograms were generated from the data and combined using
the software package IMOD (Kremer et al., 1996). The resulting
density dataset has voxel size dx ¼ ð1:28 ? 0:08Þ nm and overall
size of approximately 2100 ? 1900 ? 400 nm3.The largest rectan-
gular subset representing an ordered structure without grain
boundaries has size 850 ? 1280 ? 400 nm3; this subset is referred
to as subset ‘‘L’’. The analyses below are carried out on this subset
‘‘L’’ and also on a smaller subset, called ‘‘S’’, of approximate size
640 ? 640 ? 400 nm3.
The grey-scale density of the electron-tomography dataset was
smoothed with a Gaussian kernel of width r ¼ 6 nm. It was seg-
mented by the converging active contour method (CAC (Sheppard
et al., 2004)), followed by removal of small isolated clusters such
that the chitin and the void phase form a pair of single-connected
components. (The CAC method uses a combination of watershed
Fig. 1. Spatial structure of the chitin phase of C. rubi. The right fraction of the three-
dimensional body represents a subset of the tomographic data, segmented to yield
50% volume fraction, suitably rotated. The left side represents a solid body bounded
by Schoen’s triply-periodic Gyroid minimal surface. Also shown is a single srs
network tracing the centres of the void phase (orange) and projections onto the
planes perpendicular to one of the threefold rotation and threefold screw axes
(bottom) and one of the fourfold rotation and fourfold screw axes (top).
2
G.E. Schröder-Turk et al./Journal of Structural Biology xxx (2011) xxx–xxx
YJSBI 5942No. of Pages 6, Model 5G
2 February 2011
Please cite this article in press as: Schröder-Turk, G.E., et al. The chiral structure of porous chitin within the wing-scales of Callophrys rubi. J. Struct. Biol.
(2011), doi:10.1016/j.jsb.2011.01.004

Page 3

173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
and active contour methods. Initial low and high phase seed re-
gions were chosen; these regions were then evolved according to
a speed function dependent on gradient and intensity.)
The sample porosity cannot be precisely determined from the
3D density dataset: the obtained volume fraction depends on the
choice of segmentation parameters. We note however that this
uncertainty does not weaken our structural analysis below, since
both the quality of the segmentation and the variations of the
interface position from the corresponding parallel surface to the
Gyroid minimal surface are similar for all intermediate parameter
values of the segmentation parameter. Therefore, a precise esti-
mate for the volume fraction / of the chitin phase cannot be ex-
tracted. A rough estimate, determined by the minimal and
maximal volume fraction beyond which the segmentation does
not yield an ordered connected phase, is 15% < / < 70%.
A triangulation of the interface between the chitin and the void
phase, as given by the segmented data, is obtained by the Marching
Cubes algorithm (Lorensen and Cline, 1987) followed by mesh dec-
imation. We then estimated a best fit of the resulting interface to a
‘‘parallel–Gyroid’’ surface as follows. The Gyroid minimal surface,
discovered by Alan Schoen (Schoen, 1970), bisects space into a pair
of labyrinths; filling one labyrinth with chitin results in a pattern
with 50% porosity. A one-parameter (x say) family of parallel–
Gyroid surfaces form by displacing the Gyroid interface an equal
distance x at all points along the surface normal of each point.
The volume fraction of the parallel–Gyroid formed by a displace-
ment of x is determined by the Steiner formulae /ðxÞ ¼
1=2 þ A0x=a3ð1 ? 2pvx2=3Þ with A ¼ 3:09 and v ¼ ?8 for cubic lat-
tice parameter a if r=a < 0:19, (Schröder et al., 2003; Schröder-Turk
et al., 2007). We chose the parallel–Gyroid interface within that
family whose porosity best matched the selected porosity. The
parameters of a transformation (consisting of three rotations, a
translation and a rescaling of the coordinates by a factor) were
determined that minimised the average square distance (1/
N)PN
face, with the usual distance function, dðpiÞ ¼ jpi? fðpiÞj. The
optimal orientation, or rather an approximation thereof, is deter-
mined by a sequence of Monte-Carlo-like small random moves
(translations, rescaling, rotation of the whole dataset) combined
with a visual choice of a suitable starting configuration.
Following this rigid-body transformation of the tomographic
data representing the chitin interface, it was visually evident that
the structure resembled closely the corresponding Gyroid inter-
face, for all analysed segmentation parameters that gave reason-
able porosities /, see Fig. 1. Further, a representation of the
underlying net of the chitin phase by its thinned centred skeleton
(its medial axis computed by distance ordered homotopic thin-
ning) is in good agreement with the srs net that lines each channel
of the Gyroid or its parallel relatives, see Fig. 2.
The width of the distribution of distances d(pi) to the corre-
sponding parallel Gyroid interface, shown in Fig. 3, affords a quan-
titative measure for the fidelity of the match between the porous
chitin matrix and the Gyroid geometry. Perfect congruence would
imply an infinitely sharp distribution; the fact that the distribution
is approximately given by a Gaussian distribution, with maximal
likelihood for the points to be on the corresponding parallel Gyroid
interface and rapid decay of this probability away from this inter-
face, offers firm quantitative evidence that the porous chitin struc-
ture is given by the Gyroid I4132 structure. The fact that the
deviations from the Gyroid structure are Gaussian is commensu-
rate with these deviations being caused by noise or other small-
scale deviations. The agreement is similarly good for all porosities
analysed. While analysis on the smaller subset ‘‘S’’ yields the clos-
est match, the match of the larger subset ‘‘L’’ remains very
convincing.
i¼1½dðpiÞ?2between the N vertices piof the interface triangula-
tion to the nearest point fðpiÞ on the model parallel–Gyroid inter-
239
240
241
242
243
244
245
246
247
248
249
250
This structural stability, regardless of porosity is striking.
Though we provide detailed analysis of only a small region of the
wing-scale of a single specimen, the analysis allows us to conclude
that the structure based on a parallel–Gyroid interface is formed in
C. rubi. Lower magnification views of a single wing-scale (of a dif-
ferent specimen) show that there are significant variations in
porosity in different regions of the scale, varying approximately
over the range of values explored in our quantitative analysis,
see Fig. 4.
In addition, it is clear that these crystalline domains are rather
small, extending only over a few unit cells in any direction. Often,
the angles between neighbouring domains are small, and distinct
Fig. 2. Voxelised Medial axis of the chitin phase (blue) superposed on the single srs
graph of symmetry I4132. (a) View along a fourfold direction. (b) View along a
threefold axis (animated versions of these images are provided in the supplemen-
tary material). (For interpretation of the references to colour in this figure legend,
the reader is referred to the web version of this article.)
G.E. Schröder-Turk et al./Journal of Structural Biology xxx (2011) xxx–xxx
3
YJSBI 5942 No. of Pages 6, Model 5G
2 February 2011
Please cite this article in press as: Schröder-Turk, G.E., et al. The chiral structure of porous chitin within the wing-scales of Callophrys rubi. J. Struct. Biol.
(2011), doi:10.1016/j.jsb.2011.01.004

Page 4

251
252
253
254
255
structures within the grain boundaries are not seen; the chitin net-
work appears to traverse distinct ‘‘grains’’ uninterrupted, though
significant misorientations between neighbouring domains have
been reported in other studies (Michielsen and Stavenga, 2008).
Typical domain sizes in a high-porosity region of a single
256
257
258
wing-scale analysed to date are between 5 and 10 lm; each do-
main gives slightly different optical activity, homogenising some-
what the optical activity of the scale (Fig. 5).
259
3. Chirality
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
A remarkable feature of the lm-scale chitin framework is its
chirality. The srs net that describes the channel array of the struc-
ture is inherently handed. A recent study has suggested that both
enantiomers are to be found in wing scales of the related lycaenid
C. dumetorum (Saranathan et al., 2010). This suggestion was based
on the observation of both right- and left-handed helices in micro-
graphs. However, their suggestion is inconclusive in our view, gi-
ven the presence of both left- and right-handed helices in a
single enantiomer of srs (parallel to the <100> and <111> axes
of the cubic lattice with slightly different radii and pitches equal
to the lattice parameter and 3
respectively). The possibility of a single enantiomer of chitin with-
in individual wing-scales, within a single butterfly or indeed within
the species can therefore not be excluded on current evidence; fur-
ther studies are essential to resolve this intriguing question.
ffiffiffip
=2 times the lattice parameter,
275
4. Closing
276
277
278
279
280
281
282
283
284
285
286
287
The finding that the spatial structure of C. rubi wing scales is a
chiral photonic crystal is significant for a variety of reasons and this
ultrastructural chirality is in principle independent of the molecu-
lar-scale chirality of chitin. First, the material is optically active,
that is, it rotates the polarisation of incoming light. This is particu-
larly interesting in the context of photonic crystals, as it has been
argued that such structures could result in polarisation bandgaps:
frequency bands in which one polarisation state can be transmitted
through the crystal, while the other is reflected (Poladian et al.,
2009). Detailed modelling of the photonic features of this chitin
structure confirm the presence of partial band gaps for circularly
polarised light (Saba et al., in preparation).
0
0.2
0.4
0.6
0.8
1
1.2
-0.2 -0.1 0
D/a
0.1 0.2
P [a.u.]
S 27%
S 35%
S 50%
S 56%
L 57%
Fig. 3. Distribution of distances from the chitin interface in the optimally oriented
tomographic dataset to the nearest point on the corresponding Gyroid surface. The
lattice parameter, also determined by the Monte Carlo technique described in the
text, is a = (311 ± 5) nm. The largest possible deviation from the minimal Gyroid
interface (applicable to / = 50%) is given by the maximal domain size 0.23 a of the
Gyroid minimal surface (Schröder et al., 2003).
Fig. 4. Scanning electron micrographs of different regions within a single wing-
scale of C. rubi: (left) 10,000? magnification with a 5 lm scale bar; (right) 25,000?
magnification with a 2 l scale bar. Note the differences in porosity and the presence
of smaller domains, separated by small-angle grain boundaries (the SEM instru-
ment used was a Hitachi 4300S/N FESEM).
Fig. 5. Transmission electron micrograph of a small region within a single wing-
scale of C. rubi (2 lm scale bar). This wing-scale has been thinned by ion milling and
the viewing direction here is perpendicular to the plane of the scale. A number of
distinct domains are imaged, each with its chitin framework pattern distinctly
misoriented with respect to its neighbours. Domain boundaries have been marked
with white lines to aid visualisation. Inset: The variation of optical activity within
the wing-scale, likely corresponding to domains like those imaged by TEM, can be
seen in this optical micrograph (reflected light) of a single scale lying flat. The scale
bar is 10 lm.
4
G.E. Schröder-Turk et al./Journal of Structural Biology xxx (2011) xxx–xxx
YJSBI 5942 No. of Pages 6, Model 5G
2 February 2011
Please cite this article in press as: Schröder-Turk, G.E., et al. The chiral structure of porous chitin within the wing-scales of Callophrys rubi. J. Struct. Biol.
(2011), doi:10.1016/j.jsb.2011.01.004

Page 5

288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
Even more intriguing are the implications for metamaterials.
Recent intense interest in this field has been driven by the theoret-
ical possibility of designing structures to produce a range of
extraordinary effects, including negative refractive index materials,
electromagnetic cloaking and manipulation of the near-field (such
as superlensing and hyperlensing). Conventional approaches to
this problem involve designing structures to manipulate the mag-
netic and electrical response of the material (specifically the per-
mittivity l and permeability ?) to induce negative refractive
index. However, the refractive index of highly chiral materials
can be negative though both l and ? remain positive (Pendry,
2004; Zhou et al., 2009; Zhang and Cui, 2007). Combined with
the availability of chemical templating mechanisms to convert chi-
tin structures into e.g. rutile (TiO2) structures (Weatherspoon et al.,
2008) butterfly scales may provide an intermediate method for
generation of chiral inorganic photonic crystals with lattice size
corresponding approximately 300 nm while no suitable synthetic
self-assembly mechanism for this length scale is available.
The formation mechanism of this extraordinarily complex chi-
tin framework in vivo is of interest on many fronts. Biological stud-
ies suggest that the chitin slowly polymerises in the larval stage of
the butterfly, guided by mutual folding of the smooth endoplasmic
reticulum (SER) and plasma lipid–protein membranes to give a
Gyroid-like pattern (Ghiradella, 1994, 2008; Saranathan et al.,
2010). This ‘‘cubic membrane’’ folding geometry has been detected
in many membrane organelles, across virtually all kingdoms of life
(Landh, 1995; Almsherqi et al., 2006; Hyde et al., 1997). The same
structure, realised at a much smaller length-scale, is well-known in
hard atomic and molecular as well as soft molecular materials
(Hyde et al., 2008), including lyotropic liquid crystals of lipids
(and lipid–protein mixtures) in water (Larsson, 1989), where its
presence is a signature of molecular self-assembly into a 2D layer
subject to bending energy (Helfrich and Rennschuh, 1990; Hyde,
1990), with a strongly preferred local membrane (Gaussian) curva-
ture and a preference for uniform channel sizes (Hyde et al., 1997;
Schröder-Turk et al., 2006). Much coarser patterns, but still cubic,
membranes have been observed in butterfly larvae (Ghiradella,
1994). It is therefore reasonable to conclude that the chitin net-
work emerges via self-assembly of the SER membranes, which then
template the harder chitin matrix, qualitatively similar to the syn-
thetic route to formation of mesoporous materials.
Despite the clear links between this biological material and con-
densed materials formed by self-assembly in vitro, aspects of the
genesis of this self-assembled structure remain poorly understood.
In particular, no clear explanation has been given for the stability
of such a highly swollen Gyroid pattern, whose lattice parameter
is two orders of magnitude greater than typical dimensions in li-
pid–water or lipid–protein–water mesophases (Larsson, 1989).
Electron micrographs indicate that the SER membrane – possibly
accompanied by the plasma membrane – folds as a coherent stack
of multiple bilayers, in contrast to the usual single bilayer charac-
teristic of cubic bicontinuous mesophases (Ghiradella, 1994); Sara-
nathan et al. (2010) suggest that single SER and plasma membrane
bilayers condense and fold in concert into the Gyroid morphology.
The enhanced membrane rigidity associated with a stack of two
bilayers compared with a single bilayer goes some way towards
explaining the crystallinity of this massively swollen structure, as
follows. Since the crystalline Gyroid geometry results from mini-
misation of membrane bending energy, the enhanced modulus of
bending rigidity associated with the double bilayer may explain
the enhanced swelling of the crystalline pattern without melting.
A second aspect of this structure requires further investigation.
The chitin framework contains a single, chiral srs net, in distinction
to the pair of (enantiomeric) srs nets that line Gyroid channels.
Straightforward templating of chitin within the pair of aqueous
channels formed by a multilayer stack folded onto the Gyroid, or
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
within the space between parallel membranes, should result in a
pair of interwoven (left- and right-handed) frameworks or an achi-
ral sheet. What gives rise to this symmetry-breaking, resulting in a
chiral structure? We do not yet know, and any explanation must
await more careful studies of the specific chirality, if any, of the
chitin matrix. It is tempting to suggest that the chitin oligomeric
precursors themselves preferentially grow within a single aqueous
channel. Chitin oligomers are themselves chiral, chiral aggregation
has been seen in a number of biological chitin materials
(Giraud-Guille et al., 2004) and chitin suspensions in water are
known to spontaneously form (chiral) cholesteric mesophases
(Revol and Marchessault, 1993). A similar mechanism has been
employed in the lab to prepare single chiral srs mesoporous net-
works. In that case, chiral (sucrose) precursors were absorbed
and then polymerised within MCM-48 (an amorphous mesoporous
silica with the Gyroid structure); the carbonaceous materials that
remained after removal of the silica was itself chiral, with symme-
try I4132, indicative of a single srs network (Ryoo et al., 1999). A
similar mechanism at work in C. rubi would likely imply just a sin-
gle enantiomer of chitin in the wing-scales, a scenario whose
likelihood is in our view at present unclear. This scenario is,
though, not essential. A second example of a chiral srs mesoporous
network, again templated from MCM-48, but made of platinum,
shows that a chiral species is not needed to form srs (Terasaki
et al., 2002). We note in this case, however, that either enantiomer
of this Pt network is equally likely to form, though the origin
of the single network remains unclear. This example shows, how-
ever, that the formation of a single srs need not imply a single
enantiomer.
Clearly, full understanding of the genesis of this remarkable chi-
tin material must await further studies. However, the clear similar-
ities to synthetic self-assembled meso-scale materials suggest
concrete future directions to explore in order to achieve an ulti-
mate goal of in vitro self-assembly of these pattern for further pho-
tonics research.
389
Acknowledgments
390
391
392
393
394
395
396
397
398
We thank Emine Korkmaz, Filip Braet, Tony Romeo and Ian
Kaplin, Julie Cairney (Electron Microscopy Unit, University of
Sydney). We also thank Adrian Sheppard (Australian National
University) and all other developers of the software package
mango. S.W. acknowledges a Travel and Access Program of the
Australian MicroscopyandMicroanalysis
(AMMRF) for funding her trip to the University of Queensland,
where the tomography was performed, and particularly thanks
Jamie Riches for his invaluable assistance while she was there.
ResearchFacility
399
References
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
Akahane, Y., Asano, T., Song, B., Noda, S., 2003. High-Q photonic nanocavity in a two-
dimensional photonic crystal. Nature 425, 944–947.
Almsherqi, Z., Kohlwein, S., Deng, Y., 2006. Cubic membranes: a legend beyond the
flatland of cell membrane organization. J. Cell Biol. 173, 839–844.
Argyros, A., Manos, S., Cox, G., Dwarte, D., Large, M., 2002. Electron tomography and
computer visualisation of a 3-dimensional ‘‘photonic’’ crystal in a butterfly
wing-scale. Micron 33, 483–487.
Bermel, P., Luo, C., Zeng, L., Kimerling, L.C., Joannopoulos, J.D., 2007. Improving thin-
film crystalline silicon solar cell efficiencies with photonic crystals. Opt. Express
15, 16986–17000.
Ghiradella, H., 1984. Microscopic anatomy of invertebrates. Volume 11A of Series
Insecta. Chapter Structure of Iridescent Lepidopteran Scales: Variations on
Several Themes. Wiley-Liss, pp. 257–287.
Ghiradella, H., 1985. Structure and development of iridescent lepidopteran scales:
the papilionidae as a showcase family. Ann. Ent. Soc. Am. 2, 252–264.
Ghiradella, H., 1994. Structure of butterfly scales: patterning in an insect cuticle.
Microsc. Res. Tech. 27, 429–438.
Ghiradella, H., Radigan, W., 2005. Development of butterfly scales. II. Struts, lattices
and surface tension. J. Morphol. 150, 279–297.
Giraud-Guille, M., Belamie, E., Mosser, G., 2004. Organic and mineral networks in
carapaces, bones and biomimetic materials. C. R. Palevol 3, 503–513.
G.E. Schröder-Turk et al./Journal of Structural Biology xxx (2011) xxx–xxx
5
YJSBI 5942 No. of Pages 6, Model 5G
2 February 2011
Please cite this article in press as: Schröder-Turk, G.E., et al. The chiral structure of porous chitin within the wing-scales of Callophrys rubi. J. Struct. Biol.
(2011), doi:10.1016/j.jsb.2011.01.004

Page 6

421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
Gong, Y., Ellis, B., Shambat, G., Sarmiento, T., Harris, J., Vuckovic, J., 2010. Nanobeam
photonic crystal cavity quantum dot laser. Opt. Express 18, 8781–8789.
Ha, N., Ohtsuka, Y., Jeong, S., Nishimura, S., Suzaki, G., Takanishi, Y., Ishikawa, K.,
Takezoe, H., 2008. Fabrication of a simultaneous red-green-blue reflector using
single-pitched cholesteric liquid crystals. Nat. Mater. 7, 43–47.
Helfrich, W., Rennschuh, H., 1990. Landau theory of the lamellar-to-cubic phase
transition. In: Colloque de Physique C7-1990, pp. 189–195.
Hyde, S., 1990. Curvature and the global structure of interfaces in surfactant–water
systems. In: Colloque de Physique C7-1990, pp. 209–228.
Hyde, S., Andersson, S., Larsson, K., Blum, Z., Landh, T., Lidin, S., Ninham, B., 1997.
The Language of Shape, first ed. Elsevier Science B.V., Amsterdam.
Hyde, S., O’Keeffe, M., Proserpio, D., 2008. A short history of an elusive yet
ubiquitous structure in chemistry, materials and mathematics. Angew. Chem.
Int. Ed. 47, 7996–8000.
Ingram, A., Parker, A., 2008. A review of the diversity and evolution of photonic
structures in butterflies, incorporating the work of John Huxley (the Natural
History Museum, London from 1961 to 1990). Phils. Trans. R. Soc. B 363,
2465–2480.
Joannopoulos, J., Johnson, S., Winn, J., Meade, R., 2008. Photonic Crystals: Molding
the Flow of Light. Princeton University Press, Princeton.
Kertész, K., Bálint, Z., Vértesy, Z., Márk, G.I., Lousse, V., Vigneron, J.P., Rassart, M.,
Biró, L.P., 2006. Gleaming and dull surface textures from photonic-crystal-type
nanostructures in the butterfly Cyanophrys remus. Phys. Rev. E 74, 021922.
Kremer, J., Mastronarde, D., McIntosh, J., 1996. Computer visualization of three-
dimensional image data using imod. J. Struct. Biol. 116, 71–76.
Landh, T., 1995. From entangled membranes to eclectic morphologies: cubic
membranes as subcellular space organizers. FEBS Lett. 369, 13–17.
Larsson, K., 1989. Cubic lipid–water phases: structures and biomembranes aspects.
J. Phys. Chem. 93, 7304–7314.
Lorensen, W., Cline, H., 1987. Marching cubes: a high resolution 3d surface
construction algorithm. In: Computer Graphics (Proceedings of SIGGRAPH ’87),
pp. 163–169.
Michielsen, K., De Raedt, H., Stavenga, D., 2010. Reflectivity of the gyroid
biophotonic crystals in the ventral wing scales of the green hairstreak
butterfly, Callophrys rubi. J. R. Soc. Interface 7, 765–771.
Michielsen, K., Stavenga, D., 2008. Gyroid cuticular structures in butterfly wing
scales: biological photonic crystals. J. R. Soc. Interface 5, 85–94.
Morris, R., 1975. Iridescence from diffraction structures in the wing scales of
Callophrys rubi, the green hairstreak. J. Entomol. A 49, 149–154.
O’Keeffe, M., Peskov, M., Ramsden, S., Yaghi, O., 2008. The reticular chemistry
structure resource (rcsr) database of, and symbols for, crystal nets. Acc. Chem.
Res. 41, 1782–1789.
Pendry, J., 2004. A chiral route to negative refraction. Science 306, 1353–1355.
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
Poladian, L., Wickham, S., Lee, K., Large, M.C., 2009. Iridescence from photonic
crystals and its suppression in butterfly scales. J. R. Soc. Interface 6, S233–S242.
Prum, R., Quinn, T., Torres, R., 2006. Anatomically diverse butterfly scales all
produce structural colours by coherent scattering. J. Exp. Biol. 209, 748–765.
Revol, J., Marchessault, R., 1993. In vitro chiral nematic ordering of chitin
crystallites. Int. J. Biol. Macromol. 15, 329–335.
Ryoo, R., Joo, S.H., Jun, S., 1999. Synthesis of highly ordered carbon molecular sieves
via template-mediated structural transformation. J. Phys. Chem. B 103,
7743–7746.
Saba, M., Thiel, M., Turner, M., Hyde, S., Gu, M., Grosse-Brauckmann, K., Neshev, D.,
Mecke, K., Schröder-Turk, G., in preparation. Circular dichroism in biomimetic
cubic chiral nets.
Saranathan, V., Osuji, C.O., Mochrie, S.G.J., Noh, H., Narayanan, S., Sandy, A.,
Dufresne, E.R., Prum, R.O., 2010. Structure, function, and self-assembly of single
network gyroid (I4132) photonic crystals in butterfly wing scales. PNAS 107,
11676–11681.
Schoen, A., 1970. Infinite Periodic Minimal Surfaces without Self-Intersections.
Technical Report. NASA.
Schröder, G., Ramsden, S., Christy, A., Hyde, S., 2003. Medial surfaces of hyperbolic
structures. Eur. Phys. J. B 35, 551–564.
Schröder-Turk, G., Fogden, A., Hyde, S., 2006. Bicontinuous geometries and
molecular self-assembly: comparison of local curvature and global packing
variations in genus-three cubic, tetragonal and rhombohedral surfaces. Eur.
Phys. J. B 54, 509–524.
Schröder-Turk, G., Fogden, A., Hyde, S., 2007. Local v/a variations as a measure of
structural packing frustration in bicontinuous copolymer mesophases, and
prediction of an alternating Im?3m (I-WP) phase in block-copolymers with
polydispersity. Eur. Phys. J. B 59, 115–126.
Sheppard, A., Sok, R., Averdunk, H., 2004. Techniques for image enhancement and
segmentation of tomographic images of porous materials. Physica A 339,
145–151.
Terasaki, O., Liu, Z., Ohsuna, T., Shin, H., Ryoo, R., 2002. Microsc. Microanal. 8, 35–39.
van Renesse, R. (Ed.), 1997. Optical Document Security. Artech House. Boston,
London (Chapter 12).
Vukusic, P., Sambles, J., 2003. Photonic structures in biology. Nature 424, 852–855.
Weatherspoon, M., Cai, Y., Crne, M., Srinivasarao, M., Sandhage, K., 2008. 3d rutile
titania-based structures with morpho butterfly wing scale morphologies.
Angew. Chem. 120, 8039–8041.
Zhang, C., Cui, T.J., 2007. Negative reflections of electromagnetic waves in a strong
chiral medium. Appl. Phys. Lett. 91, 194101.
Zhou, J., Dong, J., Wang, B., Koschny, T., Kafesaki, M., Soukoulis, C.M., 2009. Negative
refractive index due to chirality. Phys. Rev. B 79, 121104.
Q1
506
6
G.E. Schröder-Turk et al./Journal of Structural Biology xxx (2011) xxx–xxx
YJSBI 5942No. of Pages 6, Model 5G
2 February 2011
Please cite this article in press as: Schröder-Turk, G.E., et al. The chiral structure of porous chitin within the wing-scales of Callophrys rubi. J. Struct. Biol.
(2011), doi:10.1016/j.jsb.2011.01.004