Conformational and Orientational Transformation of Silk Proteins in

Biomacromolecules 2008, 9, 2399–2407

2399

Conformational and Orientational Transformation of Silk Proteins in the Major Ampullate Gland of Nephila clavipes Spiders Thierry Lefe`vre,† Simon Boudreault,‡ Conrad Cloutier,‡ and Michel Pe´zolet*,† De´partement de Chimie-CERSIM-CREFSIP, and De´partement de Biologie, Universite´ Laval, Pavillon Alexandre-Vachon, Que´bec G1V 0A6, Canada Received April 11, 2008; Revised Manuscript Received June 23, 2008

The orientational and conformational transformation of the native liquid silk into a solid fiber in the major ampullate gland of the spider Nephila claVipes has been studied by Raman spectromicroscopy. The spectra show that the conformation of silk proteins in the glandular sac contains several secondary structure elements, which is consistent with intrinsically unfolded proteins. A few R-helices are also present and involve some alanine residues located in the polyalanine segments of the spidroin sequence. The conversion of the silk solution in the major ampullate gland appears to be a two-state process without intermediate states. In the first and second limbs of the duct, silk is isotropic and spidroins are generally native-like. β-Sheets start to develop between the second and the third limb of the duct, suggesting that early β-sheets are generated by shear forces. However, most of the β-sheets are formed between the draw down taper and the valve. The early β-sheets formed upward of the draw down taper might play the role of nucleation sites for the subsequent β-sheet aggregation. The alignment of the polypeptides chains occurs near the valve, revealing that orientational and conformational changes do not occur simultaneously. Extensional flow seems to be the driving force to produce the orientational order, which in turn is associated with the formation of the major part of the β-sheets. The slow evolution of the spidroin conformation up to the draw down taper followed by the rapid transformation between the drawn down taper and the valve may be important to achieve the optimal structure of the final fiber.

Introduction The conversion of the native liquid silk contained in the abdominal glands of spiders into insoluble fibers that are highly adapted to perform specific tasks is a fascinating phenomenon of Nature that is still not completely elucidated. The spinning process, in conjunction with the primary structure of the silk proteins, determines the structural organization of silk, which in turn determines its properties. There is, therefore, a great interest to understand the details of the spinning process, especially for fibers such as the dragline thread because this biomaterial displays remarkable mechanical properties.1,2 The dragline silk is secreted by the major ampullate gland that give rise to a hierarchically organized,3–10 semicrystalline1,2,11 material. The crystalline phase is made of short polyalanine segments that adopt the β-sheet conformation, whereas the amorphous phase is formed of glycine-rich protein sequences that most probably adopt 31-helices, turns, β-sheets, and other secondary structures.12–16 The crystalline β-sheets are strongly aligned along the fiber axis, whereas the remaining of the polypeptide chains are poorly oriented.13,16–18 The major ampullate (MA) gland of orb-weaving spiders, apart from the dragline, produces the radial threads, the mourning threads, as well as the frame of webs. It is a highly specialized secretory system divided into three main constitutive parts, which are from anterior to posterior: the tail, the sac, and the duct (Figure 1). The tail is a long corrugated tubular region where the majority of the proteins that form the fiber, the soTo whom correspondence should be addressed. Phone: 418-656-2481. Fax: 418-656-7916. E-mail: [email protected]. † De´partement de Chimie-CERSIM-CREFSIP. ‡ De´partement de Biologie.

Figure 1. Typical image of the MA gland of N. clavipes. All tissues surrounding the gland and the duct were removed, conserving only the anterior tissue attaching the tail to the body wall. The different parts of the glands are indicated. The numbers refer to the locations that were probed in the experiments shown in Figure 2, 4, and 6, although a different gland than that illustrated was used to record the Raman spectra.

called spidroins, are synthesized. The sac is a storage ampulla that contains the spinning dope. The distal part of the sac ends with a funnel that makes the transition toward a long tapered duct. The latter is composed of three limbs folded on themselves into a flat S shape and is acknowledged to be the area where the native liquid silk is transformed into a fiber. The duct ends with the spigot that leads the silk to the outside.

10.1021/bm800390j CCC: $40.75  2008 American Chemical Society Published on Web 08/15/2008

2400

Biomacromolecules, Vol. 9, No. 9, 2008

Figure 2. Microscope images of in situ silk exposed by dissection (white arrows) in the third limb of the MA gland of N. clavipes: (A) Middle of the third limb (point #09 in Figure 1); (B) between the valve and the draw down taper (point #11 in Figure 1); and (C) just beyond the valve (point #13 in Figure 1). All images were taken at 10× magnification.

In the literature, the sac is anatomically subdivided into two zones labeled A and B. Spidroins are secreted in the A-zone, whereas other proteins, most probably glycoproteins that seem to be involved in the coating of the fiber, are secreted in the B-zone.19,20 In the sac, the spinning solution is found under the form of spherical droplets of a few micrometers in diameter that elongate along the flow direction near the funnel.21 The native conformation of the spidroins corresponds to that of unfolded proteins22–24 containing random structural sequences, segments with a 31-helix (polyproline II-like) conformation,24 and a few R-helices.24 Silk in the ampulla and in the first and second limbs is in a liquid-crystalline state,20,25 which seems to be critical for the spinning process. The measured shear viscosity of the spinning dope is consistent with the liquidcrystalline nature of the silk and corresponds to a shear-thinning material.26,27 The shear viscosity is predicted to decrease as silk flows through the canal of the duct so that it makes the viscous silk solution ”spinnable” by the spider.26

Lefèvre et al.

It was recognized at least 40 years ago that the conversion of the fluid silk into a solid thread is not an extrusion process and that tension, due to a falling spider’s body weight, its displacement, or active pulling by its hind legs, is the driving force at the origin of the formation of the silk fiber.28 Intraabdominal body pressure would be only responsible for the continuous supply of liquid silk into the duct.29 However, shear forces along the luminal wall of the duct have also been proposed to play a major role in the spinning process.20,30 Shear forces and extensional flow may both potentially induce the formation of β-sheets and align the polypeptide chains. Experimentally, the extensional viscosity of the silk solution contained in the sac measured as a function of the applied strain tends to increase infinitely as a result of molecular orientation and water evaporation.26 The transformation of the silk has been initially localized at the valve, a cuticular thickening of the lining of the third limb of the duct wall of orb-weaving spiders located near the base of the spinneret or spigot (Figure 1). The valve may have different functions.28,29,31 It has been proposed to regulate the amount of fluid supplied to the spigot and to control the thread thickness (restriction die) and the force required for its extraction by acting as an anchoring point (clamping function).28,29 The restriction die function has recently been questioned, but the clamp action has been confirmed.32 In addition, it has been suggested that the valve may be used to restart the spinning mechanism in case of a breakage of the fiber.32 The start of β-sheets formation has recently been found in a more anterior region of the duct than the valve itself.33 The diameter of the tapered duct decreases as a two-stage hyperbolic curve from the funnel up to a zone called the draw down taper (which is in limb 3, see Figure 1), where the diameter decreases more rapidly, following a two-stage exponential function.21 The draw down taper in Nephila claVipes (N. claVipes) is located approximately 2.5 mm proximally from the valve. When confocal fluorescence microscopy of silk fixed and stained with Congo red is used, it has been found that the β-sheets start to develop from the draw down taper.33 The formation of β-sheets can be induced by a mechanical stress on the silk dope located just before this region of the duct.33 Despite several progresses, many questions remain unanswered relative to the molecular details and the location of the entire transformation of native silk into the final thread. In particular, it is not known how the conformational and orientational conversions occur and whether these phenomena are concomitant. It also has to be determined whether the conversion is a multistep process that displays possible intermediate states and whether significant changes occur in the first and second limbs. These questions are addressed here on the major ampullate (MA) gland of the spider N. claVipes by using polarized Raman spectromicroscopy. The combined analysis of several conformation-sensitive bands has allowed the detection of very small changes in the protein secondary structure occurring in the first two limbs of the duct. Larger modifications of the molecular orientation and conformation were observed in the third limb. A detailed analysis of the Raman spectra in the native silk before spinning is also presented.

Experimental Section N. ClaVipes spiders were collected in Florida. They were raised in the laboratory in 20 × 50 × 60 cm cages at 58 ( 5% relative humidity (RH) and 24 ( 2 °C. The spiders were fed four times a week with small crickets and with 3 drops of 10% w/v glucose solution per week. Because the experiments were carried out over 8 months, data were

Silk Proteins of Nephila clavipes Spiders


2401

Figure 3. Raman spectra of the native (wet) silk and dry silk probed in the area of the sac (B zone) of the MA gland of a N. clavipes spider. The “wet” silk corresponds to the fluid dope that was expelled from the ampulla of the gland (immersed in phosphate buffer) after making a small incision in the sac.

obtained on several spiders sacrificed at different stages of maturity and no difference was observed between mature on nonmature spiders. Prior to dissection, the spiders were allowed to walk for at least 60-80 cm, while trailing silk behind them. Spiders were then anesthetized with CO2 to immobilize them. It has been shown that the anesthesia of spiders with CO2 can alter the mechanical properties of the silk fibers obtained by forced drawing due to the effect of the anesthesia on the muscular network of the MA gland.34,35 However, to our knowledge, no study has shown any effect of anesthesia on the structure of the static silk dope contained in the duct of an inactive gland. Nevertheless, to minimize potential effects of anesthesia on the direct environment of the MA gland, the time period of anesthesia was kept as short as possible (3-4 min). Spiders were then dissected and rinsed with a phosphate buffer saline. The MA glands were carefully extracted, deposited on glass slides or polystyrene Petri dishes, and immersed in the same buffer to stabilize the environmental conditions of the glands. Although the protein conformation can be investigated directly in the sac of the MA gland,24 the epithelium of the duct wall is too thick to allow a direct Raman observation of the silk through it. Therefore, the epithelium was gently excised at predetermined points to directly expose the silk dope to the laser beam for probing its conversion toward the final fiber as it progresses along the duct. To obtain detailed spectroscopic information on silk, spectra with a very high signal-to-noise ratio are needed. To this end, spectra were recorded in the dry state rather than in the wet state because the Raman signal is significantly smaller in the latter case. As shown below, it appeared that the molecular structure of the spidroins is not affected by drying so that dehydration represents a simple means to obtain a snapshot of the events occurring within the duct. Some typical images of silk regions probed in the course of this work are shown in Figure 2. A great care has been taken to probe loci in the duct such that the silk material was minimally perturbed during the dissection procedure (i.e., especially that silk had not been stretched). Despite these precautions, it appeared that the location of the changes may exhibit some variability that is not related to the dissection procedure. As a matter of fact, we have observed that the two contralateral MA glands of a same spider do not necessarily process the silk fully symmetrically, that is, the spectral changes do not occur precisely at the same location along the ducts of the two glands. Such observations have already been noted by Work.36

In addition, variability is also observed between spider individuals as may be expected. Therefore, the exact location of the molecular modifications may vary to some extent, but the main tendencies given in the present study are reproducible. Raman spectra were recorded at 22.0 ( 0.5 °C and 20 ( 5% RH using a LABRAM 800HR Raman spectrometer (Horiba Jobin Yvon, Villeneuve d’Ascq, France) coupled to an Olympus BX 30 fixed stage microscope. The excitation light source was the 514.5 nm line of an Ar+ laser (Coherent, INNOVA 70C Series Ion Laser, Santa Clara, CA). The laser beam was focused with 50× or 100× long working distance objectives, generating an intensity at the sample of approximately 5-10 mW. The confocal hole and the entrance slit of the monochromator were generally fixed at 200 and 100 µm, respectively. Data were collected by a one-inch open electrode Peltier-cooled CCD detector (1024 × 256 pixels). For polarization measurements, a half-wave plate (Melles Griot) was used to set the polarization of the incident laser beam. A polarizer was placed before the entrance slit of the monochromator to allow the detection of the polarized scattered light. A broadband quarter-wave plate was placed after the polarizer to eliminate the polarization dependence of the grating. To follow the molecular orientation, two polarized spectra, labeled Ixx and Izz, were recorded. The indices mean that both the incident and the scattered radiation are polarized in the X or the Z direction, respectively, where Z corresponds to the “fiber” axis (i.e., the flow direction) and X is the perpendicular direction.37 The spectra were corrected to account for the polarization dependence of the instrument using isotropic samples such as liquid chloroform. The measurement time of a single spectrum was typically around 30 s. No sign of sample deterioration was observed under these experimental conditions. The spectra were only corrected for a slight fluorescence background over the 400-1800 cm-1 spectral range using a polynomial baseline. No smoothing was applied on the spectra. Wavenumber shifts of the spectra obtained from different experiments were corrected using the tyrosine band at 1615 cm-1. Fourier deconvolution in the amide I and amide III regions was performed using the method of Griffith and Patiente38 using a gamma factor of 11 and smoothed at 90% with a Bessel function. All spectral manipulations were performed using GRAMS/AI 7.0 (ThermoGalactic, Salem, NH).

2402


Lefèvre et al.

Figure 4. Original (A) and difference (B) spectra of the silk dope of the MA gland of N. clavipes recorded at 14 different positions along the gland and spinning tube from the B section of the sac to the spigot. Numbers correspond to locations shown in Figure 1. In A, all spectra are normalized with respect to the area measured over the whole spectral range. In B, the first spectrum of A (before spinning) has been subtracted from all other spectra.

Results and Discussion Native Protein Conformation before Spinning. Figure 3 shows the comparison of the Raman spectra of the native silk dope and the dry silk dope probed in the region of the sac (Bzone) of the MA gland of a N. claVipes spider. The spectra display bands due to the silk proteins (spidroins) and from other constituents. Some bands that are not due to proteins are observed in the spectra of Figure 3. These bands have been irregularly detected for different N. claVipes individuals and at different positions throughout the whole gland. By careful examination of the concomitant intensity variations of certain of these bands, two different compounds labeled C1 and C2 have been identified. The first one (C1) displays bands around 1550 and 1155 cm-1 that are assigned to the sCdCs and dCsCd stretching vibrations of carotenoids, respectively.39 Compound C2 exhibits

two major peaks at 1526 and 1390 cm-1. Although the complete spectrum of the pure component is lacking, the 1390-cm-1 Raman line reveals that C2 is highly characteristic of an isoquinoline compound.40 This interpretation is consistent with Holl and Henze41 who have studied the yellow pigments present in the web threads of Nephila spiders and noted the presence of xanthurenic acid, an isoquinoline compound. Among the bands due to the silk proteins, some of them are highly conformation-sensitive and arise mainly from the polypeptide backbone. They are found at 1657 cm-1 (amide I), 1260 cm-1 (amide III), 1102 cm-1 (CsC stretching and CH3 rocking), and 525 cm-1 (alanine residues). The amide I band is very similar in the spectra of the silk in the native and dry states. The slightly larger bandwidth observed for silk in the native state is due to the overlapping contribution of the water OsH bending vibration. The amide III region and the ”skeletal” (CsC


stretching and CH3 rocking) one are free of bands arising from water. Because these two bands of the spinning dope are identical in both states, the protein conformation appears to be conserved upon dehydration of the dope. The asymmetry and the large width of the amide I and amide III bands show that the native silk protein contains several conformational elements, which is typical of an intrinsically unfolded protein as already discussed.22–24 Fourier self-deconvolution (data not shown) indeed reveal three components at ∼1635, 1654, and ∼1685 cm-1 in the amide I region and two components at 1251 and 1265 cm-1 in the amide III region. The components at 1654 and 1265 cm-1 are due to R-helix,42 whereas the other components are assigned to unordered chain segments.16,43,44 Random and R-helical conformations have been previously found in the spidroins in solution by vibrational circular dichroism (VCD) spectroscopy.24 The presence of a few R-helices in MA silk before spinning is also supported by the weak band at 525 cm-1. This band is due to the CO in-plane bending and CNCR deformation modes45,46 and is characteristic of polyalanine sequences in the R-helix conformation as found in synthetic polypeptide poly-L-alanine46,47 and in the cocoon silk dope of the silkworm Samia cynthia ricini.48 Because this band is dramatically less intense for the spider MA silk dope contained in the sac compared to poly-L-alanine, the amount of R-helices formed by the polyalanine segments of silk spidroins is small. It can also be noted that the intensity of the 525 cm-1 band is smaller in the wet than in the dry state. Therefore, although the presence of this band reveals that alanine residues of the spidroins have a certain propensity to form R-helices, this tendency is diminished in a wet environment. Indeed, it has been shown by NMR spectroscopy16 that alanine residues of MaSp1 and MaSp2 mainly adopt the 31-helix (polyproline IIlike) conformation in solution. Because this type of helix is stabilized by the formation of hydrogen bonds between the amide groups and water molecules,49 it is conceivable that the absence of water in the dry state promotes the formation of R-helices in the spidroins. Indeed, in this case, the helix structure is stabilized by intramolecular hydrogen bonds between CdO and NsH peptide groups of the polypeptide chain. The band located at 939 cm-1 is also conformation-sensitive (see below). This band has been observed in muscle proteins50 and tropomyosin51 that are recognized as containing high amounts of R-helices. This band may also represent a R-helix marker for silk proteins. The detailed assignment of the major peaks of Figure 3, including those that are less or not sensitive to the spidroin conformation (mainly side-chain vibrations) is given in Table 1. Conformational and Orientational Changes During the Spinning Process. Figure 4A shows the Raman spectra of the silk dope of the MA gland of N. claVipes recorded at different positions along the spinning apparatus, from the sac to the spigot. The different locations probed are shown in Figure 1. It is important to stress that the spectra recorded at different locations along the duct were all obtained from a single gland. Major spectral modifications can be observed along the duct, especially in the conformation-sensitive regions of the spectra (amide I, amide III, skeletal, and 525 cm-1 bands). The amide I band of silk narrows and the peak maximum shifts from 1657 to 1668 cm-1 downward from the sac to the spigot. The amide III band also narrows and shifts from 1260 to 1242 cm-1. Finally, the symmetric and narrow line at 1102 cm-1 decreases in intensity at the expense of two new bands at 1094 and 1068 cm-1. All these changes are representative of the loss of part


2403

Table 1. Position and Assignment of the Major Raman Bands of N. clavipes Major Ampulate Silk band positiona (cm-1) 525 643 829 and 851 904 939 1002 (1094 and 1068)c 1028 1102 1163 1175 1155 1207 1260 (1242 and 1228)c 1305 1335 1390 1416 1452 1526 1550 1603 1615 1657 (1668)c

assignmentb Ala (CO in-plane bend + CNCR deformation) R-helix42,46 Tyr (CsC ring twist)60 Fermi resonance of Tyr52 Ala (CsN stretching + (CNCR) deformation)42,45,46 R-helix50,51 Phe (CsC breathing of benzene ring) skeletal (CRsCβ stretching + CH3 rocking), C1 (methyl rocking of conjugated chain) NsCR stretching and HR bending45 Tyr60 C1 (CsC stretching of conjugated chain)39 Phe, Tyr60 amide III (NsH bend + CsN stretching) Ala (HR bend)45 Ala (HR bend)45 C2 (CsC stretching of isoquinoline)40 COO- stretching CH3 asymmetric bend, CH2 stretching42 C1 (CdC stretching of conjugated chain)39 C2 Phe, Tyr-protonated form (in-plane ring stretching) Tyr (in-plane ring stretching, ionized form) amide I (CdO stretching)

a Position may vary by a few wavenumbers depending on the location in the duct due to conformation changes. b For combination modes only the main contributions are given. c Values in parenthesis correspond to the solid MA silk in the β-sheet conformation.

of the initial native structure for the benefit of the formation of pleated β-sheets. No trace of a new secondary structure other than β-sheet is found, suggesting that some of the native conformational elements are retained in the amorphous phase of the final fiber. These segments may only undergo alignment and dehydration without conformational change during the spinning process. In addition to these spectral modifications, the 525 cm-1 band gradually disappears as silk is probed proximally to the spigot, indicating that the R-helix segments are lost and most probably converted to β-sheets. The spectra recorded beyond the valve do not show any trace of the band at 525 cm-1, thus indicating that no residual polyalanine R-helices remain in the fiber. One can also note the decrease in intensity of the band located at 939 cm-1, an observation already reported for the synthetic polypeptide poly-L-alanine when its conformation is converted from R-helix to β-sheets.47 Figure 4A shows an increase of the intensity of the Raman line due to alanine located at 904 cm-1, which has also been observed by Frushour and Koenig for β-poly-L-alanine and which is then characteristic of the formation of β-sheets.47 The intensity ratio of the line at 851 cm-1 with respect to that at 829 cm-1 has been shown to be an indicator of the hydrophobicity of the environment of the tyrosine hydroxyl groups.52 The decrease in intensity at 851 cm-1 in Figure 4A as silk is probed downward from the sac to the spigot reflects the fact that tyrosine side chains are in an increasingly more hydrophobic environment toward the end of the spinning apparatus. This is related to the loss of water molecules as silk goes from a liquid solution to the solid fiber state.

2404


To highlight the spectral changes observed in Figure 4A, the first spectrum of the series was subtracted from all other spectra (Figure 4B). The amide I region displays the increase of the component due to β-sheets at 1668 cm-1 and the loss of unordered structures at 1632 and 1696 cm-1. The amide III region is characterized by the appearance of two bands located at 1226 and 1238 cm-1 due to vibrational modes with A and B2 symmetry of the β-sheet structure, respectively.45 The band at 1163 cm-1 is a combination mode principally due to the NsCR stretching and HR bending of the amide groups in the β-form.45 The skeletal vibration mode region near 1100 cm-1 is particularly interesting because there is no overlapping between the two conformations. It shows nicely the conversion from the native structure represented by the line at 1102 cm-1 to the β-sheets reflected by the components at 1094 and 1068 cm-1. Finally, the difference spectra also reveal the presence of several isosbestic points throughout the entire spectral region, showing that the conformational conversion follows a two-state process. This means that there is no intermediate state during the spinning process, the native structure being directly converted into β-sheets. Figure 5 shows the polarized Raman spectra recorded at three different positions in the MA gland of the spider N. claVipes with the polarization of the incident and scattered light either parallel (Izz) or perpendicular (Ixx) to the flow direction within the duct. Figure 5A shows the polarized spectra of the silk contained in the sac of the gland (B-zone). They are typical of the native state as described above. As can be seen, the two spectra have identical intensities over the whole spectral range showing no preferential orientation. The spinning dope at this stage is thus isotropic. The polarized spectra recorded in the first and second limb (not shown) are very similar to those of Figure 5A, although a very small amount of β-sheets has been observed for some specimens. The spectra recorded just upward of the valve are reproduced in Figure 5B. From the position and shape of the amide I, amide III, and skeletal bands, silk appears as a mixture of spidroins that adopt both the native and the β-sheet conformations. The reduced but nonzero intensity at 525 cm-1 is consistent with this finding. These spectra show that part of the conformational transition has occurred. However, the Ixx and Izz spectra are identical, revealing that silk is still isotropic at this stage. Therefore, in the spinning process, the formation of β-sheets clearly precedes the molecular alignment. Figure 5C shows the polarized spectra recorded distal to the valve. The spectra are now strongly dominated by β-sheet bands (amide I, amide III, and skeletal modes), the 525 cm-1 helical band has disappeared, and the Ixx and Izz spectra have different shapes and intensities, showing that the proteins are highly oriented. These spectra are in fact very close to those of the final solid fibers collected in the webs of spiders (not shown). Such spectra have been discussed elsewhere37 and they essentially show that the polypeptide chains, the β-sheets, in particular, are preferentially aligned along the fiber axis. To follow more quantitatively the conformation and the molecular orientation of the proteins along the silk production pathway, the positive area of the β-sheet component and the intensity ratio R ) Izz/Ixx of the amide I band have been measured for different positions along the duct, from the funnel to the spigot. For an isotropic system, R should be equal to one. If R < 1, the Raman tensors associated with the CdO bonds are mainly aligned perpendicular to the fiber axis, whereas for R > 1, the Raman tensors are mostly oriented parallel to the fiber axis.

Lefèvre et al.

Figure 5. Polarized Raman spectra obtained (A) in the B-zone of the sac (point #01 in Figure 1), (B) just ahead of the valve (point #12 in Figure 1), and (C) distal to the valve (point #13 in Figure 1) of the MA gland of N. clavipes.

As seen from Figure 6, the conformational and orientational transformations are not in phase. In the first two limbs, there are essentially no β-sheets and the spectra are typical of native silk. The presence of small amounts of β-sheets has sometimes been observed in the first limb of the duct. Most often, a significant proportion of β-sheets have been observed near the second loop of the duct, between the distal region of the second limb and the proximal region of the third limb. These β-sheets are most likely formed as a result of shear forces against the luminal wall of the duct during the silk flow. However, a major conformational change occurs in the distal half-section of the third limb, near and beyond the draw down taper. The majority of the β-sheets are formed just before the valve, although their content still continues to increase slightly just after the valve up to the spigot. From these results, it appears that a large proportion of the β-sheets are formed in the distal section of



2405

Figure 6. Evolution of the conformation and the orientation of the silk proteins as a function of the location in the gland and duct from the sac to the spigot. The numbers correspond to locations shown in Figure 1. The conformation has been probed by measuring the positive amide I band (see Figure 4) corresponding to the β-sheet component, A(1668), whereas the qualitative orientation parameter, R ) Ixx/Izz, has been used to follow orientational conversion. The first, second, and third limbs are delimited by gray areas. Some particular locations of the duct are identified by arrows. The full lines that cross the data points are only guides for the eye.

the third limb, indicating an important role played by the extensional flow in the conformational transformation (see below). It cannot be excluded that the small amount of β-sheets formed initially in the first two limbs and those formed in the proximal part of the third limb by shear forces are required for the final extensional flow to be fully efficient. In other words, the first β-sheets formed in the duct before a tension is applied on silk may be crucial to promote the formation of the high amount of β-sheets found in the final fiber. These β-sheets would act as nucleation sites for the subsequent β-aggregation of spidroins and they may actually determine (accelerate) the kinetics of β-sheet formation. The decrease of pH and change of the mineral environment of the spinning solution along the duct may also contribute to the optimization of the spinning process.53,54 The amount of these β-sheet nuclei formed during the flow of silk in the canal of the duct might be limited by the slow reduction of the duct diameter. As a matter of fact, the hyberbolic shape of the duct has been suggested to be responsible for a low and uniform shear stress, preventing a premature coagulation of the silk.20,21 The presence of these early protein nuclei is consistent with the fact that the aggregation process of proteins in general and amyloids in particular follows a nucleation/propagation process.55 The hypothesis of a nucleation-dependent aggregation mechanism has already been proposed for the spinning process of Bombyx mori cocoon silk based on results obtained by a circular dichroism spectroscopic study of regenerated silk fibroin in solution.56 The formation of the β-sheet crystallites of silk then appears very similar to the crystallization of polymers. The intensity ratio R shows that silk is isotropic in the first and second limbs of the MA gland, as well as in the proximal part of the third limb. Molecular orientation starts to develop

after a significant amount of β-sheets are formed, generally near the middle of the third limb, although it appears sometimes in more distal regions. Therefore, it seems that the flow of the spinning dope along the duct does not induce molecular alignment. Development of protein orientation occurs in a short region of the duct and is always found very near or after the valve, suggesting that extensional flow is the main driving force to induce molecular orientation. The orientational parameter R is smaller than 1, showing that the carbonyl groups mainly align perpendicular to the flow direction, that is, the protein backbones preferentially orient along the flow direction. It is likely that the formation of the β-sheets during extensional flow is the consequence of the stretching of the polypeptide chains, in the same way that the crystallization of polymers can be induced by drawing.57 This comparison emphasizes the similarity between the aggregation mechanism of MA spidroins during natural spinning and the strain-induced crystallization of synthetic polymers. The formation of β-sheets when spiders draw their silk can further be rationalized by the fact that the tension-induced stretching of the polypeptide chains destabilizes the native conformation of the spidroins leading directly to β-sheet formation. The reason for the direct conversion into β-sheet seems to lie in the fact that β-pleated sheet is the most stable, and then the preferred, secondary structure of proteins.58 Finally, due to a lower solubility of the intermolecular β-sheets, the β-aggregation of silk proteins induced by the elongational flow may simultaneously induce the phase separation of water. Water may then be efficiently removed from the lumen by the epithelium in the distal part of the spinning tubule.59

2406


Conclusions Raman spectromicroscopy turns out to be efficient to investigate the molecular changes of the silk proteins along the MA spinning apparatus of spiders without the use of fixative agents of biological tissues such as glutaraldehyde and without resin embedding. Nevertheless, the spectra obtained here were not recorded in vivo because the experimental procedure used to access the silk material inside the duct involves several manipulations, including anesthesia and dissection of the spider, excision of the gland from accessory tissues, immersion in a buffer, and local removal of small pieces of the duct’s epithelium. One should thus keep in mind that even though the results presented here are reproducible, they are representative of the silk dope contained in glands that have undergone these manipulations. It would certainly be very interesting, to generalize the conclusions presented here, to make similar observations using other characterization techniques, experimental conditions, and spider species. In this work, both the orientation and the conformation of the proteins in the spinning dope in the duct have been characterized. The data obtained in the course of this work have shown that, taken together, the amide I, amide III, and skeletal bands are sensitive to very small protein conformational changes, and even very small amounts of β-sheets can be detected in silk by Raman spectromicroscopy. The high sensitivity of this technique has allowed us to detect small modifications that occur in the first limb, second limb, and the first half of the third limb of the duct but that the main molecular changes occur distally, between the draw down taper and the valve. The transformation appears to occur without intermediate states, the native conformation being directly converted to β-sheets. All the R-helices initially present in the native state disappear upon spinning. Orientational and conformational transformations are not synchronous, as the β-sheets are formed before the development of molecular alignment. The slow initial conformational conversion of the silk protein between the sac and the draw down taper, followed by the rapid transformation between the drawn down taper and the valve, may both be important to achieve the optimal conformation and level of orientation of the final fiber. Acknowledgment. Funding for the Raman spectrometer was obtained through a grant from the Canadian Foundation for Innovation. This work was also supported by the Natural Sciences and Engineering Research Council (NSERC) of Canada and the Fonds Que´bećois de Recherche sur la Nature et les Technologies (FQRNT). The authors express their gratitude to François Lapointe and Jean-Fançois Rioux-Dube´ for their valuable technical support.

References and Notes (1) Denny, M. W. SilkssTheir properties and functions. The Mechanical Properties of Biological Materials, Society for Experimental Biology, ed.; Cambridge University Press: Cambridge, 1980; Vol. XXXIV, pp 246-272. (2) Gosline, J. M.; DeMont, M. E.; Denny, M. W. The structure and properties of spider silk. EndeaVour 1986, 10, 37–43. (3) Frische, S.; Maunsbach, A. B.; Vollrath, F. Elongate cavities and skincore structure in Nephila spider silk observed by electron microscopy. J. Microsc. (Oxford) 1998, 189, 64–70. (4) Hernandez Cruz, D.; Rousseau, M. E.; West, M. M.; Pe´zolet, M.; Hitchcock, A. P. Quantitative mapping of the orientation of fibroin β-sheets in B. mori cocoon fibers by scanning transmission X-ray microscopy. Biomacromolecules 2006, 7, 2247–2257. (5) Miller, L. D.; Putthanarat, S.; Eby, R. K.; Adams, W. W. Investigation of the nanofibrillar morphology in silk fibers by small angle X-ray scattering and atomic force microscopy. Int. J. Biol. Macromol. 1999, 24, 159–165.

Lefèvre et al. (6) Oroudjev, E.; Soares, J.; Arcidiacono, S.; Thompson, J. B.; Fossey, S. A.; Hansma, H. G. Segmented nanofibers of spider dragline silk: atomic force microscopy and single-molecule force spectroscopy. Proc. Nat. Acad. Sci. U.S.A. 2002, 99, 6460–6465. (7) Putthanarat, S.; Stribeck, N.; Fossey, S. A.; Eby, R. K.; Adams, W. W. Investigation of the nanofibrils in silk fibers. Polymer 2000, 41, 7735– 7747. (8) Riekel, C.; Vollrath, F. Spider silk fibre extrusion: Combined wideand small-angle X-ray microdiffraction experiments. Int. J. Biol. Macromol. 2001, 29, 203–210. (9) Sapede, D.; Seydel, T.; Forsyth, V. T.; Koza, M. M.; Schweins, R.; Vollrath, F.; Riekel, C. Nanofibrillar structure and molecular mobility in spider dragline silk. Macromolecules 2005, 38, 8447–8453. (10) Sponner, A.; Unger, E.; Grosse, F.; Weisshart, K. Differential polymerization of the two main protein components of dragline silk during fibre spinning. Nat. Mater. 2005, 4, 772–775. (11) Termonia, Y. Molecular modeling of spider silk elasticity. Macromolecules 1994, 27 (25), 7378–7381. (12) Eles, P. T.; Michal, C. A. A DECODER NMR study of backbone orientation in Nephila claVipes dragline silk under varying strain and draw rate. Biomacromolecules 2004, 5 (3), 661–665. (13) Ku¨mmerlen, J.; van Beek, J. D.; Vollrath, F.; Meier, B. H. Local structure in spider dragline silk investigated by two-dimensional spindiffusion nuclear magnetic resonance. Macromolecules 1996, 29 (8), 2920–2928. (14) Valluzzi, R.; Szela, S.; Avtges, P.; Kirschner, D.; Kaplan, D. Methionine redox-controlled crystallization of biosynthetic silk spidroin. J. Phys. Chem. B 1999, 103 (51), 11382–11392. (15) van Beek, J. D.; Hess, S.; Vollrath, F.; Meier, B. H. The molecular structure of spider dragline silk: Folding and orientation of the protein backbone. Proc. Nat. Acad. Sci. U.S.A. 2002, 99 (16), 10266–10271. (16) Lefevre, T.; Rousseau, M.-E.; Pezolet, M. Protein secondary structure and orientation in silk as revealed by Raman spectromicroscopy. Biophys. J. 2007, 92, 2885–2895. (17) Hinnan, M. B.; Lewis, R. V. Isolation of a clone encoding a second dragline silk fibroinsNephila claVipes dragline silk is a two-protein fiber. J. Biol. Chem. 1992, 267 (27), 19320–19324. (18) Simmons, A. H.; Michal, C. A.; Jelinski, L. W. Molecular orientation and two-component nature of the crystalline fraction of spider dragline silk. Science 1996, 271 (5245), 84–87. (19) Sponner, A.; Vater, W.; Monajembashi, S.; Unger, E.; Grosse, F.; Weisshart, K. Composition and hierarchical organization of a spider silk. PLoS One 2007, (10), e998. (20) Vollrath, F.; Knight, D. P. Liquid crystalline spinning of spider silk. Nature 2001, 410, 541–548. (21) Knight, D. P.; Vollrath, F. Liquid crystals and flow elongation in a spider’s silk production line. Proc. R. Soc. London, Ser. B 1999, 266, 519–523. (22) Dicko, C.; Knight, D.; Kenney, J. M.; Vollrath, F. Structural conformation of spidroin in solution: a synchrotron radiation circular dichroism study. Biomacromolecules 2004, 5, 758–767. (23) Hijirida, D. H.; GianDo, K.; Michal, C.; Wong, S.; Zax, D.; Jelinski, L. 13C NMR of Nephila claVipes major ampullate silk gland. Biophys. J. 1996, 71, 3442–3447. (24) Lefe`vre, T.; Leclerc, J.; Rioux-Dube´, J.-F.; Buffeteau, T.; Paquin, M.C.; Rousseau, M.-E.; Cloutier, I.; Auger, M.; Gagne´, S. M.; Boudreault, S.; Cloutier, C.; Pe´zolet, M. Conformation of spider silk proteins in situ in the intact major ampullate gland and in solution. Biomacromolecules 2007, 8, 2342–2344. (25) Willcox, P. J.; Gido, S. P.; Muller, W.; Kaplan, D. L. Evidence of a cholesteric liquid crystalline phase in natural silk spinning process. Macromolecules 1996, 29, 5106–5110. (26) Kojic´, N.; Bico, J.; Clasen, C.; McKinley, G. H. Ex vivo rheology of spiders. J. Exp. Biol. 2006, 209, 4355–4362. (27) Chen, X.; Knight, D. P.; Vollrath, F. Rheological characterization of Nephila spidroin solution. Biomacromolecules 2002, 3, 644–648. (28) Wilson, R. S. Control of drag-line in certain spiders. Am. Zool. 1969, 9, 103–111. (29) Wilson, R. S. The control of dragline spinning inthe garden spider. Q. J. Microsc. Sci. 1962, 104, 557–571. (30) Xie, F.; Zhang, H.; Shao, H.; Hu, X. Effect of shearing on formation of silk fibers from regenerated Bombyx mori silk fibroin aqueous solution. Int. J. Biol. Macromol. 2006, 38, 284–288. (31) Wilson, R. S. The structure of the dragline control valves in the garden spider. Q. J. Microsc. Sci. 1962, 103, 549–555. (32) Vollrath, F.; Knight, D. P. Structure and function of the silk production pathway in the spider Nephila edulis. Int. J. Biol. Macromol. 1999, 24, 243–249.

Silk Proteins of Nephila clavipes Spiders (33) Knight, D. P.; Knight, M. M.; Vollrath, F. Beta transition and stressinduced phase separation in the spinning of spider dragline silk. Int. J. Biol. Macromol. 2000, 27, 205–210. (34) Madsen, B.; Vollrath, F. Mechanics and morphology of silk drawn from anesthetized spiders. Naturwissenschaften 2000, 87, 148–153. (35) Pérez-Rigueiro, J.; Elices, M.; Plaza, G. R.; Real, J. I.; Guinea, G. V. The influence of anaesthesia on the tensile properties of spider silk. J. Exp. Biol. 2006, 209, 320–326. (36) Work, R. W. Mechanisms of major ampullate silk fiber formation by orb-web-spinning spiders. Trans. Am. Microsc. Soc. 1977, 96, 170– 189. (37) Rousseau, M. E.; Lefevre, T.; Beaulieu, L.; Asakura, T.; Pezolet, M. Study of protein conformation and orientation in silkworm and spider silk fibers using Raman microspectroscopy. Biomacromolecules 2004, 5 (6), 2247–2257. (38) Griffith, P. R.; Patiente, G. L. Introduction to spectral deconvolution. Trends Anal. Chem. 1986, 5, 209–215. (39) Rimai, L.; Kilponenn, R. G.; Gill, D. Excitation profile of laser Raman spectra in the resonance region of two carotenoid pigments in solution. J. Am. Chem. Soc. 1970, 92, 3824–3825. (40) Wait, S. C.; McNerney, J. C. Vibrational spectra and assignments for quinoline and isoquinoline. J. Mol. Spectrosc. 1970, 34, 56–77. (41) Holl, A.; Henze, M. In Pigmentary constituents of yellow threads of Nephila webs, Proceedings of the XI Europaïsches Arachnologishes Colloquium, Technische Universita¨t, Berlin, Germany; 1988; Haupt, J., Ed.; European Society of Arachnology, Technische Universita¨t: Berlin, 1988; p 350. (42) Krimm, S.; Bandekar, J. Vibrational spectroscopy and conformation of peptides, polypeptides, and proteins. AdV. Protein Chem. 1986, 38, 181–364. (43) Maiti, N. C.; Apetri, M. M.; Zagorski, M. G.; Carey, P. R.; Anderson, V. E. Raman spectroscopic characterization of secondary structure in natively unfolded proteins: R-Synuclein. J. Am. Chem. Soc. 2004, 126, 2399–2408. (44) Painter, P. C.; Koenig, J. L. The solution conformation of poly(Llysine). A Raman and infrared spectroscopic study. Biopolymers 1976, 15, 229–240. (45) Moore, W. H.; Krimm, S. Vibrational analysis of peptides, polypeptides, and proteins. II. β-Poly(L-alanine) and β-poly(L-alanylglycine). Biopolymers 1976, 15, 2465–2483. (46) Rabolt, J. F.; Moore, W. H.; Krimm, S. Vibrational analysis of peptides, polypeptides, and proteins. 3. R-Poly(L-alanine). Macromolecules 1977, 10 (5), 1665–1074.


2407

(47) Frushour, B. G.; Koenig, J. L. Raman spectroscopic study of mechanically deformed poly-L-alanine. Biopolymers 1974, 13, 455– 474. (48) Rousseau, M.-E.; Beaulieu, L.; Lefevre, T.; Paradis, J.; Asakura, T.; Pezolet, M. Characterization by Raman spectroscopy of the straininduced conformational transition in fibroin fibers from the silkworm Samia cynthia ricini. Biomacromolecules 2006, 7, 2512–2521. (49) Pappu, R. V.; Rose, G. D. A simple model for polyproline II structure in unfolded states of alanine-based peptides. Protein Sci. 2002, 11, 2437–2455. ´ ; enard, D.; Caille´, J.-P. Raman spectroscopy (50) Pezolet, M.; Pigeon, M.; M of cytoplasmic muscle fiber protein. Biophys. J. 1988, 53, 319–325. (51) Frushour, B. G.; Koenig, J. L. Raman spectroscopic study of tropomyosin denaturation. Biopolymers 1974, 13, 1809–1819. (52) Siamwiza, M.; Lord, R. C.; Chen, M. C. Interpretation of the doublet at 850 and 830 cm-1 in the Raman spectra of tyrosyl residues in proteins and certain model compounds. Biochemistry 1975, 14, 4870– 4876. (53) Dicko, C.; Vollrath, F.; Kenney, J. M. Spider silk protein refolding is controlled by changing pH. Biomacromolecules 2004, 5, 704–710. (54) Knight, D. P.; Vollrath, F. Changes in element composition along the spinning duct in a Nephila spider. Naturwissenschaften 2001, 88, 179– 182. (55) Jarrett, J. T., Jr. Seeding “one-dimensional crystallization” of amyloid: A pathogenic mechanism in alzheimer’s disease and scrapie. Cell 1993, 73, 1055–1058. (56) Li, G.; Zhou, P.; Shao, Z.; Xie, X.; Chen, X.; Wang, H.; Chunyu, L.; Yu, T. The natural silk spinning process. A nucleation-dependent aggregation mechanism. Eur. J. Biochem. 2001, 268, 6600–6606. (57) Morioka, T.; Kakiage, M.; Takeshi, Y.; Komoto, T.; Higushi, Y.; Kamiya, H.; Arai, K.; Murakami, S.; Uehara, H. Oriented crystallization induced by uniaxial drawing form poly(tetrafluoroethylene) melt. Macromolecules 2007, 40, 9413–9419. (58) Perczel, A.; Hudaky, P.; Palfi, V. Dead-end street of protein folding: Thermodynamic rationale of amyloid fibril formation. J. Am. Chem. Soc. 2007, 129, 14959–14965. (59) Vollrath, F.; Knight, D. P.; Hu, X. W. Silk productrion in a spider involves acid bath treatment. Proc. R. Soc. London, Ser. B 1998, 265, 817–820. (60) Fisher, W. B.; Eysel, H. H. Polarized Raman spectra and intensities of aromatic amino acids phenylalanine, tyrosine and tryptophan. Spectrochim. Acta 1992, 48A, 725–732.

BM800390J

Conformational and Orientational Transformation of Silk Proteins in

Recommend Documents