Chitin-based scaffolds are an integral part of the skeleton of the marine demosponge Ianthella basta

Abstract

The skeletons of demosponges, such as Ianthella basta, are known to be a composite material containing organic constituents. Here, we show that a filigree chitin-based scaffold is an integral component of the I. basta skeleton. These chitin-based scaffolds can be isolated from the sponge skeletons using an isolation and purification technique based on treatment with alkaline solutions. Solid-state ¹³C NMR, Raman, and FT-IR spectroscopies, as well as chitinase digestion, reveal that the isolated material indeed consists of chitin. The morphology of the scaffolds has been determined by light and electron microscopy. It consists of cross-linked chitin fibers approximately 40–100 nm in diameter forming a micro-structured network. The overall shape of this network closely resembles the shape of the integer sponge skeleton. Solid-state ¹³C NMR spectroscopy was used to characterize the sponge skeleton on a molecular level. The ¹³C NMR signals of the chitin-based scaffolds are relatively broad, indicating a high amount of disordered chitin, possibly in the form of surface-exposed molecules. X-ray diffraction confirms that the scaffolds isolated from I. basta consist of partially disordered and loosely packed chitin with large surfaces. The spectroscopic signature of these chitin-based scaffolds is closer to that of α-chitin than β-chitin.

1. Introduction

Chitin has been known to be part of the skeletal structure of various invertebrates for many years. Odier (1823) demonstrated that the exoskeleton of insects is composed of chitin. Arthropods are the most abundant species occurring on earth and chitin provides the chemical basis for their skeletal structures. For a long time, however, chitin has not been recognized to be part of the skeletal formations of sponges (Porifera). Sponges – the first known Metazoans – are the phylogenetically most ancient invertebrates. In sponges, chitin was only found in the inner layer of so-called gemmulae. These microbodies are produced by fresh water sponges prior to their seasonal disappearance under extreme environmental conditions (Simpson, 1984). However, chitin was recently detected within skeletal formations of Verongula gigantea and Aplysina sp. demosponges (Ehrlich et al., 2007a) as well as several glass sponge species (Hexactinellida) (Ehrlich et al., 2007b, 2008; Ehrlich and Worch, 2007). These initial observations indicated an important role of chitin as a skeletal element in sponges and stimulated our present investigations concerning the presence of chitin in other sponge species (especially further members of the order Verongida apart from Aplysina and Verongula species). Chitin as a skeletal material in sponges is of special interest as its presence proves the existence of this important biopolymer several hundred million years before the appearance of chitin in arthropod skeletons.

Verongid sponges form a cohesive and distinct order of marine sponges (Erwin and Thacker, 2007). Although clearly delineated from other orders, the intraordinal relationships among verongid sponges are rather complicated and poorly understood. The skeleton makes up the bulk of the sponge (Bergquist, 1978). A thick skin-like dermis as well as the presence of spherulous cells among the secretory cell content are common to all species. The object of the present study is the marine sponge Ianthella basta Hyatt, 1875 (see Fig. 1A). It is a characteristic representative of the Ianthellidae family, one of the largest families among the verongid sponges. This family consists of 12 species belonging to three genera (Ianthella, Anomoianthella, and Hexadella). It is distinguished from other Verongida sponge families by the presence of eurypylos choanocyte chambers (Erwin and Thacker, 2007). Ianthella sponges are easily identified by their large size (typically 0.5–1 m diameter), their remarkable colour, the fan-shaped morphology of the body, as well as by their characteristic skeleton. This skeleton can be dried and endures in beach-stranded species (Bergquist and Kelly-Borges, 1995). Ianthella sponges are found in the Indian and Pacific oceans. For example, they occur abundantly in South East Asia, Australia, Lord Howe Island, New Caledonia, West Central Pacific, as well as the region between Guam and the West coast of the USA (Hooper and Wiedenmayer, 1994).

Fig. 1.

(A) Underwater photograph of a living marine demosponge Ianthella basta. The diameter of the sponge corresponds to about 1 m. (B) Photographs of the freeze-dried I. basta skeleton (top); a sample after NaOH treatment step 3 (middle) and subsequent H₂O₂ purification (bottom). (C) SEM images of a freeze-dried I. basta skeleton sample (top) and of an isolated scaffold after treatment step 3 (bottom).

The skeleton of demosponges is known to be a composite material containing various organic constituents. So-called spongin fibers are often found within these skeletal formations. In addition, demosponge skeletons can exhibit mineralized parts containing calcium carbonate or amorphous silica. It was, however, previously stated (Bergquist, 1978) that Verongida sponges do not exhibit a mineralized skeleton. Instead, they were reported to have a collagenous mesohyl stabilized by spongin fibers with a granulated interior region (“pith”) and a laminated exterior “bark”. Apart from spongin, various other organic compounds are found in sponges. The biotechnological and pharmaceutical potential of these compounds is well known: molecules with cytotoxic, antifouling, antitumoral, antibiotic, antiviral, enzyme-inhibitory, and anti-inflammatory activities, as well as favorable properties concerning the treatment of Alzheimer’s disease have been isolated from sponges (Faulkner, 2001). Ianthella basta contains brominated macrocyclic alkaloids, so-called bastadins. Bastadins represent a unique class of natural products based on tyrosine-derived dipeptides (Kazlauskas et al., 1980, 1981; Pordesimo and Schmitz, 1990; Greve et al., 2008) with interesting medicinal properties (Franklin et al., 1996; Aoki et al., 2006; Kotoku et al., 2008). For example, they are cytotoxic against cancer cells. Carotenoid sulfates, bastaxantins, and desulfated bastaxanthiols could also be isolated from I. basta (Hertzberg et al., 1983, 1989). Furthermore, special proteoglycans found in sponges are involved in cell adhesion (Humphreys, 1963; Fernàndez-Busquets and Burger, 2003).

Three different types of chitin are known in nature, namely α-, β-, and γ-chitin (Lotmar and Picken, 1950; Rudall, 1963). Chitin recently isolated from Aplysina sp. as well as Verongula gigantea was found to be α-chitin (Ehrlich et al., 2007a). β-Chitin occurs in the extracellular fibers of diatoms such as Thalassiosira sp. and Cyclotella sp. (McLachlan et al., 1965; Blackwell et al., 1967; Herth and Barthlott, 1979). The crystal structures of α- and β-chitin are well known (Carlström, 1957; Minke and Blackwell, 1978; Gardner and Blackwell, 1975). Chitin has also been studied extensively in the past by solid-state ¹³C NMR spectroscopy (Peter et al., 1984; Schaefer et al., 1987; Fukamizo et al., 1986; Tanner et al., 1990; Andrade et al., 2002; Schmidt-Rohr and Mao, 2002; Cárdenas et al., 2004; Kono, 2004; Kameda et al., 2005; Webster et al., 2006) as well as by Raman and Infrared (IR) spectroscopy (Pearson et al., 1960; Focher et al., 1992; Cárdenas et al., 2004; Yamaguchi et al., 2005). Therefore, these well-established spectroscopic techniques were used within the present work in order to comprehensively characterize the chitin-based scaffold in I. basta skeletons and to monitor its isolation by the alkali-based treatment proposed by Ehrlich et al. (2007a).

2. Experimental

2.1. Sample collection and preparation

Samples were collected at Western Shoals in Apra Harbor (Guam) in fall 2008 and spring 2009. The sponges were collected underwater from depths of 5–12 m by scuba diving. The collected species were visually inspected underwater with respect to their intactness. No signs for fouling or the presence of other invertebrates such as epibionts could be observed for the collected species. Sponges were kept in large Ziploc bags during collection and stored in a cooler during transport to the Guam marine laboratories. The sponges where again carefully inspected with respect to the intactness of their skeletons and the presence of algae or invertebrates using a stereomicroscope. Contaminants or damages were not observed for the collected species. After this inspection, the sponges were frozen at −20 °C and freeze-dried prior to their transport to the bioanalytical laboratories in Dresden (Germany).

2.2. Isolation of chitin-based scaffolds

The isolation of the chitin-based scaffolds was performed according to the following treatment steps:

• Step 1: The skeletons were cut into pieces of 10–30 cm² and treated with 20% acetic acid at 37 °C for 12 h. Afterwards, the samples were rinsed three times with distilled water.

• Step 2: The samples were treated with 2.5 M NaOH at 37 °C for 24 h. Afterwards, the samples were rinsed three times with distilled water.

• Step 3: Once more, the samples were treated with 2.5 M NaOH at 37 °C – now for 48 h – resulting in total NaOH treatment time of 72 h. Afterwards, the samples were rinsed three times with distilled water.

For final purification, the samples were optionally cleaned in 35% H₂O₂ at 37 °C.

It should be noted that the NaOH concentration of 2.5 M corresponds to 4% v/v. This is well below the critical concentration of 25–30% v/v where the transformation of β-chitin into α-chitin starts to take place (Noishiki et al., 2003). Therefore, the crystal structure of the chitin-based scaffolds was not influenced by the extraction procedure. A commercially available crab α-chitin sample was purchased from FLUKA. Monomeric N-acetyl-D-glucosamine was purchased from Sigma–Aldrich. β-Chitin was prepared from the diatom species Thalassiosira rotula as follows: The diatom species T. rotula was collected by Prof. M. Sumper (Regensburg, Germany) in the North Sea during an expedition at the German North Sea coast in July 2006. A 20 l culture of T. rotula was grown following the procedures described by Gröger et al. (2009). The culture was harvested and treated with sodium dodecyl sulfate (SDS) and ethylene diamine tetraacetic acid (EDTA) in order to remove loosely attached organic material following the procedures described by Kröger et al. (1999). Finally, the sample was freeze-dried for one day and afterwards dried in a vacuum oven at 40 °C for three days. Solid-state ¹³C NMR analysis confirmed the high crystallinity and purity of the obtained β-chitin (see below). Note that β-chitin is able to swell in contrast to α-chitin. The solid-state ¹³C NMR spectrum of β-chitin was measured in the hydrated state and agrees very well with the spectra published previously (Tanner et al., 1990).

2.3. Solid-state ¹³C NMR spectroscopy

Solid-state ¹³C NMR experiments were performed on a Bruker Avance 300 spectrometer operating at 75.47 MHz for ¹³C using a commercial double resonance 4 mm MAS NMR probe. Ramped ¹H–¹³C cross-polarization (CP; Pines et al., 1973; Metz et al., 1994) was used (contact time: 4 ms, number of scans: 40,000, repetition time: 2 s). SPINAL ¹H-decoupling (Fung et al., 2000) was applied during signal acquisition. The temperature inside the rotor was calibrated using the ²⁰⁷Pb chemical shift of lead nitrate as a temperature reference (Ferguson and Haw, 1995). The chemical shifts were referenced relative to tetramethylsilane (TMS).

2.4. Raman and FT-IR spectroscopy

Raman spectra were recorded on a Kaiser HoloLab Series 2000 microscope coupled to an f/1.8 Holospec spectrograph (Kaiser Optical Systems, Ann Arbor MI, USA) with a liquid nitrogen cooled CCD camera (Roper Scientific, Trenton NJ, USA). A Toptica XTRA laser (Toptica Photonics AG, Gräfelfing, Germany) with an excitation wavelength of 785 nm was used. The spectra were baseline-corrected using a multi-point linear baseline (at 350, 800, 1700, 1960, 2600, 2750, and 3050 cm⁻¹). Infrared spectra were recorded on a Nicolet 210 FT-IR Spectrometer. About 1–2 mg of the sample was mixed with 400 mg KBr and compressed. 100 scans were recorded at a spectral resolution of 2 cm⁻¹. All spectra were baseline corrected with a two-point linear baseline (at 845 and 1890 cm⁻¹).

3. Results and discussion

Images of the skeleton of the marine demosponge I. basta after different isolation steps are shown in Fig. 1B and C (cf. Experimental section). The alkali-based treatment reveals a scaffold which closely resembles the shape of the I. basta skeleton. It consists of cross-linked fibers forming a two-dimensional network with almost quadratic chambers. After extraction steps 2 and 3, the material still contained a certain amount of pigments giving rise to the red color (Fig. 1B). These pigments were removed by subsequent soaking in H₂O₂ for 10 min. After this final purification, clean white scaffolds remain (Fig. 1C). SEM investigations show that the scaffolds isolated by NaOH treatment from I. basta skeletons consist of a rather loosely packed and deeply fissured material (see Figs. 2 and 3). The X-ray diffractograms of the scaffolds obtained after NaOH treatment exhibit only a few, broad reflections at positions characteristic for α-chitin (see Supplementary material). In combination with the observations demonstrated in Figs. 2 and 3, this behaviour indicates a nanocrystalline organization of the material. Similar observations were made for the nanocrystalline squid pen β-chitin (Fan et al., 2008) which exhibits similarly broad X-ray reflections. For the I. basta skeleton after the third isolation step, the average particle size is estimated to be of the order of tens of nanometers. It is, therefore, impossible to determine the crystal structure of the isolated material by X-ray diffraction. Therefore, the structural characterization of the isolated I. basta scaffold must rely on spectroscopic techniques such as solid-state ¹³C NMR or optical spectroscopy. These techniques do not require crystalline samples.

Fig. 2.

SEM images showing the micro- and nano-organization of a selected chitin fiber from the I. basta skeleton after step 3. The image at the bottom is taken from the disrupted fiber shown in the top image (white arrow).

Fig. 3.

SEM image of an alkali-treated sample showing the nano-organization of the I. basta chitinous fibres (characteristic nanofibrils are marked with white arrows).

The ¹³C {¹H} CP-MAS NMR spectra of the skeleton samples after the different treatment steps are shown in Fig. 4. After step 1, the sample still contains a multitude of overlapping signals indicating the presence of various organic constituents. However, only seven major signals remain after steps 2 and 3 (24 and 72 h NaOH-based treatment, respectively). As can be seen from the pronounced asymmetry, the signal centered around ca. 75 ppm is the superimposition of two signals. The signals remaining after NaOH treatment occur at the characteristic chemical shifts of α-chitin. This observation shows that the isolated scaffolds mainly consist of chitin, and is also confirmed by Raman and FT-IR spectroscopy (see Supplementary material). Chitin consists of poly-N-acetyl-D-glucosamine, and the chemical shifts of the C-1 and C-4 carbon atoms sensitively reflect the polymerization of N-acetyl-D-glucosamine (Webster et al., 2006). ¹³C chemical shifts of 93.2 ppm (C-1) and 74.6 ppm (C-4) are found for the monomer (see Fig. 5). In contrast, 104.4 ppm (C-1) and 83.3 ppm (C-4) are observed for α-chitin (see Table 1). It should be noted that the latter two signals show up at 104.4 ppm (C-1) and 83.3 ppm (C-4) for I. basta in the freeze-dried samples as well as after the subsequent extraction steps. These chemical shift values are characteristic for α-chitin. In contrast, no significant signal intensity occurs at 93.2 ppm. The presence of larger amounts of monomeric N-acetyl-D-glucosamine can, therefore, be ruled out for all I. basta samples under study.

Fig. 4.

¹³C {¹H} CP-MAS NMR spectra (293 K) of the I. basta skeleton samples measured after the different isolation steps (see Section 2). The spectrum of the α-chitin sample is shown for comparison. Dashed vertical lines indicate the chemical shifts of the signals in α-chitin.

Fig. 5.

¹³C {¹H} CP-MAS NMR spectra (293 K) of the I. basta skeleton samples measured after the different isolation steps (see Section 2) and monomeric N-acetyl-D-glucosamine. Dashed vertical lines indicate chemical shifts which are particularly indicative. Note that the signals due to C-2 and C=O in the monomer exhibit a splitting due to the same effect as observed for β-chitin (see Fig. 6 and related text).

Table 1.

Chemical shifts and line widths of the various solid-state ¹³C NMR signals observed for the chitin samples under study. Chemical shifts of asymmetric or split signals (C=O, C-2 in β-chitin) determine the center of gravity of the corresponding signals. Line widths are the full widths measured at half maximum height of the signals. Experimental errors are ±0.2 ppm for the chemical shifts and ±5 Hz for the line widths.

Sample	C=O		C-1		C-4		C-5		C-3		C-6		C-2		CH3
	δ (ppm)	Δν1/2 (Hz)	δ (ppm)	Δν1/2 (Hz)	δ (ppm)	Δν1/2 (Hz)	δ (ppm)	Δν1/2 (Hz)	δ (ppm)	Δν1/2 (Hz)	δ (ppm)	Δν1/2 (Hz)	δ (ppm)	Δν1/2 (Hz)	δ (ppm)	Δν1/2 (Hz)
I. basta	174.2	270	104.4	170	83.3	215	76.0	195	73.6	115	61.3	345	55.3	180	23.2	110
α-chitin (reference)	174.2	270	104.4	125	83.3	185	76.0	145	73.6	100	61.2	260	55.3	125	23.1	75
β-chitin (T. rotula)	175.8	115	105.3	30	84.4	25	75.4	40	73.0	40	59.8	70	55.6	125	22.7	40

Residual impurities such as pigments were still present after the third isolation step (see Fig. 1B). To remove these impurities, the sample was further cleaned using 35% H₂O₂. Fig. 6 exhibits the solid-state ¹³C NMR spectrum of α-chitin, β-chitin, and the chitin-based scaffold isolated from I. basta after H₂O₂ purification. It should be noted that the spectra measured for the α- and β-chitin reference samples agree very well with previously published spectra (see for example Tanner et al., 1990; Cárdenas et al., 2004; Kono, 2004; Kameda et al., 2005). After H₂O₂ treatment, the solid-state ¹³C NMR spectrum of the I. basta skeleton agrees even more with that of α-chitin than the spectrum after step 3 of the NaOH-based isolation procedure (cf. Fig. 4). Within the experimental error, the signals of the chitin-based I. basta scaffold are found at the characteristic positions of α-chitin (cf. also Table 1). In contrast, several carbon atoms in I. basta chitin exhibit significantly different chemical shifts than observed for β-chitin. This is particularly true for C=O, C-1, C-4, and C-6 (see Fig. 6 and Table 1). It is therefore concluded that the structure of the chitin-based scaffold isolated from I. basta is closer to α-chitin than β-chitin. The native I. basta skeleton contains about 7 wt.% of chitin as could be determined by measuring the amount of chitin isolated from a freeze-dried skeleton sample of known weight.

Fig. 6.

¹³C {¹H} CP-MAS NMR spectra of β-chitin from T. rotula, α-chitin from crabs, and an I. basta scaffold after NaOH treatment and subsequent H₂O₂ purification measured at 293 K. Inserts: Selected signals measured at room temperature and at 248 K. Dashed vertical lines indicate the chemical shifts of the signals in α-chitin.

An interesting effect is observed for the carbonyl ¹³C NMR signal in the α-chitin spectrum. The signal is broad and asymmetric which has previously been explained by the existence of two unresolved, overlapping signals (Kameda et al., 2005). These two signals are caused by single and double hydrogen bonded carbonyl groups within the hydrogen-bond network of α-chitin. At elevated temperatures, the two signals merge into a single symmetric line due to rapid conformational exchange processes (Kameda et al., 2005). In order to slow down this exchange process, the samples were cooled to 249 K. Cooling resulted in the observation of two separate signals (see Fig. 6). The intensity ratio of the two lines at 172.8 and 176.1 ppm was approximately 3:2. The C=O and C-2 signals of β-chitin, as well as N-acetyl-D-glucosamine exhibits a temperature-independent splitting of ca. 50–60 Hz due to the residual dipolar coupling¹ with the neighboring quadrupolar ¹⁴N nucleus (Tanner et al., 1990) as can be seen in Figs. 5 and 6. However, this splitting is much smaller than the temperature-dependent splitting of ca. 250 Hz observed at 248 K for C=O in α-chitin (see Fig. 6), which is caused by the aforementioned hydrogen-bond influence (Kameda et al., 2005). The characteristic asymmetry is not observed for I. basta chitin after extraction step 3. The C=O signal in I. basta chitin occurs at an isotropic chemical shift of 174.2 ppm. The latter value agrees well with the weighted average of 174.1 ppm for the two signals observed in α-chitin at 248 K. This weighted average corresponds to the center of gravity of the asymmetric C=O signal for α-chitin at 293 K. In contrast, a significantly different value of 175.5 ppm is observed for the center of gravity of the C=O signal in β-chitin. These observations indicate that the hydrogen-bond network characteristic for highly crystalline α-chitin is perturbed in I. basta chitin. This may be due to the presence of minor amounts of impurities such as pigments. Nevertheless, the average C=O chemical shift observed for I. basta agrees well with α-chitin as do the other signals.

Crystalline β-chitin exhibits extremely narrow solid-state ¹³C NMR signals (see Tanner et al., 1990; cf. also Fig. 6 and Table 1). In contrast, the signals of the α-chitin sample are considerably broader. The signals in the I. basta chitin spectrum are even broader than in the crab α-chitin sample (see Fig. 6 and Table 1). Moreover, the C-1 signal is highly asymmetric. It should be noted that squid pen β-chitin also exhibits a solid-state ¹³C NMR spectrum with broadened signals compared with highly crystalline β-chitin samples (see, e.g., Tanner et al., 1990 and references therein). Tanner and coworkers explained this effect as arising from the poorly crystalline state of squid pen chitin which was found to be caused by a significant amount of surface-exposed polysaccharide chains. This interpretation was supported by recent X-ray diffraction analyses (Fan et al., 2008). The diameter of squid pen β-chitin fibrils is only 2–3 nm. The chitin-based I. basta scaffolds obviously exhibit a similar nanoscale organization. The surface of the material is rough and deeply fissured, as can be seen in the corresponding SEM images, and the X-ray diffraction patterns are broadened (see above). It is likely that this organization of the chitin results in a large amount of surface-exposed polysaccharide chains. These observations also explain the broadened solid-state ¹³C NMR signals and may be a further reason for the aforementioned perturbation of the hydrogen-bond network. In addition, the native sponge skeleton contains other organic material such as proteins and bromotyrosine-related compounds which were removed from the skeleton and probably hydrolysed by the NaOH treatment (see Fig. 1). In the native sponge skeleton, these constituents are intercalated into the chitin-based scaffold or may even be cross-linked with the chitin, thus preventing it from forming unperturbed crystalline chitin. Similar results were reported for chitin isolated from beetle larva. Intercalated catechol compounds lead to an amorphous structure of the chitin-based cuticle (Zhang et al., 2000).

Chitinase digestion experiments (see Supplementary material) again confirmed the chitinuous nature of the isolated I. basta scaffolds. Furthermore, these experiments revealed another interesting phenomenon. In contrast to the freeze-dried skeletons, purified samples were readily digested in chitinase solution. Previously, it was suggested that the skeletal fibers of sponges may be involved in the sequestration and accumulation of brominated compounds (Turon et al., 2000; Teeyapant et al., 1993). Tabudravu et al., (2002) reported that brominated tyrosine-derived compounds isolated from the verongid sponge Aplysinella rhax inhibit bacterial chitinase. This behaviour may explain why the freeze-dried skeleton samples withstand the chitinase test. Possibly, bromotyrosine-related compounds present in the skeleton inhibit the chitinase thus protecting the chitin-based scaffolds of Verongida sponges. Therefore, these compounds may be crucial for the survival of Verongida sponges in the marine environment.

The results of Raman and infrared spectroscopic studies are shown in Figs. 7 and 8, respectively. The band assignment is based on previous publications (Pearson et al., 1960; Focher et al., 1992; Cárdenas et al., 2004; Yamaguchi et al., 2005). For comparison, spectra of α- and β-chitin are shown in addition to the spectra of the I. basta scaffold after NaOH treatment and H₂O₂ purification. The spectra of I. basta agree very well with the spectra of α-chitin (see also Tables 2 and 3), while characteristic differences to the β-chitin spectra were observed. In both, the Raman and IR spectra, β-chitin exhibits narrower bands than I. basta chitin. Furthermore, the relative intensities of several bands are different. For example, β-chitin exhibits a narrow Raman band at 995 cm⁻¹ which is almost absent and strongly broadened in I. basta chitin. A significant shift is also observed for the Raman band at 1413 cm⁻¹ in I. basta chitin. This band occurs at 1420 cm⁻¹ in β-chitin. Infrared spectra of β-chitin reveal two additional bands for CH_x deformations at about 1455 and 1374 cm⁻¹; these bands were not observed in I. basta. Additionally, bands in the region of 1300–1200 cm⁻¹ further confirmed the similarity of I. basta chitin to α-chitin: The C–O–C and C–O stretching vibration region (1200–950 cm⁻¹) of β-chitin exhibits a greater number of narrower bands with different intensities than I. basta chitin. Another characteristic marker is the CH deformation of the β-glycosidic bond. This band shifts from 890 cm⁻¹ in β-chitin to 895 cm⁻¹ in α-chitin and I. basta chitin. Therefore, optical spectroscopy confirms the solid-state ¹³C NMR spectroscopic finding that the chitin-based scaffolds isolated from I. basta are closer to α-chitin than β-chitin.

Fig. 7.

Raman spectra of β-chitin from T. rotula, α-chitin from crabs, and an I. basta scaffold after NaOH and H₂O₂ treatment.

Fig. 8.

Infrared absorption spectra of β-chitin from T. rotula, α-chitin from crabs, and I. basta chitin after NaOH and H₂O₂ treatment. Dashed vertical lines are drawn to highlight characteristic differences between α- and β-chitin. The band due to the β-glycosidic bond occurs at 895 cm⁻¹ in α-chitin and at 890 cm⁻¹ for β-chitin. The amide I band splits into two components at ca. 1630 and 1660 cm⁻¹ due to the influence of hydrogen bonding. Note the intensity differences observed between α-and β-chitin.

Table 2.

Wavenumbers and assignment of the bands observed in the Raman spectra of the chitin samples shown in Fig. 7.

I. basta Wavenumber/cm−1	α-Chitin Wavenumber/cm−1	β-Chitin (T. rotula) Wavenumber/cm−1	Assignment
895	895	890	CHx deformation (o. o. p.)
	918 weak	919
950	952	952
		1000	C–O–C/C–O stretching
1057	1058	1048
1108	1108	1102
		1132
1146	1147	1147
1202	1203	1201	Amide III
		1243
1264	1264	1268
		1311
1326	1326	1326
		1360
1375	1375	1376	CHx deformation
1411	1413	1417
1449	1448	1447

Table 3.

Wavenumbers and assignment of the bands observed in the FT-IR spectra of the chitin samples shown in Fig. 8.

I. basta Wavenumber/cm−1	α-Chitin Wavenumber/cm−1	β-Chitin (T. rotula) Wavenumber/cm−1	Assignment
897	895	890	CH deformation (o.o.p.), C1 axial (β bond)
920*	919		CHX deformation (o.o.p.)
		926
951	953	954
973*	976*	971
		993*	C–O–C/C–O stretching
1013*	1012*	1014*
1027	1028	1025
		1032*
		1041*
		1064
1071	1073	1078
		1103*
		1109
1115	1115
		1122
		1136
1159	1157	1154
1204	1205	1202	Amide III
1230*	1235	1237
		1246
1263	1261	1265
		1305*
1310	1312	1312
		1330
		1364*	CHX deformation
		1374
1378	1378	1379
1415	1415	1420*
1430	1429	1432
		1456
1559	1558	1562	Amide II
1633*	1630	1630	Amide I
1662	1660	1660*

^*= shoulder, o.o.p. = out of plane.

The observation of chitin forming an integral part of sponge skeletons is an intriguing finding. Is this chitin of exogenous or endogenous nature? Previously, it was reported that chitin in several sponges may be caused by a variety of micro-invertebrates harboured by the sponges (Dauby and Jeuniaux, 1986) such as polychaete worms (Haplosyllis basticola) that colonize the fan-like sponge I. basta. Indeed, H. basticola species were observed (Sarda et al., 2002) inside the characteristic small inhalant pores (ostia), oscules, and internal channels of I. basta (Hooper and Wiedenmayer, 1994). These worms do not cause wounding or other damage to the sponges (Sarda et al., 2002). However, our field and laboratory observations on fresh I. basta material as well as light and electron microscopic studies did not yield any hints as to the presence of foreign invertebrates nor were remnants of their skeletons/cuticles found in the I. basta skeleton samples under study. Additionally, no evidence for bacterial or fungal activities could be observed. Instead, our studies prove the presence of a filigree chitin-based scaffold as an integral constituent of the sponge skeleton. This scaffold exactly resembles the shape of the sponge skeleton (see Fig. 1A and B).

In summary, it can be concluded that our solid-state ¹³C NMR, Raman, and FT-IR spectroscopic as well as electron microscopic studies unequivocally prove that the scaffolds isolated from the marine demosponge I. basta are chitin-based networks made up of cross-linked fibers of ca. 40–100 nm diameter (see Fig. 1B and C). These fibers consist of loosely packed chitin with rough and deeply fissured surfaces. In agreement with this observation, the X-ray diffractogram of I. basta chitin is strongly broadened. The solid-state ¹³C NMR signals of the isolated chitin-based scaffolds were also broader than in the crystalline α-chitin sample indicating a high amount of disordered polysaccharide chains, possibly in the form of surface-exposed molecules. Taking all these results together, our investigations show that the uniquely structured, two-dimensional chitin-based scaffolds found in I. basta skeletons consist of a special type of partially disordered chitin. The structure of this material is closer to α-chitin than to β-chitin. Since the processing of chitin is technologically difficult, such chitin-based sponge scaffolds may be of interest for various applications, in particular because it is possible to generate the required amount of material from natural sources, e.g. marine ranching or primmorph cultivation of sponges.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jsb.2009.06.018.