Stable free radicals in insect cuticles: Electron spin resonance spectroscopy reveals differences between melanization and sclerotization

Archives of Biochemistry and Biophysics 453 (2006) 179–187 www.elsevier.com/locate/yabbi

Stable free radicals in insect cuticles: Electron spin resonance spectroscopy reveals diVerences between melanization and sclerotization Hartmut Kayser a,¤, Cornelia G. Palivan b a b

Department of Biology I, University of Ulm, Albert-Einstein-Allee 11, 89081 Ulm, Germany Department of Chemistry, University of Basel, Klingelbergstrasse 80, 4056 Basel, Switzerland Received 24 May 2006, and in revised form 5 July 2006 Available online 25 July 2006

Abstract Insect cuticles (exuviae; cast skins) were examined for the Wrst time by ESR spectroscopy for the presence of stable free radicals, as found in melanins. All cuticles, except those from a locust albino strain, irrespective of the presence of melanin, provided single-line signals of varied g-values and linewidths. The ESR signals of melanins, isolated or in cuticles, were characterized by g-values <2.004 and small linewidths in the range of 4–6 G, while sclerotized cuticles, lacking melanin, showed g-values >2.004 and broad linewidths of 5–11 G. The melanin spectra were comparable to those reported for eumelanins with indol-based monomers. Minor signals ascribed to pheomelanins were found in several probes. The ‘sclerotin’ spectra were broader and displayed unresolved hyperWne structure in some cases. As for melanins, the location and environment of the radicals in cuticles giving rise to the two types of ESR spectra could not be assigned. Changes in the radical environment due to insecticide or solvent treatment can be detected by ESR spectroscopy. © 2006 Elsevier Inc. All rights reserved. Keywords: Electron spin resonance spectroscopy; Stable free radicals; Melanin; Cuticle; Scleotization; Insects

Melanins are amorphous, irregular, polymeric pigments widely occurring in living organisms [1]. While melanins are commonly known to be associated with skin and hair color in humans and other mammalians, various types of melanins are also present in the inner ear, eyes, brain, bird feathers, hoof horn, plants, fungi, and the ink sac of cuttleWsh [2–4]. Melanin pigments are classiWed into two groups: eumelanins, the indole-type arising from enzymatic oxidation of tyrosine through 3,4-dihydroxyphenylalanine (dopa)1, and pheomelanins formed by enzymatic oxidation of tyrosine in the presence of cysteine with cysteinyldopa as intermediate (or related sulfhydryl compounds) [5,6]. Formed from diVerent precursors and hence resulting in diVerent polymers the colors of melanins vary from the pitch black to brown and even red. *

Corresponding author. Fax: +49 731 50 22581. E-mail address: [email protected] (H. Kayser). 1 Abbreviations used: dopa, 3,4-dihydroxyphenylalanine; gyromagnetic factor.

g-factor,

0003-9861/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.abb.2006.07.001

Research on melanins has revealed an unusual variety of properties. Formed by oxidative polymerization of various phenolic compounds, melanins can act as both oxidants and reductants, showing a high activity in binding drugs and metal ions [3,4,7–10]. The chemical properties of melanins such as high molecular weight, insolubility in organic solvents and water render them diYcult to usual physicochemical or histochemical analysis. Because these pigments uniquely contain a stable population of organic free radicals of o-semiquinone type, ESR spectroscopy is a suitable method for studying them [11,12]. Thus a set of qualitative ESR criteria has been established for the identiWcation of these types of free radicals [13]. ESR spectra of eumelanins and pheomelanins diVer in lineshape. ESR spectra of eumelanins are single lines, whereas the spectra of pheomelanins reveal also hyperWne splitting [6]. Therefore, the analysis of the ESR lineshape permits to determine the type of melanin in the biological samples under study [14]. Melanins have also been reported from insects, where they are typically located in cuticles and hence contribute to

180

H. Kayser, C.G. Palivan / Archives of Biochemistry and Biophysics 453 (2006) 179–187

the insects’ coloration. However, since appropriate insect material is commonly not available in necessary amount, these pigments have been chemically characterized in only few species, and hence the terms ‘melanins’ or ‘melanization’ are commonly used to describe just black pigmentation [15]. In vertebrate cells, melanins have been well described as osmiophilic granules at the ultrastructural level. Similar results have been obtained also from the integument of some insects, while in other cases of visual black pigmentation, no comparable granular material was found [16]. The present study examines the use of ESR spectroscopy to identify and characterize melanin-based pigmentation of insect cuticles (exuviae as cast at molting) from various orders. The ESR spectra of melanins prepared from human hair as well as from insect cuticles by acid hydrolysis were compared with those obtained from intact cuticles that displayed black pigmentation. Unexpectedly, also cuticles without any visible ‘melanin’ pigments, initially expected to provide melanin-free controls, also exhibited ESR signals of comparable intensity. However, these ESR signals, diVered signiWcantly from those of isolated melanins and melanin-containing cuticles. Thus, ESR spectroscopy not only proved to be a useful tool to characterize melanin-type pigmentation of intact insect cuticles, but also allowed to discriminate between true melanins and melanin-like polymers in non-melanized sclerotized cuticles, based on the gyromagnetic factor and linewidth values of their ESR spectra. Apart of some early and tentative experiments by Mason and colleagues [17], the present study is the Wrst comprehensive one applying ESR spectroscopy to characterize cuticles from a broad range of insects.

subsequently washed several times with water until neutrality. Next, the cuticles were air-dried and crushed in a mortar to obtain as small as possible particles. This coarse cuticle powder was Wlled into standard NMR tubes (4 mm diameter) at quantities of 40–90 mg/tube. Less material was available in only a few probes of melanins. All procedures were carried out at room temperature and under ambient light. To prepare melanins, cuticles were subjected to a series of washing steps with 0.1 N NaOH, water, 0.1 N HCl, water, ethanol, and Wnally acetone. The washed cuticles were dried at 70 °C and crushed in a mortar. The cuticle material was next hydrolyzed in 6 N HCl at 110 °C under nitrogen to prevent autoxidation of released catechols that could give rise to melanin formation, as observed in initial experiments. The HCl was renewed four to six times every 12 h until the supernatants were fairly colorless. These extracts were discarded. The remaining dark material was washed with acetone, which remained colorless with most samples. When the acetone extract was brown (pupal cuticles from Inachis io; larval and pupal cuticles from Pieris brassicae), it was diluted with water and acidiWed with HCl to pH < 1 resulting in a dark precipitate that was collected by centrifugation and dried. The dark pigment thus obtained was soluble in 0.5 N NaOH. The remaining HCl-resistant dark material was dissolved in 0.5 N NaOH and dark pigment was precipitated upon acidiWcation, as above. The yield of this dark pigment, called melanin here, varied between 24% in the black larval cuticles from I. io (nymphalid butterXy) and 1% in the brown puparia from Calliphora vicina (blowXy). Electron spin resonance (ESR) spectroscopy

Materials and methods Preparation of cuticles and of melanins All ‘cuticles’ examined were exuviae representing the distal part of the cuticles that are shed during molting of larvae and pupae, respectively. In the present work, the more commonly known term ‘cuticles’ is used throughout. Puparia are sclerotized exuviae from last instar Xy larvae undergoing pupation within the puparium. Insects were obtained from in-house stocks, commercial breeders, Syngenta Crop Protection and Novartis Animal Health. For the insecticide study, the Xy larvae were feed diet supplemented with 0.25 mg/L cyromazine and 0.015 mg/L dicyclanil, respectively. The cuticles were mechanically cleaned from any contaminating material under a binocular microscope. In case of puparia, the pupal cuticles (after emergence of the Xies) and the dead Xies (after insecticide treatment), respectively, were completely removed. The cuticles (except of those from Hymenopterans) were then washed with usually three changes of either water or 0.1 N HCl or 0.1 N NaOH or acetone under magnetic stirring and suspension in an ultrasonic bath. Samples treated with HCl or NaOH were

ESR measurements were performed using X-band spectrometers (Brucker ESP300 or Brucker ElexSys E500). The samples were measured without further preparation at room temperature and in contact with atmospheric oxygen. The ESR parameters gyromagnetic factor (g-factor) and linewidth () were obtained using DPPH (,-diphenyl-picryl hydrazyl) as internal standard and peak-to-peak distance of the spectrum, respectively. The value of the double integral of the ESR spectrum divided by the receiver gain value and the mass of the sample, J¤ was used to compare the concentration of paramagnetic centers in the various samples. The area under the absorption curve was obtained by using double integration of the ESR spectrum. All samples were measured in the same experimental conditions: modulation amplitude (1 G), microwave power (2 mW), time constant (20 ms); only the receiver gain was varied for an optimum measurement of the ESR spectra. Results and discussion The examined samples comprised cuticles, as shed at each molt (i.e., exuviae), from larvae and pupae representing several insects orders, as well as melanins isolated from

H. Kayser, C.G. Palivan / Archives of Biochemistry and Biophysics 453 (2006) 179–187

human hair or insect cuticles. Pupal cuticles from moths and Xies, for example, are usually not melanized in the sense that they do not show black pigmentation; they typically look chestnut-brown said to be due to sclerotization, a hardening process using dopamine derivatives as key crosslinking agents [18,19]. Since insect cuticles have not yet been subject of any detailed study by ESR, we compared the signals from cuticles with those obtained from typical melanins that were studied in the present work or reported in the literature. We use the term ‘melanin’ for alkali-soluble black products obtained by acid hydrolysis of hairs and cuticles, respectively [20]. The ESR data (g-factor; linewidths ; normalized double integral of ESR signal, J¤, as a measure of spin density) from all samples are compiled in Table 1, and a two-dimensional plot of the values of g-factors and linewidths is presented in Fig. 1. As it can be seen in this Figure, there are two diVerent clusters of ESR data: the Wrst one, cluster A, deWned by a g-factor <2.004 and relatively small linewidth values ( < 6 G), and a second one, cluster B, with g-factors >2.004 and larger linewidths ( > 6 G). General characterization of the ESR signals The ESR signals did not change signiWcantly in intensity over months or even years, as found with several samples, indicating high stability of the free radical population in melanins and cuticles as well. The shape of the ESR spectra of the various insect probes can be classiWed into two categories, as exempliWed in Fig. 2: one, characterized by single line spectrum (either with a small line width of  6 6 G, or with large linewidth  > 6 G), and a second one, showing hyperWne pattern. The single line ESR spectra with small linewidth and having a Lorenzian lineshape indicate that in these free radicals the unpaired electrons are located nearby nuclei with I D 0 (such as carbons, oxygens, sulphur, etc.), where there is no hyperWne interaction. The broad shape of ESR spectra results from unresolved hyperWne pattern due to the interaction of the unpaired electrons with neighboring nuclei having I 5 0, and from strong dipolar interactions between these unpaired electrons. In these samples shorter distances between the unpaired electrons are expected determining greater spin–spin interactions that are responsible for the line broadening close to a Gaussian line [21]. The presence of resolved hyperWne pattern in some ESR spectra is indicating that in these free radicals the unpaired electrons are located nearby nuclei with I 5 0, such as nitrogens, for example. Examples will be discussed below. EVect of diVerent washing media on the ESR signals The ESR spectra of cuticles were aVected to some extend by the media initially examined to prepare clean samples. Those media (water, acetone, 0.1 N HCl, 0.1 N NaOH) were studied in the houseXy (Musca domestica) and in several moths (Manduca sexta, Spodoptera littoralis, Heliothis

181

virescens), as speciWed in Table 1. No clear-cut eVects on either g-factor, linewidth or spin density (see below) were found in samples from the three moths species. In the houseXy, however, consistent results were obtained with independent preparations. Washing with acetone produced an ESR spectrum with low linewidths ( t 6.5–6.6 G), while treatment with HCl signiWcantly broadened the signal ( t 9.6–9.7 G). Water and NaOH produced medium eVects. Remarkably, washing the puparia with HCl resulted in the lowest spin densities. The highest spin densities were obtained with cuticles washed with water, slightly less with NaOH; both media were characterized by medium linewidth values. In all other samples, the cuticles were washed with water only. If not stated otherwise, all results refer to this water-washed material. ESR signals from isolated melanins There are two diVerent types of ESR spectra characterizing melanins according to the nature of the free radicals. The ESR spectra of eumelanins consist of a single narrow ESR line with a 4–6 G linewidth and a g-factor value close to 2.004 [4,11,12]. The pheomelanins show a diVerent ESR spectrum, characteristic for an immobilized radical with hyperWne splitting to nitrogen. This could be associated with the presence of semiquinonimine radicals in the pigment polymer [5]. N14 hyperWne couplings in  radicals are generally axial and strongly anisotropic (Apar À Aperp); g values are also axial. Therefore the spectrum of the immobilized radical is dominated by z components, as previous shown for cysteinyldopa (2Az t 32 G) where the free radicals are partly nitrogen-centered, as semiquinonimine type. The reported gyromagnetic factors encompass a broad range of values: g-factor values in the range of 2.51–2.56 occur in cysteinyldopa- and cysteinyldopamine-melanins, while co-oxidation of dopa or dopamine results in lower gfactors of 2.0037–2.0046 [22]. All ESR signals recorded in the present study are in the range of low g-factors; values >2.005 that would indicate a cysteinyl-derived polymer have not been found in any of these samples. It has been previously shown that the free radical content of human hair as well as of eumelanin and pheomelanin fractions depends on the source, water content or pH [12,14]. Therefore, in our comparison we will refer only to the present hair samples measured under the same conditions as the insect cuticles. The ESR spectrum obtained from our human black hair sample showed hyperWne structure with Az t 32 G (Fig. 3). However, as the ratio between the intensity of the central peak and the side wings is diVerent from 1.3, a value speciWc for a pheomelanin-type ESR spectrum [5], a mixture of eumelanin and pheomelanin type or a mixed polymer has to be considered for this sample of black hair. The gyromagnetic factor and the linewidth of our brown hair sample are slightly higher compared to the black hair sample, but still within the range usually found for melanins (Fig. 3).

182

H. Kayser, C.G. Palivan / Archives of Biochemistry and Biophysics 453 (2006) 179–187

Table 1 Overview of the ESR data of examined samples Type/source

Dev. Stage

Linewidth  (G)

Treatmenta

g-value 2,0035 2,0039 2,0045 2,0036 2,0036 2,0036 2,0035 2,0034 2,0043

4.0 4.8 7.5 5.0 4.7 4.5 4.8 4.7 5.6

4.57 4.90 2.79 4.07 2.45 3.01 3.42 2.87 2.23

Spin density J¤

Isolated melanins Human black hair Human brown hair Manduca sexta Eudia pavonia Pieris brassicae

Pupa Larva Larva

Inachis io

Larva

Araschnia levana

Larva

HCl/NaOH/HCl HCl/NaOH/HCl HCl/NaOH/HCl HCl/NaOH/HCl Acetone sediment Acetone supernatant Acetone sediment Acetone supernatant HCl/NaOH/HCl

Pupa Pupa Pupa Pupa Pupa (abdomina) Pupa (proboscis) Pupa Pupa Pupa Pupa Larva, black mutant Pupa, black mutant Pupa, black mutant Pupa1 Pupa2 Pupa Pupa Pupa Pupa Pupa1 Pupa2 Pupa2 Pupa2 Pupa2

Acetone1 Water1 NaOH1 HCl1 Water1 Water1 Acetone2 Water2 NaOH2 HCl2 Water Water1 Water2 Water Water Acetone Water NaOH HCl Water Acetone Water NaOH HCl

2,0046 2,0045 2,0046 2,0043 2,0045 2,0046 2,0044 2,0044 2,0044 2,0043 2,0037 2,0045 2,0044 2,0044 2,0044 2,0048 2,0047 2,0049 2,0048 2,0043 2,0046 2,0047 2,0046 2,0047

8.0 7.8 7.0 8.0 7.2 7.7 8.2 7.7 7.0 7.8 4.5 7.2 8.0 11.0 10.5 7.9 8.0 7.5 8.75 8.2 8.5 8.5 7.25 8.7

0.39 0.03 0.09 0.09 0.11 0.09 0.09 0.10 0.16 0.07 0.07 0.09 0.06 0.22 0.33 0.22 0.22 0.25 0.20 0.07 0.23 0.25 0.27 0.20

Pupa, light Pupa, dark Larva Pupa, light Pupa, dark Larva, without spikes Larva, spikes + head caps

Water Water Water Water Water Water Water

2,0045 2,0044 2,0038 2,0047 2,0042 2,0038 2,0039

7.3 6.0 5.2 7.25 6.0 5.0 4.9

0.03 0.06 0.07 0.06 0.08 0.21 0.14

Puparium1 Puparium1 Puparium1 Puparium1 Puparium2 Puparium2 Puparium2 Puparium2 Puparium2 Puparium2 (elongated) Puparium3 (normal) Puparium3 (elongated) Puparium3 (elongated) Puparium3 (elongated) Puparium4 (elongated)

Acetone1 Water1 NaOH1 HCl1 Acetone2 Water2 NaOH2 HCl2 Water Cyromazine/water Dicyclanil/water Dicyclanil/water Dicyclanil/water Cyromazine/water Cyromazine/water

2,0043 2,0043 2,0042 2,0043 2,0042 2,0043 2,0042 2,0044 2,0047 2,0047 2,0046 2,0048 2,0048 2,0048 2,0045

6.5 7.55 7.77 9.7 6.6 7.4 7.63 9.6 8.5 8.3 7.5 7.5 7.5 8.12 8.0

0.11 0.20 0.18 0.08 0.17 0.33 0.26 0.06 0.25 0.11 0.19 0.24 0.27 0.21 0.09

Cuticles Lepidoptera—moths Manduca sexta

Cerura vinula Heliothis virescens

Spodoptera littoralis

Lepidoptera—butterXies Pieris brassicae

Inachis io

Diptera Musca domestica

H. Kayser, C.G. Palivan / Archives of Biochemistry and Biophysics 453 (2006) 179–187

183

Table 1 (continued) Linewidth  (G)

Type/source

Dev. Stage

Treatmenta

g-value

Lucilia sericata

Puparium Puparium, elongated Puparium, elongated Puparium Puparium, light Puparium, dark

Water Cyromazine/water Cyromazine/water Water Water Water

2,0046 2,0044 2,0047 2,0047 2,0047 2,0047

6.5 7.4 6.7 9.0 7.25 6.9

0.15 0.02 0.19 0.15 0.28 0.25

Pupa Larva Larva

Untreated Untreated Untreated

2,0038 2,0036 2,0046

5.8 5.3 6.25

0.08 0.11 0.12

Dark thorax, last larva Albino thorax, last larva

Water Water

2,0043 5.3 No signal obtained

Calliphora vicina Protophormia terraenovae Hymenoptera Athalia rosae Monohadnus monticola Orthoptera Locusta migratoria

Spin density J¤

0.01

J¤ Spin densities were normalized by dividing the double integral of the ESR spectrum by the receiver gain and the mass of the sample. a Samples marked with diVerent superscript numbers refer to diVerent batches of insects or treatments of cuticles (see columns). For details see text.

Fig. 1. 2D plot of g-factors and linewidths obtained from the ESR spectra of all examined cuticles and melanins. The sample sources are speciWed in Table 1.

In the present study, black insect pigments identiWed as melanins on the basis of their ESR spectra, were isolated by acid hydrolysis from pupal cuticles of M. sexta, and from larval cuticles of Eudia pavonia, P. brassicae, I. io, and Araschnia levana, all representing Lepidopteran species (moths and butterXies; see Table 1). With the exception of those from M. sexta and A. levana, the three other samples exhibited ESR spectra similar to those of hair melanin (with gfactor values near 2.0035 and linewidths around 5 G). Therefore in these three samples the free radicals can be considered as of typical eumelanin-type. In case of M. sexta and A. levana, the g-factor values were signiWcantly higher (>2.004). This was also true for the linewidth in the M. sexta sample ( t 7.6 G), but not in the case of A. levana ( D 5.5 G), which was similar to typical values for hair melanin (Fig. 4). The two-dimensional plot of g-factor and linewidths values from all ESR spectra, shown in Fig. 1, visualizes these diVerences. The result of A. levana is sur-

Fig. 2. Examples of ESR spectra with diVerent lineshape: (a) larva from P. brassicae (water washed); (b) pupal cuticles from H. virescens (water washed); (c) pupal cuticles from M. sexta (acetone washed); (c¤) enlarged central part of the ESR spectrum from (c). The arrows indicate hyperWne pattern.

prising as the larvae visually appear as black as those of I. io, for example, while the isolated melanin products are of obvious diVerent nature.

184

H. Kayser, C.G. Palivan / Archives of Biochemistry and Biophysics 453 (2006) 179–187

trum of M. sexta ‘melanin’, isolated from pupal cuticles, is similar to the ESR spectra of washed-only ‘sclerotized’ cuticles from this and other insects, as will be discussed below. ESR signals of cuticles from various insects

Fig. 3. ESR spectra of isolated human melanins: (a) black hair; (b) brown hair. The arrows indicate the pheomelanin type pattern.

Lepidoptera: moths Cuticle melanins that, in isolated form, provided typical ESR spectra as described above, were also studied in situ, i.e. at their original cuticular site in the exuviae. Larvae of the black mutant of M. sexta produce large amounts of black pigment in the cuticle due to deWciency of juvenile hormone [23]. This black pigment has also been shown to incorporate radiolabelled dopa and dopamine, consistent with its melanin nature [24]. Moreover, an electron microscopic study identiWed numerous electron-dense granules, reminiscent to melanins, in the outmost part of the cuticle of black larvae [16]. The present work conWrmed these earlier results, as cuticles from black larvae exhibited an ESR spectrum with a g-factor and linewidth value comparable to those of isolated melanins (Table 1; Fig. 5). It may be noted that pupal cuticles from the black mutant strain of M. sexta were comparable to those from the wild strain in their ESR spectra, as discussed below. The ESR data obtained with chestnut-brown pupal cuticles from various moths (M. sexta, S. littoralis, H. virescens, and C. vinula), though apparently scattered in the 2D-plot (see Fig. 1), are all located in cluster B, deWned by g-factors >2.004 and linewidths  > 6 G. Examples of spectra form H. virescens and M. sexta are presented in Fig. 2b and c, the latter one exhibiting intense hyperWne structure, shown in Fig. 2c¤. The C. vinula samples is characterized by the largest line width ( V 10–11 G), but still unresolved hyperWne interaction (Fig. 6a). In this case, a pheomelanin type signal should be considered in addition to the eumelanin

Fig. 4. ESR spectra of isolated insect melanins: (a) M. sexta, pupa; (b) A. levana, larva. The arrows indicate the pheomelanin type pattern.

The melanin with the most exceptional ESR spectrum was the one isolated from the chestnut-brown pupal cuticles of M. sexta (Fig. 4a), which do not show any visible black coloration that might indicate the presence of melanin. It is known that these cuticles use N--alanyldopamine as major component of sclerotization [18,19]. Therefore, this melanin-like hydrolysis product may diVer from true melanins in its precursor monomers or their way of polymerization or linkage to other cuticle components. It unlikely represents an artefact that is derived from catechols released from the cuticles upon acid hydrolysis, since no such products were observed in control experiments with added dopa or dopamine. Remarkably, the ESR spec-

Fig. 5. ESR spectra from M. sexta: (a) melanin-containing larval cuticle from black mutant; (b) melanin-free pupal cuticle. Both samples were washed with water.

H. Kayser, C.G. Palivan / Archives of Biochemistry and Biophysics 453 (2006) 179–187

185

(‘golden’) pupae of I. io, (Table 1). In pupae of both insects, P. brassicae and I. io, the black pigment appears in the light microscope as a non-granular homogeneous dark area located beneath the epicuticle [16,25]. Despite being comparable at the microscopic level, results from radiolabelling and chemical studies (I. io only) argue for quite diVerent black pigments: the black pupal pigment of I. io is a true melanin [25], that from P. brassicae pupae speciWcally uses dopamine and -alanine [16], as mentioned above. In conclusion, biochemically diVerent melanin-like polymers may produce comparable ESR signals.

Fig. 6. ESR spectra with extreme parameters: (a) C. vinula, pupa (water washed)—linewidth of 11; (b) H. virescens, pupa (NaOH washed)—g-factor of 2.0049. g represents the gyromagnetic factor.

type signal, to explain the broadening of the overall shape of the ESR spectrum [10]. It is possible that other types of copolymers contribute to this very large ESR spectrum. Of all examined cuticles, those from H. virescens exhibited the highest g-factor (up to 2.0049, largely independent of the washing medium; cf. Table 1), which is very close to values expected for cysteinyldopa- and cysteinyldopamine-melanins (Fig. 6b; see also Fig. 2b). Lepidoptera: butterXies Typical eumelanin-type ESR spectra (g-factor <2.004; linewidth <6 G; see Fig. 1) were also obtained from the black larval cuticles and their isolated pins of I. io as well as of the larval cuticles of P. brassicae (Fig. 2a) that display black patches on an otherwise transparent cuticle. While numerous electron-dense melanin-like granules have been detected by electron microscopy in the outer part of the cuticle in I. io, not any pigment-related structures have been identiWed in P. brassicae [16]. In the light microscope, the black cuticle pigment of P. brassicae larvae is represented as a homogenous dark non-granular, rather diVuse zone. This observation indicates that melanins, even when they exhibit similar ESR characteristics, may be diVerently represented at the microscopic level. Although the black-spotted pupal cuticles of P. brassicae are visually similar to the larval ones, their ESR spectra were strikingly diVerent: the signal from pupal cuticles has a broader linewidth ( 7 6 G) and a higher g-factor (>2.004), compared to that from larval cuticles (Table 1). This may reXect a remarkable biosynthetic diVerence: the black larval pigment patches in P. brassicae are labelled by dopa, while the black pupal pigment speciWcally incorporates dopamine and -alanine [16]. Dark and light pupae of this insect produced comparable ESR spectra. Similar results have been obtained with cuticles from dark and light

Diptera The ESR data for the chestnut-brown puparial cuticles from the four Xy species examined (M. domestica, Lucilia sericata, Protophormia terraenovae, and C. vicina) are all located in cluster B, together with those from moths and butterXies of similar coloration (Fig. 1). Fly cuticles were of additional interest since the insecticide cyromazine and its structural analogue dicyclanil are known to aVect dipteran puparia with lethal outcome [26]. The two compounds, which produce highly characteristic puparia of elongated shape but fairly unchanged color, did not signiWcantly aVect the overall characteristics of the ESR spectra, according to studies in two Xies (M. domestica and L. sericata). Remarkably, however, the ESR spectra of all cuticles of elongated shape from M. domestica, obtained in independent experiments by treatment with both insecticides, showed enlarged ESR signals due to external wings (especially in the low-Weld part of the spectrum) resulting from unresolved hyperWne interaction (Fig. 7). The broadening of the ESR spectrum after insecticide treatment becomes evident by a comparison with the normal spectrum

Fig. 7. Insecticide eVect on ESR spectra of puparia from M. domestica: (a) untreated control; (b) elongated form after dicyclanil treatment; (c) elongated form after cyromazine treatment. All samples were washed with water. Asterisks indicate non-resolved hyperWne pattern, which enlarge the lateral wings of the ESR spectrum.

186

H. Kayser, C.G. Palivan / Archives of Biochemistry and Biophysics 453 (2006) 179–187

obtained from those cuticles that were not aVected in their shape by the applied insecticide dose. Broad ESR signals and unresolved hyperWne structure was also found in another dipteran insect, L. sericata, after treatment with cyromazine. Hymenoptera Two species of this insect order became available for the present study. With Athalia rosae, ESR spectra similar to those from eumelanins were obtained from the black larval and pupal cuticles suggesting the presence of true melanin at both developmental stages (Fig. 8). By contrast, the light larval cuticles from M. monticola provided an ESR spectrum (not shown) comparable to that from non-melanized cuticles, as found in samples from Xies and moths. Consequently, the two hymenopteran species become allocated to diVerent clusters of ESR data (Fig. 1). Orthoptera Cuticles from the locust, L. migratoria, the single orthopteran insect studied, strongly vary in the extension of their dark patches, depending on population density and other factors [27]. The present, medium-dark sample showed an ESR spectrum with a g-factor >2.004 G, in contrast to the melanins, but with a linewidth ( t 5.3 G) as small as that obtained with melanins (either isolated or at their cuticular site). These ESR data are close to those of ‘melanized’ (i.e., black pigmented) pupal cuticles from the examined butterXies (P. brassicae and I. io), as can be seen in Fig. 1. Therefore, we suppose that in this case the ESR spectrum is also due to the eumelanin-type. The pale-white and comparably soft cuticles from an albino strain of this locust [28] did not show any ESR signal suggesting that the usually color-producing dopa-based mechanisms of cuticle sclerotization are not operative in albino forms. This raises the question how suYcient stiVening of the albino cuticle is achieved after molting to allow locomotion and feeding. In the comparable case of a mutant fruit Xy, Ceratitis capitata, white puparia result from a defective transport of precursors to the cuticle [29].

Fig. 8. ESR spectra from A. rosae: (a) pupal cuticle; (b) larval cuticle. These samples were not washed.

Spin densities obtained from the integral of the ESR spectra Besides the gyromagnetic (g) factor and the linewidth values, the double integral of the ESR spectra normalized to the sample weight and receiver gain, J¤, were calculated (see Table 1). As the spin density of the sample is proportional to the area under the resonance absorption curve, we used the double integral of the ESR spectra to compare the amount of free radicals across the samples. As expected, the highest values for J¤ were obtained for melanins, isolated either from hair or cuticles. All these normalized double integrals of ESR spectra values, J¤, were >2, with the highest one of »5 for human brown hair. For cuticles, J¤ values were <0.4, ranging between 0.01 and 0.39 (0.15 on average) and with no apparent relation to the presence of typical narrow eumelanin ESR signals versus broad signals with unresolved or resolved hyperWne structure. On the other hand, in the case of cuticles, where light as well as dark forms occur and could be separately measured, the visually darker pigmented ones contained 1.5- to 2-fold more spins compared to the light forms. This can well be seen in pupae of P. brassicae and I. io. Remarkably, cuticles of the deep black larvae of I. io with ESR signals belonging to cluster A (representing true melanins) contained twice as much spin density compared to the dark grey, but not black pupae with ESR data clustering in B (see Fig. 1). Overall, there is an obvious relation between spin density, largely independent of its chemical basis, and the intensity of dark pigmentation. This became most evident from the examination of an albino strain of the locust, mentioned above, where the completely colorless cuticles provided no ESR signal, while they were obviously suYciently hardened for normal growth. In this case, mechanical support may come only from the formation of chitin, which is not known to produce free radicals. Overview of results The present apparently Wrst ESR study of insect cuticles, using cast skins (exuviae), yielded several unexpected results. 1. Cuticles from larvae and pupae from a wide range of species contain stable free radicals characterized by a single ESR line with or without hyperWne structure. 2. Presumed to represent negative controls, non-melanized (chestnut-brown) cuticles, as those from Xy puparia and moths pupae, unexpectedly were found as rich in stable free radicals as those containing true (black) melanins. 3. Isolated melanin and melanin in situ, i.e. deposited within the cuticle, provided comparable ESR spectra. Thus, melanin can be easily identiWed from the ESR data of cuticles. 4. The g-factor and linewidths values of all ESR spectra recorded form two clusters, as displayed in a 2D plot, which can be related to the type of cuticle darkening and hardening: one represents isolated melanins and melanin-containing cuticles, and the other one describes sclerotized (but visually non-melanized) cuticles. The lat-

H. Kayser, C.G. Palivan / Archives of Biochemistry and Biophysics 453 (2006) 179–187

ter type may generally involve N--alanine as a precursor, as is known from some of the examined insects. 5. The clustering of ESR data does not reXect the visible color tone of the cuticle, meaning that black pigmentation does not necessarily indicate the presence of true melanin. 6. Spin densities of both types of cuticles (melanized and non-melanized, respectively) are of comparable levels, but more than one order of magnitude below those of true melanins. 7. ESR spectroscopy is able to detect slight changes in the radical environment arising from various impacts on cuticle biosynthesis or structure, such as insecticide or solvent exposure. Final considerations This study revealed that ESR spectroscopy allows to discriminate between two types of post-molting processes of cuticle hardening, commonly described as sclerotization, and of darkening, referred to as melanization, in a nondestructive way using relatively small amounts of material. As in the case of chemically deWned melanins, however, the characteristics of the ESR spectra of cuticles do not allow a clear assignment of the stable free radicals. The single line ESR spectra with small linewidth are obtained from free radicals were the unpaired electrons could be located nearby nuclei with I D 0, such as carbon, oxygen or sulphur nuclei, while the broadening of the ESR spectra might be due to dipolar interaction or unresolved hyperWne interaction with nitrogens (for example from the amino group of -alanyldopamine). The observation that cuticles apparently devoid of melanin provide signals that look similar to those of melanins indicates common features of melanization and sclerotization. Both processes use dopa as a precursor that undergoes further oxidation and quinoid coupling to form polymers like melanins, and that also forms links to other cuticle components, such as proteins and chitin [18,19,30,31]. The type(s) of monomers, which are mainly dopa, dopamine, N--alanyldopamine or Nacetyldopamine, and their way of polymerization and further linkage is likely to aVect the radical’s environment and thus g-factor value and shape of the ESR spectra. The native nature and location of the melanin-like material that was obtained by hydrolysis of sclerotized, not visibly melanized, cuticles, as from M. sexta pupae, remains unknown. It unlikely originated as an artefact of hydrolysis; rather it may occur dispersed within the cuticle, possibly as by-product of sclerotization due to the use of common precursors. Acknowledgments We thank Prof. G. Gescheidt for providing ESR facilities at the University of Basel, where most samples were

187

measured. This study is based on early experiments kindly performed by Dr. Rist at the former Ciba-Geigy (Basel). References [1] G. Prota, Melanins and Melanogenesis, Academic Press, San Diego, 1992. [2] P.A. Riley, Int. J. Biochem. Cell Biol. 29 (1987) 1235–1239. [3] B.C. Cope, L. Hopegood, R.J. Lathman, R.G. Linford, J.D. Reilly, M.R.C. Symons, F.A. Taiwo, J. Mater. Chem. 8 (1998) 43–45. [4] B.B. Adhyaru, N.G. Akhmedov, A.R. Katrutzky, C.R. Bowers, Magn. Reson. Chem. 41 (2003) 466–474. [5] R.C. Sealy, J.S. Hyde, C.C. Felix, I.A. Menon, G. Prota, Science 217 (1982) 545–547. [6] R.C. Sealy, J.S. Hyde, C.C. Felix, I.A. Menon, G. Prota, H.M. Swartz, S. Persad, H.F. Haberman, Proc. Natl. Acad. Sci. USA 79 (1982) 2885–2889. [7] T. Shina, T. Sarna, H.M. Swartz, A. Stroppolo, R. Gerbasi, L. Zecca, Free Radic. Biol. Med. 23 (1997) 110–119. [8] E. Buszman, B. Pilawa, T. Witoszynska, M. Latocha, T. Wilczok, Appl. Magn. Reson. 24 (2003) 401–407. [9] B. Pilawa, E. Chodurek, T. Wilczok, Appl. Magn. Reson. 24 (2003) 417–422. [10] M. Matuszczyk, E. Buszman, B. Pilawa, T. Witoszynska, T. Wilczok, Chem. Phys. Lett. 394 (2004) 366–371. [11] M.S. Blois, A.B. Zahlan, J.E. Mailing, Biophys. J. 4 (1964) 471–490. [12] R.C. Sealy, C.C. Felix, J.S. Hyde, H.M. Swartz, in: W.A. Pryor (Ed.), Free Radicals in Biology, vol. IV, 1980, pp. 209–259. [13] W.S. Enoch, M.J. Nilges, H.M. Swartz, Pigment Cell Res. 6 (1993) 91– 99. [14] M. Pasenkiewicz-Gierula, R.C. Sealy, Biochim. Biophys. Acta 884 (1986) 510–516. [15] H. Kayser, in: G.A. Kerkut, L.I. Gilbert (Eds.), Comprehensive Insect Physiology, Biochemistry and Pharmacology, vol. 10, Pergamon Press, Oxford, 1985, pp. 367–415. [16] I. Kayser-Wegmann, H. Kayser, in: K. Scheller (Ed.), The Larval Serum Proteins of Insects, Thieme, Stuttgart, 1983, pp. 151–167. [17] H.S. Mason, D.J.E. Ingram, B. Allen, Arch. Biochem. Biophys. 86 (1960) 225–230. [18] T.L. Hopkins, K.J. Kramer, Ann. Rev. Entomol. 37 (1992) 273–302. [19] S.O. Andersen, in: L.I. Gilbert, K. Iatrou, S.S. Gill (Eds.), Comprehensive Molecular Insect Science, vol. 4, Elsevier, Oxford, 2005, pp. 145– 170. [20] W. Fogal, G. Fraenkel, J. Insect Physiol. 15 (1969) 1437–1447. [21] B. Bilinska, B. Pilawa, Z. Zawada, E. Wylegala, T. Wilczok, A.E. Dontsov, N.L. Sakina, M.A. Ostrovsky, V.B. Ilyasova, Spectrochim. Acta A 58 (2002) 2257–2264. [22] A. Dzierzega-Lecznar, B. Pilawa, K. Stepien, T. Wilczok, Nukleonika 42 (1997) 343–351. [23] L. Safranek, L.M. Riddiford, J. Insect Physiol. 21 (1975) 1931– 1938. [24] M. Hori, K. Hiruma, L.M. Riddiford, Insect Biochem. 14 (1984) 267– 274. [25] A. Maisch, D. Bückmann, J. Insect Physiol. 33 (1987) 396–402. [26] K.C. Binnington, A. Retnakaran, in: K.C. Binnington, A. Retnakaran (Eds.), Physiology of The Insect Epidermis, CSIRO, Melbourne, 1991, pp. 307–334. [27] M.P. Pener, Adv. Insect Physiol. 23 (1991) 1–79. [28] E. Hasegawa, S. Tanaka, Jpn. J. Ent. 62 (1994) 315–324. [29] P. Wappner, K.J. Kramer, T.L. Hopkins, M. Merritt, J. Schaefer, L.A. Quesada-Allué, Insect Biochem. Mol. Biol. 25 (1995) 365–373. [30] M. Sugumaran, Pigment Cell Res. 15 (2002) 2–9. [31] K.J. Kramer, M.R. Kanost, T.L. Hopkins, H. Jiang, Y.C. Zhu, R. Xu, J.L. Kerwin, F. Turecek, Tetrahedron 57 (2001) 385–392.