Adsorption of bovine serum albumin onto synthetic Fe-doped geomimetic chrysotile.

Downloaded from http://rsif.royalsocietypublishing.org/ on June 3, 2015

rsif.royalsocietypublishing.org

Adsorption of bovine serum albumin onto synthetic Fe-doped geomimetic chrysotile Alessio Adamiano1,2, Isidoro Giorgio Lesci1, Daniele Fabbri1,2 and Norberto Roveri1

Research Cite this article: Adamiano A, Lesci IG, Fabbri D, Roveri N. 2015 Adsorption of bovine serum albumin onto synthetic Fe-doped geomimetic chrysotile. J. R. Soc. Interface 12: 20150186. http://dx.doi.org/10.1098/rsif.2015.0186

Received: 3 March 2015 Accepted: 7 May 2015

Subject Areas: environmental science, chemical biology, biogeochemistry Keywords: synthetic Fe-doped chrysotile, analytical pyrolysis/gas chromatography – mass spectrometry, chrysotile– bovine serum albumin adduct, 2,5-diketopiperazines, Fe surface interaction

1

Department of Chemistry ‘G. Ciamician’, Alma Mater Studiorum University of Bologna, via Selmi 2, 40126 Bologna, Italy 2 Centro Interdipartimentale di Ricerca in Scienze Ambientali Sede di Ravenna, Universita` di Bologna, via S. Alberto 163, 48100 Ravenna, Italy Synthetic stoichiometric and Fe-doped geomimetic chrysotile nanocrystals represent a reference standard to investigate the health hazard associated with mineral asbestos fibres. Experimental evidence suggests that the generation of reactive oxygen species and other radicals, catalysed by iron ions at the fibre surface, plays an important role in asbestos-induced cytotoxicity and genotoxicity. In this study, structural modification of bovine serum albumin (BSA) adsorbed onto synthetic chrysotile doped with different amounts of Fe has been investigated by Fourier transform infrared spectroscopy (FT-IR), thermal gravimetric analysis (TGA) and analytical pyrolysis coupled with gas chromatography–mass spectrometry. FT-IR data evidenced a marked increase in disordered structures like random coil and b-turn of BSA–nanocrystal adduct with 0.52 wt% of Fe doped. The TGA profile of the BSA revealed that its interaction with the synthetic chrysotile surface was strongly affected by the substitution of Fe into the chrysotile structure. The 2,5-diketopiperazine yields, formed upon thermal degradation of the polypeptide chain (pyrolysis–gas chromatography), changed when the BSA was adsorbed on the nanofibres. In general, results suggested that minute amount (less than 1 wt%) of Fe doping in chrysotile affected the protein–nanofibre interactions, supporting the role that this element may play in asbestos toxicity. The catalytic role of iron and the consequent unfolding of protein due to the structural surface modification of nanofibres were also evaluated.

1. Introduction Author for correspondence: Isidoro Giorgio Lesci e-mail: [email protected]

Asbestos is a commercial term including some magnesium silicates which crystallize in fibrous forms, including chrysotile, Mg3Si2O5(OH)4, by far the most widespread material of this type [1]. These fibres, belonging to the serpentine minerals, are constituted of stacking layers of silica (Si2O5)n22n sheets linked to octahedral brucite sheets [Mg3O2(OH)4]nþ2n. Health hazards associated with asbestos and its deleterious environmental effects are well documented in the medical and general health literature [2]. The molecular mechanisms underlying the fibrogenic and tumorigenic effects of asbestos are not yet fully understood, but chemical–physical factors related to surface reactivity and level of contaminants seem to be involved [3,4]. In natural chrysotile (Mg3Si2O5(OH)4), iron is a contaminant (about 0.5–2%, w/w), replacing isomorphously magnesium and silicon that may have important health implications. The role of Fe in the toxicity of natural chrysotile fibres has been widely investigated in the past [3,5–7], while more recent experimental evidences supported the view that the generation of reactive oxygen species and other radicals, catalysed by iron ions at the fibre surface, is implied in asbestos-induced cytotoxicity and genotoxicity [8]. The available data also suggest that the Fe content of asbestos, as well as redox active Fe associated with or mobilized from the surface of the fibres, is important in generating HO. The fibre properties that are related to their cytotoxicity [9,10] and mutagenic responses [11] are strongly affected by the surface chemical adsorption of biological molecules and macromolecules, such as proteins, cell-membrane lipids and nucleic acids [12]. Alterations in these essential cellular components can modify

& 2015 The Author(s) Published by the Royal Society. All rights reserved.


2. Experimental Reagents were from Sigma-Aldrich S.r.l., 0.06 M phosphate buffer pH 7.4 was from Riedel-de Haen Sigma-Aldrich S.r.l. and BSA was from CarloErba.

2.1. Synthetic chrysotile nanofibres Synthesis of the chrysotile nanocrystals used in this study was previously described by Falini et al. [27]. Shortly, a hydrothermal synthesis reactor (Parr Stirred ‘Mini’ reactor model 4652) with a 500 cm3 movable vessel constructed with ‘alloy C-276’ (a metal alloy that contains 6.5 wt% Fe) was used to carry out the hydrothermal reaction of SiO2—with Aerosil 380 as a powder with a surface area of about 380 m2 g21 (Eigenmann and Veronelli S.p.A.)—and MgCl2 in an aqueous NaOH solution up to pH 13 and at temperature of 3008C on the saturated vapour pressure curve (82 atm) with a run duration of 24 h. The fibrous precipitate removed from the solution was repeatedly washed with deionized water before being dried for 3 h at 1508C. In order to synthesize the series of Fedoped chrysotile samples, a gel mixture of SiO2, FeCl3 and MgCl2 in an aqueous solution was prepared. The pH of the gel mixture, which contained an Si/(Mg þ Fe) molar ratio in the range of 0.6–0.7, was adjusted to 13 by means of an aqueous NaOH solution. Concentrations of MgCl2 and FeCl3 ranging from 9.75 up to 10 mM were used. The precipitate removed from the solution was repeatedly washed with deionized water before being dried for 3 h at 1508C.

2.2. Preparation of bovine serum albumin –chrysotile nanofibre adducts Samples of adsorbed BSA were prepared by mixing 40 mg of inorganic phase with 10 ml of protein dissolved at 5 mg ml21 in phosphate buffer (0.010 M Na2HPO4, 0.140 M NaCl, 0.003 M KCl) pH 7.4 in 10 ml polyethylene tubes. The mixture was rotated end-over-end at 378C for 16 h. The solid BSA– nanofibre adduct was recovered by centrifuging at 12 700g for 3 min, washed three times with water, freeze-dried at 2608C under vacuum (3 mbar) for 12 h and stored at –208C prior to analysis. The supernatant solution was assayed for protein content by means of UV spectroscopy in order to quantify the albumin not adsorbed on the nanofibres. The amount

2

J. R. Soc. Interface 12: 20150186

BSA was adsorbed onto synthetic geomimetic chrysotile nanocrystals (or nanofibres) with different Fe-doping extents (none, 0.53 and 1.78 wt%). The amount of Fe-doping was chosen in order to simulate the observed amount of Fe in natural chrysotile [1]. The secondary structure of adsorbed BSA in the adducts was investigated by Fourier transform infrared (FT-IR) spectroscopy. Thermal gravimetric analysis (TGA) was applied in order to gather information on the extent of surface interactions. The thermal behaviour of adsorbed BSA was further investigated by analytical pyrolysis coupled to gas chromatography–mass spectrometry (GC–MS) which provided the molecular identity of the degradation products. As lung disease occurs after long periods of exposure to asbestos, we believe that the interaction of the fibres with proteins, particularly albumin, may contribute to the long latency of the disease. The main purpose of this study is to clarify how the structural changes of the surface, induced by iron, influence the interaction with biological system.


the cell functions and hence drive the cell to either neoplastic transformation or apoptosis. In order to gather information on the role of iron in these protein–asbestos surface interactions, the study of model protein–iron-doped mineral systems would be useful. Given the multiplicity of factors that make a study on naturally occurring minerals complex, geoinspired chrysotile nanocrystals have been synthesized as a unique phase with definite structure, morphology and chemical composition to be used as a standard reference sample for the investigation of the molecular interaction between chrysotile fibres and biological systems [13,14]. The geoinspired synthetic chrysotile fibres with stoichiometric composition Mg3Si2O5(OH)4, in contrast to the natural ones, do not exert any significant cytotoxic effect [15]. Moreover, the interaction of this synthetic chrysotile with bovine serum albumin (BSA) and human serum albumin (HSA) has been investigated [16–18]. With regard to chrysotile fibres doped with iron, they have been synthesized as single-tube nanocrystals with a central hole diameter of 7 nm and wall thickness of about the same size [19]. Chrysotile nanotubes show structural modifications as a function of the Fe doping extent affecting their morphological aggregation. This morphological modification has been observed when Fe preferentially replaces Si resembling the lizardite structure [20]. Fe can replace both Mg and Si, differently modifying the chrysotile structure as a function of the Fe doping extent and the Fe doping process appears strongly affected by the presence of metallic Fe in the synthetic environment. In fact, octahedral coordinated Fe has been observed in all the substitution range in Fe-doped chrysotile synthesized in absence of metallic Fe. On the contrary, tetrahedral coordinated Fe inducing a flattening of the chrysotile structure appears prevalent in respect of octahedral coordinated Fe in highly Fe-doped chrysotile synthesized when metallic Fe is available in the synthetic environment [19,21– 23]. The position of the sites responsible for catalytic and redox activity of Fe of asbestos is currently unknown, even if the fibre surface appears mainly to control the Fe reactivity. In fact, Fe3þ is stable at neutral pH and has a very low water solubility [24]. Lesci [25] observed that the exclusive substitution of Mg with Fe (at low amount of iron) increases the mismatch between the octahedral– tetrahedral spacing along the b-axis, while this mismatch is reduced by the contemporary Fe replacement to Mg and Si, as observed for Fe doped chrysotile at higher amounts of iron. These results have allowed one to relate the effect of Fe doping on the surface chemical–physical characteristics of both synthetic and mineral chrysotile. While there are few studies on the interaction of BSA/HSA onto geoinspired stoichiometric chrystotile (see above), no investigations have been made concerning the effect of Fe substitution in the chrysotile structure in modifying the protein–surface interaction. In this study, we have used albumin because this is an important and abundant plasma protein that has been extensively used as a marker in vivo, especially in lung diseases [26]. Two amounts of iron were selected in order to simulate the observed amount of Fe in natural chrysotile. Furthermore, on the basis of a recent study about the propriety of the chrysotile surface structure when iron replaces Mg in octahedral and/or Si in tetrahedral sites [25], we could predict a progressive effect with increasing Fe content. For example, we observed an increase of z-potential with increasing iron concentration that may contribute to the electrostatic interaction with the protein.


2.3. Fourier transform infrared analysis

2.4. UV spectroscopy The concentration of BSA solutions was determined by spectrophotometric analysis, by recording UV–visible spectra between 200 and 300 nm with a Perkin Elmer Cary 5 UV-Vis-NIR spectrophotometer against a blank buffer solution. The measurements were performed using a 1 cm quartz cell. The extinction coefficient of BSA at 280 nm in phosphate buffer pH 7.4 was 0.65 M21 cm21. The amount of adsorbed protein was calculated as the difference between the concentrations of the initial solution and that of the supernatant after solid separation.

The elemental distribution in the synthetic samples was obtained using inductively coupled plasma (ICP) atomic emission spectrometry (AES). The ICP-AES measurements were carried out with a Varian Liberty Model 200 analyser to allow the rapid determination of 15 elements in the wavelength range of 179–800 nm. ICP-AES analysis was carried out on sample solutions prepared according to an acidic dissolution of the samples inside hermetically sealed Teflon holders processed in a microwave mineralizer Milestone Model MLS 1200. For the sample preparation for ICP-AES analysis, the dissolution of synthetic chrysotile samples was performed following two steps: (i) 1 ml of 48 wt% HF and 5 ml of 48 wt% HNO3 were added to 40 mg of sample and processed in the mineralizer for 30 min operating at 250 W and (ii) in order to obtain the formation of BF42 complexes, 20 ml of 1.8 wt% Li2B4O7 was added to the preparation followed by reprocessing in the mineralizer for 15 min at 250 W. Prior to analysis the sample solutions were diluted up to 100 ml in volume with doubly distilled water.

2.7. Pyrolysis and sample treatment Pyrolysis was performed on pure BSA, BSA –stoichiometric chrysotile and BSA-Fe-doped chrysotile adducts. The apparatus employed for off-line pyrolysis experiments was described elsewhere [36]. Briefly, it consists of a pyrolysis chamber fitted for a pyroprobe model 1000 (CDS Analytical Inc.) equipped with a resistively heated platinum filament. A sample holder quartz tube containing an exactly weighed amount of sample (about 7– 10 mg) was introduced into the platinum coil and the probe was in turn inserted into the pyrolysis chamber. The apparatus was flushed with a nitrogen stream at 200 ml min21 prior to pyrolysis. The exit of the pyrolysis chamber was connected through a Tygonw tube to a cartridge for air monitoring containing a XAD-2 resin as adsorbent (orbo-43) purchased from Supelco. In order to collect higher amounts of pyrolysis products, several batch pyrolyses were performed with the same cartridge on the same BSA–chrysotile test samples. After pyrolysis, the cartridge was eluted with 5 ml of acetonitrile [36]. The solution was collected, concentrated under a gentle nitrogen stream and spiked with 0.1 ml of 250 mg l21 of sarcosine anhydride solution in acetonitrile prior to GC –MS. An aliquot of the acetonitrile used to elute the adsorbent cartridge was placed in a vial with 0.1 ml of 1-benzo-3-oxo-piperazine at 250 mg l21 as silylation internal standard. The obtained solution was then spiked with 60 ml of bis(trimethylsilyl)trifluoroacetamide containing 1% trimethylchlorosilane. The vial was placed in a heater at 608C for 2 h prior to GC –MS.

2.8. Gas chromatography –mass spectrometry 2.5. Differential thermal analysis –thermal gravimetric analysis Differential thermal analysis (DTA) –TGA was carried out by using a Polymer Thermal Science STA 1500 instrument. The weights of the samples were in the range of 5–10 mg. Heating was performed in a platinum crucible under a nitrogen flow (100 cm3 min21) at a rate of 108C min21 up to 7008C.

Sample solutions were injected under splitless conditions into the injector port of an Agilent 6850 gas chromatograph connected to an Agilent 5975 quadrupole mass spectrometer. Analytes were separated by a DB-5HT (Agilent Technology) fused-silica capillary column (stationary phase poly(5% diphenyl/95% dimethyl)siloxane, 30 m, 0.25 mm i.d., 0.25 mm film thickness) using helium as carrier gas (at constant pressure, 33 cm s – 1 linear velocity at 2008C). The underivatized solutions

3


FT-IR spectra were recorded on a Thermo Nicolet 380 FT-IR spectrometer. Each powdered sample (approx. 1 mg) was mixed with about 100 mg of anhydrous KBr. The mixtures were pressed into 7 mm diameter discs. Pure KBr discs were used as background. The infrared spectra were registered from 4000 to 400 cm21 at 4 cm21 resolution. Fourier self-deconvolution and second derivative resolution enhancement were applied to narrow the widths of infrared bands and increase the separation of the overlapping components. The resolution enhancement resulting from self-deconvolution and the second derivative is such that the number and position of the component bands to be fitted are determined. The curve-fitting was carried out employing Bruker OPUS peak software (v. 4.0). The number of bands was entered into the program along with their respective positions and half-heights. The program iterates the curve-fitting process to achieve the best Gaussian-shaped curves that fit the protein spectrum. A best fit is determined by the root mean square of differences between the original protein spectrum and the sum of all individual resolved bands. The assignment of component bands in amide I has been done according to the literature data [16,28–33]. The percentages of each secondary structure were calculated from the integrated areas of the component bands. Amide I bands centred between 1656 and 1652 cm21 are considered to be characteristic of a-helical structures. Fragments in a nonordered (random) conformation are associated with a broad IR band at 1644–1648 cm21. Infrared bands between 1620 and 1640 cm21 are assigned to b-strands by many authors [34]. Bands at 1695 and 1625 cm21 are typical of b-sheets, while bands at 1688, 1683, 1681 and 1616 cm21 are attributed to turns [35].

2.6. Inductively coupled plasma atomic emission spectrometry analysis


of adsorbed protein was calculated as the difference between the concentrations of the initial solution and that of the supernatant. The obtained amounts of BSA adsorbed on the nanofibres were 10.06 wt %, 9.45 wt% and 8.15 wt %, respectively, for Fe 0 wt%, Fe 0.52 wt% and 1.78 wt% samples.


4

arb. units 1700

1680

1660

1640

1620

1600

1580

arb. units

(b)

1720

1700

1680

1660

1640

1620

1720

1700

1680

1660

1640

1620

1720

1700 1680 1660 1640 wavenumber (cm–1)

1620

(c)

(d)

arb. units

The secondary structure of the adduct surface was investigated in the amide I (1700–1600 cm21) region. Fourier self-deconvolution and secondary derivative were applied to estimate number, position and width of the component bands. Fourier self-deconvolution of BSA before adsorption is compared with that of stoichiometric chrysotile and Fechrysotile after interaction with BSA. An evident BSA structural modification in amide I region can be easily appreciated after adsorption onto inorganic phase surface (figure 1). From the second derivative of the bands, the actual half-width can be estimated, based on which a curve-fitting process was iterated to achieve the best fitted curves. The component bands were assigned on the basis of the literature data [16,28–33]. From the integrated areas of the component bands, the percentages of each secondary structure for BSA as such and adsorbed on synthetic nanofibres were assigned (table 1). The protein adhesion on inorganic surface leads to a significant reduction in b-sheet as a function of the increasing Fe amounts in the chrysotile structure, with the relative contribution of b-sheet decreasing from 44 wt% (native BSA) to 11 wt% undoped adduct BSA–nanocrysotile, 7% and 4% for 0.52 and 1.78 wt% Fe doped adducts, respectively. On the contrary, the occurrences of random coil regions and b-turns were predominant in the case of BSA adsorbed onto Fe-doped chrysotile. In particular, the content of b-turns increased from 15 to 48% following adsorption onto the nanofibre, and a further increase to 58% in the case of the 0.52% Fedoped nanofibre (table 1). b-Turns are mostly composed of hydrophilic amino acids, and therefore these amino acids could be considered the most efficient agents for the surface interaction [37]. A remarkable effect of Fe doping was the increase in the proportion of random conformation, even though no significant differences were observed between the two concentrations.

arb. units

3.1. Fourier transform infrared analysis

Figure 1. FT-IR analysis of BSA (a), BSA– chrysotile adducts at 0% (b), 0.52% (c) and 1.78% (d ) of Fe content. Clearly, the lowest iron loading (0.52 wt%) gives to chrysotile fibres the potency to reduce appreciably the a-helix. Any further insertion of iron ions in the chrysotile structure determines a decrease in the interaction with a-helix and protein adhesion on inorganic surface and may be due to several factors, for instance, covalent or electrostatic chemical bonds, hydrogen and van der Waals bonds, hydrophobic


3. Results and discussion

(a)


were separated with the following temperature programmes: from 508C to 3008C (held 5 min) at 58C min21, with GC injector port maintained at 2608C. The silylated solutions were separated with the following temperature programmes: from 1008C (held 5 min) to 3108C at 58C min21, with GC injection port maintained at 2808C. GC–MS interface and quadrupole were maintained at 280 and 2508C, respectively, for all the analyses. Mass spectra were recorded in the full scan acquisition mode under electron ionization (70 eV) at 1 scan s21 in the 35–650 m/z range. The mass amount of each target analyte evolved from pyrolysis (Wganalyte) was calculated from the peak area in the total ion current chromatogram and the peak area of the appropriate internal standard. All the analytes were quantified using a unitary response factor with respect to cyclo(Gly-Leu) determined from the analysis of a single calibration solution. Yield (%) values were calculated by the formula Wganalyte/Wgads.BSA 100, where Wgads.BSA stands for the amount of adsorbed BSA on synthetic chrysotile surface determined by means of UV spectroscopy (see above). Owing to the low sample amount of BSA adducts, the precision was determined by triplicate analysis of pure BSA. The obtained relative standard deviations ranged from 3 to 47% (11% on average) and were assumed to be the same for the pyrolysis of adsorbed BSA.


Table 1. Fourier self-deconvolution and secondary derivative of FT-IR peaks in figure 1. Values are reported as secondary structure percentage detected in samples. a-helix (%)

b-sheet (%)

BSA lyophilized powder stoichiometric BSA – chrysotile adducts

38 + 2 37 + 2

44 + 1 11 + 1

3+1 4+1

15 + 1 48 + 2

BSA-Fe (0.52 wt%) chrysotile adducts BSA-Fe (1.78 wt%) chrysotile adducts

18 + 2 31 + 2

7+2 4+1

17 + 3 14 + 3

58 + 2 51 + 2

3.2. Thermal gravimetric analysis The thermogravimetric results of pure BSA are presented in the range from 1308C to 5008C in figure 2. The data show a four-step thermal degradation characterized by a high weight loss (approx. 69%) due to the protein degradation process at about 2278C, 3138C, 3218C and 4458C, respectively. Figure 3 shows the thermograms of BSA adduct with stoichiometric chrysotile and chrysotile doped with Fe at 0.52 wt% and 1.78 wt%. From TGA, the amount of BSA adsorbed onto the nanofibre can be calculated from the weight loss in the range 200–5508C. The amount of albumin coated onto synthetic nanofibres was determined in triplicate and the results were 10.38 + 0.03 wt% for stoichiometric synthetic chrysotile, and 9.58 + 0.03 wt% and 8.49 + 0.02 wt% in the case of Fedoped chrysotile at 0.52 wt% and 1.78 wt%, respectively. Apparently, the amounts of adsorbed protein on the chrysotile surface decreased on increasing Fe amounts, the same trend being observed by UV analysis on the supernatant solutions. In fact, UV spectroscopy analysis showed that the levels of BSA adsorption were 10.06 wt%, 9.46 wt% and 8.15 wt% for Fe 0 wt%, Fe 0.52 wt% and Fe 1.78 wt% samples, respectively. It is worth remarking that adsorption experiments were performed in triplicate and the observed differences were statistically significant ( p , 0.05, two tailed t-test). The temperature of maximum weight loss increased from 304 to 308 then 3138C, as the content of Fe increased from 0, to 0.52 and 1.78%, respectively. Thus, the temperature of degradation of the protein adsorbed onto the chrysotile

120

b-turn (%)

320.92°C

100 0.6 68.71% derivative weight (%/°C)

weight (%)

313.47°C

80 60 40 227.42°C

0.2

445.14°C

20 0 130

0.4

230

330 temperature (°C)

430

Figure 2. TGA analysis of pure bovine serum albumin (BSA). surface increased on increasing Fe amounts suggesting a larger protein alteration on increasing the Fe doping extent. The higher interaction between nanofibre and BSA could be due to the distortions of the crystalline cell when iron substitutes both tetrahedral and octahedral sites that modify the electric surface feature of the nanofibre [25], inducing a strong electrostatic interaction with the BSA. This hypothesis is supported by the surface characterization features reported in [25] where the value of the z-potential slowly increases in positive direction with the increasing of the iron content of the sample (about þ27.8 mV and þ32 mV for samples containing iron at 0.52 and 1.78% respectively). This occurrence is possibly due to the partial dissolution of the (FexOy)nþ species and the consequent specific adsorption of the hydrolytic cationic complexes at the surface of the sample.

3.3. Analytical pyrolysis The identity and quantity of products evolved from the thermal degradation of adsorbed BSA were investigated by analytical pyrolysis followed by GC –MS of the collected pyrolysate, directly or after trimethylsilylation. Typical examples of GC –MS traces of the original pyrolysates and after trimethylsilylation are reported in figures 4 and 5, respectively. As expected, the pyrograms are featured by typical thermal degradation products of the proteins. The most significant pyrolysis products which could be clearly attributed to specific amino acids were selected in this study. The identification and the precursor amino acids are reported in tables 2 and 3. The initial region of chromatogram is featured by the elution of compounds deriving from the lateral chain fragmentation of aromatic amino acids. At higher elution times, the GC trace is characterized by peaks due to cyclic


and hydrophilic interactions. The relative predominance of any one of them is closely related to the protein structure and the physico-chemical features of the inorganic surface. As reactivity properties of surface are related to neighbouring metal atoms [23,38,39], the reduction of protein adhesion at the higher amount of iron could be related to some changes in the coordinative state of trivalent iron, which is known to replace both Mg octahedral structural ions and Si tetrahedral ones [23]. Furthermore, Lesci et al. [25] observed that Fe exclusively substituted to Mg (at low amount of iron) increases the mismatch between the octahedral–tetrahedral spacing along the b-axis, while this mismatch is reduced by the contemporary Fe replacement to Mg and Si, as observed for Fe-doped chrysotile at higher amount of iron. The hypothesized catalytic role of iron due to the structural surface modification of nanofibres was also supported by the data of TGA and analytical pyrolysis analysis. It seems likely that, owing to the iron-induced superficial modification of nanofibres, the interaction between BSA molecules and the nanofibres is strengthened, resulting in the unfolding of proteins and their spreading on the nanofibre surface.

random (%)


sample

5


0.3

100

6


9.580%

95

10.38%

0.2

0 Fe % 0.52 Fe % 1.87 Fe %

85

304.40°C

80

0.1

307.86°C


weight (%)

90

derivative weight (%/°C)

8.486%

313.27°C 370.31°C

75

378.55°C

70 150

0 550

350 temperature (°C)

Figure 3. TGA of bovine serum albumin – chrysotile adducts at different Fe percentages.

1 15

4

is TMS

23 10 1516 8 is 20

counts

counts

6

26 25

4

31 40

10.00

20.00 min

cyclo Pro-Leu TMS

13

30.00

Figure 4. GC– MS trace (total ion) of the solution obtained after off-line pyrolysis at 5008C of BSA – chrysotile adduct doped at 0% Fe. Peak numbers correspond to compounds listed in table 2.

dipeptides (2,5-diketopiperazines, DKPs) formed upon thermal cyclization of adjacent amino acids in the protein backbone. The identified DKPs [36] are listed in tables 2 and 3. All the listed pyrolysis products were detected in the pyrolysates of the native protein and of its chrysotile adducts, irrespective of the Fe content. Although the molecular pyrolytic pattern of BSA is qualitatively similar in all the investigated samples, quantitative differences were observed. The yields of evolved pyrolysis products are reported in figures 6 and 7 for DKPs and lateral chain fragmentation products, respectively. Cyclo(Pro-Leu) is the most abundant DKP with yield of 2.7% (10% RSD) for pure BSA. The yield of cyclo(Pro-Leu) decreased when BSA is adsorbed onto undoped chrysotile and a further decrease was observed for doped nanofibres. A similar trend was observed for cyclo(Pro-Val). The yields of other DKPs were much lower and differences were less evident. However, in

10.00

9

22

20.00 min

30.00

Figure 5. GC– MS trace (total ion) of the silylated solution obtained after offline pyrolysis at 5008C of BSA adsorbed on chrysotile Fe-doped at 1.78%. Peak numbers correspond to compounds listed in table 3.

general adsorbed BSA produced higher DKP yields than pure BSA especially when containing Gly (silylated DKPs). The formation of pyrolysis products originated from the thermal degradation of the lateral chain was enhanced when BSA was adsorbed onto the nanofibres as evidenced by the higher yields of BSA –chrysotile adducts in comparison to solely BSA. The higher yields were often observed in the case of BSA adsorbed onto chrysotile doped with Fe at 0.52 wt%. Several studies present in the literature reported a decrease in a-chain and an increase in random structure as a consequence of the surface interaction between proteins and nanoparticles [16,17,21,47]. For synthetic chrysotile nanofibres, this change was proved to be related to the coating extent of the protein on the nanofibre by Artali et al. [17], who recorded a great change in protein conformation (HSA) for passing from a low (0.3 mg ml21) to a medium (1.0 mg ml21) surface coating extent. High surface coating extent (2.5 mg ml21), which in turn corresponds to high amount of protein adsorbed onto chrysotile nanofibres, appeared to be related to b-turn increase


Table 2. Compounds tentatively identified in the pyrolysates of BSA–chrysotile adducts and the amino acid precursors. Numbers refer to peak labels in figure 4. compound

1

phenol

2 3

4-methylphenol benzylnitrile

4 5

retention time (min)

origin

references

66, 94

Tyr

[40]

11.27 12.40

77, 79, 107, 108 90, 99, 116, 117

Tyr Phe

[40] [40]

benzenepropanenitrile piperidine

14.02 14.74

91, 92, 131 62, 82, 98

Phe Ala

[40] [41]

6 7

indole phenol, 5-methoxy-2-methyl

14.83 15.87

89, 90, 91, 117 110, 125, 136, 167

Trp Tyr

[40] [42]

8

methylindole

16.18

77, 96, 130, 131

Trp

[41]

9 10

pyrrolizine isoquinoline-1-ethyl

17.54 18.99

84, 97, 140, 167 129, 132, 156, 157

His Trp

[40] [41]

11 12

pyrimidinedione derivative imidazole derivative

19.39 19.74

98, 140, 153, 170 123, 138, 151, 166

Ala Trp

[41] [41]

13

quinoline-8-ethyl

20.93

128, 129, 154, 156

Trp

[41]

14 15

diketodipyrrole cyclo(Pro-Ala)

21.59 21.64

65, 93, 130, 186 70, 97, 125, 168

Pro-Pro Pro-Ala

[43] [44]

16 17

cyclo[Pro-Ala) cyclo[Pro-Gly)

22.08 22.46

70, 97, 125, 168 83, 98, 111, 154

Pro-Ala Pro-Gly

[44] [44]

18 19

cyclo[Gly-Leu) cyclo(Pro-Val)

22.60 23.49

56, 85, 99, 114, 127, (170) 70, 72, 125, 154, (196)

Gly-Leu Pro-Val

[44] [44]

20

cyclo(Pro-Val)

23.96

70, 72, 125, 154, (196)

Pro-Val

[44]

32 21

cyclo(Leu-Leu) cyclo(Leu-Leu)

25.42 25.54

86, 98, 113, 170, 226 86, 98, 113, 170, 226

Leu-Leu Leu-Leu

[36] [36]

22 23

cyclo(Pro-Pro) cyclo(Pro-Ile)

25.72 25.22

70, 96, 138, 166, 194 70, 86, 125,154, (210)

Pro-Pro Pro-Ile

[44] [36]

24

cyclo(Pro-Ile)

25.39

70, 86, 125,154, (210)

Pro-Ile

[36]

25 26

cyclo(Pro-Leu) cyclo(Pro-Leu)

25.57 25.76

70, 86, 125, 154, (210) 70, 86, 125, 154, (210)

Pro-Leu Pro-Leu

[44] [44]

27 28

cyclo(Pro-Lys) cyclo(Pro-PyroGlu)

26.50 27.09

70, 125, 154, 166, 208 70, 96, 124, 152, 180, 208

Pro-Lys Pro-Glu

[36] [36]

29

cyclo(Pro-Lys) derivative

27.29

70, 125, 154, 166, 208

Pro-Lys

[36]

30 31

cyclo(Pro-Lys) derivative cyclo(Phe-Ala)

28.76 28.97

70, 84, 152, 180, 208 44, 91, 127, 218

Pro-Lys Phe-Ala

[36] [45]

32 33

pyrazole-4-carboxaldehyde cyclo(Phe-Ala)

28.97 29.50

228 44, 91, 127, 218

His Phe-Ala

[42] [45]

34 35

cyclo(Pro-Met) cyclo(Pro-Met)

30.88 31.14

70, 139, 154, 167, 228 70, 139, 154, 167, 228

Pro-Met Pro-Met

[44] [44]

36

cyclo(Val-Phe)

30.71

91, 113, 127, 155, 246

Val-Phe

[46]

37 38

cyclo(Val-Phe) cyclo(Phe-Leu)

31.33 32.38

91, 113, 127, 155, 246 91, 113, 141, 204, 260

Val-Phe Phe-Leu

[46] [46]

39 40

cyclo(Phe-Leu) cyclo(Pro-Phe)

32.76 32.50

91, 113, 141, 204, 260 70, 91, 125, 153, 244

Phe-Leu Pro-Phe

[46] [47]

41

cyclo(Pro-Phe)

33.28

70, 91, 125, 153, 244

Pro-Phe

[47]

9.45

and b-sheet decrease, while no modifications in the a-helix band ranges were recorded. In addition, the TGA profiles of BSA–chrysotile adducts reported in figure 3 clearly show a shift of the characteristic peak of BSA towards higher

temperature at increasing Fe amount in the nanofibres. These results suggest that, despite a decrease of the coating extent in the chrysotile surface at increasing Fe concentration, a stronger interaction occurred between BSA and chrysotile surfaces.


mass spectra


no.

7


3.0

8

BSA


NF 0% 2.5

NF 0.52%

2.0

1.5


yields (wt% from BSA)

NF 1.78%

1.0

0.5

0

14 15 17 18 19 22 23 25 27 28 31 32 33 35 37 39 1* 3* 6* 8* 9* 10*14*

Figure 6. Yields (wt% from BSA) of DKPs evolved from the pyrolysis of native BSA and BSA adsorbed onto chrysotile nanofibres undoped and doped with Fe. Numbers correspond to compounds listed in tables 2 and 3, respectively, for DKPs detected in the native and in the silylated form (*).

Table 3. Compounds tentatively identified in the pyrolysates of BSA– nanochrysotile adducts after silylation. Numbers refer to peaks in figure 5. no.

compound

retention time (min)

mass spectra

origin

references

1

cyclo(Ala-Ala) diTMS

11.04

156, 171, 255, 271, 286

Ala-Ala

[36]

2 3

cyclo(Ala-Ala) diTMS cyclo(Gly-Ala) diTMS

11.42 12.31

156, 171, 255, 271, 286 129, 142, 241, 257, 272

Ala-Ala Gly-Ala

[36] [36]

4 5

parabanic acid diTMS 1H-indole, 1-TMS

12.74 13.52

73, 100, 147, 243, 258 73, 79, 174, 189

Glu Trp

— —

6 7

cyclo(Ala-Val) diTMS cyclo(Ala-Val) diTMS

13.80 14.26

156, 197, 257, 299, 314 156, 197, 257, 299, 314

Ala-Val Ala-Val

[36] [36]

8

cyclo(Gly-Gly) diTMS

14.37

73, 100, 147, 243, 258

Gly-Gly

[36]

9 10

cyclo(Gly-Val) TMS cyclo(Ala-Leu) diTMS

14.9 15.15

142, 170, 257, 285, 300 156, 170, 297, 313

Gly-Val Ala-Leu

[36] [36]

11 12

cyclo(Ala-Leu) diTMS acetonitrile, 2-phenyl N-TMS

15.27 15.44

156, 170, 297, 313 73, 105, 116, 190, 205

Ala-Leu Phe

[36] —

13

2-piperidinecarboxylic acid O-TMS

15.78

73, 147, 156, 258, 273

Ala

—

14 15

cyclo(Gly-Leu) diTMS pyroglutamic acid

16.38 17.72

156, 257, 271, 299, 314 73, 147, 156, 258 (273)

Gly-Leu Glu

[45] —

16 17

benzene amine derivative indole derivative

19.66 24.66

73, 195, 210, 251 73, 247, 321, 336

Phe Trp

— —

18

tyrosine derivative

24.75

73, 147, 179, 294, 309

Tyr

—

Adsorption onto a charged surface rather than a neutral one emphasizes the importance of electrostatic interactions [32]. At pH ¼ 7.4 (adsorption condition), BSA was negatively charged (Z potential of 25.1 mV) [48] while the chrysotile surface was supposed to be positively charged [49]. FT-IR results indicated a great increase in random coil and b-turn, consistent with a decrease of the a-chains, for all the BSA –chrysotile adducts (table 1 and figure 1) and

with respect to pure BSA. Conformational transition resulting in a higher b-turn content was also observed for the interaction between BSA and different clay surfaces [32]. The effect was evident for synthetic chrysotile samples doped with Fe, and particularly for the sample at 0.52% of Fe content for which the greatest increase in BSA random structures was recorded. Genotoxic and cytotoxic evaluations carried out on Fe-doped geoinspired chrysotile have


18

14


16

9

BSA NF 0 NK 0.52%

12 10 8 6 4 2 0 1

2

3

4

5

6

7

8

9 10 11 12 13 4* 12* 13* 15* 16* 17* 18*

Figure 7. Yields (wt% from BSA) of pyrolysis products from the lateral chain fragmentation of native BSA and BSA adsorbed onto chrysotile undoped and doped with Fe. Numbers correspond to compounds listed in tables 2 and 3, respectively, for analytes detected in the native and in the silylated form (*).

highlighted that the generation of reactive oxygen species and other radicals catalysed by Fe ions in the chrysotile fibres is potentiated when Fe ions are organized into specific chrysotile crystallographic sites, having coordination state able to activate free radical generation [8]. In a previous study, it was demonstrated that at the low amount of Fe content in the nanocrystals both biological and radical activity were more significant than the nanocrystal with higher Fe content [8,23]. This higher reactivity could be due to the higher percentage of Fe replacing Mg in octahedral position. The native conformation of a protein is tightly controlled by the shape complementarily of the hydrophobic residues that allow close packing of the cores [37] while b-turns, which consist predominantly of hydrophilic amino acid residues, are concentrated near the protein surface. This finding, together with the obtained FT-IR results, indicated that the BSA molecules are adhered on the nanofibre surface with many hydrophobic groups pointing outward. The interaction of BSA with chrysotile was supported by the inspection of the pyrolysis products arising from specific amino acids. The formation of pyrolysis products was inhibited (lower yields) or favoured (higher yields) once the BSA was adsorbed onto the nanofibres. The effect was apparently dependent on the nature of amino acid as both effects were observed. Several works describing the mechanism of DKP formation during pyrolysis have been reported in the literature [45,46,50,51]. The thermochemical path responsible for their formation seems to be influenced by temperature and the type of amino acids involved; however, the effect of interaction with mineral matrix was seldom investigated [52]. Being formed from the cyclization of the backbone it could be assumed that the initial conformation of the protein plays an important role in the formation of DKPs as well as the steric hindrance of the amino acid lateral chain. Apparently, the cyclization of sterically hindered amino acids (e.g. Pro-Leu) was reduced, while that of less hindered amino

acids (e.g. Gly) enhanced especially in the case of nanofibres doped at the 0.52% level. The thermal degradation of the side chains was seemingly favoured in the adsorbed BSA especially with doped nanofibres. The highest differences were observed for pyroglutamic acid. Polar amino acids are supposed to interact differently with the brucite layer compared to neutral amino acids [53]. In general, these effects on the yields of evolved pyrolysis products could be explained by an increase in random structures, especially of b-turns, that favour the interaction of the polypeptide chain with the chrysotile surface.

4. Conclusion In this study, we have compared the interaction of BSA with chrysotile nanofibres having different amounts of iron, without the interference of other variables (such as differences in structure, size, tubular morphology and exposed surface, or physico-chemical modifications). Our results suggest that the role of iron in asbestos –BSA interaction is more complex than that previously suggested. Iron may behave not simply as a catalytic site generating unfolding of the protein at the fibre surface but could also influence in a more complex way the interaction between chrysotile nanofibres and protein. The interaction between BSA and Fe-doped synthetic chrysotile nanofibres has been investigated by means of FTIR, TGA and pyrolysis coupled to GC –MS. An inverse correlation between the nanofibre Fe percentage and the amount of adsorbed BSA was recorded. TGA showed that the higher is the Fe percentage inside the nanofibres the stronger is the interaction between BSA and synthetic chrysotile. The existence of specific interactions was further confirmed by the molecular analysis of the thermal degradation products, whose formation was altered once the protein was adsorbed onto the fibre. Differences in the yields of cyclic dipeptides


yields (wt% from BSA)

NF 1.78%


the surface, but a stronger electrostatic effect occurs, probably due to the distortions of the crystalline cell when iron substitutes both tetrahedral and octahedral sites, modifying the electric surface feature of the nanofibre [25], inducing a strong electrostatic interaction with the BSA. While the results here presented cannot be directly extended to the interpretation of asbestos toxicity due to metal contamination, they give further support to the role of iron in pivoting the structural changes of globular proteins interacting with the surface of the mineral.

all the references were found by Web of Science, PubMed and Scopus.

Authors’ contributions. The authors meet all of the following criteria: (i) substantial contributions to the analysis and interpretation of data, (ii) drafting the article and revising it critically for important intellectual content and (iii) final approval of the version to be published.

Competing interests. We declare we have no competing interests. Funding. The authors thank the Italian ‘Ministero dell’Istruzione, MIUR’ (PRIN 2007 no. prot. 2007498XRF_005 and PRIN 2009RR5KCE_004), the University of Bologna ‘Alma Mater Studiorum’, CIRCMSB (Inter University Consortium for the Research on the Metal Chemistry in Biological Systems), Chemical Center S.r.l. and LEBSC S.r.l. for technical and financial support.

References 1.

Ross M. 1981 The geological occurrences and health hazards of amphibole and serpentine asbestos. Mineral. Rev. Mineral 9A, 279 –323. 2. Schreier H. 1989 Asbestos in the natural environment. New York, NY: Elsevier. 3. Kamp DW, Weitzman SA. 1999 The molecular basis of asbestos induced lung injury. Thorax 54, 638–652. (doi:10.1136/thx.54.7.638) 4. Fubini B, Otero C. 1999 Chemical aspects of the toxicity of inhaled mineral dusts. Chem. Soc. Rev. 28, 373–381. (doi:10.1039/a805639k) 5. Kamp DW, Graceffa P, Prior WA, Weitzman SA. 1992 The role of free radicals in asbestos-induced diseases. Free Radic. Biol. Med., 12, 293– 315. (doi:10.1016/0891-5849(92)90117-Y) 6. Hardy JA, Aust AE. 1995 Iron in asbestos chemistry and carcinogenicity. Chem. Rev. 95, 97 –118. (doi:10.1021/cr00033a005) 7. Quinlan TR, Marsh JP, Janssen YMW, Borm PA, Mossman BT. 1994 Oxygen radicals and asbestosmediated disease. Environ. Health Perspect. 102, 107–110. (doi:10.1289/ehp.94102s10107) 8. Gazzano E et al. 2007 Iron-loaded synthetic chrysotile: a new model solid for studying the role of iron in asbestos toxicity. Chem. Res. Toxicol. 20, 380–387. (doi:10.1021/tx600354f ) 9. Davis JM, Addison J, Bolton RE, Donaldson K, Jones AD, Smith T. 1986 The pathogenicity of long versus short fiber samples of amosite asbestos administered to rats by inhalation and intraperitoneal injection. Br. J. Exp. Pathol. 67, 415–430. 10. Miller K. 1978 The effects of asbestos on macrophages. Rev. Toxicol. 5, 319–354. (doi:10. 3109/10408447809081010)

11. Valerio F, De Ferrari M, Ottaggio L, Repetto E, Santi L. 1980 Cytogenetic effects of Rhodesian chrysotile on human lymphocytes in vitro. Mutat. Res. 122, 397 –402. (doi:10.1016/0165-7992(83)90026-X) 12. Kamp DW, Panduri V, Weitzman SA, Chandel N. 2002 Asbestos-induced alvelolar epithelial cell apoptosis: role of mithochondrial disfunction caused by iron-derived free radicals. Mol. Cell. Biochem. 234/235, 153–160. (doi:10.1023/ A:1015949118495) 13. Roveri N, Falini G, Foresti E, Fracasso G, Lesci IG, Sabatino P. 2006 Geoinspired synthetic chrysotile nanotubes. J. Mater. Res. 21, 2711 –2725. (doi:10. 1557/jmr.2006.0359) 14. Pierini F, Foresti E, Fracasso G, Lesci IG, Roveri N. 2010 Potential technological applications of synthetic geomimetic nanotubes. Isr. J. Chem. 50, 484 –499. (doi:10.1002/ijch.201000062) 15. Gazzano E, Foresti E, Lesci IG, Tomatis M, Riganti C, Fubini B, Roveri N, Ghigo D. 2005 Different cellular responses evoked by natural and stoichiometric synthetic chrysotile asbestos. Toxicol. Appl. Pharmacol. 206, 356–364. (doi:10.1016/j.taap. 2004.11.021) 16. Falini G, Foresti E, Lesci IG, Lunelli B, Roveri N, Sabatino P. 2006 Bovine serum albumin interaction with chrysotile: spectroscopic and morphological studies. Chem. Eur J. 12, 1968 –1974. (doi:10.1002/ chem.200500709) 17. Artali R, Del Pra A, Foresti E, Lesci IG, Roveri N, Sabatino P. 2008 Adsorption of human serum albumin on the chrysotile surface: a molecular dynamics and spectroscopic investigation. J. R. Soc. Interface 5, 273–283. (doi:10.1098/rsif.2007.1137)

18. Sabatino P, Casella L, Granata A, Iafisco M, Lesci IG, Monzani E, Roveri N. 2007 Synthetic chrysotile nanocrystals as a reference standard to investigate surface induced serum albumin structural modifications. J. Colloid Interface Sci. 314, 389– 397. (doi:10.1016/j.jcis.2007.05.081) 19. Foresti E, Hochella MF, Kornishi H, Lesci IG, Madden AS, Roveri N, Xu H. 2005 Morphological and chemical/physical characterization of Fe-doped synthetic chrysotile nanotubes. Adv. Funct. Mater. 15, 1009 –1016. (doi:10.1002/adfm.200400355) 20. Foresti E, Fornero E, Lesci IG, Rinaudo C, Zuccheri T, Roveri N. 2009 Asbestos health hazard: a spectroscopic study of synthetic geoinspired Fe-doped chrysotile. J. Hazard. Mater. 162, 1494– 1506. (doi:10.1016/j.jhazmat.2008.06.066) 21. Piperno S, Kaplan-Ashiri I, Cohen SR, Popovitz-Biro R, Wagner HD, Tenne R, Foresti E, Lesci IG, Roveri N. 2007 Characterization of geoinspired and synthetic chrysotile nanotubes by atomic force microscopy and transmission electron microscopy. Adv. Funct. Mater. 17, 3332– 3338. (doi:10.1002/ adfm.200700278) 22. Borghi E, Occhiuzzi M, Foresti E, Lesci IG, Roveri N. 2010 Spectroscopic characterization of Fe-doped synthetic chrysotile by EPR, DRS and magnetic susceptibility measurements. Phys. Chem. Chem. Phys. 12, 227–238. (doi:10.1039/B915182F) 23. Turci F, Tomatis M, Lesci IG, Roveri N, Fubini B. 2011 The iron-related molecular toxicity mechanism of synthetic asbestos nanofibers: a model study for high-aspect-ratio nanoparticles. Chem. Eur. J. 17, 350– 358. (doi:10.1002/chem. 201001893)


Data accessibility. No supporting data were produced for this paper and

10


derived from backbone cyclization and products arising from side chain cracking were observed from solely BSA and BSA adsorbed onto the nanofibres. The differences depended on the nature of the amino acid involved and the content of Fe. Although clear trends were not apparent, it was evident that tiny amounts of Fe doping (less than 1%) had a significant effect on the thermal stability of amino acids, and in particular peptide sequences leading to DKPs. Since DKPs are formed by the cyclization of the backbone, the secondary structure may play an important role. FT-IR analysis recorded an increase in b-turn and random coil fraction inside BSA adsorbed on nanofibre surface, showing an increase in the disordered structure larger for the adducts at 0.52% of Fe. In fact, as shown by TGA/DTA, the thermal stability of the proteins adsorbed on the chrysotile surface increased on increasing Fe amounts, highlighting a better adhesion of the protein which increased the Fe doping extent. The obtained results in this study suggest two modalities of interaction of the surface of chrysotile with the BSA. In fact, a lower amount of iron (0.52 wt%) corresponds to a greater catalytic effect of the surface of chrysotile in respect to the protein, in the sense of the unfolding of the a-helix. On the contrary, at higher concentration of the iron in chrysotile we observe a reduced catalytic effect of


35.

36.

38.

39.

40.

41.

42.

43.

44. Stankiewicz BA, Hutchins JC, Thompson R, Briggs DEG, Evershed RP. 1997 Assessment of bog-body tissue preservation by pyrolsyis-gas chromatography/mass spectrometry. Rapid Commun. Mass Spectrom. 11, 1884 –1890. (doi:10.1002/ (SICI)1097-0231(199711)11:17,1884::AID-RCM 62.3.0.CO;2-5) 45. Johnstone RAW, Povall TJ. 1975 Methods of peptide sequencing. Part I. Conversion of oligopeptides into cyclic dipeptides: a gas chromatographic-mass spectrometric study. J. Chem. Soc. Perkin Trans. 1 1297– 1300. (doi:10.1039/p19750001297) 46. Ginz M, Engelhardt UH. 2001 Identification of new diketopiperazines in roasted coffee. Eur. Food Res. Technol. 213, 8– 11. (doi:10.1007/s002170100322) 47. Chen YH, Liou SE, Chen CC. 2004 Two-step mass spectrometric approach for the identification of diketopiperazines in chicken essence. Eur. Food Res. Technol. 218, 589– 597. (doi:10.1007/s00217-0040901-x) 48. Salis A, Bostrom M, Medda L, Cugia F, Barse B, Parsons DF. 2011 Measurements and theoretical interpretation of points of zero charge/potential of BSA protein. Langmuir 27, 11 597–11 604. (doi:10. 1021/la2024605) 49. Ozeki S, Takano I, Shimizu M, Kaneko KJ. 1989 z-potential of synthetic chrysotile asbestos in aqueous simple salt solutions. J. Colloid Interface Sci. 132, 523–531. (doi:10.1016/00219797(89)90266-X) 50. Smith GG, Reddy GS, Boon JJ. 1988 Gas chromatographic-mass spectrometric analysis of the Curie-point pyrolysis products of some dipeptides and their diketopiperazines. J. Chem. Soc. Perkin Trans. 2 203 –211. (doi:10.1039/p29880000203) 51. Ledesma EB, Kalish MA, Nelson PF, Wornat MJ, Mackie JC. 2006 Formation and fate of PAH during the pyrolysis and fuel-rich combustion of coal primary tar. Fuel 79, 1801 –1814. (doi:10.1016/ S0016-2361(00)00044-2) 52. Fabbri D, Adamiano A, Torri C. 2010 Pyrolysis of proteins in the presence of active inorganic solids. In 18th European Biomass Conf. and Exhibition, Lyon, France, 3– 7 May 2010, pp. 1498–1500. Florence, Italy: ETA-Florence Renewable Energies. 53. Rimola A, Sodupe M, Ugliengo P. 2009 Affinity scale for the interaction of amino acids with silica surfaces. J. Phys. Chem. C 113, 5741–5750. (doi:10. 1021/jp811193f)

11


37.

proteins. Adv. Protein Chem. 38, 181 –364. (doi:10. 1016/S0065-3233(08)60528-8) Bramanti E, Benedetti E. 1996 Determination of the secondary structure of isomeric forms of human serum-albumin by a particular frequency deconvolution procedure applied to Fouriertransform IR analysis. Biopolymers 38, 639–653. (doi:10.1002/(SICI)1097-0282(199605)38: 5,639::AID-BIP8.3.0.CO;2-T) Fabbri D, Adamiano A, Falini G, De Marco R, Mancini I. 2012 Analytical pyrolysis of dipeptides containing proline and amino acids with polar side chains. Novel 2,5-diketopiperazine markers in the pyrolysates of proteins. J. Anal. Appl. Pyrolysis 95, 145 –155. (doi:10.1016/j.jaap.2012.02.001) Word JM, Lovell SC, Richardson JS, Richardson DC. 1999 Asparagine and glutamine: using hydrogen atom contacts in the choice of side-chain amide orientation. J. Mol. Biol. 285, 1735 –1747. (doi:10. 1006/jmbi.1998.2401) Zecchina A, Spoto G, Ghiotti G, Garrone E. 1994 Cr2þ ions grafted to silica and silicate surfaces: FTIR characterization and ethylene polymerization activity. J. Mol. Catal. 86, 423– 446. (doi:10.1016/ 0304-5102(93)E0158-D) Fubini B, Bolis V, Cavenago A, Garrone E, Ugliengo P. 1993 Structural and induced heterogeneity at the surface of some silica polymorphs from the enthalpy of adsorption of various molecules. Langmuir 9, 2712–2720. (doi:10.1021/la00034a034) Chiavari G, Galletti G. 1992 Pyrolysis-gas chromatography/mass spectrometry of amino acids. J. Anal. Appl. Pyrolysis 24, 123 –137. (doi:10.1016/ 0165-2370(92)85024-F) Knicker H, Roãô JC, Hatcher PG, Minard RD. 2001 Identification of protein remnants in insoluble geopolymers using TMAH thermochemolysis/ GC+MS. Organic Geochem. 32, 397–409. (doi:10. 1016/S0146-6380(00)00186-8) Bracewell JM, Robertson GW. 1984 Quantitative comparison of the nitrogen-containing pyrolysis products and amino acid composition of soil humic acids. J. Anal. Appl. Pyrolysis 6, 19 –29. (doi:10. 1016/0165-2370(84)80002-9) Kurata S, Ichikawa K. 2008 Identification of small bits of natural leather by pyrolysis gas chromatography/mass spectrometry. Bunseki Kagaku 57, 563– 569. (doi:10.2116/bunsekikagaku. 57.563)


24. Selikoff J, Lee DHK. 1978 Asbestos and disease, pp. 358–375, ch. 14. New York, NY: Academic Press. 25. Lesci IG, Balducci G, Pierini F, Soavi F, Roveri N. 2014 Surface features and thermal stability of mesoporous Fe doped geoinspired synthetic chrysotile nanotubes. Micropor. Mesopor. Mater. 197, 8 –16. (doi:10.1016./j.micromeso.2014.06.002) 26. Hackett TL, Scarci M, Zheng L, Tan W, Treasure T, Warner JA. 2010 Oxidative modification of albumin in the parenchymal lung tissue of current smokers with chronic obstructive pulmonary disease. Respiratory Res. 11, 180–190. (doi:10.1186/14659921-11-180) 27. Falini G, Foresti E, Gazzano M, Gualtieri AF, Leoni M, Lesci IG, Roveri N. 2004 Tubular-shaped stoichiometric chrysotile nanocrystals. Chem. Eur. J. 10, 3043 –3049. (doi:10.1002/chem.200305685) 28. Iafisco M, Sabatino P, Lesci IG, Prat M, Rimondini L, Roveri N. 2010 Conformational modifications of serum albumins adsorbed on different kinds of biomimetic hydroxyapatite nanocrystals. Colloids Surf. B Biointerfaces 81, 274– 284. (doi:10.1016/j. colsurfb.2010.07.022) 29. Jakobsen RJ, Wasacz FM. 1990 Infrared spectrastructure correlations and adsorption behavior for helix proteins. Appl. Spectr. 44, 1478 –1489. (doi:10.1366/0003702904417724) 30. Carrasquillo KG, Aponte-Carro JC, Alejandro A, Diaz Toro D, Griebenow K. 2001 Reduction of structural perturbations in bovine serum albumin by non-aqueous microencapsulation. J. Pharm. Pharmacol. 53, 115– 120. (doi:10.1211/00223 57011775091) 31. Fu K, Griebenow K, Hsieh L, Klibanov AM, Langer R. 1999 FTIR characterization of the secondary structure of proteins encapsulated within PLGA microspheres. J. Controlled Release 58, 357 –366. (doi:10.1016/S0168-3659(98)00192-8) 32. Servagent-Noinville S, Revault M, Quiquampoix H, Baron M-H. 2000 Conformational changes of bovine serum albumin induced by adsorption on different clay surfaces: FTIR analysis. J. Colloid Interface Sci. 221, 273–283. (doi:10.1006/jcis.1999.6576) 33. Pelton JT, McLean LR. 2000 Spectroscopic methods for analysis of protein secondary structure. Anal. Biochem. 277, 167– 176. (doi:10.1006/abio. 1999.4320) 34. Krimm S, Bandekar J. 1986 Vibrational spectroscopy and conformation of peptides, polypeptides, and

Adsorption characteristics of bovine serum albumin onto alumina with a specific crystalline structure.

Adsorption of human serum albumin onto highly orientated pyrolytic graphite surface studied by atomic force microscopy.

Albumin adsorption onto surfaces of urine collection and analysis containers.

Interfacial thermodynamics of protein adsorption and ion co-adsorption. III. Electrochemistry of bovine serum albumin adsorption on silver iodide.

Morphological effect of gold nanoparticles on the adsorption of bovine serum albumin.

Water Interface: A Neutron Reflection Study.

Interpreting the adsorption of serum albumin and lactoglobulin onto ZnS nanopaticles: effect of conformational rigidity of the proteins.

Effects of surface functionalization on the adsorption of human serum albumin onto nanoparticles - a fluorescence correlation spectroscopy study.

Trehalose covalently conjugated to bovine serum albumin.

ATP binding to bovine serum albumin.

Bovine serum albumin interacts with bacterial luciferase.

Brdicka currents observed with bovine serum albumin and completely reduced bovine serum albumin in the presence of urea.

Facile synthesis of hairy core-shell structured magnetic polymer submicrospheres and their adsorption of bovine serum albumin.

Effects of polythiophene surface structure on adsorption and conformation of bovine serum albumin: a multivariate and multitechnique study.

Nanoparticle-protein interactions: a thermodynamic and kinetic study of the adsorption of bovine serum albumin to gold nanoparticle surfaces.

Applicability of the extended Derjaguin-Landau-Verwey-Overbeek theory on the adsorption of bovine serum albumin on solid surfaces.

Adsorption of bovine serum albumin on Zr co-sputtered a-C(:H) films: Implication on wear behaviour.

Complexes of xylan and synthetic polyelectrolytes. Characterization and adsorption onto high quality unbleached fibres.

Competitive adsorption of magnesium and calcium ions onto synthetic and biological apatites.

Kinetics of Thermal Denaturation and Aggregation of Bovine Serum Albumin.

Immunological properties of peptic fragments of bovine serum albumin.

Charge--transfer interaction involving sulfhydryl groups of bovine serum albumin.

Binding of commercial diuretics with bovine serum albumin.

Photochemistry of modified proteins. Benzophenone-containing bovine serum albumin.