Toxicology in Vitro 28 (2014) 1396–1401
Contents lists available at ScienceDirect
Toxicology in Vitro journal homepage: www.elsevier.com/locate/toxinvit
In vitro permeation of platinum and rhodium through Caucasian skin A. Franken a,⇑, F.C. Eloff a, J. Du Plessis b, C.J. Badenhorst a, A. Jordaan c, J.L. Du Plessis a a
North-West University, Potchefstroom Campus, Private Bag X6001, Potchefstroom 2520, South Africa Centre of Excellence for Pharmaceutical Sciences, North-West University, Private Bag X6001, Potchefstroom 2520, South Africa c Laboratory for Electron Microscopy CRB, North-West University, Private Bag X6001, Potchefstroom 2520, South Africa b
a r t i c l e
i n f o
Article history: Received 1 April 2014 Accepted 19 July 2014 Available online 30 July 2014 Keywords: Platinum group metals (PGMs) Skin sensitisation Franz diffusion cells Caucasian skin Skin permeation
a b s t r a c t During platinum group metals (PGMs) refining the possibility exists for dermal exposure to PGM salts. The dermal route has been questioned as an alternative route of exposure that could contribute to employee sensitisation, even though literature has been focused on respiratory exposure. This study aimed to investigate the in vitro permeation of platinum and rhodium through intact Caucasian skin. A donor solution of 0.3 mg/ml of metal, K2PtCl4 and RhCl3 respectively, was applied to the vertical Franz diffusion cells with full thickness abdominal skin. The receptor solution was removed at various intervals during the 24 h experiment, and analysed with high resolution ICP-MS. Skin was digested and analysed by ICP-OES. Results indicated cumulative permeation with prolonged exposure, with a significantly higher mass of platinum permeating after 24 h when compared to rhodium. The mass of platinum retained inside the skin and the flux of platinum across the skin was significantly higher than that of rhodium. Permeated and skin retained platinum and rhodium may therefore contribute to sensitisation and indicates a health risk associated with dermal exposure in the workplace. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction In South Africa the platinum group metals (PGMs) mining and refining industry is one of the largest components in the mining sector (Chamber of Mines, 2012). The mining of PGMs include the precious metals platinum (Pt), palladium (Pd), rhodium (Rh), ruthenium (Ru), iridium (Ir) and osmium (Os) (Ravindra et al., 2004). These precious metals have excellent catalytic properties and are resistant to corrosion, resulting in these metals being valuable in various industries, increasing the demand for these metals (Iavicoli et al., 2012). The mining of PGMs and the subsequent smelting, refining or recycling processes results in workers being potentially exposed to these metals, in various forms, on a daily basis. Refinery workers are often simultaneously exposed to the salts of platinum, rhodium and palladium (Kielhorn et al., 2002). Secondary industries such as catalyst manufacturers, electronic industries and jewellery fabrication utilise these precious metals leading to potential occupational exposure in industries other than refineries (Kielhorn et al., 2002; Cristaudo et al., 2005). One of the hazards during metal refining and handling of metals is worker ⇑ Corresponding author. Tel.: +27 (0) 182992437; fax: +27 (0) 182991053. E-mail addresses: [email protected] (A. Franken), [email protected] (F.C. Eloff), [email protected] (J. Du Plessis), cas.badenhorst@ angloamerican.com (C.J. Badenhorst), [email protected] (A. Jordaan), [email protected] (J.L. Du Plessis). http://dx.doi.org/10.1016/j.tiv.2014.07.007 0887-2333/Ó 2014 Elsevier Ltd. All rights reserved.
dermal exposure, as verified by Du Plessis et al. (2010, 2013) indicating dermal exposure to nickel and/or cobalt in base metal refineries. No published information is available on the extent of worker dermal exposure to PGM salts, and no in vitro information is available on the possibility of skin permeation. The potent allergenicity of halide platinum salts is well documented, however the allergenicity of rhodium salts is less known (Cleare et al., 1976; Calverley et al., 1995). Platinum salts are well known respiratory sensitisers, although the skin sensitisation potential thereof is less documented. The National Institute for Occupational Safety and Health (NIOSH) pocket guide to chemical hazards identifies the skin as an exposure route and a target organ to soluble platinum as salts, with symptoms including dermatitis and sensitisation of the skin (NIOSH, 2013). Platinum salt sensitivity occurs when haptens combine with serum proteins to form antigens, and is known as a type 1, immunoglobulin IgE-mediated response. Symptoms of platinum salt sensitisation include respiratory symptoms such as tightness in the chest and wheezing, and also dermal symptoms such as eczema and urticaria (Kiilunen and Aitio, 2007). Soluble rhodium salts are considered to be sensitisers, with studies reporting allergic reactions including respiratory symptoms, urticaria and contact dermatitis in occupationally exposed subjects (Santucci et al., 2000; Goossens et al., 2011). NIOSH (2013) acknowledges the skin as a possible exposure route for soluble rhodium; however no possible skin symptoms are
A. Franken et al. / Toxicology in Vitro 28 (2014) 1396–1401
listed. In the past rhodium was believed to be allergologically safe, and was used to plate other metals to avoid dermatitis in persons sensitised to nickel and cobalt. However two case studies reported dermatitis of workers using rhodium solutions in the jewellery industry (Bedello et al., 1987; De La Cuadra and Grau-Massanées, 1991). De La Cuadra and Grau-Massanées (1991) concluded that rhodium as a salt may be a potential sensitiser, but not as a metal. Studies have concluded that the compounds that are capable of eliciting reactions are a small group of charged compounds containing a reactive ligand system, such as a chloride or bromide. Therefore it is the charge and overall reactivity of a compound that is directly related to its allergenicity (Cleare et al., 1976). During the refining process as well as in secondary industries these metals are found in numerous chemical forms including the most commonly found halide complexes (salts) (Bolm-Audorff et al., 1992; Linnett and Hughes, 1999; Cristaudo et al., 2005). Previous studies acknowledge platinum as a sensitiser, manifesting in respiratory and dermal symptoms although these studies focused on respiratory exposure (Calverley et al., 1995; Linnett and Hughes, 1999; Merget et al., 2000). The literature suggests sensitisation to occur primarily through respiratory exposure, not taking the dermal route of exposure into consideration. Maynard et al. (1997) however suggested that dermal exposure to these metals could be an alternative route contributing to sensitisation. Their results indicated that respiratory exposure to platinum were below the eight hour and short term occupational exposure limits (OEL), yet exposure resulted in sensitisation to soluble platinum in refinery workers. Numerous in vitro studies have proven that metals are able to permeate through intact human skin, and permeation increases in damaged skin (Larese Filon et al., 2007, 2009). These studies included nickel, cobalt and chromium; however no in vitro studies are available for PGMs. This study aimed to establish if a platinumand rhodium salt are able to permeate through intact Caucasian skin by making use of an in vitro diffusion system.
1397
2.2. Preparation of skin membranes Human abdominal full thickness skin was obtained after surgical abdominoplasty procedures with the approval of the NorthWest University Ethics Committee (NWU-00114-11-A5). Informed consent was given by the donors. The skin was stored in a freezer at 18 °C for a maximum period of four months. A study done by Harrison et al. (1984) reported that freezing the skin did not affect the absorption of water by comparing frozen and fresh skin segments. On the day of the experiment, the skin was allowed to thaw, and all subcutaneous fat was removed. The skin was cut into circles with a diameter of 2.4 cm and clamped between the two parts of the Franz cell. Skin integrity was tested before and after each experiment using a Tinsley LCR Data bridge (Model 6401). The measurement was taken at 1 kHz in the parallel equivalent circuit using two stainless steel electrodes (Fasano and Hinderliter, 2004). Cells with an electrical resistance below 10 kX were rejected, and cells with similar resistance values within one skin donor were chosen for diffusion studies. Skin was obtained from Caucasian women between the ages of 37 and 52. In total skin from four different donors were used, skin from two donors were used to establish platinum permeation, and skin from two donors were used to establish rhodium permeation. 2.3. In vitro diffusion system Vertical Franz diffusion cells with a receptor compartment of 2 ml were used to perform the percutaneous permeation studies (Franz, 1975). The receptor compartment was maintained at a temperature of 37 °C to simulate the normal temperature below the skin. The physiological solution used as receptor solution was continuously stirred using a magnetic stirrer bar. Each piece of skin was clamped between the two compartments of the Franz cell with a stainless steel clamp. The mean exposed skin area was 1.066 cm2, and the thickness of the skin was measured as less than 1 mm. 2.4. Platinum permeation
2. Materials and methods 2.1. Chemicals All chemicals were analytical grade. Ammonia (32%), hydrochloric acid (37%) and potassium di-hydrogen phosphate were purchased from Merck, South Africa; sodium chloride, urea, lactic acid (88–92%) and disodium hydrogen phosphate were purchased from Sigma Aldrich, South Africa. Nitric acid (65%) and hydrochloric acid (33%) were purchased from De Bruyn Spectroscopic Solutions, South Africa. Hydrogen peroxide (50%) and acetone were purchased from Associated Chemical Enterprises, South Africa. Potassium tetrachloroplatinate and rhodium trichloride were obtained from Johnson Matthey and sponsored by Anglo American Technical Solutions, South Africa. Type 1 water used for the solutions was produced with a Millipore purification pack system (milliQ water). A 0.9% sodium chloride solution was used to test the integrity of the skin. The physiological solution (receptor solution) was prepared by dissolving 4.76 g Na2HPO4, 0.38 g KH2PO4 and 18 g NaCl into 2 L of milliQ water, the final pH was adjusted to 7.35. The synthetic sweat solution used as the donor solution was prepared with 0.5% NaCl, 0.1% lactic acid and 0.1% urea in milliQ water. For the platinum donor solution 0.03233 g potassium tetrachloroplatinate (K2PtCl4) was dissolved in 50 ml synthetic sweat where after the pH was adjusted to 6.5 with ammonia. For the rhodium donor solution 0.04306 g of rhodium (III) chloride (RhCl3) was dissolved in 50 ml synthetic sweat and the pH was adjusted to 6.5. A final concentration of 0.3 mg/ml of each metal was used as the donor solution.
The receptor compartment was filled with 2 ml of physiological (receptor) solution, and the donor compartment was filled with 1 ml of the K2PtCl4 dissolved in synthetic sweat. Ten cells were used as experimental cells, and 10 cells as blanks. The receptor solution was removed at 6, 8, 10, 12, 14 and 24 h and placed into clearly marked vials. During each removal the receptor compartment was rinsed with an additional 2 ml of physiological solution, and this rinsing solution was added to the original sample removed. The receptor compartment was rinsed to ensure that all the platinum in this compartment was removed at the specific time interval. After completion of the experiment the donor solution was removed and placed into a vial, and the compartment was rinsed four times with 1 ml of synthetic sweat each time to remove all remaining platinum from the skin. The donor rinsing solution was added to the original donor solution for analysis. After testing the skin integrity the cell was dismantled and the skin was placed into a vial for analysis. 2.5. Rhodium permeation The above mentioned experiment was repeated by dissolving RhCl3 in synthetic sweat as donor solution. 2.6. Blanks The blank cells were treated the same as the other cells with the exception that no platinum or rhodium was added to the synthetic sweat placed in the donor compartment.
1398
A. Franken et al. / Toxicology in Vitro 28 (2014) 1396–1401
2.7. Skin content evaluation The skin pieces were removed from the vial and weighed for skin digestion purposes. The vials were rinsed with acetone to remove any residue from the vial. The rinsed solution was placed into a beaker with the skin where the acetone was evaporated using a hot plate. Nitric acid and hydrogen peroxide were added to the skin mixture in separate stages and evaporated to destroy the organic material. Hydrochloric acid was added to the beaker to ensure all the platinum was in a soluble state. The final solution was diluted to 10 ml with hydrochloric acid, which was analysed by Inductively Coupled Plasma-Optical Emission Spectrometry (ICP–OES) (Spectro Arcos) to establish the concentration of the metals remaining in the skin. 2.8. Standardisation experiments Various standardisation experiments were conducted to establish the appropriate concentration for the donor solution; these experiments included receptor solution removals at one, two and four hours. Analyses of these results predominantly indicated values close to or below the detection limit of analysis. These experiments also indicated that exposure up to six hours did not contribute to the steady state flux through the skin, and therefore removals at these time intervals were excluded from subsequent experiments. 2.9. Analytical measurements The receptor solutions were analysed with high resolution Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) (Thermo Scientific Element XR) to establish the concentration of platinum and rhodium in the solutions. An internal standard of 10 lg/L rhenium was added to all the solutions to compensate for volume inaccuracies. For platinum the analysis was done using the Pt195 isotope in low resolution mode, whereas the Rh103 isotope was used in high resolution mode for rhodium. The instrument was calibrated with Pt or Rh calibration standards that were matrix matched with a range of 0, 5, 10, 20, 50, and 100 ng/L solutions containing platinum, rhodium and the internal standard rhenium to provide the initial calibration curves. If necessary, the calibration was extended to 1, 2, 5, 10, 20 or even 50 lg/L if the sample concentrations required it. The lower limit of detection for platinum was 30 ng/L and 20 ng/L for rhodium. The concentration of platinum and rhodium in the donor solutions and in the solutions resulting from the skin digestion were analysed by ICP-OES (Spectro Arcos). An internal standard of 4 mg/L yttrium was added to the solutions to compensate for volume inaccuracies. The wavelengths used for determination was 265.945 nm and 214.423 nm for platinum, and 343.489 nm for rhodium. The instrument was calibrated with a standard solution of 0, 1, 2, 3, 4, 5, 10 and 20 mg/L to provide the calibration curves. 2.10. Transmission electron microscopy (TEM) Solutions of K2PtCl4 and RhCl3 dissolved in synthetic sweat (donor solutions) were characterised with a FEI Tecnai G2 high resolution transmission electron microscope operated at 200 kV. Micrographs were captured and particles measured using DigitalMicrographTM (Gatan Inc.) computer software. 2.11. Data analysis The concentration of platinum and rhodium (ng/cm3) in the receptor solution was converted to the total mass that permeated through the exposed skin area (ng/cm2) and then plotted against
time as the cumulative mass of metal that permeated per area of skin. Flux permeation was calculated from the steady-state region of the graph after curve-fitting. Eq. (1) was developed by Díez-Sales et al. (1991) to describe the mass of substance crossing a membrane at a given time:
" !# 1 t 1 2 X ð1Þn Dn2 p2 t QðtÞ ¼ AKhC m D 2 exp 2 h 6 p n¼1 n2 h
ð1Þ
where Q(t) is the mass that passed through the membrane at a given time t, A is the actual surface diffusion area, K is the partition coefficient between the skin and donor solution, h is the membrane thickness, Cv is the concentration in the donor solution and D is the diffusion coefficient of the permeant in the skin. As t approaches infinity the exponential term becomes negligible and the equation can be simplified to Eq. (2):
t 1 QðtÞ ¼ AKhC m D h 6
ð2Þ
Since K, D and h are unknown, the products Kh and D/h2 were replaced with a and b which were calculated by fitting Eq. (2) to the data by using Easyplot Version 4.0.5 (Spiral Software, Aerious Ltd.). The permeability coefficient (kp) and flux (J) were calculated using Eqs. (3) and (4) respectively:
kp ¼
KD ð¼ abÞ h
J ¼ kp C m
ð3Þ ð4Þ
Lag time was calculated as the intercept of the steady-state region with the x-axis. Average blank concentrations were subtracted from experimental data. Data analysis was performed using Statistica Version 11 (Statsoft Inc.). Data was not normally distributed, and therefore Box– Cox transformed (Box and Cox, 1964). Data is reported as mean ± standard error of means (SEM). The difference between independent Box–Cox transformed data was assessed by means of the t-test with separate variance. A p value of 60.05 was considered as statistically significant. 3. Results Platinum and rhodium were able to permeate through intact Caucasian skin, and permeation increased cumulatively over time. After 8 h of skin exposure, permeation of platinum into the receptor solution was 0.53 ± 0.18 ng/cm2 and rhodium permeation was 0.2 ± 0.05 ng/cm2; whereas after 12 h of skin exposure permeation was 0.97 ± 0.3 ng/cm2 and 0.47 ± 0.1 ng/cm2 respectively. After 24 h the permeation increased more than two fold to 2.57 ± 0.57 ng/cm2 and 1.11 ± 0.29 ng/cm2 respectively. The flux of platinum across the skin of 0.12 ± 0.02 ng/cm2/h was significantly higher (p = 0.015) than the flux of rhodium of 0.05 ± 0.01 ng/cm2/h. After 24 h of skin exposure to these metals, 1459.47 ± 99.42 ng/cm2 of platinum was retained inside the skin, which was significantly higher (p < 0.001) than that of rhodium (757.04 ± 70.43 ng/cm2). TEM investigation estimated the size of platinum and rhodium particles to be similar with an average diameter of 9.7 ± 0.7 nm (n = 8) and 9.6 ± 0.9 nm (n = 6) respectively, considering the difficulty in identifying one particle as multiple particles agglomerated together. The diameter of the agglomerates were calculated with an average diameter of 42.9 ± 7.6 nm for platinum (n = 12) and 52.5 ± 8.3 nm (n = 9) for rhodium. However when observing the agglomerates the platinum agglomerates with a diameter of
A. Franken et al. / Toxicology in Vitro 28 (2014) 1396–1401
30–40 nm were more abundant, whereas the rhodium agglomerates with a diameter of >50 nm were more abundant.
4. Discussion In this study the skin permeation of a platinum and rhodium salts using intact full thickness human skin was investigated. The results indicated that both platinum and rhodium were able to permeate through intact Caucasian skin and a cumulative increase in permeation over time. Results indicated that the cumulative mass of platinum that permeated through the skin after eight, 12 and 24 h was consistently higher when compared to rhodium permeation. In occupational settings usual work shifts last for eight hours, however some shift work requires workers to work for 12 h. Results indicate that the extended working hours from eight to 12 h will increase permeation by 83% for platinum and 135% for rhodium with continued skin contamination after the additional four hours of exposure. In the event of poor personal hygiene, where skin contamination is not carefully removed by the workers, the permeation of these metals may continue even after working hours, which is considered as the ‘take-home’ effect. Results indicated an increase in skin permeation of 385% and 455% for platinum and rhodium exposure respectively between eight and 24 h of exposure (See Fig. 1). If the skin is not decontaminated (washed) these metals remain on the skin, and will be able to permeate until it is properly removed. As a worst-case scenario workers may return to work with already contaminated skin and further skin contamination may occur, increasing dermal exposure, and skin permeation of the metal salts. These results emphasise the importance of good personal hygiene after the work shift, as well as providing appropriate facilities for decontamination in the workplace. The percentage of metal that permeated through the skin after 24 h of exposure was calculated taking the applied concentration into account. These results indicated low overall permeation with platinum permeation of 2.3 104% and rhodium permeation of 1 104% when compared to the initial concentration of 0.3 mg/ ml applied to the skin. Although overall permeation was low, it is important to note that permeation through the skin occurred, and that these metals were also retained inside the skin. The mass of metals retained in the skin was calculated using the full thickness skin, without distinguishing between the epidermis and stratum corneum. After 24 h the mass of platinum retained inside the skin was almost double the mass of rhodium, expressed in percentage of 2.24% and 1.16% for platinum and rhodium respectively. There is a 10,000 fold increased percentage retained inside the skin than the percentage that permeated through the skin (See Table 1). These metals are retained inside the skin, and may continue to act as haptens and form antigens in the skin, contributing to dermal effects even after skin contamination was removed. Skin was obtained from four Caucasian donors, and the authors acknowledge the possibility of variability between donors as a limitation of the study and results should be interpreted accordingly. Flux describes the rate of diffusion of the metal through the skin over time (McDougal and Boeniger, 2002). Flux provides information that is useful for risk assessment purposes since it gives an indication of the mass that will permeate through the skin per hour of exposure. Flux can therefore be compared for different substances. With the in vitro method, an infinite dose was applied to the skin to simulate the highest flux possible if the skin was loaded with high concentrations of the substance, therefore representing the worst-case permeation when dermal exposure is in excess (McDougal and Boeniger, 2002). From these results it is evident that platinum flux across the skin is more than double the flux of rhodium, indicating faster permeation of platinum through the skin.
1399
The molecular size of substances may influence permeation through the skin as absorption is inversely related to molecular weight, however this influence is negligible for molecules with a molecular weight of less than 500 g/mol (Dalton) (Barry, 2007). Absorption of substances through intact human skin declines rapidly for molecules with a molecular weight over 500 g/mol (Dalton) (Bos and Meinardi, 2000). Taking the molecular weight of potassium tetrachloroplatinate (415.09 g/mol) and rhodium (III) chloride (209.26 g/mol) into account, the size of the entire molecule is not likely to be a contributing factor to the rate of permeation. However during TEM analysis smaller agglomerates of platinum were more frequently observed when compared to the agglomerates of rhodium, suggesting that the smaller agglomerates of platinum contributed to the faster permeation reported for Pt (See Fig. 2). The follicular route may provide a significant permeation pathway for these metals, as suggested by Knorr et al. (2009). In contrast to molecular size, the charge of a molecule will influence permeation through the skin. Skin permeation of charged molecules is normally very poor because of their high polarity (Hadgraft et al., 1989). With substances that are ionisable, the quantity of charged and uncharged species is dependent on pH, and the transport of these charged species through the skin is less rapid than uncharged species (Smith, 1990). If these metal salts dissociate completely, K2PtCl4 would lead to Pt2+, whereas RhCl3 would result in Rh3+. The difference in charge could be a contributing factor to the faster permeation of platinum through the skin. However dissociation of these metals in synthetic sweat at a pH of 6.5 is unknown and during TEM analysis agglomerates with suspected chloride attachments as well as agglomerates without the salts were observed. Further research is suggested to gain insight into the dissociation of these metal salts in sweat. The dermal route was previously suggested as an alternative route leading to sensitisation in refinery workers where skin contamination occurred, however both eight hour and short term respiratory exposure levels measured were below the OEL (Maynard et al., 1997). The in vitro skin permeation results confirmed that these metal salts were able to permeate through the skin and thereby could contribute to sensitisation. Literature on platinum and rhodium sensitisation in the workplace list symptoms associated with the skin, such as dermatitis, eczema, urticaria and contact dermatitis; however these studies report these symptoms as a consequence of respiratory sensitisation (Bedello et al., 1987; De La Cuadra and Grau-Massanées, 1991; Calverley et al., 1995; Merget et al., 2000; Cristaudo et al., 2005). The probability of dermal exposure contributing to these symptoms was not observed. Considering results revealing that platinum and rhodium are able to permeate through the skin, and notable percentages thereof were retained inside the skin, the probability exists that
Fig. 1. Cumulative mass of platinum (n = 15, Human skin from 2 donors) and rhodium (n = 11, Human skin from 2 donors) permeated per area (mean and SEM).
1400
A. Franken et al. / Toxicology in Vitro 28 (2014) 1396–1401
Fig. 2. Platinum agglomerate (A) bar 10 nm; and rhodium agglomerate (B) bar 20 nm visualised by TEM in the donor solution. Estimated individual particles as inserts.
Table 1 Platinum and rhodium concentrations in the receptor solution, retained inside the skin and permeation (mean ± SEM). Receptor solution
Pt Rh p
Skin
Permeation
Cumulative mass permeated in 24 h (ng/cm2)
Percentage diffused after 24 h of exposure (%)
Mass retained inside the skin (ng/cm2)
Percentage in the skin after 24 h (%)
Flux (ng/ cm2)
Lag time (h)
2.57 ± 0.57 1.11 ± 0.29 0.016⁄
2.3 104 ± 0.5 104 1.0 104 ± 0.3 104 0.016
1459.47 ± 99.42 757.04 ± 70.43
Contents lists available at ScienceDirect
Toxicology in Vitro journal homepage: www.elsevier.com/locate/toxinvit
In vitro permeation of platinum and rhodium through Caucasian skin A. Franken a,⇑, F.C. Eloff a, J. Du Plessis b, C.J. Badenhorst a, A. Jordaan c, J.L. Du Plessis a a
North-West University, Potchefstroom Campus, Private Bag X6001, Potchefstroom 2520, South Africa Centre of Excellence for Pharmaceutical Sciences, North-West University, Private Bag X6001, Potchefstroom 2520, South Africa c Laboratory for Electron Microscopy CRB, North-West University, Private Bag X6001, Potchefstroom 2520, South Africa b
a r t i c l e
i n f o
Article history: Received 1 April 2014 Accepted 19 July 2014 Available online 30 July 2014 Keywords: Platinum group metals (PGMs) Skin sensitisation Franz diffusion cells Caucasian skin Skin permeation
a b s t r a c t During platinum group metals (PGMs) refining the possibility exists for dermal exposure to PGM salts. The dermal route has been questioned as an alternative route of exposure that could contribute to employee sensitisation, even though literature has been focused on respiratory exposure. This study aimed to investigate the in vitro permeation of platinum and rhodium through intact Caucasian skin. A donor solution of 0.3 mg/ml of metal, K2PtCl4 and RhCl3 respectively, was applied to the vertical Franz diffusion cells with full thickness abdominal skin. The receptor solution was removed at various intervals during the 24 h experiment, and analysed with high resolution ICP-MS. Skin was digested and analysed by ICP-OES. Results indicated cumulative permeation with prolonged exposure, with a significantly higher mass of platinum permeating after 24 h when compared to rhodium. The mass of platinum retained inside the skin and the flux of platinum across the skin was significantly higher than that of rhodium. Permeated and skin retained platinum and rhodium may therefore contribute to sensitisation and indicates a health risk associated with dermal exposure in the workplace. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction In South Africa the platinum group metals (PGMs) mining and refining industry is one of the largest components in the mining sector (Chamber of Mines, 2012). The mining of PGMs include the precious metals platinum (Pt), palladium (Pd), rhodium (Rh), ruthenium (Ru), iridium (Ir) and osmium (Os) (Ravindra et al., 2004). These precious metals have excellent catalytic properties and are resistant to corrosion, resulting in these metals being valuable in various industries, increasing the demand for these metals (Iavicoli et al., 2012). The mining of PGMs and the subsequent smelting, refining or recycling processes results in workers being potentially exposed to these metals, in various forms, on a daily basis. Refinery workers are often simultaneously exposed to the salts of platinum, rhodium and palladium (Kielhorn et al., 2002). Secondary industries such as catalyst manufacturers, electronic industries and jewellery fabrication utilise these precious metals leading to potential occupational exposure in industries other than refineries (Kielhorn et al., 2002; Cristaudo et al., 2005). One of the hazards during metal refining and handling of metals is worker ⇑ Corresponding author. Tel.: +27 (0) 182992437; fax: +27 (0) 182991053. E-mail addresses: [email protected] (A. Franken), [email protected] (F.C. Eloff), [email protected] (J. Du Plessis), cas.badenhorst@ angloamerican.com (C.J. Badenhorst), [email protected] (A. Jordaan), [email protected] (J.L. Du Plessis). http://dx.doi.org/10.1016/j.tiv.2014.07.007 0887-2333/Ó 2014 Elsevier Ltd. All rights reserved.
dermal exposure, as verified by Du Plessis et al. (2010, 2013) indicating dermal exposure to nickel and/or cobalt in base metal refineries. No published information is available on the extent of worker dermal exposure to PGM salts, and no in vitro information is available on the possibility of skin permeation. The potent allergenicity of halide platinum salts is well documented, however the allergenicity of rhodium salts is less known (Cleare et al., 1976; Calverley et al., 1995). Platinum salts are well known respiratory sensitisers, although the skin sensitisation potential thereof is less documented. The National Institute for Occupational Safety and Health (NIOSH) pocket guide to chemical hazards identifies the skin as an exposure route and a target organ to soluble platinum as salts, with symptoms including dermatitis and sensitisation of the skin (NIOSH, 2013). Platinum salt sensitivity occurs when haptens combine with serum proteins to form antigens, and is known as a type 1, immunoglobulin IgE-mediated response. Symptoms of platinum salt sensitisation include respiratory symptoms such as tightness in the chest and wheezing, and also dermal symptoms such as eczema and urticaria (Kiilunen and Aitio, 2007). Soluble rhodium salts are considered to be sensitisers, with studies reporting allergic reactions including respiratory symptoms, urticaria and contact dermatitis in occupationally exposed subjects (Santucci et al., 2000; Goossens et al., 2011). NIOSH (2013) acknowledges the skin as a possible exposure route for soluble rhodium; however no possible skin symptoms are
A. Franken et al. / Toxicology in Vitro 28 (2014) 1396–1401
listed. In the past rhodium was believed to be allergologically safe, and was used to plate other metals to avoid dermatitis in persons sensitised to nickel and cobalt. However two case studies reported dermatitis of workers using rhodium solutions in the jewellery industry (Bedello et al., 1987; De La Cuadra and Grau-Massanées, 1991). De La Cuadra and Grau-Massanées (1991) concluded that rhodium as a salt may be a potential sensitiser, but not as a metal. Studies have concluded that the compounds that are capable of eliciting reactions are a small group of charged compounds containing a reactive ligand system, such as a chloride or bromide. Therefore it is the charge and overall reactivity of a compound that is directly related to its allergenicity (Cleare et al., 1976). During the refining process as well as in secondary industries these metals are found in numerous chemical forms including the most commonly found halide complexes (salts) (Bolm-Audorff et al., 1992; Linnett and Hughes, 1999; Cristaudo et al., 2005). Previous studies acknowledge platinum as a sensitiser, manifesting in respiratory and dermal symptoms although these studies focused on respiratory exposure (Calverley et al., 1995; Linnett and Hughes, 1999; Merget et al., 2000). The literature suggests sensitisation to occur primarily through respiratory exposure, not taking the dermal route of exposure into consideration. Maynard et al. (1997) however suggested that dermal exposure to these metals could be an alternative route contributing to sensitisation. Their results indicated that respiratory exposure to platinum were below the eight hour and short term occupational exposure limits (OEL), yet exposure resulted in sensitisation to soluble platinum in refinery workers. Numerous in vitro studies have proven that metals are able to permeate through intact human skin, and permeation increases in damaged skin (Larese Filon et al., 2007, 2009). These studies included nickel, cobalt and chromium; however no in vitro studies are available for PGMs. This study aimed to establish if a platinumand rhodium salt are able to permeate through intact Caucasian skin by making use of an in vitro diffusion system.
1397
2.2. Preparation of skin membranes Human abdominal full thickness skin was obtained after surgical abdominoplasty procedures with the approval of the NorthWest University Ethics Committee (NWU-00114-11-A5). Informed consent was given by the donors. The skin was stored in a freezer at 18 °C for a maximum period of four months. A study done by Harrison et al. (1984) reported that freezing the skin did not affect the absorption of water by comparing frozen and fresh skin segments. On the day of the experiment, the skin was allowed to thaw, and all subcutaneous fat was removed. The skin was cut into circles with a diameter of 2.4 cm and clamped between the two parts of the Franz cell. Skin integrity was tested before and after each experiment using a Tinsley LCR Data bridge (Model 6401). The measurement was taken at 1 kHz in the parallel equivalent circuit using two stainless steel electrodes (Fasano and Hinderliter, 2004). Cells with an electrical resistance below 10 kX were rejected, and cells with similar resistance values within one skin donor were chosen for diffusion studies. Skin was obtained from Caucasian women between the ages of 37 and 52. In total skin from four different donors were used, skin from two donors were used to establish platinum permeation, and skin from two donors were used to establish rhodium permeation. 2.3. In vitro diffusion system Vertical Franz diffusion cells with a receptor compartment of 2 ml were used to perform the percutaneous permeation studies (Franz, 1975). The receptor compartment was maintained at a temperature of 37 °C to simulate the normal temperature below the skin. The physiological solution used as receptor solution was continuously stirred using a magnetic stirrer bar. Each piece of skin was clamped between the two compartments of the Franz cell with a stainless steel clamp. The mean exposed skin area was 1.066 cm2, and the thickness of the skin was measured as less than 1 mm. 2.4. Platinum permeation
2. Materials and methods 2.1. Chemicals All chemicals were analytical grade. Ammonia (32%), hydrochloric acid (37%) and potassium di-hydrogen phosphate were purchased from Merck, South Africa; sodium chloride, urea, lactic acid (88–92%) and disodium hydrogen phosphate were purchased from Sigma Aldrich, South Africa. Nitric acid (65%) and hydrochloric acid (33%) were purchased from De Bruyn Spectroscopic Solutions, South Africa. Hydrogen peroxide (50%) and acetone were purchased from Associated Chemical Enterprises, South Africa. Potassium tetrachloroplatinate and rhodium trichloride were obtained from Johnson Matthey and sponsored by Anglo American Technical Solutions, South Africa. Type 1 water used for the solutions was produced with a Millipore purification pack system (milliQ water). A 0.9% sodium chloride solution was used to test the integrity of the skin. The physiological solution (receptor solution) was prepared by dissolving 4.76 g Na2HPO4, 0.38 g KH2PO4 and 18 g NaCl into 2 L of milliQ water, the final pH was adjusted to 7.35. The synthetic sweat solution used as the donor solution was prepared with 0.5% NaCl, 0.1% lactic acid and 0.1% urea in milliQ water. For the platinum donor solution 0.03233 g potassium tetrachloroplatinate (K2PtCl4) was dissolved in 50 ml synthetic sweat where after the pH was adjusted to 6.5 with ammonia. For the rhodium donor solution 0.04306 g of rhodium (III) chloride (RhCl3) was dissolved in 50 ml synthetic sweat and the pH was adjusted to 6.5. A final concentration of 0.3 mg/ml of each metal was used as the donor solution.
The receptor compartment was filled with 2 ml of physiological (receptor) solution, and the donor compartment was filled with 1 ml of the K2PtCl4 dissolved in synthetic sweat. Ten cells were used as experimental cells, and 10 cells as blanks. The receptor solution was removed at 6, 8, 10, 12, 14 and 24 h and placed into clearly marked vials. During each removal the receptor compartment was rinsed with an additional 2 ml of physiological solution, and this rinsing solution was added to the original sample removed. The receptor compartment was rinsed to ensure that all the platinum in this compartment was removed at the specific time interval. After completion of the experiment the donor solution was removed and placed into a vial, and the compartment was rinsed four times with 1 ml of synthetic sweat each time to remove all remaining platinum from the skin. The donor rinsing solution was added to the original donor solution for analysis. After testing the skin integrity the cell was dismantled and the skin was placed into a vial for analysis. 2.5. Rhodium permeation The above mentioned experiment was repeated by dissolving RhCl3 in synthetic sweat as donor solution. 2.6. Blanks The blank cells were treated the same as the other cells with the exception that no platinum or rhodium was added to the synthetic sweat placed in the donor compartment.
1398
A. Franken et al. / Toxicology in Vitro 28 (2014) 1396–1401
2.7. Skin content evaluation The skin pieces were removed from the vial and weighed for skin digestion purposes. The vials were rinsed with acetone to remove any residue from the vial. The rinsed solution was placed into a beaker with the skin where the acetone was evaporated using a hot plate. Nitric acid and hydrogen peroxide were added to the skin mixture in separate stages and evaporated to destroy the organic material. Hydrochloric acid was added to the beaker to ensure all the platinum was in a soluble state. The final solution was diluted to 10 ml with hydrochloric acid, which was analysed by Inductively Coupled Plasma-Optical Emission Spectrometry (ICP–OES) (Spectro Arcos) to establish the concentration of the metals remaining in the skin. 2.8. Standardisation experiments Various standardisation experiments were conducted to establish the appropriate concentration for the donor solution; these experiments included receptor solution removals at one, two and four hours. Analyses of these results predominantly indicated values close to or below the detection limit of analysis. These experiments also indicated that exposure up to six hours did not contribute to the steady state flux through the skin, and therefore removals at these time intervals were excluded from subsequent experiments. 2.9. Analytical measurements The receptor solutions were analysed with high resolution Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) (Thermo Scientific Element XR) to establish the concentration of platinum and rhodium in the solutions. An internal standard of 10 lg/L rhenium was added to all the solutions to compensate for volume inaccuracies. For platinum the analysis was done using the Pt195 isotope in low resolution mode, whereas the Rh103 isotope was used in high resolution mode for rhodium. The instrument was calibrated with Pt or Rh calibration standards that were matrix matched with a range of 0, 5, 10, 20, 50, and 100 ng/L solutions containing platinum, rhodium and the internal standard rhenium to provide the initial calibration curves. If necessary, the calibration was extended to 1, 2, 5, 10, 20 or even 50 lg/L if the sample concentrations required it. The lower limit of detection for platinum was 30 ng/L and 20 ng/L for rhodium. The concentration of platinum and rhodium in the donor solutions and in the solutions resulting from the skin digestion were analysed by ICP-OES (Spectro Arcos). An internal standard of 4 mg/L yttrium was added to the solutions to compensate for volume inaccuracies. The wavelengths used for determination was 265.945 nm and 214.423 nm for platinum, and 343.489 nm for rhodium. The instrument was calibrated with a standard solution of 0, 1, 2, 3, 4, 5, 10 and 20 mg/L to provide the calibration curves. 2.10. Transmission electron microscopy (TEM) Solutions of K2PtCl4 and RhCl3 dissolved in synthetic sweat (donor solutions) were characterised with a FEI Tecnai G2 high resolution transmission electron microscope operated at 200 kV. Micrographs were captured and particles measured using DigitalMicrographTM (Gatan Inc.) computer software. 2.11. Data analysis The concentration of platinum and rhodium (ng/cm3) in the receptor solution was converted to the total mass that permeated through the exposed skin area (ng/cm2) and then plotted against
time as the cumulative mass of metal that permeated per area of skin. Flux permeation was calculated from the steady-state region of the graph after curve-fitting. Eq. (1) was developed by Díez-Sales et al. (1991) to describe the mass of substance crossing a membrane at a given time:
" !# 1 t 1 2 X ð1Þn Dn2 p2 t QðtÞ ¼ AKhC m D 2 exp 2 h 6 p n¼1 n2 h
ð1Þ
where Q(t) is the mass that passed through the membrane at a given time t, A is the actual surface diffusion area, K is the partition coefficient between the skin and donor solution, h is the membrane thickness, Cv is the concentration in the donor solution and D is the diffusion coefficient of the permeant in the skin. As t approaches infinity the exponential term becomes negligible and the equation can be simplified to Eq. (2):
t 1 QðtÞ ¼ AKhC m D h 6
ð2Þ
Since K, D and h are unknown, the products Kh and D/h2 were replaced with a and b which were calculated by fitting Eq. (2) to the data by using Easyplot Version 4.0.5 (Spiral Software, Aerious Ltd.). The permeability coefficient (kp) and flux (J) were calculated using Eqs. (3) and (4) respectively:
kp ¼
KD ð¼ abÞ h
J ¼ kp C m
ð3Þ ð4Þ
Lag time was calculated as the intercept of the steady-state region with the x-axis. Average blank concentrations were subtracted from experimental data. Data analysis was performed using Statistica Version 11 (Statsoft Inc.). Data was not normally distributed, and therefore Box– Cox transformed (Box and Cox, 1964). Data is reported as mean ± standard error of means (SEM). The difference between independent Box–Cox transformed data was assessed by means of the t-test with separate variance. A p value of 60.05 was considered as statistically significant. 3. Results Platinum and rhodium were able to permeate through intact Caucasian skin, and permeation increased cumulatively over time. After 8 h of skin exposure, permeation of platinum into the receptor solution was 0.53 ± 0.18 ng/cm2 and rhodium permeation was 0.2 ± 0.05 ng/cm2; whereas after 12 h of skin exposure permeation was 0.97 ± 0.3 ng/cm2 and 0.47 ± 0.1 ng/cm2 respectively. After 24 h the permeation increased more than two fold to 2.57 ± 0.57 ng/cm2 and 1.11 ± 0.29 ng/cm2 respectively. The flux of platinum across the skin of 0.12 ± 0.02 ng/cm2/h was significantly higher (p = 0.015) than the flux of rhodium of 0.05 ± 0.01 ng/cm2/h. After 24 h of skin exposure to these metals, 1459.47 ± 99.42 ng/cm2 of platinum was retained inside the skin, which was significantly higher (p < 0.001) than that of rhodium (757.04 ± 70.43 ng/cm2). TEM investigation estimated the size of platinum and rhodium particles to be similar with an average diameter of 9.7 ± 0.7 nm (n = 8) and 9.6 ± 0.9 nm (n = 6) respectively, considering the difficulty in identifying one particle as multiple particles agglomerated together. The diameter of the agglomerates were calculated with an average diameter of 42.9 ± 7.6 nm for platinum (n = 12) and 52.5 ± 8.3 nm (n = 9) for rhodium. However when observing the agglomerates the platinum agglomerates with a diameter of
A. Franken et al. / Toxicology in Vitro 28 (2014) 1396–1401
30–40 nm were more abundant, whereas the rhodium agglomerates with a diameter of >50 nm were more abundant.
4. Discussion In this study the skin permeation of a platinum and rhodium salts using intact full thickness human skin was investigated. The results indicated that both platinum and rhodium were able to permeate through intact Caucasian skin and a cumulative increase in permeation over time. Results indicated that the cumulative mass of platinum that permeated through the skin after eight, 12 and 24 h was consistently higher when compared to rhodium permeation. In occupational settings usual work shifts last for eight hours, however some shift work requires workers to work for 12 h. Results indicate that the extended working hours from eight to 12 h will increase permeation by 83% for platinum and 135% for rhodium with continued skin contamination after the additional four hours of exposure. In the event of poor personal hygiene, where skin contamination is not carefully removed by the workers, the permeation of these metals may continue even after working hours, which is considered as the ‘take-home’ effect. Results indicated an increase in skin permeation of 385% and 455% for platinum and rhodium exposure respectively between eight and 24 h of exposure (See Fig. 1). If the skin is not decontaminated (washed) these metals remain on the skin, and will be able to permeate until it is properly removed. As a worst-case scenario workers may return to work with already contaminated skin and further skin contamination may occur, increasing dermal exposure, and skin permeation of the metal salts. These results emphasise the importance of good personal hygiene after the work shift, as well as providing appropriate facilities for decontamination in the workplace. The percentage of metal that permeated through the skin after 24 h of exposure was calculated taking the applied concentration into account. These results indicated low overall permeation with platinum permeation of 2.3 104% and rhodium permeation of 1 104% when compared to the initial concentration of 0.3 mg/ ml applied to the skin. Although overall permeation was low, it is important to note that permeation through the skin occurred, and that these metals were also retained inside the skin. The mass of metals retained in the skin was calculated using the full thickness skin, without distinguishing between the epidermis and stratum corneum. After 24 h the mass of platinum retained inside the skin was almost double the mass of rhodium, expressed in percentage of 2.24% and 1.16% for platinum and rhodium respectively. There is a 10,000 fold increased percentage retained inside the skin than the percentage that permeated through the skin (See Table 1). These metals are retained inside the skin, and may continue to act as haptens and form antigens in the skin, contributing to dermal effects even after skin contamination was removed. Skin was obtained from four Caucasian donors, and the authors acknowledge the possibility of variability between donors as a limitation of the study and results should be interpreted accordingly. Flux describes the rate of diffusion of the metal through the skin over time (McDougal and Boeniger, 2002). Flux provides information that is useful for risk assessment purposes since it gives an indication of the mass that will permeate through the skin per hour of exposure. Flux can therefore be compared for different substances. With the in vitro method, an infinite dose was applied to the skin to simulate the highest flux possible if the skin was loaded with high concentrations of the substance, therefore representing the worst-case permeation when dermal exposure is in excess (McDougal and Boeniger, 2002). From these results it is evident that platinum flux across the skin is more than double the flux of rhodium, indicating faster permeation of platinum through the skin.
1399
The molecular size of substances may influence permeation through the skin as absorption is inversely related to molecular weight, however this influence is negligible for molecules with a molecular weight of less than 500 g/mol (Dalton) (Barry, 2007). Absorption of substances through intact human skin declines rapidly for molecules with a molecular weight over 500 g/mol (Dalton) (Bos and Meinardi, 2000). Taking the molecular weight of potassium tetrachloroplatinate (415.09 g/mol) and rhodium (III) chloride (209.26 g/mol) into account, the size of the entire molecule is not likely to be a contributing factor to the rate of permeation. However during TEM analysis smaller agglomerates of platinum were more frequently observed when compared to the agglomerates of rhodium, suggesting that the smaller agglomerates of platinum contributed to the faster permeation reported for Pt (See Fig. 2). The follicular route may provide a significant permeation pathway for these metals, as suggested by Knorr et al. (2009). In contrast to molecular size, the charge of a molecule will influence permeation through the skin. Skin permeation of charged molecules is normally very poor because of their high polarity (Hadgraft et al., 1989). With substances that are ionisable, the quantity of charged and uncharged species is dependent on pH, and the transport of these charged species through the skin is less rapid than uncharged species (Smith, 1990). If these metal salts dissociate completely, K2PtCl4 would lead to Pt2+, whereas RhCl3 would result in Rh3+. The difference in charge could be a contributing factor to the faster permeation of platinum through the skin. However dissociation of these metals in synthetic sweat at a pH of 6.5 is unknown and during TEM analysis agglomerates with suspected chloride attachments as well as agglomerates without the salts were observed. Further research is suggested to gain insight into the dissociation of these metal salts in sweat. The dermal route was previously suggested as an alternative route leading to sensitisation in refinery workers where skin contamination occurred, however both eight hour and short term respiratory exposure levels measured were below the OEL (Maynard et al., 1997). The in vitro skin permeation results confirmed that these metal salts were able to permeate through the skin and thereby could contribute to sensitisation. Literature on platinum and rhodium sensitisation in the workplace list symptoms associated with the skin, such as dermatitis, eczema, urticaria and contact dermatitis; however these studies report these symptoms as a consequence of respiratory sensitisation (Bedello et al., 1987; De La Cuadra and Grau-Massanées, 1991; Calverley et al., 1995; Merget et al., 2000; Cristaudo et al., 2005). The probability of dermal exposure contributing to these symptoms was not observed. Considering results revealing that platinum and rhodium are able to permeate through the skin, and notable percentages thereof were retained inside the skin, the probability exists that
Fig. 1. Cumulative mass of platinum (n = 15, Human skin from 2 donors) and rhodium (n = 11, Human skin from 2 donors) permeated per area (mean and SEM).
1400
A. Franken et al. / Toxicology in Vitro 28 (2014) 1396–1401
Fig. 2. Platinum agglomerate (A) bar 10 nm; and rhodium agglomerate (B) bar 20 nm visualised by TEM in the donor solution. Estimated individual particles as inserts.
Table 1 Platinum and rhodium concentrations in the receptor solution, retained inside the skin and permeation (mean ± SEM). Receptor solution
Pt Rh p
Skin
Permeation
Cumulative mass permeated in 24 h (ng/cm2)
Percentage diffused after 24 h of exposure (%)
Mass retained inside the skin (ng/cm2)
Percentage in the skin after 24 h (%)
Flux (ng/ cm2)
Lag time (h)
2.57 ± 0.57 1.11 ± 0.29 0.016⁄
2.3 104 ± 0.5 104 1.0 104 ± 0.3 104 0.016
1459.47 ± 99.42 757.04 ± 70.43
