Copper dinitrate

Dermal absorption

Administrative data

Endpoint:

dermal absorption in vitro / ex vivo

Type of information:

experimental study

Adequacy of study:

weight of evidence

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: Already evaluated by the Competent Authorities for Biocides and Existing Substance Regulations.

Data source

Reference

Reference Type:

publication

Title:

Simultaneous Absorption of Copper and Zinc through Human Skin in vitro.

Author:

Pirot, F., Panisset, F., Agache, P. & Humbert, P.

Year:

1996

Bibliographic source:

Skin Pharmacol. 9: 43-52

Materials and methods

Principles of method if other than guideline:

This was a non-regulatory study carried out to investigate the simultaneous absorption of copper and zinc sulphates or chlorides through ex vivo
sliced human skin. No guidelines are available to address this objective. This summary addresses only those sections of the report that relate to copper salts.

GLP compliance:

no

Test material

Test material information

Constituent 1

Reference substance name:

Cu2+ as copper sulphate (CuSO4.5H2O)

IUPAC Name:

Cu2+ as copper sulphate (CuSO4.5H2O)

Constituent 2

Reference substance name:

Cu2+ as copper chloride (CuCl2.2H2O)

IUPAC Name:

Cu2+ as copper chloride (CuCl2.2H2O)

Constituent 3

Reference substance name:

Zn2+ as zinc sulphate (ZnSO4.7H2O)

IUPAC Name:

Zn2+ as zinc sulphate (ZnSO4.7H2O)

Constituent 4

Reference substance name:

Zn2+ as zinc chloride (ZnCl2)

IUPAC Name:

Zn2+ as zinc chloride (ZnCl2)

Details on test material:

Lot/batch number: Not available
Purity: Not stated
Stability: Not stated
Four preparations were tested:
(A) 5% (w/w) copper sulphate and 5% (w/w) zinc sulphate in white petrolatum.
(B) 5% (w/w) copper sulphate and 5% (w/w) zinc sulphate in carboxypolymethylene gel.
(C) 5% (w/w) copper chloride and 5% (w/w) zinc chloride in white petrolatum.
(D) 5% (w/w) copper chloride and 5% (w/w) zinc chloride in hydroxypropylmethylcellulose gel.
Formulae A and B contained 1.27% Cu and 1.14% Zn. Formulae C and D contained 1.86% Cu and 2.40% Zn.

Radiolabelling:

no

Test animals

Species:

human

Sex:

female

Details on test animals or test system and environmental conditions:

Skin samples used for the permeation studies was obtained from three women (37, 40 and 45 years old). The samples of skin were obtained from
either breast or abdomen during surgery.

Administration / exposure

Vehicle:

petrolatum

Duration of exposure:

At the 72nd hour, the white petrolatum ointment remaining on the skin surface was washed off with toluene. The hydrogels were washed off with
distilled water. The whole epidermis was removed from the dermis with forceps, dried for 24 hours at 105°C, weighed, and dissolved in nitric acid
(68%) for 69 hours. The mixtures were then passed through a 0.2 µm nitro-cellulose filter before determination of the copper and zinc.
To estimate the metal content of fresh tissue, the dry weight of the stratum corneum and viable epidermis and dermis were estimated at 90 and 28% of fresh weight, respectively; the density of both fresh tissues equal to unity and the volume per cm2 surface area of fresh epidermis and dermis 5 mm3 and 36 mm3, respectively.

Doses:

20 mg/cm2 of a formula

Details on study design:

Skin samples were sliced with a dermatome to an average thickness of 410 µm as assessed by high frequency (25 MHz) B-scan ultrasound imaging. All skin samples were randomised to avoid source effect. Samples were then mounted in Franz-type static diffusion cells with a 3.14 cm2 surface area, and 20 mg/cm2 of a formula was applied to the outer skin surface.

Details on in vitro test system (if applicable):

The amounts of Cu and Zn deposited within each formula are presented in Table 1. The receptor fluid consisted of isotonic saline containing 5% human albumin, 5000 IU/l penicillin G, 5 mg/l streptomycin, and 1.25 mg/l amphotericin B. During the experiment, it was continuously stirred with a magnetic rod and maintained at 33°C. It was totally withdrawn after 1.5, 3, 6, 12, 24, 48 and 72 hours and kept for analysis. The receptor chamber was immediately refilled with fresh receptor fluid to maintain a constant acceptor volume (9 ml). Five cells were used for each formula. After tenfold dilution, the collected receptor fluids were subjected to flame atomic absorption spectrometry for cu and Zn quantification at 324.7 and 213.9 nm, respectively, with an acetylene-air flame.

Results and discussion

Absorption in different matrices:

Copper in skin
Dry tissue: The ratio of copper concentrations present in the epidermis & dermis of dry, untreated control skin was 7:1 (Table 3). The copper content of both skin layers rose substantially during the course of the study, the increase was most noticable in the epidermis with formula B and in the dermis with formula D.
 
Fresh epidermis: The estimated basal concentration of copper within fresh epidermis was twice as great as that within fresh dermis (Table 4). By the end of the study, copper concentrations in the epidermis had increased strongly. Furthermore, the epidermis still contained much more copper than the dermis, with the exception of formula D.
The epidermal load of copper ranged from 2.5 to 6.3 times the cumulative amount found in receptor fluid over the 72 hours of the study. There was no clear-cut relationship with either fluxes or permeability constants.
 
Fresh dermis: Copper sulphate in hydrogel formula (B) resulted in very low copper storage within the dermis (0.2 µg cm-2). With all three remaining formulae, the amount of copper stored was equal to, lower than, or higher than the amounts collected in receptor fluid over 72 hours (A, C and D, respectively).

Total recovery:

In the fresh receptor fluid, copper concentration as a part of human serum albumin was 605 ± 5 µg/l (5.4 µg per receptor chamber volume). This amount was subtracted from samples to determine the absorbed amounts.
 
Copper permeation from CuSO4was hardly affected by the vehicle (formulae A and B,Figure 1). By contrast, permeation from CuCl2was greater with petrolatum than with hydrogel (formulae C and D). By the 72ndhour, the collected amounts of copper remained below 6% of the applied doses (Figure 2). During the experiment, copper fluxes were very low and variable (Figure 3). They all showed a peak between 1.5 and 6 hours followed by a steady state, with the exception of formula B, where a slow and permanent decrease was observed. The steady-state fluxes were calculated for the following periods: 12-72 hours (formula A); 3-24 hours (formula C); 6-24 hour(formula D) and tentatively 12-48 hours (formula B).
 
The steady-state fluxes were used to calculate the apparent permeability coefficients for copper, using the applied concentrations of copper (1.27% in sulphate formula and 1.86% in gel formulae), with 0.82 and 1 for densities of petrolatum and gel, respectively (Table 2).

Any other information on results incl. tables

pH:

A slight decrease in pH was observed following gel application (formulae B and D), possibly due to simultaneous percutaneous absorption of the ions, or of the undissociated salts. Absorption from petrolatum formulations (A and C) did not result in any change of receptor fluid pH.

Table 1. Deposited amounts of copper and zinc (mg/cm²) with each of the 4 formulae

 

Formula	Counter-ion	Vehicle	Copper	Zinc
A	S04	petrolatum	297±39	268±37
B	S04	hydrogel1	257±19	230±17
C	Cl2	petrolatum	385±54	497±70
D	Cl2	hydrogel 2	373±4	481±5

Table 2. Copper: K o/w, flux at steady state and K_p

 

Formula	Counter-ion	Vehicle	K o/w x 10^6	Flux mg cm^-2h^-1x 10^3	Kp cm h^-1x 10^6
A	S04	petrolatum	0.85±0.07	56±54^a	3.25±3.14^b
B	S04	hydrogel 1	0.85±0.07	64±80^a	4.54±5.67^b
C	Cl2	petrolatum	59.70±4.77	404±310	16.03±12.28
D	Cl2	hydrogel 2	59.70±4.77	47±20a	2.26±0.97^b

a Difference in flux versus formula C (p < 0.05).

b Difference in K_pversus formula C (p < 0.001).

Table 3. Copper levels (mg/g dry tissue) in human skin after a 72-hour topical application

 

Formula	Counter-ion	Vehicle	Epidermis	Dermis
A	S04	petrolatum	9.25±2.29^a,b	0.32±0.31^c
B	S04	hydrogel 1	4.10±1.17^a	0.02±0.01^c
C	Cl2	petrolatum	19.09±6.79	0.81±0.70^c
D	Cl2	hydrogel 2	8.20±1.24^a,b	3.73±3.79
Control			0.05±0.04^a	0.007±0.005^c

                                                    Mean±standard deviation of 5 determinations.

a  Difference with formulation C (p < 0.0001).

b  Difference with control (p < 0.0001).

c  Difference with formulation D (p < 0.01).

Table 4. Estimated amounts of copper found at 72h within fresh tissue, and of total absorbed amounts (mg/cm²).

 

(1) Formula	(2) Counter-ion	(3) Vehicle	(4) Fresh-epidermis	(5) Fresh dermis	(6) Receptor Fluid 72 h	(7) Total absorbed (5+6)
None			0.147	0.071
A	S04	petrolatum	27	3.2	4.3	7.5
B	S04	hydrogel 1	12	0.2	4.0	5.1
C	Cl2	petrolatum	56	8.2	22.1	30.3
D	Cl2	hydrogel 2	24	37.6	5.5	43.1

Discussion:

Over 72 hours, the cumulative amount of copper recovered in receptor fluid (saline containing 5% serum albumin) was very low (5.5 µg/cm²or less), except for copper in petrolatum (22 µg/cm²). Accordingly, the corresponding K_pvalues remained in the range of 10^-6cm/h. While the amount of copper applied in the experiment was in excess, less than 6% of the total was absorbed. The time course of transcutaneous copper fluxes always showed a peak between 1.5 and 6 hours, followed by immediate or delayed subsidence. This pattern suggests a decrease in either the partition coefficient or the diffusion coefficient (or in both), in relation to the sorption of the permeant within the stratum corneum.

 

No relationship was found between the K_pand either the K_owor the type of vehicle.

In the epidermis, copper storage was found to be equal to or greater than the 72-hour absorbed cumulative amount; storage in the dermis was generally lower than the 72-hour absorbed cumulative amount. This finding is in accordance with the well-known high storage capacity of the stratum corneum. Within the epidermis as a whole, storage was generally larger from petrolatum formulations, suggesting the influence of lower partition coefficients. Furthermore, chlorides were stored in generally larger amounts, possibly in relation to their higher K_ow.

 

Within the dermis, chloride was retained in substantial amounts, mostly when the vehicle was aqueous. This suggested a relation to the higher cumulative quantity found in the receptor fluid following use of an aqueous vehicle, as compared with sulphate formulations. The contrary was observed with sulphates, although they were poorly retained.

 

As a whole, the added amounts of the permeant found within the dermis and collected from the receptor fluid represented the amount actually absorbed. Table 4 shows that some figures increased greatly. For example, copper absorption is multiplied by 1.7 and 7.8 for formulae A and D, respectively.

Applicant's summary and conclusion

Conclusions:

Measured amounts of copper in receptor fluid were low, accounting for 1.45%, 1.56%, 5.74% and 1.47% of the applied dose for preparations A
(petrolatum), B (hydrogel 1), C (petrolatum) and D (hydrogel 2), respectively. Measured Kowvalues for copper were also very low, at 0.85 x 106± 0.07 as sulphate and 59.7 x 106± 4.77 as chloride.

Executive summary:

Materials and methods:

A study was carried out to investigate the simultaneous absorption of copper and zinc sulphates or chlorides through ex-vivo sliced human skin taken from three females aged 37 – 45 years old. The study was not designed to follow an internationally accepted guideline, and was not carried out in compliance with GLP.

The following four preparations were tested: (A) 5% copper sulphate and 5% zinc sulphate in white petrolatum; (B) 5% copper sulphate and 5% zinc sulphate in carboxypolymethylene gel; (C) 5% copper chloride and 5% zinc chloride in white petrolatum; (D) 5% copper chloride and 5% (w/w) zinc chloride in hydroxypropylmethylcellulose gel. Formulae A and B contained 1.27% Cu and 1.14% Zn. Formulae C and D contained 1.86% Cu and 2.40% Zn.

 

Skin samples were sliced with a dermatome to an average thickness of 410 µm and randomised. Samples were then mounted in Franz-type static diffusion cells with a 3.14 cm²surface area, and 20 mg/cm²of a preparation was applied to the outer skin surface. The receptor fluid, which consisted of isotonic saline containing 5% human albumin, was continuously stirred at 33°C. After 1.5, 3, 6, 12, 24, 48 and 72 hours the receptor fluid was withdrawn and kept for analysis. The receptor chamber was then refilled. Five cells were used for each preparation. After tenfold dilution, the receptor fluids were subjected to flame atomic absorption spectrometry (AAS) for Cu and Zn quantification at 324.7 and 213.9 nm.

After 72 hours, the white petrolatum ointment was washed off the skin surface with toluene. Hydrogels were washed off with distilled water. The whole epidermis was removed from the dermis, dried for 24 hours at 105°C, weighed, and dissolved in nitric acid. The mixtures were then passed through a 0.2 µm nitro-cellulose filter before determination of the copper and zinc. The dry weight of the stratum corneum and viable epidermis and dermis were estimated at 90 and 28% of fresh weight, respectively; the density equal to unity and the volume per cm²surface area of fresh epidermis and dermis 5 mm³and 36 mm³, respectively.

The pH of aqueous formulations and receptor fluids was measured. The pH of Formulae B (sulphates) and D (chlorides) were 1.6 ± 0.01 and 4.43 ± 0.01, respectively.

 

K_owvalues were determined by the modified shake flask method. 0.5 g of metal salt was mixed with 5 ml water and 5 ml 1-octanol for 12 hours at room temperature. The solutions were then centrifuged for 10 minutes at 4000 rpm. Metal concentrations in each phase weredetermined by AAS. The experiment was repeated 3 times for each salt.

 

Apparent permeability coefficients of copper and zinc were obtained with sulphate and chloride as counter-ions in the petrolatum and hydrogels, and the metal levels within skin were tested using the non-parametric Kruskal-Wallis test. K_owvalues were compared using the non-parametric Mann-Whitney test. Absorbed amounts were compared using analysis of variance. The comparison of multiple means used Fisher’s least significant difference procedure.

 

Results and Discussion:

Copper concentration in fresh receptor fluid was 605 ± 5 µg/l. This was subtracted from samples to determine the absorbed amounts.

 

Copper permeation from CuSO₄was hardly affected by the vehicle, whereas permeation from CuCl₂was greater with petrolatum than with hydrogel. Over 72 hours, the cumulative amount of copper recovered in receptor fluid was very low (4.3, 4.0 and 5.5 µg/cm²for preparations A, B and D, respectively), except for copper in petrolatum (preparation C), for which 22 µg/cm²was recovered. These amounts correspond to 1.45%, 1.56%, 5.74% and 1.47% of the applied doses for preparations A, B, C and D, respectively. Copper fluxes were very low and variable, all showing a peak between 1.5 and 6 hours followed by a steady state, with the exception of formula B, where a slow and permanent decrease was observed. This pattern suggests a decrease in either the partition coefficient or the diffusion coefficient (or in both), in relation to the sorption of the permeant within the stratum corneum (measured K_owvalues were very low at 0.85 x 10⁶± 0.07 for copper as sulphate and 59.7 x 10⁶± 4.77 for copper as chloride).

 

A slight decrease in pH was observed following gel application, possibly due to simultaneous percutaneous absorption of the ions, or of the undissociated salts. Absorption from petrolatum formulations did not result in any change of receptor fluid pH.

 

The ratio of copper concentrations present in the epidermis & dermis of dry, untreated control skin was 7:1. The copper content of both skin layers rose substantially during the course of the study. The increase was most noticable in the epidermis with formula B and in the dermis with formula D. The basal concentration of copper within fresh epidermis was twice as great as that within fresh dermis. By the end of the study, copper concentrations in the epidermis had increased strongly. Furthermore, the epidermis still contained much more copper than the dermis, with the exception of formula D. The epidermal load of copper ranged from 2.5 to 6.3 times the cumulative amount found in receptor fluid over the 72 hours of the study. There was no clear-cut relationship with either fluxes or permeability constants. These findings accord with the high storage capacity of the stratum corneum. Within the epidermis as a whole, storage was generally larger from petrolatum formulations, suggesting the influence of lower partition coefficients. Furthermore, chlorides were stored in generally larger amounts, possibly in relation to their higher K_ow.

 

Within fresh dermis, copper sulphate in hydrogel formula resulted in very low copper storage (0.2 µg cm^-2). With all three remaining formulae, the amount of copper stored was equal to, lower than, or higher than the amounts collected in receptor fluid over 72 hours (A, C and D, respectively).

 

Conclusion:

Measured amounts of copper in receptor fluid were low, accounting for 1.45%, 1.56%, 5.74% and 1.47% of the applied dose for preparations A (petrolatum), B (hydrogel 1), C (petrolatum) and D (hydrogel 2), respectively. Measured K_owvalues for copper were also very low, at 0.85 x 10⁶± 0.07 as sulphate and 59.7 x 10⁶± 4.77 as chloride.