Main

Menkes disease (MD) is an X-linked recessive disorder of copper transport caused by mutations in a gene encoding copper-transporting P-type ATPase (ATP7A). ATP7A is expressed in most cell types, although not in hepatocytes (1), and is located primarily in the trans-Golgi network, where it transports copper in from the cytosol (2). Under conditions in which copper levels are elevated, ATP7A relocalizes to the plasma membrane to expel copper from the cell. Therefore, ATP7A dysfunction causes severe problems in multiple aspects of copper trafficking, such as intestinal absorption, renal excretion, and passage through the blood–brain barrier (3).

Attempts at therapy for MD have focused on copper replacement. Subcutaneous injection of copper-histidine ameliorates the natural history of MD when started in early infancy (4,5). However, penetration of copper into the brain remains insufficient due to impaired copper transport after the blood–brain barrier becomes mature, whereas copper abnormally accumulates in the kidneys and the intestine (6). To overcome the insufficient copper distribution to the brain, diethyldithiocarbamate, a lipid-soluble chelator, has been proposed in combination with copper replacement therapy. This compound significantly increases copper levels in the brain, although the copper accumulation in the kidney remains high (6,7).

Copper-pyruvaldehyde bis(N4-methylthiosemicarbazone) (Cu-PTSM; Figure 1 ) labeled with copper radioisotopes is a radiopharmaceutical used in positron emission tomography imaging of the brain and heart (8,9). Cu-PTSM is an organic complex that is neutral and highly lipophilic, so it readily passes through the cell membrane. When Cu-PTSM reaches the cell, it is irreversibly reduced and cleaved to liberate copper, which is incorporated into intracellular copper pools (10,11). These unique characteristics of Cu-PTSM suggested that nonradiolabeled Cu-PTSM could be used as a carrier for copper.

Figure 1
figure 1

Chemical structures of copper-pyruvaldehyde bis(N4-methylthiosemicarbazone) complex (Cu-PTSM). Cu-PTSM is a stable, neutral lipophilic compound (left) and therefore diffuses through the cell membrane. When Cu-PTSM is reduced within cells, it becomes labile and releases copper (right).

Full size image

A wide variety of animal models for MD have been developed (1,12). Among them, macular mice are one of the closest models to the classical form of MD; hemizygous males exhibit MD-related symptoms and require copper injections early in life to survive (13,14,15). In this study, we investigated the copper-trafficking efficacy of Cu-PTSM to the critical organs in macular mice.

Results

Acute Toxicity Studies in Normal C3H Mice

Saline, dimethyl sulfoxide (DMSO), cupric chloride (CuCl2), or Cu-PTSM was injected subcutaneously into normal male C3H mice (15–17 d old, n = 14–16 in each group) at the same doses as used in macular mouse treatment in this study. In blood sampled at 24 h after injection, there was no significant difference in laboratory values among the groups at the doses used ( Table 1 ).

Table 1 Acute effects of CuCl2, Cu-PTSM, and solvents on hematological and biochemistry data in normal C3H mice
Full size table

Body and Organ Weights of CuCl2- and Cu-PTSM-Treated Macular Mice

The bodies and organs of CuCl2- and Cu-PTSM-treated macular mice were weighed and compared with normal male littermate mice at 4 wk of age (n = 8 in each group). In the control group, body weight reached 20.6 ± 2.0 g vs. 10.4 ± 2.5 and 13.3 ± 3.3 g in the CuCl2- and Cu-PTSM-treated groups, respectively. The body weights in the CuCl2-treated and Cu-PTSM-treated groups were significantly lower than the controls, whereas there was no significant difference between the CuCl2- and Cu-PTSM-treated groups. Figure 2 shows the organ weights of the control, CuCl2-, and Cu-PTSM-treated mice at 4 wk of age. The cerebrum weights of the CuCl2- and Cu-PTSM-treated groups were significantly lower than those of the controls, whereas the cerebellum weights did not differ among the groups. The kidney weights of the CuCl2- and Cu-PTSM-treated groups were slightly lower than those of the controls.

Figure 2
figure 2

Organ weights of control male mice and macular mice treated with cupric chloride (CuCl2) or copper-pyruvaldehyde bis(N4-methylthiosemicarbazone) (Cu-PTSM). Weights of the control (black bar), CuCl2-treated (white bar), and Cu-PTSM-treated mice (gray bar) are shown as means ± SD (in mg, n = 8). *P < 0.05, CuCl2-treated or Cu-PTSM-treated groups vs. control group. No significant difference was detected between CuCl2-treated and Cu-PTSM-treated groups in any organ.

Full size image

Organ Copper Concentrations in CuCl2- and Cu-PTSM-Treated Macular Mice

Figure 3 summarizes the copper concentrations in the major organs (n = 8). Copper levels in the cerebrum and cerebellum were significantly lower in both CuCl2- and Cu-PTSM-treated groups than in controls. No significant difference in copper levels was found between the cerebrum or cerebellum of CuCl2-treated and those of Cu-PTSM-treated groups. Conversely, the copper levels in the kidneys of CuCl2- and Cu-PTSM-treated groups were significantly higher than in those of the controls, but the copper level in the Cu-PTSM-treated group was nearly half that of the CuCl2-treated group (P < 0.001).

Figure 3
figure 3

Copper concentrations in organs of control male mice and macular mice treated with cupric chloride (CuCl2) or copper- pyruvaldehyde bis(N4-methylthiosemicarbazone) (Cu-PTSM). Measured values in the control (black bar), CuCl2-treated (white bar), and Cu-PTSM-treated mice (gray bar) are shown as means ± SD (in ng/mg wet weight, n = 8). *P < 0.05, CuCl2-treated or Cu-PTSM-treated group vs. control group;§P < 0.05, CuCl2-treated vs. Cu-PTSM-treated group.

Full size image

Effect of Cu-PTSM on Cytochrome Oxidase Activity in the Cerebral and Cerebellar Cortex

Figure 4a,b shows the histological appearance of the cerebral cortex and the cerebellar hemispheric cortex stained for cytochrome oxidase (CO) histochemistry. In the control mice, layer IV of the cerebral cortex showed the densest staining among the layers. The macular mice treated with CuCl2- or Cu-PTSM showed less intense staining than the controls. However, the tissue staining in the Cu-PTSM-treated group was denser than that in the CuCl2-treated group. Similarly, the molecular layer of the cerebellar cortex in the Cu-PTSM-treated group showed stronger staining than that of the CuCl2-treated group. No obvious difference in the tissue staining was observed among the granular layers of the three experimental groups.

Figure 4
figure 4

Results of cytochrome oxidase (CO) histochemistry in the cerebral (a) and cerebellar sections (b). In the cerebral cortex, layer IV of the somatosensory region was measured for CO activity (c). In the cerebellar hemispheric cortex, the molecular (d) and granular layers (e) were measured separately. Values are means ± SD (n = 8) in μmol/min/g of tissue. *P < 0.05, cupric chloride (CuCl2)-treated or copper-pyruvaldehyde bis(N4-methylthiosemicarbazone) (Cu-PTSM)-treated group vs. control;§P < 0.05, CuCl2-treated vs. Cu-PTSM-treated group. Cont., control.

Full size image

Figure 4c–e summarizes the results of CO activity measurements (n = 8). Optical density was measured in layer IV of the cortex. Although the levels of CO activity in both the CuCl2- and Cu-PTSM-treated groups were lower than that in the control group, the activity in the Cu-PTSM-treated group was significantly higher than that in the CuCl2-treated group. Similarly, CO activity in the cerebellar cortex molecular layer of the Cu-PTSM-treated group was significantly higher than that in the CuCl2-treated group. No significant difference was found in the granular layers among the three experimental groups.

Effects of CuCl2 or Cu-PTSM Treatment on Major Organs in Macular Mice

Table 2 shows laboratory values at 4 wk in mice treated with CuCl2 or Cu-PTSM. No significant differences were detected in hematological data among CuCl2- or Cu-PTSM-treated mice and their normal male siblings. There were no significant differences in biochemistry data for liver and renal function among the groups. Urinary β2-microglobulin, a biomarker for renal tubular function, did not rise in CuCl2- or Cu-PTSM-treated mice.

Table 2 Effects of CuCl2 or Cu-PTSM treatment on hematological, biochemistry, and urinary data in macular mice
Full size table

Discussion

The use of Cu-PTSM results in a relatively higher level of copper transport into the brain as bis(thiosemicarbazone) complexes (16,17). Cu-PTSM diffusively penetrates the cell membrane, and copper is liberated within cells through a reductive reaction (11). The principal source of Cu-PTSM reduction is NADH dehydrogenase in complex I of the mitochondrial electron transport chain (18); reaction with intracellular sulfhydryl groups is an additional reduction mechanism (19). Copper carried by Cu-PTSM shows less selective distribution among various organelles, appearing in the nucleus, mitochondria, and microsomes, as well as in cytosol, thereby entering normal cellular pools for copper ions (10). In this study, macular mice, which would die within 2 wk after birth without treatment, survived with Cu-PTSM treatment, implying that copper carried by Cu-PTSM is used in systemic copper-dependent metabolism.

The copper distribution among organs in the Cu-PTSM-treated mice may be correlated with the homing pattern of Cu-PTSM (16,20). In a radiopharmaceutical study, 2–5% of the injected dose reached the brain, whereas large amounts of Cu-PTSM were transferred to the liver. In Cu-PTSM-treated mice, copper levels in the brain remained low, and copper in the liver was not much reduced. Therefore, a large amount of Cu-PTSM is likely to be trapped by the liver. Combined use of Cu-PTSM with a drug-delivery system to reduce liver trapping may increase copper delivery to the brain accordingly.

ATP7A expression in the brain becomes abundant in the early postnatal period, when neocortical neurons and endothelial cells show dense staining for ATP7A (21). In the brindled mouse, another MD model with impaired ATP7A function, dense copper deposition is found in endothelial cells from early infancy (22). Therefore, ATP7A plays a pivotal role in the passage of copper across the blood–brain barrier, especially in the developing brain. In addition, copper delivery from glia to neurons is also mediated by ATP7A (23). Unusual copper deposits in these cells were counted in the gross copper levels of the brain in our study, but these deposits may not actually be used by neurons.

As CO requires copper for its activity, cerebral CO activity in MD is reduced and causes mitochondrial malfunction (24,25). With effective delivery of copper into the brain across the blood–brain barrier, cuproenzymes, including CO in neurons, would exhibit increased activity. As glia contribute minimally to overall CO activity, changes in optical density of CO histochemistry indicate the changes in neuronal CO activity (26). In our study, CO activity in the cerebral and cerebellar cortex in Cu-PTSM-treated mice was significantly higher than in CuCl2-treated mice. This suggests that copper carried by Cu-PTSM was delivered more effectively to neurons, to activate CO, than copper carried by CuCl2. Cu-PTSM can diffuse across cell membranes to reach neurons because of its high lipophilicity. In addition, the release of copper by NADH dehydrogenase is facilitated by high NADH levels (18,27). As the NADH/NAD+ ratio in cells with CO malfunction is high (28), Cu-PTSM may effectively relieve copper deficits in the encephalic neurons of the macular mouse.

Cu-PTSM is a safe radiopharmaceutical for use in human clinical studies; a single shot of Cu-PTSM had no effect on hematological or biochemical values in humans (20). However, at higher doses, the toxicity of Cu-PTSM should be considered, such as might be the case with copper replacement therapy. PTSM without chelated Cu is not toxic (median lethal dose (p.o.) > 4 g/kg in mice) (29), whereas nonradiolabeled Cu-PTSM exerts acute toxicity with a median lethal dose (i.v.) of 26 mg/kg in rats and causes a transient increase in alanine aminotransferase value in surviving animals (30). The histopathological evaluation revealed that the lung was the primary target organ for Cu-PTSM, which may reflect Cu-mediated oxidative injury. Although Cu-PTSM at the dose used in this study seemed not to affect the functions of the major organs in the mice, the extent of toxic effects of Cu-PTSM is largely unknown in humans and requires further investigation.

Mild renal tubular dysfunction can occur in a subset of patients with MD; urinary β2-microglobulin, a low-molecular-weight protein absorbed at the proximal tubule, increases with age, whereas other biochemical indicators of renal function remain normal (31,32). Dysfunction is caused by significant copper accumulation in the renal tubule. In macular mice, copper accumulated at high levels in the renal tubules during 1 mo of treatment with CuCl2 (6). In our study, copper accumulation in the Cu-PTSM-treated group was lower than that in the CuCl2-treated group. By radiopharmacology, the renal uptake of Cu-PTSM is relatively low, with a trace level of urinary secretion, despite high blood flow to the kidneys; this may explain the relatively lower copper accumulation with Cu-PTSM. In this study, urinary β2-microglobulin secretion did not increase in either the CuCl2-treated or Cu-PTSM-treated groups. However, in considering long-term copper supplementation for MD, the decrease in renal copper accumulation with Cu-PTSM may be beneficial in terms of copper-mediated oxidative injury.

In conclusion, although the copper-trafficking efficacy of Cu-PTSM was limited, Cu-PTSM effectively increased neuronal CO activity. The reduction in copper accumulation in the kidney would seem to be advantageous. These findings support further study of therapeutic potential of Cu-PTSM for MD.

Methods

Drugs

Cu-PTSM was synthesized by Fuji Molecular Planning (Yokohama, Japan), as described by Green et al. (8), with the use of elemental copper instead of radioactive copper. Solvents applicable for Cu-PTSM are limited because of its strong lipophilicity. In this study, DMSO (Sigma, St Louis, MO) was used because of its excellent ability to dissolve Cu-PTSM (33). CuCl2 (Wako, Tokyo, Japan) was dissolved in saline.

Animals

Normal C3H male mice, 15–17 d old, were used in acute toxicity studies for CuCl2, Cu-PTSM, and DMSO. In a copper-supplement study in the MD model mice, male hemizygous macular mice and normal littermate controls were used. Macular mice arose from the C3Hf inbred strain and possess a single base change (4223T→C, exon 22) in the Atp7a gene (14,15). All procedures involving animals were approved by the Animal Care and Use Committees of the Tohoku University School of Medicine and Teikyo University School of Medicine.

Acute Toxicity Studies

Saline (0.01 ml), DMSO (0.01 ml), CuCl2 (25 μg in 0.01 ml saline), or Cu-PTSM (57 μg in 0.01 ml DMSO) was injected subcutaneously into 15–17-d-old normal male C3H mice. The doses of CuCl2 or Cu-PTSM contain the same amount of copper, corresponding to that in subsequent experiments in macular mice. Then, 24 h later, the mice were anesthetized with ether, and blood was sampled. The hematological parameters were determined using a veterinary hematology analyzer (LC-152; Horiba, Kyoto, Japan). Blood chemistry parameters were measured using a biochemistry analyzer (Hitachi 7180; Hitachi, Tokyo, Japan).

Copper Supplement in Macular Mice

Macular mice were treated with subcutaneous injections of CuCl2 or Cu-PTSM solution postnatally on days 4, 10, and 17 with a microsyringe. The doses of CuCl2 and Cu-PTSM on the first injection day (day 4) were 20 and 46 μg, respectively, whereas those on the second and third injection days (days 10 and 17) were 25 and 57 μg, respectively. The doses of Cu-PTSM contained the same amount of copper as in the CuCl2 given in each injection.

At 4 wk of age, control and surviving CuCl2- and Cu-PTSM-treated mice were deeply anesthetized with ether and decapitated. The organs were weighed, and samples of organ tissues (~50 mg) were kept in an acid-washed perfluoroalkoxy vessel to measure tissue copper levels. Simultaneously, the unfixed cerebrum and cerebellum were quickly embedded in optimal cutting temperature compound (Sakura Finetechnical, Tokyo, Japan) and frozen on dry ice for cytochrome oxidase histochemistry. All samples were then kept at −80 °C until analysis.

In separate experiments, urine was collected from another set of CuCl2- and Cu-PTSM-treated mice, as well as from their male siblings, at 4 wk of age. Mice were then anesthetized, and blood was sampled. Urinary β2-microglobulin was measured by enzyme-linked immunosorbent assay (USCN Life Science, Wuhan, China), and samples were used for hematological and biochemistry analysis as described above.

Tissue Copper Levels

Tissue copper levels were measured as described in previous reports, using an inductively coupled plasma-mass spectrometer (ELAN DRC-e; PerkinElmer SCIEX, Concord, ON, Canada) (34). Briefly, the tissue samples were digested with 2 ml of high-purity HNO3 (Wako) at 120 °C for 2 h and dried. The processed samples were dissolved completely with 2% HNO3 to a final volume of 10 ml. Indium (115In) solution was routinely added to the samples as an internal standard to normalize analyte signal intensities for quantitative analysis. Then, the copper (65Cu) concentration was measured with an inductively coupled plasma-mass spectrometer.

CO Histochemistry in the Brain

Duplicate series of sections from each brain were prepared for CO histochemistry using the protocol described previously (35,36), with several modifications. Briefly, the frontal lobes of the brains were coronally sectioned (20 μm thick) in a cryostat microtome at −20 °C and mounted on silanized slides. A set of CO activity standards was also sectioned (20 μm thick) and mounted on each slide. The slides were then kept refrigerated at −80 °C until staining with diaminobenzidine. The sections were incubated at 37 °C for 2 h in the dark in a staining medium containing 120 ml of 0.1 mol/l phosphate buffer (pH 7.4), 60 mg of 3,3′-diaminobenzidine (Sigma), 18 mg of cytochrome c (Sigma), 3 mg of catalase (Sigma), and 4.8 g of sucrose (Wako). The sections were then dehydrated in ethanol, cleared in xylene, and coverslipped.

Photographs of the brain and standard block slices stained for CO histochemistry were obtained in 8-bit grayscale mode with a microscope (BX-61; Olympus, Tokyo, Japan) equipped with a digital camera (DP-70; Olympus). Regions of interest in stained slices were determined using a mouse brain atlas (37). Layer IV of the cerebral cortex somatosensory region and the molecular and granular layers of the hemispheric cortices of the cerebellum were measured. Grayscale values of these regions of interest were measured by densitometry using a custom-made program developed with LabVIEW (National Instruments, Austin, TX) and an image processing extension (VISION; National Instruments); the pixel values in the regions of interest and control were averaged and transformed to CO activity using a Kodak grayscale tablet with known optical density values. Cracks in a slice were programmatically removed from the analysis. The strong correlation between staining density and regional CO activity has been validated previously by Darriet et al. (38) and Gonzalez-Lima and Jones (39).

Preparation of CO Activity Standards

Standards with known CO activity were developed to quantitatively determine enzyme activity in the histochemical material. Six 4-wk-old rats were deeply anaesthetized with ether and perfused with cooled 0.1 mol/l phosphate buffer, and the brains were extracted and homogenized in ice. Half of the homogenized brain paste was heated in an incubator at 60 °C for 5 h to inactivate CO activity, whereas the other half was kept at −20 °C. Three mixtures of active and inactive pastes were made with three different mixture ratios. Each paste was frozen in a silicon tube (3 mm internal diameter) at −80 °C to make cylindrical blocks of paste (3 mm outer diameter×10 mm length). These blocks were embedded together in optimal cutting temperature compound and kept at −80 °C. A small amount (~6 mg) of each mixed paste was separately pretreated in cooled n-dodecyl β-D-maltoside buffer for 15 min, and CO activity was measured using a spectrophotometric method (cytochrome c oxidase activity assay kit; BioChain, Hayward, CA) at room temperature (25 °C).

Statistical Analyses

Data are shown as the means ± SD. Statistical analyses of data, except data for CO activity, were performed by the Kruskal–Wallis test followed by the Steel–Dwass test for multiple comparisons among the groups. CO activity measurements were analyzed by one-way factorial ANOVA after confirming a normal distribution of data, followed by a post hoc Fisher’s protected least significant difference test. For all tests, values of P < 0.05 were considered to indicate statistical significance.

Statement of Financial Support

This research was supported by a Grant-in-Aid for Exploratory Research to Mitsutoshi Munakata from the Ministry of Education, Culture, Sports, Science and Technology of Japan. The authors declared no conflict of interest.