The incidences of minocycline-induced cutaneous pigmentation have been reported in up to 41% of patients with RA who take minocycline [10
]. Histochemical staining for pigmented substances (e.g., iron, melanin) has shown that minocycline-induced cutaneous pigmentation can be classified into three or four distinct types [11
]. In the present study, histochemical staining of the biopsied muscle demonstrated that the pigments comprised iron and melanin, which is consistent with type II minocycline-induced cutaneous pigmentation [11
]. Iron deposition in skeletal muscle has rarely been described in some disorders, including idiopathic hemochromatosis, transfusion induced hemosiderosis, diabetic neuropathy, Waldenstrom’s macroglobulinemia, and AIDS patients [12
]. Laboratory investigations of our patient showed no evidence of these conditions. No previous study has reported melanin deposition in human skeletal muscle, except for cases of melanoma invasion. Here, we report the first identification of minocycline-induced pigmentation (comprising iron and melanin) in skeletal muscle.
The sural nerve biopsy demonstrated the features of a chronic axonal neuropathy, with minimal pigment observed in the cytoplasm of Schwann cells. Only a few cases of minocycline-associated sensory neuropathy have been reported, and axonal-type sensory neuropathy is a common complication in RA [15
]. Thus, it is unclear whether or not the sensory neuropathy observed in our patient is related to minocycline therapy.
Tissue pigmentation due to minocycline treatment is not usually harmful [9
]. However, the striking finding in the biopsied muscle from our patient is the presence of rimmed vacuoles that had acid phosphatase-positive autophagic activity and contained numerous pigmented granules in various forms. The strong association between autophagic vacuoles and the accumulation of minocycline-induced pigments is evident. The rimmed vacuoles were positive for p62/SQSTM1 (p62) immunostaining (Figure A), which has been reported as a diagnostic marker for drug-induced AVM [16
]. In distal myopathy with rimmed vacuoles (DMRV), activation of autophagy is thought to occur via the accumulation of unfolded/misfolded proteins [2
]. No filamentous inclusion (representing the accumulation of unfolded/misfolded proteins) was seen on electron microscopy of samples from our patient. There was also no immunostaining inclusion seen under light microscopy of samples immunostained with an antibody against TAR-DNA binding protein-43 (TDP-43) (Figure B). TDP-43-positive inclusions are thought to be a common endpoint of muscle degeneration among the myopathies associated with rimmed vacuoles, such as DMRV, inclusion body myositis and oculopharyngeal muscular dystrophy [17
]. All of these conditions have been associated with the presence of filamentous inclusions in myofibers. Thus, our findings suggest that long-term minocycline treatment might cause RVM by activating autophagy through the accumulation of minocycline-induced pigments. Although no previous report has shown minocycline therapy directly linked to autophagy in human disease, two recent studies using different in vitro culture systems found that minocycline induced autophagy and inhibited the growth of tumor cell lines, including glioma cells and epidermoid cancer cells [18
]. In addition, minocycline-induced pigments are known to form complexes containing minocycline and/or its metabolites [20
], suggesting that minocycline could directly activate autophagy.
Figure 4 Immunohistochemistry of muscle biopsy. Immunohistochemistry was performed on serial paraffin sections. Sections were immersed in citrate buffer, heated in a microwave for antigen retrieval, and immunostained with antibodies against p62/SQSTM1 (1:2000; (more ...)
One year after the cessation of minocycline therapy, our patient exhibited no worsening of gait disturbance. There was little improvement in muscle weakness, however, perhaps because she had taken minocycline for an extremely long duration even after the appearance of skin pigmentation. This is an unusual condition. However, long-term minocycline treatment may be used more widely in the future, as on-going clinical trials are assessing the therapeutic use of minocycline as an anti-inflammatory agent in a broad range of disorders. Thus, we must draw attention to this adverse effect because its insidious onset, slowly progressive course, and lack of creatine kinase elevation could complicate its prompt identification in patients.
Chloroquine and hydroxychloroquine are anti-malarial drugs that are also used to treat RA and systemic lupus erythematosus. Hydroxychloroquine is used much more frequently than chloroquine, as the latter is more likely to cause irreversible retinal damage. Long-term hydroxychloroquine treatment rarely causes myopathy and sensory neuropathy; the former typically manifests with an insidious onset and normal-to-mildly-elevated creatine kinase levels, and muscle biopsy consistently reveals curvilinear bodies and muscle fiber atrophy with vacuolar changes [3
]. After the therapy is discontinued, the resolution of symptoms is slow and may be incomplete. In rare cases, long-term hydroxychloroquine treatment can also cause skin pigmentation [23
]. In all reported cases, the main histological feature of hydroxychloroquine-induced pigmentation was the presence of iron and melanin deposits in the dermis [23
]. These features are very similar to those observed in our patient, who had never taken the anti-malarial drugs, suggesting that there may be a common pathogenic mechanism underlying these conditions. Hydroxychloroquine myopathy is usually of mild to moderate severity, but a few case of severe hydroxychloroquine myopathy, involving respiratory muscles and/or cardiac myocytes, have been reported [24
]. However, our patient did not have any symptoms related to respiratory or cardiac dysfunction. Chloroquine and hydroxychloroquine accumulate within lysosomes and are thought to block autophagy by elevating intralysosomal pH and inhibiting lysosomal enzymes [26
]. However, the exact mechanism underlying myocyte toxicity is unclear. In the future, it will be particularly interesting to examine whether iron and melanin are deposited in the muscles of hydroxychloroquine myopathy patients.