Quinolones are antibacterial agents that have the potential to induce Achilles tendon disorders - such as tendinitis or even ruptures - in patients treated with these drugs. We studied the effects of ciprofloxacin on several proteins of Achilles tendons from immature dogs, 10- to 11-weeks-old. The dogs were treated orally for 5 days with 30 or 200 mg ciprofloxacin/kg body weight or with the vehicle alone. Since quinolone-like alterations in joint cartilage were observed in magnesium-deficient animals, another group was fed a magnesium-deficient diet for 6 weeks. At necropsy, tendons (n=3 from each group) were frozen and stored until analysis when they were homogenized in a lysis buffer to release a soluble fraction of the tendon proteins. Densitometric analysis of the immunoblots with anticollagen type I, anti-elastin, anti-fibronectin, and antiintegrin antibodies showed a significant reduction of all proteins. For example, collagen type I concentrations (mean +/-SD, arbitrary densitometric units) were 3190+/-217 (controls), 1890+/-468 (30mg/kg), 1695+/-135 (200mg/kg) and 2053+/-491 in the magnesium-deficient dogs. The differences between concentrations in controls and all treated groups were statistically significant (P<0.01, t-test). Similarly, compared with control samples, relative concentrations of other proteins in tendons from ciprofloxacin-treated dogs (30 mg/kg) decreased by 73% (elastin), 88% (fibronectin), and 96% (beta1 integrin) (data from low-dose group only). A very similar pattern of protein alterations was detected in samples from magnesium-deficient dogs. In conclusion, rather low doses of a fluoroquinolone or a diet-induced magnesium deficiency caused similar biochemical alterations in the soluble fraction of proteins from canine tendons. These findings support our hypothesis that quinolone-induced toxic effects on connective tissue structures are due to the magnesium-antagonistic effects of these antibacterial agents. They also indicate that patients with a latent magnesium deficiency could be at an increased risk of quinolone-induced tendon disorders.