Objective :Retinopathy and cataract formation are hallmarks of late-stage diabetes mellitus, and the primary contributors of vision loss. The etiologies of both ocular complications are multifaceted, but hyperglycemia-induced injuries stemming from systemic inflammation and oxidative stress are known to play major roles. Curcumin, a polyphenol with anti-inflammatory and anti-oxidant properties, may be suitable for treating diabetic complications, but its clinical translation has been hampered by poor bioavailability. Nanoparticles have been studied as means to improve the pharmacokinetic parameters of less tractable drugs. Here, we utilized a gambogic acid-conjugated nanoparticle system which had been shown to facilitate gastrointestinal uptake via the transferrin receptor, and enhance drug delivery to a variety of tissues. Using diabetic rats, we measured the pathological progression of this disease in the eye as well as the liver, an organ involved in insulin resistance and critical to blood glucose homeostasis. Method Curcumin was encapsulated in double-headed nanosystem by the emulsion-diffusion-evaporation method (Fig 1A). Streptozotocin-induced diabetic Sprague Dawley (SD) rats were divided into three groups: those given daily doses (by oral gavage) of plain curcumin (D+CUR-CMC, 40mg/kg), nanoparticle-curcumin (D+GA2-CUR, 20 mg/kg), or neither (untreated diabetic). Healthy SD rats formed the control group (C). Cataract scores, blood glucose, and body weight were recorded on a weekly basis through an 11-week trial. Tissues were harvested after euthanasia on the 12th week. Quantitation of protein and mRNA expression levels were carried out by Western blot and real-time PCR, respectively. Results Nanoparticle-curcumin treatment alone led to a significant delay in cataract formation by about one stage in the scoring metric. Plain curcumin, in spite of doubled dosage, had little or no effect on cataract formation (Figure 1, B and C). Interestingly, the nanoparticle-curcumin treatment did not reduce blood glucose (Fig 1D). Weight gain, which was inversely correlated with the severity of diabetes, appeared to be marginally improved by both curcumin treatments (Fig 1E). Analysis of liver tissues by Western blot showed marked increase in the levels of P38 phosphorylation (a stress signaling kinase) and P53 among the untreated and plain curcumin groups, whereas nanoparticle-curcumin treatment yielded levels comparable to the control group (Fig 2A). Similarly, real-time PCR detected significant down-regulation of anti-oxidant genes Txnrd1 and Mt1 in the untreated and plain curcumin groups that was attenuated in the nanoparticle-curcumin group. On the other hand, Atf3, a gene inducible by P53, was upregulated in the former groups (Fig 2B). Finally, analysis of GFAP (glial fibrillary acidic protein), a marker for gliosis and an early feature of diabetic retinopathy, revealed characteristic expression increases in the diabetic and plain curcumin groups. Meanwhile, nanoparticle-curcumin treatment resulted in greatly reduced glial cell expansion and penetration through the retinal layers (Fig 2D).