Coupling to a glioblastoma-directed antibody potentiates antitumor activity of curcumin.

Current therapies for glioblastoma are largely palliative, involving surgical resection followed by chemotherapy and radiation therapy, which yield serious side effects and very rarely produce complete recovery. Curcumin, a food component, blocked brain tumor formation but failed to eliminate established brain tumors in vivo, probably because of its poor bioavailability. In the glioblastoma GL261 cells, it suppressed the tumor-promoting proteins NF-κB, P-Akt1, vascular endothelial growth factor, cyclin D1 and BClXL and triggered cell death. Expression of exogenous p50 and p65 subunits of NF-κB conferred partial protection on transfected GL261 cells against curcumin insult, indicating that NF-κB played a key role in protecting glioblastoma cells. To enhance delivery, we coupled curcumin to the glioblastoma-specific CD68 antibody in a releasable form. This resulted in a 120-fold increase in its efficacy to eliminate GL261 cells. A very similar dose response was also obtained with human glioblastoma lines T98G and U87MG. GL261-implanted mice receiving intratumor infusions of the curcumin-CD68 adduct followed by tail-vein injections of solubilized curcumin displayed a fourfold to fivefold reduction in brain tumor load, survived longer, and about 10% of them lived beyond 100 days. Hematoxylin-eosin staining of brain sections revealed a small scar tissue mass in the rescued mice, indicating adduct-mediated elimination of glioblastoma tumor. The tumor cells were strongly CD68+ and some cells in the tumor periphery were strongly positive for microglial Iba1, but weakly positive for CD68. This strategy of antibody targeting of curcumin to tumor comes with the promise of yielding a highly effective therapy for glioblastoma brain tumors.

Coupling to a cancer cell-specific antibody potentiates tumoricidal properties of curcumin.

In vitro studies have shown that curcumin, a polyphenol from the culinary component turmeric, has strong anticancer properties. However, there is no consensus on its therapeutic effect in human. Our earlier experiments involving implanted murine melanoma B16F10 cells in the neck or brain of syngeneic C57BL6 mice showed that tail vein injection of curcumin blocks formation of lesions and tumor in these mice. However, such treatment was ineffective in eliminating established tumors that already occupied ≤10% of brain volume. Possible reasons include low solubility and rapid metabolism of curcumin in vivo. To increase its efficacy, we have linked curcumin through a cleavable arm to an antibody (Ab) against the melanoma surface antigen Muc18. The antibody-coupled curcumin was 230-fold more effective in eliminating B16F10 cells in vitro, and in vivo, it rapidly decimated established, B16F10-evoked brain tumors, enabling the rescued mice to live normally far beyond 90 days from implantation of cancer cells. In contrast, mice treated with Muc18 Ab alone died of brain tumor within a month. In B16F10 cells, curcumin-Ab (adduct) treatment caused a dramatic inhibition of NF-kB: a transcription factor that is constitutively activated in cancer cells. Furthermore, overexpression of NF-kB in the B16F10 cells blocked adduct-evoked stimulation of caspase-3/7 activity. Thus, by suppressing NF-kB, the curcumin adduct inhibits other downstream tumor-promoting proteins, thereby eliminating the B16F10 cells. Our study submits a novel yet generally applicable strategy of converting curcumin into a potent anticancer agent and provides a mechanistic framework for its action.