Single-walled carbon nanotubes (SWNTs) are unique in that they strongly absorb near-infrared (NIR) light, while biological systems have very low levels of absorption of NIR light. It has recently been shown that cancer cells could internalize SWNTs functionalized with a folate moiety attached and that the cells could be then be killed by exposure of the cells to NIR radiation with an 808 nm laser, with cell death caused by extensive local heating of SWNTs inside the cells. For cancer treatment, this will require that the SWNTs be delivered directly to the cancer cell surface.

To avoid the difficult problem of delivering the SWNTs to the cancer cell surface, this project focuses on developing a means for targeting the SWNTs to the tumor vasculature. It is proposed to target the SWNTs by conjugation with the human annexin V protein. Human annexin V (molecular weight of 36,000) is a monomeric protein that binds with high affinity to phophatidylserine (PS) in phospholipid bilayers. PS is the most abundant anionic phospholipid of the plasma membrane and is tightly segregated to the internal side of the plasma membrane in most cell types. Anionic phospholipids are largely absent from the external surfaces of resting mammalian cells under normal conditions. Recently, it has been found that PS is expressed on the external surface of endothelial cells that line the blood vessels in tumors but is not expressed on the outside surface of the vascular endothelium in normal organs. Thus, annexin V can be used to specifically target the endothelial cells of the tumor vasculature. The underlying hypothesis is that a SWNT-annexin V conjugate injected into the bloodstream will localize in the tumor vasculature and that exposure of the tumor to NIR radiation will cause the tumor to be selectively killed by a combination of heating and cutting off its blood supply.

The first objective of this project is to demonstrate protein attachment to fully suspended SWNTs with no loss of protein activity or loss of the NIR absorption properties of the SWNTs. This work was first carried out with the model protein horseradish peroxidase (HRP), which has a size similar to that of annexin V (40 kDa) and is also a monomeric protein. Two approaches were used for attachment of HRP: (1) adsorption on SWNTs using the sodium cholate suspension-dialysis method, and (2) covalent conjugation to carboxymethylcellulose (CMC) adsorbed to the SNWTs. Both of these methods gave good suspension of SWNTs and almost complete retention of native HRP enzymatic activity before attachment. Both methods also resulted in retention of a substantial fraction of the NIR absorption peak at 980 nm. Atomic force microscopy (AFM) images of the SWNT/protein complexes were obtained.

The sodium cholate suspension-dialysis method was applied to purified recombinant annexin V produced in our laboratory for adsorption on SWNTs. For this protein, the method resulted in complete precipitation of the SNWT/protein complex. Therefore, annexin V was covalently coupled to CMC adsorbed to the SWNTs. The results of an in vitro binding assay of SWMT/CMC/annexin V to PS will be discussed.