A group of structurally diverse natural products that are potent inhibitors of the growth of cancer cells has been identified, including cephalostatin 1, OSW-1, ritterazine B and two schweinfurthins,.

As these compounds share a characteristic pattern of sensitivity and resistance when tested with the National Cancer Institute’s panel of sixty cell lines (MCI-60), they can be assumed to have the same cellular targets, or, at least, attack targets in the same signal transduction or metabolic pathway. However, that target or pathway has remained unknown until very recently.

Now, Matthew Shair and his co-workers at Harvard University, Massachusetts, USA, working with colleagues in the Novartis Institutes for Biomedical Research, also in Massachusetts, and in Tokyo University of Pharmacy and the Life Sciences, Tokyo, Japan, has revealed the targets of these compounds to be oxysterol binding protein (OSBP) and a close homologue, OSBP-related protein 4L (ORP4L).

These proteins have not previously been implicated in cancer cell growth or survival. Based on their findings, the researchers have named the group of natural products that bind oxysterol binding protein and its homologues the ORPphilins.

Shair and his colleagues first set out to prove that the compounds shared a target and probable mechanism of action using cell lines further to the NCI-60 panel.

They established that all the compounds were selective for p21^-/- cells, and in particular that they inhibited the growth of a HCT-116 cell line with p21 deleted much more strongly than wild type HCT-116.

They then used affinity purification to separate a protein that specifically bound to OSW-1, the most abundant of these compounds, and identified it using mass spectrometry to be OSBP.

The only other protein with comparable binding affinity seemed to be ORP4L, a paralog of OSBP that is also a high-affinity receptor for the oxysterols.

Members of the 16-protein superfamily that includes OSBP and ORP4L are known to be involved in lipid processing, metabolism and signalling, and in vesicular traffic and sterol transport.

This discovery seemed to implicate these proteins further in cancer cell growth and survival. Shair and his colleagues investigated this further by, first, knocking down OSBP levels in HCT-116 p21^-/- cells and showing that this desensitised those cells to the ORPphilins, but not to other known anti-tumour compounds such as taxols. 

They also showed a strong positive correlation between binding affinity to OSBP and ORP4L and antiproliferative activity over the whole known ORPphilin family.  Non-cytotoxic concentrations of 25-hydroxycholesterol, which competes with ORPphilins for binding to these proteins, were also shown to specifically inhibit their activity.

These compounds were, however, shown not to interact with another cellular 25-hydroxycholesterol receptor known as Insig, which implies that they do not interfere with cholesterol biosynthesis.

The researchers then mutated 15 OSBP residues located both within and outside its sterol binding region, and tested these mutant proteins for sensitivity to ORPphilins. Several of the mutations were found to decrease ORPphilin activity, but not in a way that suggested that they affected compound binding.

ORPphilins did, however, seem to reduce the concentration of OSBP within the cell and to affect its sub-cellular localisation. Pulse labelling with [3H] serine in several cancer cell lines showed that low doses of ORPphilins, like 25-hydroxycholesterol, increased the synthesis of sphingomyelin and other lipids in the cells although, unlike with 25-hydroxycholesterol, this effect was reversed with high ORPphilin doses.

Taken together, these results show that the natural products here grouped together as the ORPphilins bind selectively to two, and possibly more, members of the OSBP super-family, and that this binding mediates their ability to inhibit the growth of cancer cells.

Although much work remains to be done before the mechanism of these proteins is fully understood, these results provide a new link between lipid processing and cancer development. These compounds are bound to prove useful in probing the function and mechanism of their protein targets further.

DOI: 10.1038/nchembio.625