Isoleucine and valine in prostate cancer drive cholesterol synthesis

Isoleucine and valine in prostate cancer appear to do more than provide building blocks for proteins. In tumour cells, these two essential amino acids can supply a metabolic signal that increases cholesterol production and helps sustain androgen receptor signalling when male sex hormones are scarce.

That matters because many prostate cancers depend on androgen receptor activity. A major treatment strategy is therefore to deprive tumour cells of androgen signalling. Yet some cancers eventually adapt and become castration resistant. A study published in Nature Metabolism identifies a metabolic pathway that may contribute to this adaptation.

The researchers combined experiments in prostate cancer cell lines, three-dimensional cultures, organoids and mouse xenografts with analyses of existing human tumour datasets. At the centre of the mechanism was propionyl-CoA, a small metabolic intermediate produced during the breakdown of several nutrients.

Turning amino acids into a signal

Using isotope-labelled nutrients, the team traced the origin of propionyl-CoA in prostate cancer cells. Isoleucine and valine together accounted for about 70% of the intracellular propionyl-CoA and related propionylcarnitine pool.

Propionyl-CoA, however, was not acting merely as an intermediate in energy metabolism. It can also modify proteins by donating a propionyl group to particular lysine residues, a process known as lysine propionylation.

One important target was SREBP2, a transcriptional regulator that controls genes involved in cholesterol synthesis. Propionylation of a specific site on SREBP2 made its active nuclear form more stable, allowing the cholesterol-production programme to remain active for longer.

This is particularly interesting because cholesterol normally helps shut down its own synthesis through negative feedback. The propionyl-CoA-dependent mechanism allowed SREBP2 activity to persist despite increasing intracellular cholesterol.

From cholesterol to androgens

Extra cholesterol can be especially useful to a prostate tumour because cholesterol is the starting material for steroid hormones, including androgens.

When the researchers increased propionyl-CoA, prostate cancer cells produced more cholesterol and subsequently more testosterone and dihydrotestosterone. This helped maintain androgen receptor activity. Under experimentally androgen-depleted conditions, increasing propionyl-CoA partially restored tumour-cell growth and androgen receptor signalling.

The result was a feed-forward metabolic circuit. Isoleucine and valine breakdown generated propionyl-CoA. Propionyl-CoA stabilized SREBP2 and promoted cholesterol synthesis. Cholesterol then supplied steroid production, while renewed androgen receptor activity further stimulated lipid synthesis and storage.

Several experiments supported the individual links in this chain. Removing SREBP2 prevented the propionyl-CoA-dependent increase in cholesterol. Blocking cholesterol synthesis impaired spheroid growth. Interfering with the conversion of cholesterol into steroid hormones reduced the ability of propionyl-CoA to support growth under androgen-deprived conditions.

What happened in mice

The researchers then tested whether changing the availability of isoleucine and valine could influence this pathway in vivo. Nude mice carrying xenografts derived from castration-resistant prostate cancer cells received a diet containing 25% of the standard amounts of isoleucine and valine, while other amino acids were increased to preserve total amino-acid-derived caloric intake.

Over seven weeks, this diet markedly suppressed tumour growth and reduced markers of cell proliferation. The tumour-growth experiments included 8–9 mice per group. Restricting the two amino acids also reduced the ability of tumour cells to colonize the lungs.

Human tumour data were consistent with the proposed mechanism. Advanced prostate cancers showed metabolic gene-expression changes compatible with increased propionyl-CoA production, and a metabolite associated with the propionyl-CoA pool correlated with cholesterol levels in prostate tumour tissue. These observations in patients are associative, however, and do not establish that the pathway causes disease progression in humans.

The main limitation is therefore clear: most of the mechanistic evidence comes from cultured cells, organoids and mouse models. Dietary restriction of isoleucine and valine has not been tested as a prostate cancer therapy in people. Both are essential amino acids that humans must obtain from food.

Some cell experiments also used propionylcarnitine concentrations above typical serum levels. The investigators addressed this concern by manipulating propionyl-CoA through genetic approaches as well and obtained results supporting the same pathway.

The study therefore does not show that patients with prostate cancer should avoid foods containing isoleucine or valine. Its significance is mechanistic: it identifies a pathway connecting amino-acid breakdown to cholesterol synthesis, androgen production and adaptation to androgen deprivation. If future work confirms that this pathway is important in human tumours, its individual components could become targets for new therapeutic strategies.