#114 – Eileen White, Ph.D.: Autophagy, fasting, and promising new cancer therapies

KEYPOINTS

• 01

  Eileen White found autophagy by accident, through apoptosis

  Eileen White, of the Rutgers Cancer Institute of New Jersey, did her postdoc with Bruce Stillman at Cold Spring Harbor. She was given an adenovirus oncogene and found it was a viral version of BCL-2, a human oncogene that blocks apoptosis (programmed cell death). Cells that couldn't commit suicide survived even in buffer with no nutrients. Explaining how led her lab, in the early 1990s, into autophagy, then barely studied in mammals.

• 02

  Apoptosis is a tidy, tumour-suppressing death

  Unlike necrosis, where cells burst and inflame surrounding tissue, apoptosis breaks a cell down in a controlled way and packages it into apoptotic bodies. Macrophages clear these, which limits inflammation and may help antigen presentation. p53 activates apoptosis-promoting proteins such as PUMA and NOXA, which antagonise BCL-2. Common hotspot mutations of p53 can both disable it and actively promote cancer, so the old 'loss versus gain of function' teaching is too black-and-white.

• 03

  Autophagy is the cell's recycling system

  Under stress such as starvation, cells engulf their own components in autophagosomes and send them to lysosomes to be broken down for fuel and building blocks. Exercise induces autophagy strongly, and muscle needs it because exercise damages muscle. White points out that much of a cell's mass is protein, a point David Sabatini has made, so recycling proteins is a major supply route.

• 04

  Autophagy is essential to life

  Mice lacking the autophagy genes ATG5 or ATG7 are born but die in the newborn period. When White's lab deleted autophagy in adult mice, they died of neurodegeneration within two to three months. If those mice were fasted, they were all dead within 16 hours. The brain and liver depend on autophagy for long-term protection.

• 05

  Tumours depend on autophagy more than the host does

  In mice with lung cancer driven by RAS, deleting an essential autophagy gene throughout the body killed the tumour before the mouse. A newer model from a former trainee and a UK lab can switch autophagy off and back on, and normal tissues show a remarkable capacity to recover. That suggests treatment could use a temporary window.

• 06

  Some cancers are especially autophagy-dependent

  RAS-driven cancers, BRAF V600E lung cancers and melanomas, and homologous-recombination-deficient breast cancers appear particularly sensitive. Lung cancers that have lost LKB1, the tumour suppressor that activates AMPK, can't turn on this protective response to low energy, a mouse model made by Jessie Guo, now at Rutgers.

• 07

  The paradox: autophagy prevents cancer yet feeds established tumours

  By clearing damaged components, autophagy buffers the stress that can lead to cancer. Once a tumour exists, it uses autophagy to survive. White cautions against equating caloric restriction with autophagy. Caloric restriction may limit tumours partly through other routes.

• 08

  No one knows the dose of fasting

  Attia is sure that after about seven days of water-only fasting, autophagy is 'fully cranked', and that a 12-hour gap does nothing. Where between those it switches on is unknown. That leaves his own protocols (3 days monthly, 5 or 7 days quarterly) as guesses. Measuring autophagic flux requires blocking lysosomal breakdown with bafilomycin or hydroxychloroquine and tracking the conversion of LC3-I to LC3-II, so blood-cell measurements in fasting people only give clues. George Cahill's classic 40-day fasting study showed that the first, second and third 24 hours are metabolically very different. The open question is whether 48–72 hours gets you 80% of the benefit or only 20%.

• 09

  Metabolic tracing could answer this

  White proposes labelling mice with carbon-13 or nitrogen-15 tracers to see how different tissues use nutrients during fasts of different lengths. That could define the roughly 50 metabolites or proteins that track autophagy, and later test them in humans. It's possible now, she says, in a way it wasn't 20 years ago.

• 10

  Delaying chronic disease is the real target

  Attia argues that the animal and centenarian data point to delaying the onset of chronic disease, which by his analysis accounts for 82% of US deaths over 50. He notes that for solid tumours, median survival has improved by less than about a year since 1970.