17 MARCH 2017 • VOL 355 ISSUE 6330 1127 SCIENCE sciencemag.org
mating-type information is expressed from
the HML and HMR loci (so-called hidden
mating-type loci). The expression of genes
from these loci is controlled by silencing,
whereby chromosomal regions are pack-
aged tightly into chromatin so that the
genes located there are not transcribed (4).
Sir2, the key aging regulator, is critical for
this silencing in young cells.
It was long thought that decreasing
Sir2 function in an old mother causes sterility because of a loss of silencing at the
HML and HMR loci (9). This would cause
both mating-type genes to be expressed
simultaneously, making mating impossible. However, Schlissel et al. discovered
that silencing at these loci is intact in old
cells. They used a clever strategy to build
a highly sensitive detector of loss of silencing at the HML locus. The HMLa2 gene was
replaced with a gene encoding the enzyme
Cre recombinase. The authors also inserted
a fluorescent reporter gene elsewhere in
the yeast genome. This reporter is designed
to be Cre-responsive, such that cells in
which Cre is absent (silenced) express a red
fluorescent protein and cells in which Cre
is expressed (loss of silencing) express a
green fluorescent protein. The authors did
not observe any silencing events (switch
from red to green) as the mothers aged.
This was so surprising that they expanded
the analysis to over 1500 yeast pedigrees.
In total, only 13 loss-of-silencing events
were identified, none of which correlated
with the yeast cell’s age. Moreover, nearly
all the old mothers eventually stopped dividing without showing any loss of silencing. A chemical inhibitor of Sir2 did cause
a loss of silencing, suggesting that a decline
in Sir2 activity would be sufficient to promote HML expression, but that this doesn’t
happen during aging.
To further test the hypothesis that decline
in Sir2-mediated transcriptional silencing
underlies aging yeast cell sterility, Schlissel et al. compared the genes expressed (or
repressed) in old versus young cells and
in cells in which Sir2 activity is inhibited.
Sir2 inhibition caused high expression (loss
of silencing) of HMRa1 messenger RNA
compared to old cells. However, when the
authors analyzed publicly available RNA
sequencing data sets from yeast cells lacking Sir2 and compared them to RNA sequencing results from old and young yeast
cells, they found no overlap.
What, then, is the true cause of age-
induced sterility? When haploid cells of the
a mating type (MATa) are exposed to a-fac-
tor pheromone from haploid cells of the a
mating type (MATa), they arrest their cell
cycle and start to grow a mating projection
(see the figure). Old yeast cells are less re-
sponsive to pheromone, but this has noth-
ing to do with loss of silencing at the HML
locus. Then what does cause decreased sen-
sitivity to pheromone and impaired mating
response in old cells? One way of thinking
about the problem is to consider the asym-
metry of the phenotype. Whereas old moth-
ers respond poorly to pheromone, their
daughters do so proficiently. Could there
be a factor that is asymmetrically inherited
between an old mother and her daughter
that governs the mating response? This
scenario was reminiscent of a prion-like
aggregation-prone RNA-binding protein
called whiskey 3 (Whi3). Whi3 inhibits the
G1-to-S phase transition of the cell division
cycle and is inactivated when it aggregates
in old mother cells (10). These aggregates
also permit yeast cells to “memorize” de-
ceptive mating attempts. Unlike true pri-
ons, the Whi3 “mnemon” is asymmetrically
retained in mother cells. Schlissel et al.
asked whether the asymmetric aggregation
of Whi3 in old mother cells caused age-
dependent sterility. The authors deleted a
glutamine-rich prion-like domain in Whi3,
which is required for its capacity to aggre-
gate. Remarkably, this simple manipulation
prevented old cells from losing sensitivity
to the mating pheromone. Indeed, when
the authors visualized Whi3 (by express-
ing it in yeast as a fusion with a fluorescent
protein), they observed its aggregation in
old cells. Thus, Whi3 aggregation provides
a compelling explanation for the sterility of
old age, and suggests that the selective re-
tention of aggregated Whi3 by mother cells
allows daughter cells to become responsive
The study by Schlissel et al. shows that
contrary to previous assumptions, loss of
transcriptional silencing is not responsible for sterility in old haploid yeast, and
invokes the intriguing hypothesis that aggregation of the RNA-binding protein Whi3
instead accounts for age-related sterility.
They also propose the provocative idea that
by accumulating protein aggregates, old
cells are “differentiated” relative to young
cells, which may be beneficial to the ecology of wild yeast. The study of Schlissel et
al. is exciting in terms of resolving a long-standing question in the aging field and
providing a new link between the asymmetric retention of aggregates and phenotypes classically associated with aging. j
1. L. Guarente, C. Kenyon, Nature 408, 255 (2000).
2. C. Kenyon, J. Chang, E. Gensch, A. Rudner, R. Tabtiang,
Nature 366, 461 (1993).
3. D.E.Gottschling,O.M.Aparicio, B.L.Billington,V.A.
Zakian, Cell 63, 751 (1990).
4. J. Rine, I. Herskowitz, Genetics 116, 9 (1987).
5. J. S. Smith, J. D. Boeke, Genes Dev. 11, 241 (1997).
6. M. Kaeberlein, M. McVey, L. Guarente, Genes Dev. 13, 2570
7. J. G. Woodetal .,Nature 430, 686 (2004).
8. G. Schlissel, M. K. Krzyzanowski, F. Caudron, Y. Barral,
J. Rine, Science 355, 1184 (2017).
9. T. Smeal, J. Claus, B. Kennedy, F. Cole, L. Guarente, Cell 84,
10. F.Caudron, Y.Barral, Cell 155,1244(2013).
Haploid MATa Haploid MAT;
Cell cycle arrest
a and ;;types
Less responsive to
Unable to mate
Whi3 in daughter
(responds to ;-factor)
The sterility of old age
Aging yeast mother cells accumulate aggregated Whi3, which blocks mating. Daughter cells are spared.