[Cell Cycle 4:7, 897-900; July 2005]; 2005 Landes Bioscience

The Jak/Stat Pathway

Perspective

A Novel Way to Regulate PI3K Activity

ABSTRACT

Kathrine Abell
Christine J. Watson*

Received 05/0605; Accepted 05/11/05

KEY WORDS

ON

apoptosis, Stat3, PI3Kinase, mammary gland,

transcription, regulatory subunits

OT
D

IST

Previously published online as a Cell Cycle E-publication:

http://www.landesbioscience.com/journals/cc/abstract.php?id=1837

RIB

*Correspondence to: Christine J. Watson; Department of Pathology; University of

Cambridge; Tennis Court Road; Cambridge CB2 1QP UK; Email: cjw53@mole.bio.
cam.ac.uk

UT
E

Department of Pathology; University of Cambridge; Cambridge UK

Phosphoinositide 3-kinases (PI3Ks) have been grouped into three major classes that
have different substrate specificities. Class IA PI3Ks consist of a catalytic and a regulatory
subunit and have multiple isoforms that arise from different subunit combinations. The role
of two of the small regulatory subunits, p55 and p50, is poorly understood. We have
now identified a novel function for these subunits and have shown that their expression
is specifically induced in the involuting mouse mammary gland where they are involved
in the downregulation of PI3K signalling and Akt/PKB activity. This abrogation of survival
signalling thru Akt/PKB and its downstream targets is essential for the induction of apoptosis. The switch from lactation to involution is associated with activation of the transcription factor Stat3, by the cytokine LIF. Stat3 is essential for the induction of apoptosis and,
in the absence of Stat3 or LIF, expression of the p55 and p50 subunits is abrogated.
Surprisingly, Stat3 is a direct regulator of p55 and p50 expression, as demonstrated
using ChIP assays, and therefore these subunits are not splicing isoforms as previously
thought. An important implication of our results is that the p55 and p50 small regulatory
subunits are regulated independently of the larger p85 subunit, and have an essential
role in Stat3-mediated apoptosis in mammary gland.

20

05

LA

ND

ES

BIO

SC

IEN

CE

.D

The class IA PI3Ks play a pivotal role in mediating a variety of tyrosine kinase receptorcontrolled signalling activities including proliferation, differentiation, vesicular transport,
cytoskeletal reorganisation, metabolism and inhibition of apoptosis.1 They produce the
lipid second messenger phosphatidylinositol (3,4,5)-triphosphate (PIP3), from membrane
bound PIP2, which acts via downstream effectors such as Akt/protein kinase B (PKB).
Class IA PI3K is a heterodimeric enzyme composed of one catalytic subunit (p110,
p110 or p110) and one regulatory subunit (p85, p85, p55, p55 or p50).2 The
three catalytic subunits identified thus far are encoded by three different genes. While
most tissues express the p110 and p110 catalytic subunits,3 the third catalytic subunit
p110 is expressed predominantly in leukocytes.4 Different roles for these catalytic subunits
are suggested by their expression patterns and the results of gene deletion studies.5
The regulatory subunits are more complex. Three of the subunits, p85, p55 and
p50, are derived from the same gene, pik3r1, and are thought to be splicing variants6,7
whereas p85 (pik3r2) and p55 (pik3r3) are encoded by separate genes.8,9 These five
regulatory subunits have highly homologous carboxy-terminal domains composed of two
SH2 domains surrounding a p110 binding region. In contrast, the amino-terminal
sequences are unique to each subunit. The p85 and p85 subunits each have a large
amino-terminus containing an SH3 and a bcr homology (BH) domain whereas the smaller
subunits have unique amino-terminal domains consisting of 34 (p55 and p55) and six
amino acids (p50). Thus the functional specificity of the different regulatory subunits
resides in their N-terminus (Fig. 1A).
Each regulatory subunit can bind to each of the catalytic subunits.6,10 Activation of
tyrosine kinase receptors mediates recruitment of the PI3K heterodimer to the membrane
(Fig. 2), where the catalytic subunit phosphorylates PIP2 generating PIP3. The formation
of PIP3 activates a signalling cascade which results in the phosphorylation, by PDK1, of
specific residues of Akt/PKB which is regarded as the primary downstream target of the
PI3K pathway. Akt/PKB mediates cell survival by phosphorylating pro-apoptotic factors
such as caspase-9, Bad, and Forkhead transcription factors.11 The PI3K-Akt pathway is
negatively regulated by the lipid phosphatase, phosphatase and tensin homologue deleted
on chromosome 10 (PTEN). Loss of function mutations in PTEN result in dysregulation
of the PI3K pathway and often lead to cancer.12

www.landesbioscience.com

Cell Cycle

897

Stat3 Directly Regulates Expression of PI3K Regulatory Subunits

Several different mechanisms have

been proposed to explain how the
A
B
regulatory subunits can modulate the
activity of the catalytic subunits:
(1) In the latent state, the regulatory subunit protects the catalytic
subunit from degradation, but also
inhibits its lipid kinase activity; (2)
In the stimulated state, the regulatory
subunit associates with phosphotyrosine residues on activated receptors or signalling intermediates. This
interaction not only recruits the
catalytic p110 subunit to the
membrane in close proximity to its
lipid substrates, but also releases the
inhibition of catalytic activity imposed
by initial binding to the regulatory Figure 1. (A) Schematic representation of the structures of the PI3K regulatory subunits p85, p55 and p50
subunit; and (3) a molecular excess showing the distinct amino-terminal regions and the common p110 catatytic subunit binding domain. (B)
Comparison of the relative levels of expression of the p85, p55 and p50 regulatory subunits, active phosphoof monomeric regulatory subunits tyrosine Stat3, active Akt/PKB and PTEN in the two phases of mammary gland development characterized by
competes with holoenzyme PI3K for survival signalling (lactation) and apoptosis induction (involution).
receptor binding sites. Thus, the
regulatory subunits regulate both the
location and specific activity of the PI3K catalytic subunit and can mammary gland involution where redundant milk-producing
act as both positive and negative modulators. This complex and epithelial cells undergo apoptosis following weaning.16 Molecular
dynamic interplay between the subunits thus controls the outcome changes associated with the induction of apoptosis include the
downregulation of pro-survival PI3K and Akt/PKB activity and
of PI3K signalling.
The question ariseswhy are there so many regulatory subunits? activation of the pro-apoptotic transcription factor Signal transducer
One possibility is that they have tissue-specific effects and this is and activator of transcription (Stat)3. Transgenic expression of a
supported by their different expression patterns. The p85 and constitutively active mutant of Akt/PKB inhibits the induction of
p85 subunits are ubiquitously expressed3 although p85 is mammary epithelial cell death,17 thus over-riding the pro-apoptotic
expressed at much lower levels than p85 in liver and muscle.8 signal from Stat3. It is therefore important for Akt/PKB to be downHowever, the expression of the smaller subunits is more restricted: regulated in order to shift the balance in favor of the cell death signals
p55 is highly expressed in brain and muscle,6,10 p50 in brain, from Stat3. Surprisingly, we found that PTEN was downregulated in
liver, muscle and kidney,7,6 and p55 in brain.9 Since the molecular involution.18 This suggested to us that an alternative mechanism for
ratio of regulatory to catalytic subunits is crucial for optimal activity regulating PI3K activity may operate in this context.
As mentioned above, it has been shown that changes in the reguof PI3K, too few or too many regulatory subunits compared to the
number of catalytic subunits could dramatically affect PI3K activity. latory subunit expression level can result in substantial changes in
Evidence from knockout mice supports this notion: deletion of a PI3K signalling.15 We therefore speculated that alteration in the
single allele of p85 resulted in increased PI3K signalling which was subunit composition of PI3K may be responsible for the reduced
suggested to result from the reduced competition between the p85 PI3K activity and hence the drop in pAkt/PKB during the apoptotic
subunit monomer and the p85/p110 dimer.13 Deletion of the p55 switch.
This is indeed the case. We discovered a dramatic change in the
and p50 subunits results in increased insulin sensitivity.14 Recently,
it has been shown that deletion of the p110 and p110 catalytic expression profile of the regulatory subunits at the onset of epithelial
subunits has the opposite effect on insulin sensitivity to deleting the apoptosis with substantial transcriptional upregulation of p55 and
regulatory subunits.15 It was also proposed that the p85 subunits p50 and downregulation of p85.18 In comparison, the expression
have a PI3K independent role as negative regulators of insulin levels of the three catalytic subunits were constant. The upregulation
of p55 and p50 coincided with activation of Stat3 (Fig. 1B).
signalling.
Given the importance of these regulatory subunits, surprisingly Using conditional deletion of Stat3 in mammary epithelium, we
little is know about how their expression is controlled. Although showed that in the absence of Stat3 in vivo, upregulation of p55
overexpressing the subunits in cell models, and the generation of and p50 was abrogated, levels of activated Akt were sustained and
mice deficient for individual regulatory subunits, have provided apoptosis was prevented.118 Thus, there is a correlation between
some insights into this question, there is a pressing need for more Stat3 activation and expression of the regulatory subunits. Since
sophisticated model systems to delineate the roles of individual Stat3 is a transcription factor, this raised the interesting question: are
subunits in different tissue environments. Thus, this critical question p55 and p50 direct transcriptional targets of Stat3?
This seemed unlikely given that it had previously been believed
remains largely unresolved and until now, individual roles have not
that the three pik3r1 products, p85, p55 and p50 were splicing
been ascribed unequivocally to each of the subunits.
We are interested in the control of apoptosis in a normal physio- isoforms.6,7 However, when we examined the genomic regions
logical setting where it is induced by a switch from survival to death proximal to the N-terminal encoding exons of p55 and p50, we
signalling. A powerful model for such an apoptotic switch is identified putative promoter elements that also contained consensus
898

Cell Cycle

2005; Vol. 4 Issue 7

Stat3 Directly Regulates Expression of PI3K Regulatory Subunits

Figure 2. Proposed model for the function of PI3K regulatory subunits in the downregulation of Akt/PKB survival signaling and concomitant induction of
apoptosis in response to LIF-induced Stat3 phosphorylation.

Stat3 binding motifs. We then showed, by chromatin immunoprecipitation (ChIP) assays, that Stat3 binds directly to the p55 and
p50 promoters in vivo. These data, coupled with the selective
induction of p55/p50 expression in response to Stat3, suggests
that these transcripts are transcribed from alternate promoters, and
are not splicing isoforms as previously thought. This has major
implications for the regulation of expression of these subunits and
their possible function. In this context, it is interesting to note that
Stat3 did not induce the expression of p55 and p50 in mouse
embryonic stem (ES) cells where the function of Stat3 is to promote
self-renewal of the ES cells. We suggest that this could be due to the
inaccessibility of their promoters in ES cells due to methylation or
other chromatin modifications. Subsequent expression in response
to Stat3 will depend, therefore, on epigenetic changes in differentiating cells.
We conclude that expression of the p85, p50 and p55 transcripts from the pik3r1 gene is regulated by distinct, subunit-specific promoters and that the expression of p55 and p50 is regulated
directly by Stat3. This was confirmed by overexpression of the three
pik3r1 subunits in fully differentiated mammary epithelial cells.
Expression of either p55 or p50, but not p85, resulted in
diminished levels of pAkt/PKB showing that expression of these two
subunits either alone, or in combination, can directly diminish PI3K
activity.
We propose, therefore, the existence of a novel mechanism of
apoptosis regulation in which Stat3 mediates a molecular switch in
the subunit composition of PI3K, via activation of subunit-specific
promoters, to downregulate Akt/PKB survival signalling and suggest
that this is a mechanism by which Stat3 exerts its pro-apoptotic
function in mammary epithelial cells. This intriguing result provides
an insight into the role of constitutively active Stat3 in carcinogenesis19
and it will be interesting to determine if dysregulated Stat3 mediates
www.landesbioscience.com

its oncogenic effects by failing to induce the switch in PI3K subunit

composition.
An outstanding question concerns the mechanism by which the
subunits function to reduce PI3K signalling. It is intriguing how a
sequence of only six or 34 amino acids can have such a dramatic
effect. One possibility is that they directly replace the p85 subunit
on the receptor and this interferes with downstream signalling,
perhaps by interacting with different partners. Alternatively, the
small subunits could change the localisation of the PI3K holoenzyme
and recruit it to other cellular compartments. These putative mechanisms are illustrated in Figure 2.
Is this PI3K subunit switch unique to mammary gland where it
is essential to remove the potent PI3K survival signal to allow Stat3mediated cell death? We have shown that the presence of active
Stat3 is not sufficient to elicit a subunit switch in ES cells. Stat3 only
induces apoptosis in selected systems and this may coincide with its
ability to induce expression of these subunits. In many cell types,
Stat3 has an anti-apoptotic function. Since the p55 and p50
subunits are expressed in other tissues not undergoing apoptosis,
what is their role? In a mouse model of obesity and noninsulindependent diabetes mellitus, the ob/ob mouse, it has been shown
that the levels of the p55 and p50 subunits are elevated in liver.20
Fatty livers from ob/ob mice are characterised by hyperactivation of
Stat3,21 and it is tempting to speculate that these two events are
linked.
It may be that the first mechanism of control of subunit expression
is exerted at the level of promoter accessibility. If the promoter is in
an open configuration, Stat3 can bind and induce expression of
p55 and p50. The consequences of this will depend on cellular
context. For example, we have shown that IGFBP5 is down-stream
of Stat3 in involution.22 IGFBPs regulate bioavailability of IGF-1 by
sequestration and thus block IGF-1 induced survival signalling thru

Cell Cycle

899

Stat3 Directly Regulates Expression of PI3K Regulatory Subunits

phopsphorylation of IRS-1, Akt/PKB, and the forkhead transcription factor FKHRL1.23 Other work in our laboratory indicates that
additional signalling pathways impact on levels of pAkt/PKB to
control apoptosis and so expression of the p55 and p50 subunits
alone may not be sufficient to diminish PI3K activity in all cellular
contexts.
Are there PI3K independent effects of p55/p50 subunit
expression? Indeed, do these subunits have different or redundant
functions?
Upregulation of p55/p50 could be involved in reappearance
of adipocytes during remodelling of mammary gland since reduced
adipocyte lipid contents and fat pad masses have been reported in
the p55/p50 regulatory subunit KO models. Upregulation of the
regulatory subunits has been seen in response to different types of
injuries. Thus, it is tempting to speculate that expression of the
smaller regulatory subunits is a part of the cellular stress response.
Future work in our laboratory will be directed at identifying the
mechanism by which p55 and p50 modulate PI3K activity in
mammary gland. Interest in the function and regulation of the PI3K
regulatory subunits promises to be an interesting and fruitful area of
research which will have implications for the treatment of major
diseases such as diabetes and cancer.

19. Yu H, Jove R. The STATs of cancer--new molecular targets come of age. Nat Rev Cancer
2004; 4:97-105.
20. Kerouz NJ, Horsch D, Pons S, Kahn CR. Differential regulation of insulin receptor substrates-1 and -2 (IRS-1 and IRS-2) and phosphatidylinositol 3-kinase isoforms in liver and
muscle of the obese diabetic (ob/ob) mouse. J Clin Invest 1997; 100:3164-72.
21. Torbenson M, Yang SQ, Liu HZ, Huang J, Gage W, Diehl AM. STAT-3 Overexpression
and p21 Up-Regulation Accompany Impaired Regeneration of Fatty Livers. Am J Pathol
2002; 161:155-61.
22. Chapman RS, Lourenco PC, Tonner E, Flint DJ, Selbert S, Takeda K, Akira S, Clarke AR,
Watson CJ. Suppression of epithelial apoptosis and delayed mammary gland involution in
mice with a conditional knockout of Stat3. Genes Dev 1999; 13:2604-16.
23. Marshman E, Green KA, Flint DJ, White A, Streuli CH, Westwood M. Insulin-like growth
factor binding protein 5 and apoptosis in mammary epithelial cells. J Cell Sci 2003;
116:675-82.

References
1. Cantley LC. The phosphoinositide 3-kinase pathway. Science 2002; 296:1655-7.
2. Vanhaesebroeck B, Leevers SJ, Ahmadi K, Timms J, Katso R, Driscoll PC, Woscholski R,
Parker PJ, Waterfield MD. Synthesis and function of 3-phosphorylated inositol lipids.
Annu Rev Biochem 2001; 70:535-602.
3. Shepherd PR, Withers DJ, Siddle K. Phosphoinositide 3-kinase: The key switch mechanism in insulin signalling. Biochem J 1998; 333:471-90.
4. Fruman DA, Meyers RE, Cantley LC. Phosphoinositide kinases. Annu Rev Biochem 1998;
67:481-507.
5. Vanhaesebroeck B, Ali K, Bilancio A, Geering B, Foukas LC. Signalling by PI3K isofroms:
Insights from gene-targeted mice. Trends Biochem Sci 2005; 30:194-204.
6. Inukai K, Funaki M, Ogihara T, Katagiri H, Kanda A, Anai M, Fukushima Y, Hosaka T,
Suzuki M, Shin BC, Takata K, Yazaki Y, Kikuchi M, Oka Y, Asano T. p85alpha gene generates three isoforms of regulatory subunit for phosphatidylinositol 3-kinase (PI 3-Kinase),
p50alpha, p55alpha, and p85alpha, with different PI 3-kinase activity elevating responses
to insulin. J Biol Chem 1997; 272:7873-82.
7. Fruman DA, Cantley LC, Carpenter CL. Structural organization and alternative splicing of
the murine phosphoinositide 3-kinase p85 alpha gene. Genomics 1996; 37:113-21.
8. Otsu M, Hiles I, Gout I, Fry MJ, Ruiz-Larrea F, Panayotou G, Thompson A, Dhand R,
Hsuan J, Totty N. Characterization of two 85 kd proteins that associate with receptor tyrosine kinases, middle-T/pp60c-src complexes, and PI3-kinase. Cell 1991; 65:91-104.
9. Pons S, Asano T, Glasheen E, Miralpeix M, Zhang Y, Fisher TL, Myers Jr MG, Sun XJ,
White MF. The structure and function of p55PIK reveal a new regulatory subunit for phosphatidylinositol 3-kinase. Mol Cell Biol 1995; 15:4453-65.
10. Antonetti DA, Algenstaedt P, Kahn CR. Insulin receptor substrate 1 binds two novel splice
variants of the regulatory subunit of phosphatidylinositol 3-kinase in muscle and brain.
Mol Cell Biol 1996; 16:2195-203.
11. Song G, Oujang G, Bao S. The activation of Akt/PKB signaling pathway and cell survival.
J Cell Mol Med 2005; 9:59-71.
12. Vivanco I, Sawyers CL. The phosphatidylinositol 3-kinase-Akt pathway in human cancer.
Nat Rev Cancer 2002; 2:489-501.
13. Ueki K, Fruman DA, Brachmann SM, Tseng YH, Cantley LC, Kahn CR. Molecular balance between the regulatory and catalytic subunits of phosphoinositide 3-kinase regulates
cell signaling and survival. Mol Cell Biol 2002; 22:965-77.
14. Chen D, Mauvais-Jarvis F, Bluher M, Fisher SJ, Jozsi A, Goodyear LJ, Ueki K, Kahn CR.
p50alpha/p55alpha phosphoinositide 3-kinase knockout mice exhibit enhanced insulin
sensitivity. Mol Cell Biol 2004; 24:320-9.
15. Brachmann SM, Ueki K, Engelman JA, Kahn RC, Cantley LC. Phosphoinositide 3-kinase
catalytic subunit deletion and regulatory subunit deletion have opposite effects on insulin
sensitivity in mice. Mol Cell Biol 2005; 25:1596-1607.
16. Li M, Liu X, Robinson G, Bar-Peled U, Wagner K-U, Young WS, Hennighausen L, Furth
PA. Mammary-derived signals activate programmed cell death during the first stage of
mammary gland involution. Proc Natl Acad Sci USA 1997; 94:3425-30.
17. Schwertfeger K L, Richert MM, Anderson SM. Mammary gland involution is delayed by
activated Akt in transgenic mice. Mol Endocrinol 2001; 15: 867-881.
18. Abell K, Bilancio A, Clarkson RWE, Tiffen PG, Altaparmakov AI, Burdon TG, Asano T,
Vanhaesebroeck B, Watson CJ. Stat3-induced apoptosis requires a molecular switch in
PI(3)K subunit composition. Nat Cell Biol 2005; 7:392-97.

900

Cell Cycle

2005; Vol. 4 Issue 7