Appropriate functioning of an organ is dependent upon normal growth and organization of its cell constituents, the dysregulation of which may lead to a disease state. One signaling molecule at the nexus of this regulation is Akt, a membrane-bound oncogenic kinase that affects cellular activity by phosphorylating other regulatory molecules. However, Akt can only be activated once it is properly localized to the cell membrane, a process dependent on its ability to bind the phospholipid phosphatidylinositol-3,4,5-triphosphate (PIP3; Fig. 1a). PIP3 is generated from PIP2 by the action of phosphoinositide 3-kinase (PI3K), whereas it is converted back to PIP2 by phosphatase and tensin homolog deleted on chromosome 10 (PTEN1, also known as mutated in multiple advanced cancers-1 (MMAC1)2), which was discovered more than a decade ago as a tumor suppressor. Therefore, in addition to its role in cancer, PTEN also has a pivotal role in regulating normal cell growth. Indeed, recently, some roles of PTEN in various normal physiological processes have been revealed.

Figure 1: PTEN function and phenotypes of Pten deletion.
figure 1

(a) Schematic diagram illustrating the role of PI3K and PTEN in regulating Akt activation and cell growth. Upon receptor tyrosine kinase (RTK) activation at the cell surface, PI3K in turn is activated by downstream kinases. PI3K then phosphorylates (+P) PIP2 to generate PIP3 at the cell membrane. PIP3 then binds cytosolic Akt, anchoring it to the cell membrane where it can then be activated by kinases to affect cellular function. (b) Pten deletion in various reproductive tissues (indicated in bold) results in several pathophysiological outcomes (indicated in plain text), including lethality if deleted in the whole embryo (left). In addition to Pten deletion, ovarian surface epithelium and granulosa cell tumors require the activation of β-catenin–Wnt signaling. PF, primordial follicle (the primary oocyte is surrounded by a layer of follicular epithelium); GF, growing follicle (multiple layers of granulosa cells); AF, antral follicle (ready for ovulation with a cavity inside).

Full size image

In the ovary, bidirectional communications between germ cells (oocytes) and somatic cells (granulosa cells) are crucial to oocyte maturation, follicle growth and ovulation. Oocytes are derived from primodial germ cells, oogonia, which expand through mitosis in the fetal ovary. The germ cells become oocytes once they enter meiosis at birth. They become surrounded by a layer of epithelial cells (granulosa) and are now called primordial follicles. At puberty, primordial follicles resume growth and acquire many layers of granulosa cells to reach the ovulatory stage at each cycle (Fig. 1b, right). The nature and relative contribution of each cell type governing these processes are not clearly understood.

Recent advances in cell-specific deletion of genes in a given tissue with the Cre-LoxP system allow researchers to sort out the unique function of a given gene for each cell type it is active in. For example, a recent study has shown that deletion of Pten specifically in oocytes promptly activates follicular development in mice very early in life3. This early follicular activation resulting from heightened PI3K-Akt signaling in oocytes lacking PTEN suggests that signaling from oocytes is a contributing factor in regulating follicular activities. Interestingly, another group found that PI3K-Akt–dependent hyperphosphorylation of the transcription factor Foxo3 also induces premature oocyte activation. More intriguingly, oocyte-specific deletion of Foxo3 shows similar phenotypes to those of Pten deletion, implicating that Foxo3 lies downstream of PTEN4. This irreversible premature oocyte activation leads to rapid depletion of follicle reserve, leading to a demise of ovarian function similar to premature ovarian failure in humans. It would be interesting to know whether such conditions in humans result from mutations or single nucleotide polymorphisms in this pathway.

Although these studies highlight roles of oocyte-specific PTEN in regulating follicular function, another recent study shows that granulosa cell–specific deletion of Pten increases ovulation rate and litter size and prolongs corpora luteal lifespan. More interestingly, these mice had increased fertility over the course of a six-month breeding period5.

These studies illustrate that although Pten deletion activates the PI3K-Akt pathway in both cell types, the resulting outcome is different. PTEN-deficiency in oocytes leads to premature activation of the entire pool of primodial follicles and drastically reduces the ovarian follicle reserve, thereby rapidly advancing infertility. In contrast, granulosa cells lacking PTEN have enhanced fertility5. Of note, whereas the follicular atresia that results from enhanced apoptosis was not prevented by oocytes lacking PTEN3, PTEN deficiency in granulosa cells attenuated apoptosis with increased proliferation5. This would suggest that functional manifestation of PTEN is cell-context dependent.

These differential phenotypes of PTEN deficiency in two different cell types raise interesting questions. How do PTEN-deficient activated oocytes initiate follicular growth with granulosa cell proliferation? Likewise, how do Pten-depleted granulosa cells synchronize oocyte maturation and growth to enhance fertility? This interdependence of oocytes and granulosa cells to each other involving PTEN warrants further investigation, especially the downstream signaling mechanism(s), as many of the targets of Akt action have yet to be identified in these cell types.

This point brings up another key question. What is the mechanism for the huge decline in germ cell numbers in women from about 7 million in the fetal ovary at midpregnancy to approximately 2 million at birth, followed by a further reduction to 0.3 million at seven years of age? Notably, these declines in germ cell numbers are due to accelerated apoptosis. Is it possible that PTEN activity remains transiently high during these early stages, conferring increased apoptosis, or is PTEN activity transiently silenced to allow early germ cell activation, causing their demise? These questions need to be addressed to better understand the function of PTEN in changing ovarian events, such as oogenesis, oocyte maturation, follicular growth and ovulation. One way to achieve some insights will be to examine the expression and/or activity of PTEN at crucial times of ovarian development during fetal, neonatal and adult life.

The differential behavior of female reproductive tissues to PTEN deficiency is another intriguing area of investigation. The inability of the enhanced PI3K-Akt signaling arising from PTEN deficiency in granulosa cells to produce granulosa cell tumors, or to block their differentiation to luteal cells after ovulation, is noteworthy in the context of Pten deletion in other reproductive organs (Fig. 1b). For example, uterine deletion of Pten alone unfailingly produces endometrial cancer quite early in life6. Is it possible that different cell types show a differential sensitivity to PTEN deficiency–induced tumors? Does PI3K-Akt activation after Pten deletion require the participation of other pathways to trigger tumorigenesis in specific cell types but not in others? Does the nature of the microenvironment created by neighboring cells govern the function of PTEN? Indeed, unlike endometrial cancer, granulosa cell and ovarian surface epithelial tumors are efficiently produced only if a PTEN deficiency–induced activation of the PI3K-Akt pathway is superimposed with forced activation of β-catenin–Wnt signaling7,8.

Although cell- and tissue-specific deletion of genes has illuminated many intricate physiological processes, studies using the Cre-LoxP system, amenable to an inducible tissue-specific deletion, are required to gather a comprehensive understanding of PTEN function for specific events in reproductive tissue and other organs. For example, although Pten is expressed in the uterus at the time of implantation (T.D. and S.K.D., unpublished data), systemic deletion of Pten leads to embryonic lethality9, and uterine deletion of Pten results in endometrial cancer6, precluding the study of its function in normal embryo-uterine interactions during early pregnancy. Thus, the use of a fully inducible system for deliberate regulation of gene function may reveal PTEN's unique roles in various reproductive functions during the life of an animal.