p27KIP1 and PTEN cooperate in myeloproliferative neoplasm tumor suppression in mice

PTEN acts as a phosphatase for PIP3 and negatively regulates the PI3K/AKT pathway, and p27KIP1 is a cyclin-dependent kinase inhibitor that regulates the G1 to S-phase transition by binding to and regulating the activity of cyclin-dependent kinases. Genetic alterations of PTEN or CDKN1B (p27KIP1) are common in hematological malignancies. To better understand how mutations in these two genes might cooperate in leukemogenesis, we inactivated both genes in the hematological compartment in mice. Here, we show that the combined inactivation of Pten and Cdkn1b results in a more severe myeloproliferative neoplasm phenotype associated with lower hemoglobin, enlarged spleen and liver, and shorter lifespan compared to inactivation of Pten alone. More severe anemia and increased myeloid infiltration and destruction of the spleen contributed to the earlier death of these mice, and elevated p-AKT, cyclin D1, and cyclin D3 might contribute to the development of this phenotype. In conclusion, PTEN and p27KIP1 cooperate in tumor suppression in the hematological compartment.


Background
PTEN (phosphatase and tension homolog deleted on chromosome 10) is a tumor suppressor gene located on chromosome 10q23 and is one of the most commonly mutated or deleted genes in human cancers, including acute lymphoblastic leukemia, juvenile myelomonocytic leukemia, and non-Hodgkin's lymphoma [1,2]. PTEN acts as a phosphatase for phosphatidylinositol-3,4,5trisphosphate (PIP3) and negatively regulates the phosphatidylinositol 3-kinase (PI3K)/AKT pathway [3]. The CDKN1B gene encodes p27 KIP1 , which belongs to the Cip/Kip family of cyclin-dependent kinase inhibitors. p27 KIP1 is a key regulator of the G1 to S-phase transition by inhibiting cyclinD1/CDK4 and cyclinE/CDK2 complexes [4]. Deletions and other cytogenetic aberrations involving CDKN1B have been reported in a variety of leukemias [5][6][7]. In addition, CDKN1B expression can be a useful prognostic molecular marker for acute myeloid leukemia, where low CDKN1B expression is associated with high proliferation and, therefore, with a favorable response to chemotherapy [6]. Inactivation of the tumorsuppressor gene PTEN and lack of CDKN1B expression have been detected in some kinds of cancer, including most advanced prostate cancers and lymphomas [8,9]. It has been shown that the combined loss of PTEN and p27 KIP1 is associated with tumor cell proliferation and increased risk of recurrent disease in localized prostate cancer [10]. Loss of PTEN expression is more frequent in anaplastic large-cell lymphoma, which strongly correlates with the loss of CDKN1B expression [9].
Targeted disruption of the murine Cdkn1b gene causes a gene dose-dependent increase in animal size without other gross morphologic abnormalities [11], and deletion of Pten in the hematopoietic compartment in mice promotes excessive proliferation of leukemogenic stem cells resulting in the development of myeloproliferative

Open Access
Experimental Hematology & Oncology *Correspondence: meng.liu@gu.se neoplasm (MPN) followed by acute leukemia [12]. In mice, concomitant inactivation of Pten and Cdkn1b accelerates spontaneous neoplastic transformation of prostate cancer [8]. In order to better understand the relation and clinical relevance of these two genes in the pathogenesis of hematological malignancies, we used Cre recombinase to simultaneously inactivate Pten and Cdkn1b in the hematopoietic compartment.

Results and discussion
To determine the impact of combined deficiency of PTEN and p27 KIP1 in the hematopoietic compartment, we injected pI-pC into PCM, PM, CM and Ctrl mice. Consistent with previous studies [13], all PM mice died from MPN by 98 days after pI-pC injections (median survival 62 days), whereas CM and Ctrl mice lived much longer and no MPN phenotype was observed in CM mice. However, the maximum survival of PCM mice was only 30 days (median 24 days; p < 0.001 versus PM; Fig. 1a). Two weeks after pI-pC injections, white blood cell counts were 20.8 × 10 9 cells/L in PCM mice compared with mean counts of 18.3 × 10 9 , 13.9 × 10 9 and 13.6 × 10 9 cells/L for PM, CM and Ctrl mice, respectively (Fig. 1b). However, no morphological changes and no increase in the amounts of immature cells, including myeloblasts, could be detected in the blood and bone marrow in PCM mice compared with the other three groups (Fig. 1c, e). More severe anemia and more architectural disruption of the spleen were observed in PCM mice (Fig. 1d, e). Spleen and liver weights in PCM mice increased by 2.3-5.6 and 1.2-2.4-fold, respectively, compared with PM, CM and Ctrl mice (Fig. 2a, b). Fluorescence-activated cell sorting analysis showed an increased proportion of CD11b+/Gr1+ and LSK [Lineage-negative (lin−), Sca-1+, c-Kit+] cells in the spleen of PCM mice compared to PM, CM and Ctrl mice (p < 0.05; Fig. 2c). Splenocytes from PCM mice produced more colonies compared with the other three groups (Fig. 2d). In bone marrow, there were no differences in the percentage of LSK cells (Fig. 2e). No increased colony formation in PCM mice was observed compared to PM mice when replated, and both groups had more colonies than the Ctrl mice when replated (Fig. 2f ). Taken together, the phenotype in PCM mice is severe MPN rather than acute leukemia based on the criteria for classification of hematopoietic neoplasms in mice [14]. More severe anemia and increased myeloid infiltration and destruction of the spleen likely contributed to the earlier death of PCM mice compared with PM mice.
We performed Western blot analysis to determine the knock-out efficiency and the consequences of inactivating PTEN and p27 KIP1 on downstream molecules. Deficiency of PTEN or p27 KIP1 was observed in the respective knock-out mice (Fig. 2g). It has been shown that PTEN activity leads to the induction of p27 KIP1 , which in turn can negatively regulate the transition through the cell cycle [15]. However, the association between PTEN and p27 KIP1 might be different in different kinds of tissues. A lack of convincing correlation between PTEN and p27 KIP1 has been reported for ovarian carcinomas, indicating the possible existence of p27 KIP1 -independent pathways downstream of PTEN [16]. In our study, we found that expression of p27 KIP1 was reduced in the splenocytes of PM mice. PCM mice had higher phosphorylated AKT compared to PM, CM, and Ctrl mice (Fig. 2g), and cyclin D1 and cyclin D3 expression levels were elevated in PCM and PM mice. However, the levels of phosphorylated ERK1/2 were similar in all groups of mice. Previous studies showed the synergistic activity of PI3K/mTOR and JAK2 signaling pathway in the myeloproliferative neoplasms [17,18], therefore it will be interesting to study the JAK2 activity in PTEN and p27 KIP1 knockout mice model.
In conclusion, our results show that PTEN deficiency can promote tumor progression by a decrease in p27 KIP1 levels in the hematological compartment and that PTEN and p27 KIP1 have a cooperative role in leukemia suppression. In addition, our results show that elevated phosphorylated AKT, cyclin D1, and cyclin D3 might play an important role in the progression of the severe MPN phenotype.

Animal procedures
Mice with conditional Pten fl/fl alleles (designated P) with a mixed genomic background of 129S4/SvJae and C57BL/6J were bred with Cdkn1b fl/fl mice (designated C) to generate PC mice. The mice were housed under controlled environmental conditions with free access to water and food. Illumination was on between 0600 and 1800 h. All mice were monitored daily. Groups of 4-week-old mice were injected with 400 µg polyinosinic-polycytidylic acid (pI-pC; Sigma, St Louis, MO). Blood was taken weekly and analyzed with a hematology analyzer KX-21 (Sysmex Europe, Norderstedt, Hamburg, Germany). Three weeks after injection, groups of mice were sacrificed and their tissues were harvested for further analysis. Mice were euthanized by cervical dislocation after carbon dioxide inhalation. In addition, groups of mice were kept for a survival study. If mice had ruffled fur and become listless or lost more than 10 % of their body weight, they were euthanized. All experimental protocols were approved by the regional ethical committee of the University of Gothenburg, Sweden.