Limits...
A critical function of Mad2l2 in primordial germ cell development of mice.

Pirouz M, Pilarski S, Kessel M - PLoS Genet. (2013)

Bottom Line: By the analysis of transfected fibroblasts we found that the interaction of Mad2l2 with the histone methyltransferases G9a and GLP lead to a downregulation of H3K9me2.Together, these results demonstrate the potential of Mad2l2 in the regulation of both cell cycle and the epigenetic status.The function of Mad2l2 is essential in PGCs, and thus of high relevance for fertility.

View Article: PubMed Central - PubMed

Affiliation: Department of Molecular Cell Biology, Max Planck Institute for Biophysical Chemistry, Göttingen, Germany. mehdi.pirouz@mpibpc.mpg.de

ABSTRACT
The development of primordial germ cells (PGCs) involves several waves of epigenetic reprogramming. A major step is following specification and involves the transition from the stably suppressive histone modification H3K9me2 to the more flexible, still repressive H3K27me3, while PGCs are arrested in G2 phase of their cycle. The significance and underlying molecular mechanism of this transition were so far unknown. Here, we generated mutant mice for the Mad2l2 (Mad2B, Rev7) gene product, and found that they are infertile in both males and females. We demonstrated that Mad2l2 is essential for PGC, but not somatic development. PGCs were specified normally in Mad2l2(-/-) embryos, but became eliminated by apoptosis during the subsequent phase of epigenetic reprogramming. A majority of knockout PGCs failed to arrest in the G2 phase, and did not switch from a H3K9me2 to a H3K27me3 configuration. By the analysis of transfected fibroblasts we found that the interaction of Mad2l2 with the histone methyltransferases G9a and GLP lead to a downregulation of H3K9me2. The inhibitory binding of Mad2l2 to Cyclin dependent kinase 1 (Cdk1) could arrest the cell cycle in the G2 phase, and also allowed another histone methyltransferase, Ezh2, to upregulate H3K27me3. Together, these results demonstrate the potential of Mad2l2 in the regulation of both cell cycle and the epigenetic status. The function of Mad2l2 is essential in PGCs, and thus of high relevance for fertility.

Show MeSH

Related in: MedlinePlus

Analysis of Mad2l2 function in fibroblasts.(A) Protein extracts from HA-Mad2l2 transfected NIH3T3 cells were co-immunoprecipitated (IP) by antibodies against G9a, GLP, or IgG (as negative control). Immunoblotting (IB) was performed on 20% of gel-separated immunoprecipitates (upper blot), or 1% input (lower blot) by using anti-HA antibody. (B) Reciprocally, the same protein extract was co-immunoprecipitated (IP) by antibodies against the HA-tag, or IgG (as negative control). Immunoblotting (IB) was performed on 20% of the immunoprecipitates (upper blots), or 1% input (lower blot) by using anti-G9a, anti-GLP, or anti-HA antibodies. (C) Immunocytochemistry detects a downregulation of G9a in GFP-Mad2l2 over-expressing NIH3T3 cells (arrowheads) in comparison to untransfected cells (arrows). The lower panel shows a western blot analysis of H3K9me2 levels in GFP-Mad2l2 over-expressing, FACS-sorted NIH3T3 cells versus non-transfected NIH3T3 cells. Note the efficient downregulation of H3K9me2 by Mad2l2 overexpression. (D) A representative western blot analysis of GLP, G9a, H3K9me2 and Histone H3 levels in wild type versus knockout MEFs (left panel) and quantification of the western blot bands normalized to tubulin or actin signals (right panel). (E) The effect of Mad2l2 on cell cycle parameters. HA-Mad2l2 transfected NIH3T3 fibroblasts never expressed pH 3 (0%, 0/70; e.g. cell number #1, upper panel), and always displayed Cyclin B1 in the cytoplasm (100%, 40/40; #3, middle panel). Some of the non-transfected cells entered the mitotic prophase (#2, #4) or anaphase (#5), and displayed nuclear pH 3 (#2) or nuclear Cyclin B1 (#4, #5). HA-Mad2l2 expressing cells displayed two unseparated centrosomes detectable by γTubulin (100%, 7/7; #6, lower panel). Scale bars, 20 µm (upper and middle panels), 10 µm (lower panel). (F–G) Reciprocal co-immunoprecipitation of HA-Mad2l2 and Cdk1 from HA-Mad2l2 over-expressing protein extract, using either anti-HA or anti-Cdk1 antibodies. 50% of the immunoprecipitates, or 1.5% of total cell lysate (input) were loaded. (H) Cdk1 antibody co-immunoprecipitated HA-Mad2l2 from transfected NIH3T3 cells, but not antibodies against Cyclin B1, pCdk1, and rabbit IgG. 50% of the immunoprecipitates, or 1.5% of total cell lysate (input) were loaded. (I) Recombinant GST-Mad2l2 attenuates the kinase activity of Cdk1-Cyclin B1 (2.5 mUnits) in vitro, while GST alone is not effective. Mean values of three independent experiments with duplicate measurements, and standard deviations are indicated. (J) Immunocytochemistry demonstrates the upregulation of H3K27me3 in GFP-Mad2l2 over-expressing NIH3T3 cells (arrowheads). (K) Immunocytochemistry analysis shows suppression of phosphorylation on Ezh2 at T487 (white arrowhead) in comparison to surrounding, untransfected interphase cells. The highest level of pEzh2 was detected in mitotic cell with high level of Cdk1 activity (arrow). The right panel shows a western blot analysis of pEzh2 and Ezh2 levels in FACS-sorted, GFP-Mad2l2 over-expressing NIH3T3 cells and untransfected controls. (L) A representative western blot analysis of pEzh2, Ezh2, H3K27me3, and actin levels in wild type versus knockout MEFs (left panel) and quantification of the western blot bands normalized to actin signal (right panel). Note the inhibition of Ezh2 by phosphorylation, and the concomitant decrease of H3K27me3 in the absence of Mad2l2.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3757036&req=5

pgen-1003712-g008: Analysis of Mad2l2 function in fibroblasts.(A) Protein extracts from HA-Mad2l2 transfected NIH3T3 cells were co-immunoprecipitated (IP) by antibodies against G9a, GLP, or IgG (as negative control). Immunoblotting (IB) was performed on 20% of gel-separated immunoprecipitates (upper blot), or 1% input (lower blot) by using anti-HA antibody. (B) Reciprocally, the same protein extract was co-immunoprecipitated (IP) by antibodies against the HA-tag, or IgG (as negative control). Immunoblotting (IB) was performed on 20% of the immunoprecipitates (upper blots), or 1% input (lower blot) by using anti-G9a, anti-GLP, or anti-HA antibodies. (C) Immunocytochemistry detects a downregulation of G9a in GFP-Mad2l2 over-expressing NIH3T3 cells (arrowheads) in comparison to untransfected cells (arrows). The lower panel shows a western blot analysis of H3K9me2 levels in GFP-Mad2l2 over-expressing, FACS-sorted NIH3T3 cells versus non-transfected NIH3T3 cells. Note the efficient downregulation of H3K9me2 by Mad2l2 overexpression. (D) A representative western blot analysis of GLP, G9a, H3K9me2 and Histone H3 levels in wild type versus knockout MEFs (left panel) and quantification of the western blot bands normalized to tubulin or actin signals (right panel). (E) The effect of Mad2l2 on cell cycle parameters. HA-Mad2l2 transfected NIH3T3 fibroblasts never expressed pH 3 (0%, 0/70; e.g. cell number #1, upper panel), and always displayed Cyclin B1 in the cytoplasm (100%, 40/40; #3, middle panel). Some of the non-transfected cells entered the mitotic prophase (#2, #4) or anaphase (#5), and displayed nuclear pH 3 (#2) or nuclear Cyclin B1 (#4, #5). HA-Mad2l2 expressing cells displayed two unseparated centrosomes detectable by γTubulin (100%, 7/7; #6, lower panel). Scale bars, 20 µm (upper and middle panels), 10 µm (lower panel). (F–G) Reciprocal co-immunoprecipitation of HA-Mad2l2 and Cdk1 from HA-Mad2l2 over-expressing protein extract, using either anti-HA or anti-Cdk1 antibodies. 50% of the immunoprecipitates, or 1.5% of total cell lysate (input) were loaded. (H) Cdk1 antibody co-immunoprecipitated HA-Mad2l2 from transfected NIH3T3 cells, but not antibodies against Cyclin B1, pCdk1, and rabbit IgG. 50% of the immunoprecipitates, or 1.5% of total cell lysate (input) were loaded. (I) Recombinant GST-Mad2l2 attenuates the kinase activity of Cdk1-Cyclin B1 (2.5 mUnits) in vitro, while GST alone is not effective. Mean values of three independent experiments with duplicate measurements, and standard deviations are indicated. (J) Immunocytochemistry demonstrates the upregulation of H3K27me3 in GFP-Mad2l2 over-expressing NIH3T3 cells (arrowheads). (K) Immunocytochemistry analysis shows suppression of phosphorylation on Ezh2 at T487 (white arrowhead) in comparison to surrounding, untransfected interphase cells. The highest level of pEzh2 was detected in mitotic cell with high level of Cdk1 activity (arrow). The right panel shows a western blot analysis of pEzh2 and Ezh2 levels in FACS-sorted, GFP-Mad2l2 over-expressing NIH3T3 cells and untransfected controls. (L) A representative western blot analysis of pEzh2, Ezh2, H3K27me3, and actin levels in wild type versus knockout MEFs (left panel) and quantification of the western blot bands normalized to actin signal (right panel). Note the inhibition of Ezh2 by phosphorylation, and the concomitant decrease of H3K27me3 in the absence of Mad2l2.

Mentions: The number of early PGCs is too small for biochemical and transfection approaches. Therefore, we performed a set of experiments in fibroblasts with the intention to provide evidence for a function of Mad2l2 in epigenetic and cell cycle regulation. Since the Mad2l2 protein contains a protein-binding HORMA domain Co-immunoprecipitation was applied to identify Mad2l2 interacting partners related to histone modifications (See Text S1). First, to explore a physical interaction between Mad2l2 and G9a or GLP, NIH3T3 fibroblasts were transfected with a plasmid encoding HA-Mad2l2 (See Text S1). Co-immunoprecipitation of NIH3T3 protein extract with anti-G9a, anti-GLP or anti-HA antibodies demonstrated that Mad2l2 interacts with both methyltransferases (Figure 8A, B). Transfection of NIH3T3 cells with a vector encoding a GFP-fused Mad2l2 protein showed that G9a mRNA levels were specifically downregulated in the presence of GFP-Mad2l2 (Figures S5A). G9a protein levels were always low in Mad2l2-GFP transfected cells, while untransfected cells had either high or low levels (Figures 8C). Correspondingly, the level of H3K9me2 became completely suppressed in transfected cells (Figure 8C), while levels of H3K4me2, an unrelated histone modification, remained unaffected (Figure S5B). For the analysis of loss-of-function conditions Mad2l2 deficient MEFs were prepared, and elevated levels of G9a and H3K9me2 were observed (Figure 8D). Together, these findings indicate a negative correlation between the presence of Mad2l2 and the expression and activity of the methyltransferase G9a.


A critical function of Mad2l2 in primordial germ cell development of mice.

Pirouz M, Pilarski S, Kessel M - PLoS Genet. (2013)

Analysis of Mad2l2 function in fibroblasts.(A) Protein extracts from HA-Mad2l2 transfected NIH3T3 cells were co-immunoprecipitated (IP) by antibodies against G9a, GLP, or IgG (as negative control). Immunoblotting (IB) was performed on 20% of gel-separated immunoprecipitates (upper blot), or 1% input (lower blot) by using anti-HA antibody. (B) Reciprocally, the same protein extract was co-immunoprecipitated (IP) by antibodies against the HA-tag, or IgG (as negative control). Immunoblotting (IB) was performed on 20% of the immunoprecipitates (upper blots), or 1% input (lower blot) by using anti-G9a, anti-GLP, or anti-HA antibodies. (C) Immunocytochemistry detects a downregulation of G9a in GFP-Mad2l2 over-expressing NIH3T3 cells (arrowheads) in comparison to untransfected cells (arrows). The lower panel shows a western blot analysis of H3K9me2 levels in GFP-Mad2l2 over-expressing, FACS-sorted NIH3T3 cells versus non-transfected NIH3T3 cells. Note the efficient downregulation of H3K9me2 by Mad2l2 overexpression. (D) A representative western blot analysis of GLP, G9a, H3K9me2 and Histone H3 levels in wild type versus knockout MEFs (left panel) and quantification of the western blot bands normalized to tubulin or actin signals (right panel). (E) The effect of Mad2l2 on cell cycle parameters. HA-Mad2l2 transfected NIH3T3 fibroblasts never expressed pH 3 (0%, 0/70; e.g. cell number #1, upper panel), and always displayed Cyclin B1 in the cytoplasm (100%, 40/40; #3, middle panel). Some of the non-transfected cells entered the mitotic prophase (#2, #4) or anaphase (#5), and displayed nuclear pH 3 (#2) or nuclear Cyclin B1 (#4, #5). HA-Mad2l2 expressing cells displayed two unseparated centrosomes detectable by γTubulin (100%, 7/7; #6, lower panel). Scale bars, 20 µm (upper and middle panels), 10 µm (lower panel). (F–G) Reciprocal co-immunoprecipitation of HA-Mad2l2 and Cdk1 from HA-Mad2l2 over-expressing protein extract, using either anti-HA or anti-Cdk1 antibodies. 50% of the immunoprecipitates, or 1.5% of total cell lysate (input) were loaded. (H) Cdk1 antibody co-immunoprecipitated HA-Mad2l2 from transfected NIH3T3 cells, but not antibodies against Cyclin B1, pCdk1, and rabbit IgG. 50% of the immunoprecipitates, or 1.5% of total cell lysate (input) were loaded. (I) Recombinant GST-Mad2l2 attenuates the kinase activity of Cdk1-Cyclin B1 (2.5 mUnits) in vitro, while GST alone is not effective. Mean values of three independent experiments with duplicate measurements, and standard deviations are indicated. (J) Immunocytochemistry demonstrates the upregulation of H3K27me3 in GFP-Mad2l2 over-expressing NIH3T3 cells (arrowheads). (K) Immunocytochemistry analysis shows suppression of phosphorylation on Ezh2 at T487 (white arrowhead) in comparison to surrounding, untransfected interphase cells. The highest level of pEzh2 was detected in mitotic cell with high level of Cdk1 activity (arrow). The right panel shows a western blot analysis of pEzh2 and Ezh2 levels in FACS-sorted, GFP-Mad2l2 over-expressing NIH3T3 cells and untransfected controls. (L) A representative western blot analysis of pEzh2, Ezh2, H3K27me3, and actin levels in wild type versus knockout MEFs (left panel) and quantification of the western blot bands normalized to actin signal (right panel). Note the inhibition of Ezh2 by phosphorylation, and the concomitant decrease of H3K27me3 in the absence of Mad2l2.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3757036&req=5

pgen-1003712-g008: Analysis of Mad2l2 function in fibroblasts.(A) Protein extracts from HA-Mad2l2 transfected NIH3T3 cells were co-immunoprecipitated (IP) by antibodies against G9a, GLP, or IgG (as negative control). Immunoblotting (IB) was performed on 20% of gel-separated immunoprecipitates (upper blot), or 1% input (lower blot) by using anti-HA antibody. (B) Reciprocally, the same protein extract was co-immunoprecipitated (IP) by antibodies against the HA-tag, or IgG (as negative control). Immunoblotting (IB) was performed on 20% of the immunoprecipitates (upper blots), or 1% input (lower blot) by using anti-G9a, anti-GLP, or anti-HA antibodies. (C) Immunocytochemistry detects a downregulation of G9a in GFP-Mad2l2 over-expressing NIH3T3 cells (arrowheads) in comparison to untransfected cells (arrows). The lower panel shows a western blot analysis of H3K9me2 levels in GFP-Mad2l2 over-expressing, FACS-sorted NIH3T3 cells versus non-transfected NIH3T3 cells. Note the efficient downregulation of H3K9me2 by Mad2l2 overexpression. (D) A representative western blot analysis of GLP, G9a, H3K9me2 and Histone H3 levels in wild type versus knockout MEFs (left panel) and quantification of the western blot bands normalized to tubulin or actin signals (right panel). (E) The effect of Mad2l2 on cell cycle parameters. HA-Mad2l2 transfected NIH3T3 fibroblasts never expressed pH 3 (0%, 0/70; e.g. cell number #1, upper panel), and always displayed Cyclin B1 in the cytoplasm (100%, 40/40; #3, middle panel). Some of the non-transfected cells entered the mitotic prophase (#2, #4) or anaphase (#5), and displayed nuclear pH 3 (#2) or nuclear Cyclin B1 (#4, #5). HA-Mad2l2 expressing cells displayed two unseparated centrosomes detectable by γTubulin (100%, 7/7; #6, lower panel). Scale bars, 20 µm (upper and middle panels), 10 µm (lower panel). (F–G) Reciprocal co-immunoprecipitation of HA-Mad2l2 and Cdk1 from HA-Mad2l2 over-expressing protein extract, using either anti-HA or anti-Cdk1 antibodies. 50% of the immunoprecipitates, or 1.5% of total cell lysate (input) were loaded. (H) Cdk1 antibody co-immunoprecipitated HA-Mad2l2 from transfected NIH3T3 cells, but not antibodies against Cyclin B1, pCdk1, and rabbit IgG. 50% of the immunoprecipitates, or 1.5% of total cell lysate (input) were loaded. (I) Recombinant GST-Mad2l2 attenuates the kinase activity of Cdk1-Cyclin B1 (2.5 mUnits) in vitro, while GST alone is not effective. Mean values of three independent experiments with duplicate measurements, and standard deviations are indicated. (J) Immunocytochemistry demonstrates the upregulation of H3K27me3 in GFP-Mad2l2 over-expressing NIH3T3 cells (arrowheads). (K) Immunocytochemistry analysis shows suppression of phosphorylation on Ezh2 at T487 (white arrowhead) in comparison to surrounding, untransfected interphase cells. The highest level of pEzh2 was detected in mitotic cell with high level of Cdk1 activity (arrow). The right panel shows a western blot analysis of pEzh2 and Ezh2 levels in FACS-sorted, GFP-Mad2l2 over-expressing NIH3T3 cells and untransfected controls. (L) A representative western blot analysis of pEzh2, Ezh2, H3K27me3, and actin levels in wild type versus knockout MEFs (left panel) and quantification of the western blot bands normalized to actin signal (right panel). Note the inhibition of Ezh2 by phosphorylation, and the concomitant decrease of H3K27me3 in the absence of Mad2l2.
Mentions: The number of early PGCs is too small for biochemical and transfection approaches. Therefore, we performed a set of experiments in fibroblasts with the intention to provide evidence for a function of Mad2l2 in epigenetic and cell cycle regulation. Since the Mad2l2 protein contains a protein-binding HORMA domain Co-immunoprecipitation was applied to identify Mad2l2 interacting partners related to histone modifications (See Text S1). First, to explore a physical interaction between Mad2l2 and G9a or GLP, NIH3T3 fibroblasts were transfected with a plasmid encoding HA-Mad2l2 (See Text S1). Co-immunoprecipitation of NIH3T3 protein extract with anti-G9a, anti-GLP or anti-HA antibodies demonstrated that Mad2l2 interacts with both methyltransferases (Figure 8A, B). Transfection of NIH3T3 cells with a vector encoding a GFP-fused Mad2l2 protein showed that G9a mRNA levels were specifically downregulated in the presence of GFP-Mad2l2 (Figures S5A). G9a protein levels were always low in Mad2l2-GFP transfected cells, while untransfected cells had either high or low levels (Figures 8C). Correspondingly, the level of H3K9me2 became completely suppressed in transfected cells (Figure 8C), while levels of H3K4me2, an unrelated histone modification, remained unaffected (Figure S5B). For the analysis of loss-of-function conditions Mad2l2 deficient MEFs were prepared, and elevated levels of G9a and H3K9me2 were observed (Figure 8D). Together, these findings indicate a negative correlation between the presence of Mad2l2 and the expression and activity of the methyltransferase G9a.

Bottom Line: By the analysis of transfected fibroblasts we found that the interaction of Mad2l2 with the histone methyltransferases G9a and GLP lead to a downregulation of H3K9me2.Together, these results demonstrate the potential of Mad2l2 in the regulation of both cell cycle and the epigenetic status.The function of Mad2l2 is essential in PGCs, and thus of high relevance for fertility.

View Article: PubMed Central - PubMed

Affiliation: Department of Molecular Cell Biology, Max Planck Institute for Biophysical Chemistry, Göttingen, Germany. mehdi.pirouz@mpibpc.mpg.de

ABSTRACT
The development of primordial germ cells (PGCs) involves several waves of epigenetic reprogramming. A major step is following specification and involves the transition from the stably suppressive histone modification H3K9me2 to the more flexible, still repressive H3K27me3, while PGCs are arrested in G2 phase of their cycle. The significance and underlying molecular mechanism of this transition were so far unknown. Here, we generated mutant mice for the Mad2l2 (Mad2B, Rev7) gene product, and found that they are infertile in both males and females. We demonstrated that Mad2l2 is essential for PGC, but not somatic development. PGCs were specified normally in Mad2l2(-/-) embryos, but became eliminated by apoptosis during the subsequent phase of epigenetic reprogramming. A majority of knockout PGCs failed to arrest in the G2 phase, and did not switch from a H3K9me2 to a H3K27me3 configuration. By the analysis of transfected fibroblasts we found that the interaction of Mad2l2 with the histone methyltransferases G9a and GLP lead to a downregulation of H3K9me2. The inhibitory binding of Mad2l2 to Cyclin dependent kinase 1 (Cdk1) could arrest the cell cycle in the G2 phase, and also allowed another histone methyltransferase, Ezh2, to upregulate H3K27me3. Together, these results demonstrate the potential of Mad2l2 in the regulation of both cell cycle and the epigenetic status. The function of Mad2l2 is essential in PGCs, and thus of high relevance for fertility.

Show MeSH
Related in: MedlinePlus