Huang Ling，Guo Yong-Jun，Zhao Xin-yang

(1.Beijing University of Traditional Chinese Medicine，Beijing 100029，China;2.Henan Academy of Medical Sciences，Henan Cancer Institute，Zhengzhou 450003，China;3.Department of Biochemistry and Molecular Genetics，School of Medicine，UAB Stem Cell Institute，AL 35216，USA)

Tumor development of human takes multiple steps over a long period of time.The multiple steps with DNA mutations on genes is important for maintaining genome stability.Cell cycle and differentiation and epigenetic changes give rise to funor stem cells，which produce the mass of heterogeneous tumor cells.The cancer stem cell theory indicates that cancer cells are heterogeneous with a hierarchal structure，and only the cancer stem cells have the ability of self-renewal and to initiate tumorigenesis in a series of transplantation in xenograft models［1］.Cancer stem cells derive not only from normal stem cells but also from various kinds of progenitor cells.The existence of cancer stem cells was first elegantly demonstrated in leukemia by John Dick’s group［2］，and later cancer stem cells originated from breast cancer，prostate cancer，brain cancer and other types of cancers were discovered［3-5］.How the cancer stem cell maintains its ability to self-renewal is important for the survival of tumor.In the past ten years，researchers have found that epigenetic regulations is one of the most important factors for the regulation of caner stem cell self-renewal.Among them，protein arginine methyl transferases(PRMT)are found to methylate the arginine residues on histone tails and transcription factors.

Methylation is a more stable kind of modifications compared to phosphorylation and acetylation.Technically，there is no amino acid that can really mimic methylated arginine.Therefore，mutation of arginine to another amino acid cannot really demonstrate that the methylation on that amino acid is important but rather the arginine residue per semight be important.

The members of protein arginine methyl transferases share the high sequence homology on their catalytic domains.The proteins of PRMT1 and PRMT3 and PRMT4 were crystalized［6-8］.Basically the structure of catalytic domain is similar among the three PRMTs.

PRMT1 is responsible for the most of the arginine methylation reactions in eukaryotic cells.PRMT1 exists in nucleus as well as in cytoplasm.PRMT1 is isolated as an H4 methyl transferase methylating R3 on H4 tail，and is recruited by androgen receptor to MMTV-LTR promoter［9］.PRMT1 is also recruited by p53，the gatekeeper for genome stability.In this case，PRMT1，when it is recruited by p53，methylates histone H4R3 on p53 target gene GADD45［10］.After histone is methylated at H4R3，the histone H3 acetylation by p300 is greatly enhanced by both groups on p53 and androgen receptor.Nevertheless，it is not clear how many p53 target genes really need PRMT1 for transcriptional activation besides GADD45.GADD45 is also regulated by FOXO1.Since FOXO1 is arginine methylated by PRMT1，it is interesting to know whether PRMT1 is required for FOXO1 mediated transcriptional activation of the GADD45 gene.PRMT1 is highly expressed in different kinds of tumors as shown on the oncogene database(http://oncomine.com).

RUNX1 is required for the long-term self-renewal of adult hematopoietic stem cell and for megakaryocyte differentiation and lymphocyte differentiation［11，12］.RUNX1 is fused with ETO to form a fusion protein called AML1-ETO，which is responsible for 20%of acute myeloid leukemia.RUNX1 is very often mutated in acute myeloid leukemia and myeloid dysplasia syndrome.RUNX1 activates or represses gene expression in hematopoietic stem cells and different blood lineage cells in the context dependent manner.Many kinds of coactivators and corepressors are shown to bind to RUNX1.It showed that arginine methylation plays the key roles to determine whether RUNX1 is a transcriptional activator or repressor.RUNX1 is methylated by PRMT1 when PRMT1 is recruited by RUNX1 to PU.1 enhancer and promoter regions during myeloid differentiation.Once methylated，RUNX1 cannot interact with the SIN3 repressor complex during myeloid differentiation［13］.PRMT1 is upregulated during granulocyte differentiation.We speculate that RUNX1 is a transcriptional repressor at the stage of hematopoietic stem cells and the lack of PRMT1 expression might slow down the differentiation into granulocytes.But the role of PRMT1 is more complicated.As for megakaryocyte differentiation，PRMT1 level has to be downregulated，which leads to the silence of PU.1 gene expression by the unmethylated RUNX1 protein.Persistent high expression of PRMT1 blocks human cord blood derived CD34+cells from differentiation into mature megakaryocytes［14］.We don’t know how overexpression of PRMT1 will enhance GMP at the expense of MEP formation or whether PRMT1 will block particular transcription program at later stage of megakaryocyte differentiation when the MEP is already formed.

PRMT1 interacts with the SH3 domain of the EEN gene.EEN is a fusion partner for MLL［15］.Knocking down PRMT1 reduces the colony formation ability of MLL-EEN transduced mouse stem cells/progenitor cells.When MLL is artificially fused with PRMT1 or SAM68，which is a substrate of PRMT1，the fusion proteins can transform mouse progenitor cells in vitro.Although the MLL-EEN internal deletion，which cannot interact with PRMT1，cannot cause leukemia in mouse model，the paper didn’t show whether MLL-PRMT1 can cause leukemia in mouse model.MLL-PRMT1 methylates H4R3 on the HOXA9 promoter.Presumably arginine methylation enhances histone acetylation which activates the HOXA9 transcription just like PRMT1 in p53 mediated transcriptional activation.Recently ASH2L，a subunit of MLL complex，has been shown to be methylated by PRMT1［16］.How PRMT1 regulates MLL enzymatic activity and how the methylated MLL complex cross-talks to the MLL fusion complex are guaranteed for further investigation.

PRMT1 methylates BRCal，a gene often mutated in breast cancer［17］.Methylated BRCal fine-tunes the gene network regulated by BRCal.For example，the interaction between BRCal and SP1 is enhanced by methylation.In addition to molecular biology data，the methylated form of BRCal has been found in samples of human breast cancer tissue.Estrogen receptor plays important role in breast cancer.In that scenario，PRMT1 methylates estrogen receptor at its DNA binding site in cytoplasm［18］.Methylated ERα forms complex with Src and p85，a subunit of PI3 kinase.As a result，PRMT1 activates the Akt pathway which is required for breast cancer cell growth.Therefore in breast cancer，PRMT1 is clearly an oncogene which is agreeable with the survey that PRMT1 expression is high in breast cancer cells.

PRMT1 methylates Foxo family members:FOXO1 and FOXO3［19］in HeLa cells which was established from ovarian cancer cells.The methylation sites on FOXO transcription factors are also the anchor sites for the AKT kinases.As a result of methylation，the AKT kinases cannot lodge on FOXO transcription factor thus the transcription factors cannot be phosphorylated by the Akt kinases and thus the transcription factor avoid being transported out of the nucleus to be polyubiquitylated and degraded.The paper implies that PRMT1 antagonizes the Akt-PI3 kinase pathway via Foxo1，which activates apoptosis program through genes such as Bim.The Akt pathway is constantly on in many types of cancers.By enhancing transcription for apoptosis related gene(BIM gene)via direct methylation of FOXO1，PRMT1 is a tumor suppressor gene which seems to be against what we believe since the level of PRMT1 is usually higher than normal in cancer cells.In the Yamagata etc’s paper，they used H2O2to demonstrate that PRMT1 triggers the methylation of FOXO1in response to the redox status change.It is still unknown whether the methylation of FOXO1 will change under the physiological level of redox stress or under conditions where the AKT pathway is disrupted or enhanced.The same group later showed that BCL2 antagonist BAD is methylated by PRMT1［20］.Methylation of BAD disrupt its association with the 14-3-3-β protein，thus the BAD protein localizes to mitochondria to cause apoptosis.These data further support the idea that the AKT kinase docking sites are methylated by the AKT kinases，but there are no data showing directly how the Akt pathway are affected by manipulating PRMT1 level and vice versa.

In acute myeloid leukemia，the AKT/FOXO pathway behaves the opposite to the solid tumors［21］.Here FOXO1 is constantly active in 40% of AML.Inhibiting the FOXO1 activity or activation of the AKT pathway results in the death of the MLL-AF9 leukemia cells.Since FOXO1 is methylated by PRMT1，PRMT1 inhibitor might be used to block the methylation of FOXO1，which leads to subsequent protein degradation after FOXO1 is phosphorylated by the AKT kinases.In this scenario，PRMT1 inhibitors should be tested to treat leukemia.

PRMT1 is required for maintaining chromosome integrity by methylating TRF2 and MRE11.PRMT1 methylates TRF2 in telomere shelterin complex［22］.Through TFR2 methylation，the telomere maintains the normal length.Mutation of arginine residues to lysine in TRF2 will lead to cell senescence.In agreeable with the sitespecific mutagenesis assay，knocking down PRMT1 will promote the shortening of the telomere length.Maintaining the telomere length is important for replication of chromosome.In cancer cells，telomere instability will lead to abnormal chromosome fusion，deletion etc.Abnormal shortening of telomere predispose the blood cells to acute myeloid leukemia or other preleukemic diseases as well as other types of cancers.

PRMT1 methylates MRE11which is the DNA exonuclease in DNA damage response complex(MRE11-RAD50-NBS1 complex).Methylation of MRE11 has no effect on complex formation but methylated MRE11 have higher enzymatic activity［23］.The importance of PRMT1 in maintaining genome stability is also shown in conditional PRMT1 knockout mouse［24］.The embryonic stem cells from these animals are often aneuploidy，polyploidy and with fusion on chromosome ends.

Although PRMT1 has not been found to be mutated in cancer patient samples，PRMT1 level is upregulated.We speculate that PRMT1 is a gene which oncogenes are addicted to(i.e.PRMT1 is a synthetic lethal gene with mutated oncogenes)just like HSP90 gene.With success of developing HSP90 inhibitor，we believe PRMT1 can be targeted with similar approaches.

Reference

［1］ Clarke M F，Dick J E，Dirks P B，et al.Cancer stem cells--perspectives on current status and future directions［J］.AACR Workshop on cancer stem cells.Cancer Res，2006，66(19):9339-9344.

［2］ Bonnet D，Dick J E.Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cel［J］.Nat Med，1997，3(7):730-737.

［3］ Badve S，Nakshatri H.Breast-cancer stem cells-beyond semantics［J］.The lancet oncology，2012，13(1):43-48.

［4］ Greaves M，Maley C C.Clonal evolution in cancer［J］.Nature，2012，481(7381):306-313.

［5］ Jordan C T，Guzman M L，Noble M.Cancer stem cells［J］.N Engl J Med，2006，355(25):1253-1261.

［6］ Zhang X，Cheng X.Structure of the predominant protein arginine methyltransferase PRMT1 and analysis of its binding to substrate peptides［J］.Structure，2003，11(5):509-520.

［7］ Zhang X，Zhou L，Cheng X.Crystal structure of the conserved core of protein arginine methyltransferase PRMT3［J］.Embo J，2000，19(14):3509-3519.

［8］ Yue W W，Hassler M，Roe SM，et al.Insights into histone code syntax from structural and biochemical studies of CARM1 methyltransferase［J］.EMBO J，2007，26(20):4402-4412.

［9］ Wang H，Huang Z Q，Xia L，et al.Methylation of histone H4 at arginine 3 facilitating transcriptional activation by nuclear hormone receptor［J］.Science，2001，293(5531):853-857.

［10］ An W，Kim J，Roeder R G.Ordered cooperative functions of PRMT1，p300，and CARM1 in transcriptional activation by p53［J］.Cell，2004，117(6):735-748.

［11］ Ichikawa M，Asai T，Saito T，et al.AML-1 is required for megakaryocytic maturation and lymphocytic differentiation，but not for maintenance of hematopoietic stem cells in adult hematopoiesis［J］.Nat Med，2004，10(3):299-304.

［12］ Growney J D，Shigematsu H，Li Z，et al.Loss of Runx1 perturbs adult hematopoiesis and is associated with a myeloproliferative phenotyp［J］.Blood，2005，106(2):494-504.

［13］ Zhao X，Jankovic V，Gural A，et al.Methylation of RUNX1 by PRMT1 abrogates SIN3A binding and potentiates its transcriptional activity［J］.Genes Dev，2008，22(5):640-653.

［14］ Chang Y I，Hua W K，Yao C L，et al.Protein-arginine methyltransferase 1 suppresses megakaryocytic differentiation via modulation of the p38 MAPK pathway in K562 cells［J］.J Biol Chem，2010，285(2):2595-2606.

［15］ Cheung N，Chan L C，Thompson A，et al.Protein arginine-methyltransferase-dependent oncogenesis［J］.Nat Cell Biol，2007，9(10):1208-1215.

［16］ Butler J S，Zurita-Lopez C I，Clarke S G，et al.Protein-arginine methyltransferase 1(PRMT1)methylates Ash2L，a shared component of mammalian histone H3K4 methyltransferase complexes［J］.J Biol Chem，2011，286(14):1234-1244.

［17］ Guendel I，Carpio L，Pedati C，et al.Methylation of the tumor suppressor protein，BRCal，influences its transcriptional cofactor function［J］.Plos One，2010，5(6):11379.

［18］ Le Romancer M，Treilleux I，Leconte N，et al.Regulation of estrogen rapid signaling through arginine methylation by PRMT1［J］.Mol Cell，2008，31(2):212-221.

［19］ Yamagata K，Daitoku H，Takahashi Y，et al.Arginine methylation of FOXO transcription factors inhibits their phosphorylation by Akt［J］.Mol Cell，2008，32(2):221-231.

［20］ Sakamaki J，Daitoku H，Ueno K，et al.Arginine methylation of BCL-2 antagonist of cell death(BAD)counteracts its phosphorylation and inactivation by Akt［J］.Proc Natl Acad Sci USA，2011，108(15):6085-6090.

［21］ Sykes S M，Lane S W，Bullinger L，et al.AKT/FOXO signaling enforces reversible differentiation blockade in myeloid leukemias［J］.Cell，2011，146(5):697-708.

［22］ Mitchell TR，Glenfield K，Jeyanthan K，et al.Arginine methylation regulates telomere length and stability［J］.Mol Cell Biol，2009，29(18):4918-4934.

［23］ Boisvert F M，Dery U，Masson J Y，et al.Arginine methylation of MRE11 by PRMT1 is required for DNA damage checkpoint control［J］.Genes Dev，2005，19(6):671-676.

［24］ Yu Z，Chen T，Hebert J，et al.A mouse PRMT1 null allele defines an essential role for arginine methylation in genome maintenance and cell proliferation［J］.Mol Cell Biol，2009，29(11):2982-2996.