ATM (ataxia telangiectasia mutated)

2021-04-01   Jean Loup Huret 

Honorary Associate Professor of Medical Genetics of the French universities Quinçay, FRANCE

Identity

HGNC
LOCATION
11q22.3
LOCUSID
ALIAS
AT1,ATA,ATC,ATD,ATDC,ATE,TEL1,TELO1
FUSION GENES

Abstract

Review on ATM, with data on DNA, on the protein encoded, and where the gene is implicated.

DNA/RNA

Atlas Image
ATM (11q22.3) in normal cells: PAC 1053F10 - Courtesy Mariano Rocchi, Resources for Molecular Cytogenetics.

Description

Transcript (hg38) including UTRs: chr11:108,222,832-108,369,099, size: 146,268 on plus strand; coding region: chr11:108,227,625-108,365,508 Size: 137,884 according to UCSC. ATM has 30 transcripts (splice variants). The canonical form (ATM-003) is the longest, with 64 exons: 12954 bp --> 3056 amino acids (aa). Most transcripts (9 of 11) code for short (93 to 168aa) proteins which contain the TAN domain only. One transcript (ATM-001) with 27 exons, codes for a 1,369 aa protein.

Transcription

Table 1. ATM exons and transcription - Canonical form (NextProt Exons). Coding Positions from 108,227,625 to 108,365,505 [length: 137,881 bp].
IdentifierPosition on geneLengthAmino acids
ENSE00002185659633 - 70371
ENSE000021517402301 - 238888
ENSE000037429335112 - 5213102Met1 - Lys24
ENSE000037250825293 - 5405113Lys25 - Arg62
ENSE000037448056695 - 6840146Arg62 - Arg111
ENSE0000166708813187 - 13351165Arg111 - Glu166
ENSE0000167071021470 - 21635166Glu166 - Arg221
ENSE0000173959822305 - 22543239Arg221 - Gly301
ENSE0000161729524481 - 24644164Gly301 - Gln355
ENSE0000165281526450 - 26619170Val356 - Trp412
ENSE0000165830628218 - 28589372Trp412 - Cys536
ENSE0000163873129354 - 29548195Cys536 - Ser601
ENSE0000165540630334 - 3042996Ser601 - Cys633
ENSE0000159211631331 - 31556226Cys633 - Glu708
ENSE0000177490033732 - 33857126Ile709 - Lys750
ENSE0000172383534998 - 35123126Ser751 - Lys792
ENSE0000176950936503 - 3659290Lys793 - Leu822
ENSE0000171937844688 - 44859172Ala823 - Gly880
ENSE0000359577045927 - 46126200Gly880 - Met946
ENSE0000159192348581 - 4866383Tyr947 - Ser974
ENSE0000176429648768 - 48923156Ser974 - Trp1026
ENSE0000349173850049 - 5012476Trp1026 - Glu1051
ENSE0000348412750239 - 50369131Ala1052 - Arg1095
ENSE0000360575357008 - 57125118Arg1095 - Met1134
ENSE0000161854358512 - 58685174Ser1135 - Lys1192
ENSE0000176133860227 - 60396170Val1193 - Arg1249
ENSE0000164961061744 - 61990247Arg1249 - Gln1331
ENSE0000172896665117 - 65232116Ile1332 - Gly1370
ENSE0000346919466494 - 66620127Gly1370 - Pro1412
ENSE0000358103567119 - 67318200Asp1413 - Arg1479
ENSE0000359867770136 - 70310175Arg1479 - Gln1537
ENSE0000352949770830 - 70994165Val1538 - Glu1592
ENSE0000355926472444 - 72576133Glu1593 - Asp1637
ENSE0000348040674804 - 74899 96Asp1637 - Glu1669
ENSE0000356558477231 - 77402172Glu1669 - Cys1726
ENSE0000359011579165 - 79306142Cys1726 - Lys1773
ENSE0000350806180370 - 80546177Phe1774 - Glu1832
ENSE0000347209282192 - 82369178Val1833 - Glu1892
ENSE0000347251785414 - 8550188Glu1892 - Arg1921
ENSE0000359103487677 - 87832156Arg1921 - Arg1973
ENSE0000345842789928 - 9001588Arg1973 - Gln2002
ENSE0000361455293340 - 9342889Asp2003 - Arg2032
ENSE0000355239193528 - 93630103Arg2032 - Gln2066
ENSE0000352257794890 - 95038149Ala2067 - Ser2116
ENSE0000368881997471 - 97575105Ser2116 - Arg2151
ENSE0000364838998818 - 98937 120Arg2151 - Arg2191
ENSE00003542516102827 - 103061235Arg2191 - Gln2269
ENSE00003580001103575 - 103742168Leu2270 - Ala2325
ENSE00003479049105162 - 105275114Asn2326 - Lys2363
ENSE00003599084106538 - 106755218Ala2364 - Arg2436
ENSE00003659411107731 - 107938208Arg2436 - Lys2505
ENSE00003596787108961 - 109074114Arg2506 - Asn2543
ENSE00003571052109396 - 109554159Leu2544 - Glu2596
ENSE00003483502110279 - 110417139Asp2597 - Lys2643
ENSE00003573212111403 - 11148583Lys2643 - Lys2670
ENSE00003560896112486 - 112626141Val2671 - Lys2717
ENSE00003666486113362 - 113478117Gly2718 - Lys2756
ENSE00003611442120739 - 120888150Val2757 - Met2806
ENSE00003671649123260 - 123425166Glu2807 - Val2862
ENSE00003588344124796 - 12488287Val2862 - Gly2891
ENSE00003609743131283 - 131397115Gly2891 - Arg2929
ENSE00003638878132328 - 13239164Arg2929 - Glu2950
ENSE00003502604142599 - 142735137Val2951 - Ser2996
ENSE00002195058142842 - 1466193778Ser2996 - Val3056

The ATM promotor is bi-directional and also directs the transcription of the NPAT gene.

Proteins

Atlas Image
Figure 1: ATM amino acids sequence

Description

This large polypeptide of 350 kDa and 3,056 residues bears a PI3 kinase signature within its carboxy-terminal catalytic site, but has the catalytic activity of a serine-threonine protein kinase. This motif is characteristic of a protein family of which ATM is a member - the PI-3 kinase-like protein kinases (PIKKs; Lovejoy et al., 2009; Baretić et al., 2014).
ATM can be divided in 2 parts: a N-solenoid and a FATKIN. The N-solenoid is made, from N-term, of a Spiral (aa 1-1160) and a Pincer ((aa 1161-1890), itself made of a N-pillar (aa 1161-1430), Bridge (aa 1431-1600), C-pillar (aa 1601-1680), Railing (aa 1681-1800), and Cap (aa 1801-1890). Following the N-solenoid is the FATKIN (aa 1891-3056). The FATKIN, consists of a FAT and the C-terminal kinase domain. The FATKIN can be divided into five domains: tetratricopeptide repeat domains TRD1 (aa 1903-2025), TRD2 (aa 2032 -2190), and TRD3 (aa 2195-2476); HRD (HEAT-repeats domain) (aa 2484-2612); and a kinase domain (aa 2618-3056). There are also a LST8-binding element (LBE, aa 2791-2829) and an activation loop (aa 2888-2910), see Figures 1 and 2 (Young et al., 2005; You et al., 2005; Bhatti et al., 2011; Baretic and Williams 2014; Cremona and Behrens 2014; Lau et al., 2016; Wang et al., 2016; Baretic et al., 2017; Baretic et al., 2019).
The main domains and motifs are, from N-term to C-term:
a TAN motif (aa 15-27 (located in 18-40 in UniProt)); a chromatin-association domain of ATM (amino acids 5-224); a nuclear localization signal: KRKK (aa 385-388), within aa 227u2013568; a leucine zipper (aa 1217-1239); a FAT domain (aa 1940-2566 according to UniProt; includes TRD1 to 3 and HRD, i.e. from aa 1903 to 2612 according to Baretic et al., 2017); a phosphatidylinositol 3- and 4-kinases signature 1 (aa 2716-2730 (UniProt)); a phosphatidylinositol 3- and 4-kinases signature 2 (aa 2855-2875 (UniProt)); and a FATC domain (aa 3024-3056 according to Baretic et al., 2017 and to UniProt).
The Spiral has roles in binding substrates, regulators, and adaptors. TAN is a motif which is conserved specifically in the Tel1/ATM subclass of the PIKKs (interPro). TAN motif (Tel1/ATM N-terminal or Telomere-length maintenance and DNA damage repair) contains a conserved (L/V/I)XXX(R/K)XX(E/D)RXXX(L/V/I) signature. In the case of ATM: LEHDRA TERKKEV (aa 15-27) The TAN motif plays a role in telomere length maintenance. FAT/FATC domain: The PI-kinase domain of members of the PIK-related family is made of a FAT (FRAP, ATM, TRRAP) domain and the C-terminal FATC domain (interPro). The FATC (FRAP, ATM, TRRAP C-terminal) domain is essential for the kinase activity. PI3/4-kinase (Phosphatidylinositol 3-/4-kinase, catalytic domain): Phosphatidylinositol 3-kinase (PI3-kinase) is an enzyme that phosphorylates phosphoinositides on the 3-hydroxyl group of the inositol ring (interPro). Leucine zipper: region required for dimerization mediating sequence-specific DNA-binding (interPro).
Other remarkable sites according to Prosite: (see Figure 2)
- Protein kinase C phosphorylation sites: aa 21, 39, 127, 151, 274, 305, 373, 475, 491, 554, 571, 616, 775, 791, 808, 917, 1037, 1104, 1179, 1487, 1558, 1769, 1770, 1857, 1880, 1905, 1990, 2058, 2134, 2146, 2194, 2242, 2264, 2329, 2434, 2438, 2513, 2608, 2611, 2640, 2685, 2745, 2754, 2761
- Casein kinase II phosphorylation sites: 127, 200, 274, 373, 403, 470, 515, 571, 629, 644, 646, 655, 710, 767, 837, 865, 891, 934, 1004, 1048, 1100, 1118, 1143, 1179, 1212, 1242, 1263, 1350, 1403, 1589, 1601, 1609, 1721, 1748, 1819, 1891, 1966, 1988, 1993, 2000, 2011, 2123, 2134, 2142, 2184, 2218, 2242, 2333, 2348, 2359, 2375, 2408, 2476, 2573, 2592, 2812, 2921, 2947, 2996
- cAMP- and cGMP-dependent protein kinase phosphorylation sites: 1923, 2751
- Tyrosine kinase phosphorylation sites: KcqEllnY (116-123),KtqEkgaY (296-303),RhgErtpY (447-454), KvsEtfgY (1196-1203), KevEgtsY (2117-2124), KrslEsvY (2160-2167), KksfEekY (2810-2817)
- N-glycosylation sites: 81, 272, 567, 591, 704, 765, 789, 1230, 1240, 1356, 1660, 1719, 1855, 2994, 3044
- N-myristoylation sites (role in membrane targeting): 134, 138, 301, 506, 558, 724, 774, 1016, 1302, 1456, 1458, 1672, 1817, 1925, 1980, 2020, 2063, 2342, 2369, 2678, 2917, 3019, 3023, 3029
Amino acids 90 to 97 interact with TP53, BRCA1, and STK11. The binding site for NBS1 maps to the Spiral/N-pillar interface. The interaction with ABL1 is in aa 1373-1382. CLK2, a regulator of ATM, stability binds to aa 830-1290 and aa 2680-3056.
ATM homodimer: The FATC, the LBE (aa 2791-2829), the activation loop (aa 2888-2910), and the PIKK regulatory domain form a compact arrangement that has been referred to as the FLAP. It joins the TRD3 helices (aa 2378-2476), referred to as the FLAP binding element (FLAP-BE)). FLAP and FLAP-BE can form either an open dimer with a limited intermolecular interface or a tightly packed closed dimer with a larger interface. The active site of the open dimer is compatible with substrate binding, whereas the PIKK regulatory domain blocks the active site in the closed dimer (Baretic et al., 2017).
Atlas Image
Figure 2: ATM gene and protein

Expression

ATM is expressed in all tissues.
Atlas Image
Figure 3: ATM electron microscopy structure Images are taken from PhosphoSitePlus and ModBase

Localisation

Mostly in the nucleus throughout all stages of the cell cycle.

Function

ATM is a homeostatic protein kinase with an extremely broad range of roles in various cellular circuits (Shiloh et al., 2013; Guleria et al., 2016; Shiloh, 2014; Cremona et al., 2014; Ambrose et al., 2013; Espach et al., 2015; Awasthi et al., 2016).
The PI3 kinase signature is a motif characteristic of a protein family of which ATM is a member - the PI-3 kinase-like protein kinases (PIKKs; Lovejoy et al., 2009; Baretić et al., 2014). This family also contains the MTOR protein, which regulates many signaling pathways in response to nutrient levels, growth factors and energy balance (Alayev et al., 2013; Cornu et al., 2013); the catalytic subunit of the DNA-dependent protein kinase (DNA-PKcs), which is involved in the NHEJ pathway of double strand breaks (DSB) repair and other genotoxic stress responses (Davis et al., 2014; Jette N et al., 2015), SMG1, which plays a key role in nonsense-mediated mRNA decay (Yamashita, 2013); and ATR, which responds to stalled replication forks and a variety of DNA lesions that lead to the formation of single-stranded DNA, including deeply resected DSBs (Errico et al., 2012; Maréchal et al., 2013; Awasthi et al., 2016). The redundancy, crosstalk and collaboration between the latter three PIKKs, which collectively respond to a broad spectrum of genotoxic stresses, are being extensively investigated (Lovejoy et al., 2009; Maréchal et al., 2013; Sirbu et al., 2013; Thompson, 2012; Gobbini et al., 2013; Chen et al., 2012).
It should be noted that in A-T patients, the two PIKKs that converse and cooperate with ATM in the response to genotoxic stress, ATR and DNA-PK, remain active. In view of the functional relationships between the three protein kinases, some of ATMs duties are probably carried out to a certain extent by ATR and/or DNA-PK, in A-T cells. On the other hand, the lack of a very versatile member of this trio may lead to some suboptimal responses of the other two, if they depend on the crosstalk with ATM. This interesting question is a subject of intensive research.
Homeostatic protein kinase involved in many cellular circuits. A primary role in the DNA damage response. Activated vigorously by DNA double-strand breaks and activates a broad network of responses. ATM initiates cell cycle checkpoints in response to double-strand DNA breaks by phosphorylating TP53, BRCA1, H2AX, ABL1, NFKBIA and CHEK1, as well as other targets; in certain types of tissues ATM inhibits radiation-induced, TP53-dependent apoptosis.
  • Double strand breaks. The most widely documented function of ATM, and the one associated with its most vigorous activation, is the mobilization of the complex signaling network that responds to DSBs in the DNA (Shiloh Y et al., 2013; Cremona CA et al., 2014; Awasthi P et al., 2016; Thompson LH, 2012; McKinnon PJ, 2012). DSBs are induced by exogenous DNA breaking agents or endogenous reactive oxygen species (Schieber M et al., 2014), and are an integral part of physiological processes including meiotic recombination (Borde V et al., 2013; Lange J et al., 2011) and the rearrangement of antigen receptor genes in the adaptive immune system (Alt FW et al., 2013). DSBs are repaired via nonhomologous end-joining (NHEJ), or homologous recombination repair (HRR; Shibata A et al., 2014; Chapman JR et al., 2012; Jasin M et al., 2013; Radhakrishnan SK et al., 2014). DSBs also activates the DDR, a vast signaling network that mobilizes special cell cycle checkpoints, extensively alters the cellular transcriptome, and changes the turnover, activity and function of numerous proteins that ultimately leads to modulation of numerous cellular circuits. This network is based on a core of dedicated DDR players and the ad-hoc recruitment of proteins from many other arenas of cellular metabolism, which typically undergo special, damage-induced post-translational modifications (PTMs; Shiloh Y et al., 2013; Sirbu BM et al., 2013; Thompson LH, 2012) (Goodarzi AA et al., 2013; Panier S et al., 2013; Polo SE et al., 2011).
    Once ATM mobilizes the vast DDR network in response to a DSB (McKinnon PJ, 2012; Shiloh Y et al., 2013; Bhatti S et al., 2011), its protein kinase activity is rapidly enhanced, and PTMs on the ATM molecule are induced, including several autophosphorylations and an acetylation (Shiloh Y et al., 2013; Bhatti S et al., 2011; Bakkenist CJ et al., 2003; Kozlov SV et al., 2006; Bensimon A et al., 2010; Sun Y et al., 2007; Kaidi A et al., 2013; Paull TT, 2015).
    ATM subsequently phosphorylates key players in various arms of the DSB response network (Shiloh Y et al., 2013; Bensimon A et al., 2010; Matsuoka S et al., 2007; Mu JJ et al., 2007; Bensimon A et al., 2011), including other protein kinases that in turn phosphorylate still other targets (Bensimon A et al., 2011).
  • Single-strand break repair and base excision repair. A broader, overarching role for ATM in maintaining genome stability was recently suggested in addition to mobilizing the DSB response (Shiloh Y, 2014). According to this conjecture, ATM supports other DNA repair pathways that respond to various genotoxic stresses, among them single-strand break repair (SSBR; Khoronenkova SV et al., 2015) and base excision repair (BER) - a cardinal pathway in dealing with the daily nuclear and mitochondrial DNA damage caused by endogenous agents (Wallace SS, 2014; Bauer NC et al., 2015).
    ATMs involvement in these processes is based on its ability to phosphorylate proteins that function in these pathways. In this way ATM also takes part also in resolving non-canonical DNA structures that arise in DNA metabolism, and in regulating other aspects of genome integrity such as nucleotide metabolism, the response to replication stress, and resolution of the occasional conflicts that arise between DNA damage and the transcription machinery. ATM is not critical for any of these processes in the same way it is for the DSB response, but rather contributes to their regulation (in most cases, their enhancement) when the need arises (Shiloh Y, 2014; Segal-Raz H et al., 2011; Zolner AE et al., 2011).
    This function of ATM may explain the moderate, variable sensitivity of ATM-deficient cells to a broad range of DNA damaging agents. Among them are UV radiation, alkylating agents, crosslinking agents, hydrogen peroxide, 4-Nitroquinoline 1-oxide, phorbol-12-myristate-13-acetate and topoisomerase 1 poisons (Yi M et al., 1990; Ward AJ et al., 1994; Hoar DI et al., 1976; Paterson MC et al., 1976; Smith PJ et al., 1980; Mirzayans R et al., 1989; Henderson EE et al., 1980; Scudiero DA, 1980; Jaspers NG et al., 1982; Teo IA et al., 1982; Barfknecht TR et al., 1982; Fedier A et al., 2003; Leonard JC et al., 2004; Lee JH et al., 2006; Zhang N et al., 1996; Smith PJ et al., 1989; Alagoz M et al., 2013; Katyal S et al., 2014; Speit G et al., 2000; Shiloh Y et al., 1985; Hannan MA et al., 2002).
    ATM-deficient cells also exhibit reduced efficiency in resolving TOP1 (Topoisomerase I) -DNA covalent intermediates (Alagoz M et al., 2013; Katyal S et al., 2014).
    This ongoing role of ATM is its routine function in the daily maintenance of genome stability, while its powerful role in the DSB response is reserved for when this harmful lesion interferes with the daily life of a cell. Thus, when ATM is missing, not only is there markedly reduced response to DSBs, the ongoing modulation of numerous pathways in response to occasional stresses becomes suboptimal. All of these lesions are part of the daily wear and tear on the genome that contributes to ageing.
    An additional role for ATM in genome dynamics was proposed following evidence that ATM is involved in shaping the epigenome in neurons by regulating the localization of the histone deacetylase 4 ( HDAC4; Li J et al., 2012; Herrup K et al., 2013; Herrup K, 2013), targeting the EZH2 component of the polycomb repressive complex 2 (Li J et al., 2013), and regulating the levels of 5-hydroxymethylcytosine in Purkinje cells (Jiang D et al., 2015).
  • Oxidative stress/Cellular homeostasis.
    Cytoplasmic fraction of ATM. ATMs role in cellular homeostasis is further expanded by its cytoplasmic fraction. Specifically, cytoplasmic ATM was found to be associated with peroxisomes (Watters D et al., 1999; Tripathi DN et al., 2016; Zhang J et al., 2015) and mitochondria (Valentin-Vega YA et al., 2012). In view of the evidence of increased oxidative stress in ATM-deficient cells, it has long been suspected that ATM senses and responds to oxidative stress (Gatei M et al., 2001; Rotman G et al., 1997; Rotman G et al., 1997; Barzilai A et al., 2002; Watters DJ, 2003; Takao N et al., 2000; Alexander A et al., 2010). This conjecture was validated by work from the Paull lab (Guo Z et al., 2010a), which identified an MRN-independent mode of ATM activation, differentiating it from DSB-induced activation, stimulated by reactive oxygen species (ROS) and leading to ATM oxidation (Paull TT, 2015; Guo Z et al., 2010a; Guo Z et al., 2010b; Lee JH et al., 2014). ATM was also found to be involved specifically in the protection against oxidative stress induced by oxidized low-density lipoprotein (Semlitsch M et al., 2011). It has thus assumed the role of a redox sensor (Ditch S et al., 2012; Tripathi DN et al., 2016; Krüger A et al., 2011). Recently, the first phospho-proteomic screen was carried out to identify substrates of ROS-activated ATM (Kozlov SV et al., 2016). An important arm of the ATM-mediated response to ROS extends to peroxisomes (Tripathi DN et al., 2016). Work from the Walker lab showed that ROS-mediated activation of peroxisomal ATM leads to ATM-mediated phosphorylation of LKB and subsequent activation of AMPK and TSC2, which dampens mTORC1-mediated signaling, eventually decreasing protein synthesis and enhancing autophagy (Alexander A et al., 2010; Tripathi DN et al., 2013; Zhang J et al., 2013; Alexander A et al., 2010; Alexander A et al., 2010). Further work from this lab (Zhang J et al., 2015) showed that ATM also phosphorylates the peroxisomal protein PEX5, flagging it for ubiquitylation and subsequent binding to the autophagy adapter, SQSTM1 (p62), in the process of autophagy-associated peroxisome degradation (pexophagy) - a critical process in peroxisome homeostasis (Till A et al., 2012).
    Mitochondrial fraction of ATM. Still another arm of the ATM-mediated response to oxidative stress operates in the mitochondrial fraction of ATM. ATM is thus emerging also as a regulator of mitochondrial homeostasis. Evidence is accumulating of its involvement in mitochondrial function, mitophagy, and the integrity of mitochondrial DNA (Valentin-Vega YA et al., 2012; Ambrose M et al., 2007; Eaton JS et al., 2007; Fu X et al., 2008; Valentin-Vega YA et al., 2012; DSouza AD et al., 2013; Sharma NK et al., 2014) and further work is needed to identify its substrates in mitochondria and the mechanistic aspects of its action in this arena.
  • Links between ATM and the SASP (senescence-associated secretory phenotype). Several laboratories recently described direct links between ATM and the SASP - a cardinal feature of cell senescence. Work from the Gamble lab (Chen H et al., 2015) showed that the histone variant macroH2A.1 is required for full transcriptional activation of SASP-promoting genes, driving a positive feedback loop that enhances cell senescence. This response is countered by a negative feedback loop that involves ATM activation by endoplasmic reticulum stress, elevated ROS levels or DNA damage. ATMs activity is required for the removal of macroH2A.1 from sites of SASP genes, thus leading to SASP gene repression. The Elledge lab identified a major SASP activator - the transcription factor GATA4, whose stabilization drives this process (Kang C et al., 2015). Importantly, the activation of this pathway was dependent on both ATM and ATR, as was senescence-associated activation of TP53 and CDKN2A (p16INK4a). On the other hand, the Zhang lab (Aird KM et al., 2015) recently showed that when cell senescence is induced by replication stress (e.g., following nucleotide deficiency), ATM inactivation allows the cell to bypass senescence by shifting cellular metabolism: upon ATM loss, dNTP levels rise due to up-regulation of the pentose phosphate pathway, whose key regulator, glucose-6-phosphate dehydrogenase ( G6PD) is under functional regulation by ATM (Aird KM et al., 2015; Cosentino C et al., 2011).
  • Insulin response and lipoprotein metabolism. Other metabolic arenas in which ATM involvement is gaining attention are insulin response and lipoprotein metabolism, clinically represented by the metabolic syndrome. This role of ATM in cellular physiology was recently thoroughly and convincingly reviewed (Espach Y et al., 2015). Briefly, ATM was found to participate in several signaling pathways mediated by insulin (Yang DQ et al., 2000; Miles PD et al., 2007; Viniegra JG et al., 2005; Halaby MJ et al., 2008; Jeong I et al., 2010); and heterozygosity for Atm null allele in ApoE-deficient mice was found to aggravate their metabolic syndrome (Wu D et al., 2005; Schneider JG et al., 2006; Mercer JR et al., 2010), an effect that was partly relieved by the mitochondria-targeted antioxidant MitQ (Mercer JR et al., 2012.
  • IGF1 receptor. Another pathway by which ATM may impact on cellular senescence is the dependence of IGF1R (IGF-1 receptor) expression on ATM (Peretz S et al., 2001; Goetz EM et al., 2011; Ching JK et al., 2013); the mechanism remains to be elucidated, but ATM impacts on IGF-1-mediated pathways, including those that affect cellular senescence (Luo X et al., 2014).
  • Beta-adrenergic receptor. Another series of observations assigned ATM a protective role in cardiac myocyte apoptosis stimulated by β-adrenergic receptor and myocardial remodeling. Loss of Atm in mice induced myocardial fibrosis and myocyte hypertrophy and interfered with cardiac remodeling following myocardial infarction (Foster CR et al., 2011; Foster CR et al., 2012; Foster CR et al., 2013; Daniel LL et al., 2014). The mechanistic aspects of these effects are still unclear, but ATMs apparent involvement in myocardial homeostasis might be relevant to the observation of elevated arteriosclerosis in A-T carriers (Swift M et al., 1983; Su Y et al., 2000).
  • Homology

    Phosphatidylinositol 3-kinase (PI3K)-like proteins, most closely related to ATR and the DNA-PK catalytic subunit.

    Mutations

    Note

    The cellular phenotype of A-T represents genome instability, deficient DNA damage response (DDR), and elevated oxidative stress, in addition to a premature senescence component (Shiloh et al., 1982).

    Germinal

    Various types of mutations have been described, dispersed throughout the gene, and therefore most patients are compound heterozygotes; most mutations appear to inactivate the ATM protein by truncation, large deletions, or annulation of initiation or termination, although missense mutations have been described in the PI3 kinase domain and the leucine zipper motif.
    Patients with the severe form of A-T are homozygous or compound heterozygous for null ATM alleles. The corresponding mutations usually lead to truncation of the ATM protein and subsequently to its loss due to instability of the truncated derivatives; a smaller portion of the mutations create amino acid substitutions that abolish ATMs catalytic activity (Taylor et al., 2015; Gilad et al., 1996; Sandoval et al., 1999; Barone et al., 2009) (see also http://chromium.liacs.nl/LOVD2/home.php?select_db=ATM).
    Careful inspection of the neurological symptoms of A-T patients reveals variability in their age of onset and rate of progression among patients with different combinations of null ATM alleles (Taylor et al., 2015; Crawford et al., 2000; Alterman et al., 2007). Thus, despite the identical outcome in terms of ATM function, additional genes may affect the most cardinal symptom of A-T. Other, milder types of ATM mutations further extend this variability, and account for forms of the disease with extremely variable severity and age of onset of symptoms. The corresponding ATM genotypes are combinations of hypomorphic alleles or combinations of null and hypomorphic ones. Many of the latter are leaky splicing mutations and others are missense mutations, eventually yielding low amounts of active ATM (Taylor et al., 2015; Alterman et al., 2007; Soresina et al., 2008; Verhagen et al., 2009; Silvestri et al., 2010; Saunders-Pullman et al., 2012; Verhagen et al., 2012; Worth et al., 2013; Claes et al., 2013; Méneret et al., 2014; Nakamura et al., 2014; Gilad et al., 1998).

    Somatic

    B A variety of missense somatic, biallelic mutations were identified in hematologic malignancies, most notably mantle cell lymphoma and T-prolymphocytic leukaemia.
    Missense mutations outside of the PI3 kinase and leucine zipper domains have been described among breast cancer patients, although these mutations have not been found in A-T patients. Whether these mutations contribute to breast cancer though not to ataxia-telangiectasia remains controversial.

    Implicated in

    Top note
    Germline mutations: Ataxia telangiectasia syndrome is associated with greater than 100-fold increased risk of leukemias and lymphomas
    From a total of 296 consecutive genetically confirmed A-T patients, 66 patients who developed a malignant tumor; 47 lymphoid tumors and 19 non-lymphoid tumors were diagnosed. The development of childhood tumors below 16yrs (33 lymphoid and 3 brain) in A-T patients is associated almost exclusively with the absence of ATM kinase activity. After the age of 16yrs, there were 11 lymphoid malignancies, and 13 other tumors (including 7 breast cancers) (Reiman et al. 2011).
    Two-hundred seventy-nine patients with AT were enrolled in a registry. Sixty-nine patients developed 70 malignancies. Cancer types were mainly lymphoid (4 T-cell ALLs (acute lymphoblastic leukemia), 2 B-cell ALLs, 12 Hodgkin lymphomas, 26 B-cell NHL (non-Hodgkin lymphoma), 4 T-cell NHL, 3 T-PLL (T prolymphocytic leukemia), and carcinomas (3 breast, 2 gastric, 2 liver). The median age at diagnosis of malignancy of any type was 12.5 years (10yrs for most lymphoid diseases, 24yrs for T-PLL, and 31yrs for carcinomas) (Suarez F et al. 2015).
    Somatic mutations: According to COSMIC, somatic mutations in ATM are the following: missense substitution (57%), nonsense substitution (12%), synonymous substitution (6%), frameshift deletion (6%), frameshift insertion (2%), inframe deletion (1%), complex mutation (0.2%), inframe insertion (0.02%). They are found in the following cancers: meningioma 12%; endometrium 12% (undifferentiated-dedifferentiated carcinoma 20%, endometrioid carcinoma 15%, clear cell carcinoma 11%); prostate 12%; bladder 10%; colorectal adenocarcinoma 9%; liver cancers 5% (combined hepatocellular-cholangiocarcinoma 10%, hepatocellular carcinoma 4%); stomach adenocarcinoma 7%; haematopoietic and lymphoid 7% (T cell prolymphocytic leukemia 45%, mantle cell lymphoma 20-45%, chronic lymphocytic leukaemia 12%, diffuse large B cell lymphoma 6-9%, multiple myeloma 5%, follicular lymphoma 4%, adult T cell lymphoma-leukemia 4%, ALL 3%, AML (acute myeloid leukemias) 2%, Burkitt lymphoma 1%); lung 7% (adenocarcinoma 7%, squamous cell carcinoma 6%, small cell carcinoma 4%); biliary tract 6%; ovary 5% (mixed adenosquamous carcinoma 21%, serous carcinoma 4%); pancreas 5%; oesophagus 4% ( adenocarcinoma 9%, squamous cell carcinoma 3%); breast 4%; soft tissue 3% (alveolar rhabdomyosarcoma 5%, embryonal rhabdomyosarcoma 3%, liposarcoma 3%), thyroid 3% ( anaplastic carcinoma 5%, papillary carcinoma 2%, medullary carcinoma 2%, follicular carcinoma 2%); kidney 3% (clear cell renal cell carcinoma 3%, papillary renal cell carcinoma 3%); cervix squamous cell carcinoma 3%; testis 2%; bone 2%; ( Ewing sarcoma_peripheral primitive neuroectodermal tumor 1%); central nervous system 2% (astrocytoma Grade III or IV 2%).
    PhosphositePlus give the following data: bladder: 12%; endometrial: 12%; colorectal 11%; stomach 10%; lung adenocarcinoma 9%; lung (squamous cell) 5%; prostate 4%; kidney (clear cell or chromophobe) 3%; head/neck 3%; breast 2%; glioblastoma 1%; ovary 1%; thyroid 1%.
    Translocations: In contrast, translocations/hybrid genes and fusion proteins involving ATM are extremely rarely found:
    t(6;11)(p22;q22) JARID2/ATM in lung adenocarcinoma (Yoshihara et al, 2015) PMID 25500544
    ATM (11q22.3) / ATM (11q22.3) in T-cell prolymphocytic leukemia (Bradshaw et al., 2002) and in breast invasive carcinoma.
    CUL5 (11q22.3) / ATM (11q22.3) in a cell line (Klijn et al., 2015).
    and, according to ChimerDB 3.0 and/or ChiTaRS databases: ITGB8 (7p21.1) / LOC440600 (1p13.3) in uterine corpus endometrial carcinoma, ATP6V1C2 (2p25.1) / ATM (11q22.3), CUX1 (7q22.1) / ATM (11q22.3), ASPH (8q12.3) / ATM (11q22.3) in diseases not specified.
    Entity name
    Ataxia telangiectasia
    Note
    Ataxia telangiectasia is a prototype genome instability syndrome (Perlman SL et al., 2012; Lavin MF, 2008; Crawford TO, 1998; Chun HH et al., 2004; Taylor AM et al., 1982; Taylor AM et al., 2015; Taylor AM, 1978; Butterworth SV et al., 1986; Kennaugh AA et al., 1986).
    Disease
    Ataxia telangiectasia is a progressive cerebellar degenerative disease with telangiectasia, immunodeficiency, premature aging, cancer risk, radiosensitivity, and chromosomal instability.
    Prognosis
    Prognosis is poor: median age at death: 17 years; survival rarely exceeds 30 years, though survival is increasing with improved medical care.
    Cytogenetics
    Spontaneous chromatid/chromosome breaks; non clonal stable chromosome rearrangements involving immunoglobulin superfamilly genes e.g. inv(7)(p14q35); clonal rearrangements.
    Entity name
    Gastric cancer
    Note
    ATM is frequently altered or deleted in gastric cancer.
    Germline mutations: A total of 282 patients with gastric adenocarcinoma (182 males and 100 females) were enrolled in a study. The most recurrent germline mutation was a mutation in ATM (1%) (Ji et al., 2020). A large study involving more than 600,000 cancer patients found that 0.7% patients had an ATM pathogenic variant. A higher risk for gastric cancer was estimated (OR (odds ratio), 2.97; 95% CI, 1.66-5.31) (Hall et al., 2021). In a case-control investigation of 345 gastric adenocarcinoma patients and 467 controls, the ATM rs189037 G>A polymorphism was associated with a significantly higher risk of gastric cancer, and patients with this polymorphism had lower overall survival (Tao et al., 2020).
    Somatic mutations: A panel of 543 cancer-associated genes was used to analyze genomic profiles in a cohort of 484 patients with gastric cancer. Fifty-one of the 484 (10.5%) patients carried at least one somatic mutation in an homologous recombination (HR) gene; ATM (16/484, 3.3%) was among the most frequently mutated HR genes (Fan et al., 2020). Two independent cohorts, a training set (n=524) and a validation set (n=394), of gastric cancer patients were enrolled. ATM, CHEK2, and TP53 expressions were examined. Somatic ATM loss, CHEK2 loss, and TP53 positivity were observed in 22%, 14%, and 36% of the training set, and in 17%, 12%, and 36% of the validation set. Also, patients with non-aberrant expressional levels of all 3 DNA damage response-related proteins had a more favorable outcome than others (Lee et al., 2014).
    Expression: Decreased expression and phosphorylation of ATM at serine 1981 ("S1981") were consistently found in tumors. Low level of phosphorylated ATM was significantly correlated with poor differentiation, lymph node metastasis and poor 5-year survival (Kang B et al., 2008). ATM is a target of "miR-181a" ( MIR181A1 (1q32.1) or MIR181A2 (9q33.3)). There is an inverse correlation between miR-181a and ATM protein expression in gastric cancer. Over-expression of miR-181a might be involved in development of gastric cancer by promoting proliferation and inhibiting apoptosis probably through directly targeting ATM (Zhang et al., 2014).
    Prognosis
    123 randomly assigned patients received treatment (PARP1 inhibitor/taxane vs placebo/taxane). The prevalence of low ATM expression in patients was 14%. PARP1 inhibitor/taxane is active in the treatment of patients with metastatic gastric cancer, with a greater overall survival benefit in ATM low patients (Bau et al., 2010).
    Entity name
    Salivary gland basal cell adenoma
    Note
    A missense mutation in the ATM gene (c.2572T>C, p.F858L) was seen in a basal cell adenoma with an allele frequency of 53 %, raising the possibility of a germline mutation (Wilson et al., 2016).
    48 samples of adenoid cystic carcinoma were studied. Low expression of ATM in cancer cells was significantly correlated with poor survival, while Low ATM expression in stromal fibroblasts was not significantly correlated to patient survival (Bazarsad et al., 2018)
    Entity name
    hepatocellular carcinoma and cholangiocarcinoma
    Note
    Germline/somatic mutations: A comprehensive genomic profiling identified 3,765 gene aberrations in 760 in gallbladder cancers (but no germline testing was performed). Each tumor harbored at least 1 gene aberration, with an average of 5 gene aberration per tumor About 15 genes were implicated, from TP53, was the most recurrently found, in 61% of tumors, CDKN2A in 29%, u2026 to ATM in 6%. (Abdel-Wahab et al., 2020). 131 patients with biliary tract cancers ( intrahepatic cholangiocarcinoma (64%), gallbladder adenocarcinoma (17%), extrahepatic cholangiocarcinoma (16%) and otherwise unspecified (4%)) were studied, 15 oncogenic somatic mutations in BAP1 or ATM were found. There was no case of ATM germline mutation (Maynard et al., 2020). 357 patients with liver cancer (214 with hepatocellular carcinoma, 122 with intrahepatic cholangiocarcinoma, and 21 with mixed hepatocellular-cholangiocarcinoma) were studied. 26% patients had at least one DNA damage repair (DDR) gene mutation, 15 of whom carried germline mutations (in BRCA2, BRCA1, ATM, PMS2 u2026.). The most commonly DDR genes with somatic mutations were ATM (5%) and BRCA1/2 (5%). 9 % of patients with intrahepatic cholangiocarcinoma had BRCA1/2 somatic mutations, and 6% of patients with hepatocellular carcinoma had ATM somatic mutations (Lin et al., 2019).
    Expression: Fifty pairs of hepatocellular carcinoma specimens and corresponding adjacent liver tissues were collected. The methylation frequency of the ATM promoter was significantly higher in hepatocellular carcinoma tissues than in normal liver tissues. Methylated ATM was correlated with lower ATM expression. Methylation of the ATM promoter was significantly associated with better outcome in patients with locally advanced hepatocellular carcinoma who initially received radiotherapy (Yan et al., 2020).
    Entity name
    pancreatic cancers
    Note
    Pancreatic ductal adenocarcinoma: 1 to 4% of patients with pancreatic ductal adenocarcinoma with or without a family history of pancreatic cancer have a pathogenic germline ATM variant, and 2 to 18% of pancreatic ductal adenocarcinoma samples have somatic ATM alterations, either mutations, or loss of heterozygosity. Patients with loss of ATM expression and normal TP53 expression had decreased overall survival compared to patients with loss of ATM expression and abnormal TP53 expression, indicating that ATM expression may be an important prognostic factor for survival in patients with pancreatic cancer (review in Nanda and Roberts 2020). A large study involving more than 600,000 cancer patients with various cancers found that 0.7% patients had a germline ATM pathogenic variant. A higher risk for pancreatic cancer was found (OR= 4.2) (Hall et al., 2021).
    Hereditary pancreatic cancer: 5%-10% of pancreatic ductal adenocarcinoma are hereditary pancreatic cancers. A subset of familial pancreatic cancer have germline mutations in DNA repair genes BRCA2, ATM, or PALB2 (review in Bakker and de Winter. 2012; Rustgi 2014). In a series of 166 familial pancreatic cancer probands, 2.4% carried germline heterozygous ATM mutations. In one patient, the tumor analysis showed the second allele of ATM to be mutated, suggesting that the ATM loss in this patient was driven by the classic two-hit model (Roberts et al., 2012). 638 patients with familial pancreatic cancers were selected and whole genome sequencing was performed. The highest ranked gene was ATM with 19 heterozygous premature truncating variants, followed by TET2 (9 variants), DNMT3A (7), POLN and POLQ (6 each) etc u2026 (Roberts et al., 2016). ATM loss was observed in 50 of 396 (13%) pancreatic ductal adenocarcinomas tumors, and more often in patients with a family history of pancreatic cancer (25%) than in those without (11%). ATM loss was associated with a significantly poorer outcome in patients with normal TP53 expression (Kim et al., 2014). A classification of high-to-moderate risk of pancreatic cancer in familial pancreatic cancer families according to germ-line mutation was defined: while cases with a mutation in CDKN2A, BRCA2, or PALB2 were high risk mutations, ATM, BRCA1 and mismatch genes were classified moderate risk mutation genes (Llach et al., 2020). 549 patients diagnosed with pancreatic ductal adenocarcinoma were included in another study. Germline pathogenic variants were identified in 16 genes, including ATM (11 cases, 2%). No patient with CHEK2 or ATM pathogenic variants responded to treatment with PARP1 inhibitor (Fountzilas et al., 2021). Of 708 patients with pancreatic cancer, eleven pathogenic germline mutations were identified: 3 in ATM, 1 in BRCA1, 2 in BRCA2, 1 in MLH1, 2 in MSH2, 1 in MSH6, and 1 in TP53 (Grant et al., 2015). 96 patients, in another study, were tested. Fourteen pathogenic mutations were identified: four in ATM, two in BRCA2, CHEK2, and MSH6, and one in BARD1, BRCA1, FANCM, and NBN. (Hu et al., 2016).
    E2F1 can induce the long noncoding RNA CDKN2B-AS1 expression through the ATM/E2F1 signaling pathway. CDKN2B-AS1 overexpression promotes epithelial to mesenchymal transition in pancreatic cancer cells. CDKN2B-AS1 can repress the ATM/E2F1 signaling pathway by negatively regulating the expression of CDKN2A and CDKN2B (Chen et al., 2017).
    ATM deficiency accelerates metastatic murine pancreatic ductal adenocarcinoma formation, leads to persistent DNA damage, increases chromosomal instability, and also renders murine pancreatic tumors highly sensitive to radiation (Drosos et al., 2017).
    ATM expression was knocked down using shRNA in two cell lines. ATM-deficient pancreatic cancer cells were found to be more sensitive to radiation, but not to chemotherapeutic agents than wild-type pancreatic cancer cells (Ayars et al., 2017).
    107 pancreatic neuroendocrine tumors specimens were investigated. High expression of ATM and CCNB1 (cyclin B1) was related to well-differentiated endocrine tumors. The high ATM expression group had a significantly lower recurrence rate (Shin et al., 2012)
    Prognosis
    Therapeutic implications of ATM alterations: clinical trials that target ATM-deficient tumors especially with PARP1 inhibitors are reviewed in Armstrong et al., 2019.
    Entity name
    Thyroid cancers
    Note
    Papillary thyroid carcinoma: 522 patients with papillary thyroid carcinoma and 885 control cases were studied; the CG/GG genotypes of rs1800057 were significantly associated with an increased risk of papillary thyroid carcinoma, and the AG/GG genotypes of rs189037 was inversely associated with papillary thyroid carcinoma risk. (Xu et al., 2012). A possible association between ATM rs189037 G>A polymorphisms and metastasis of papillary thyroid cancers was detected in female patients, but not in male patients (Gu et al., 2014). Of note is that ATM rs189037 G>A genetic polymorphism has also been proposed to contribute in an increased risk of head and neck and lung cancer (Bhowmik et al., 2015). In a study on 437 papillary thyroid carcinoma and 184 cancer-free controls, 3 ATM SNPs (rs373759, rs664143, and rs4585) were in strong linkage disequilibrium. When the three haplotypes (C-A-G), (T-G-T), and (C-G-T) of these three ATM SNP sites were analyzed, ATM haplotype (C-G-T) +/- was associated with a lower risk of papillary thyroid carcinoma than ATM haplotype (C-G-T) -/- (Song et al., 2014).
    Familial non-medullary thyroid cancer: Consistent ATM variants (ATM p.P1054R-rs1800057- and rs149711770) were described in families with familial non-medullary thyroid cancer (Miasaki et al., 2020). 277 cancer predisposition genes were tested in 17 families with familial non-medullary thyroid cancer. One frameshift variant and five missense variants. An ATM variant was identified in 3 instances in 2 families (Wang et al., 2019).
    Entity name
    Pheochromocytoma
    Note
    Seven of thirty-eight patients with pheochromocytoma had heritable mutations including RET (n=3), VHL (2), SDHB (1), and ATM and PDGFRA (1 each) (Xiong MJ, Osunkoya 2020).
    Entity name
    Breast cancer
    Note
    Ataxia-telangiectasia: Early reports were somewhat contradictory. In 161 families affected by ataxia-telangiectasia, ATM mutation heterozygotes were found to have double fold more breast cancer chance comparable to the normal population. This likelihood was upgraded 5-fold in women below age of 50 (Swift et al., 1991), but another study concluded that heterozygous ATM mutations do not confer genetic predisposition to early onset of breast cancer (FitzGerald et al., 1997). Of more than 1,500 mutations reported from A-T patients listed in the ATM Mutation Database (www.lovd.nl/ATM) greater than 80% are predicted to truncate the protein with no obvious clustering in specific regions of the gene (Bernstein Wecare Study Collaborative Group 2017). 443 BRCA1/2 negative familial breast cancer cases and 521 controls were studied. ATM mutations known to cause ataxia-telangiectasia were found: 12 in familial breast cancer cases and 2 in controls. 37 nonsynonymous missense variants were identified, 12 of which were present in both cases and controls, 15 were present exclusively in cases and 10 were present exclusively in controls. There was no difference in frequency between cases and controls for classical missense variants S49C, F858L, P1054R, L1420F and D1853N. The relative risk of breast cancer associated with ATM deleterious mutations was estimated to be 2.4 (Ahmed and Rahman 2006). From a total of 296 consecutive genetically confirmed A-T patients, 66 patients who developed a malignant tumor; 47 lymphoid tumors and 19 non-lymphoid tumors were diagnosed. After the age of 16yrs, there were 11 lymphoid malignancies, and 13 other tumors, including 7 breast cancers (Reiman A et al. 2011). Germline mutations are found in up to 3% of hereditary breast and ovarian cancer families. In a review of breast tumors of 21 ataxia-telangiectasia family cases, ATM-associated tumors are near-tetraploid in 70% of cases and show loss of heterozygosity (67%) at the ATM locus. Tumors arising in ATM deleterious variant carriers are not associated with increased large-scale genomic instability. ATM-associated tumors are distinct from BRCA1-associated tumors in terms of morphological characteristics and genomic alterations, and they are also distinguishable from sporadic breast tumors. 97% of ATM-associated tumors were ER+. The luminal B (ER+, PR+/-, ERBB2 (HER2)- and Ki-67 ≥20%) subtype (46%) was over-represented among tumors developed by A-T participants (Renault AL et al., 2018).
    Other germline mutations: In a meta-analysis, Byrnes et al., 2008 concluded that germline mutations in ATM, BRIP1, PALB2 and CHEK2, that are known to interact with BRCA1 and BRCA2, may be associated with a high risk of breast cancer for a subset of women. Coding regions ATM, CHEK2, PALB2 and XRCC2 were analyzed in 13,087 breast cancer cases and 5,488 controls. 1,273 variants were identified in the four genes: 785 in ATM, 165 in CHEK2, 255 in PALB2 and 68 in XRCC2. PALB2 truncating variants were associated with the highest breast cancer risk, with an estimated OR=4.7; the risk for ATM and CHEK2 truncating variants were OR=3.3 and OR=3.1 respectively. There was no association between XRCC2 truncated variants and breast cancer risk. Truncating ATM variants were more common in breast cancer cases with a family history of breast cancer, and with ER+ cases. Missense variants in ATM, CHEK2 and PALB2 did not seem to contribute to breast cancer risk (Decker et al., 2017). A large study involving more than 600,000 cancer patients with various cancers found that 0.7% patients had an ATM pathogenic variant. In particular, 7271T>G was associated with higher invasive ductal breast cancer risk than other missense and truncating ATM pathogenic variants. Low-to-moderate risks were seen for ductal carcinoma in situ, male breast cancer, and ovarian cancer. 7271T>G is associated with high risk for breast cancer, with a 3- to 4-fold risk increase. Carriers are eligible for increased breast and pancreatic cancer screening for prevention and/or early detection (Hall et al., 2021). A meta-analysis was performed concerning the association between ATM variants and the risk of breast cancer. The OR of this association was estimated at 1.7- 2.3. V2424G variant (c.7271T>G) was the most associated with breast cancer incidence (Moslemi et al., 2021)
    Expression: ATG4C is responsible for autophagic activity. Downregulation of ATM expression induces a decrease in the autophagic flux. ATM expression regulates ATG4C levels. Positive correlation between ATM and ATG4C expression was found in all subtypes of breast cancer human samples (511 breast cancer samples from TCGA), except for the basal like subtype (the 4 subtypes are: 2 estrogen receptor (ER)-positive (estrogen receptors ESR1 and ESR2)subtypes, with either low (luminal A) or high (luminal B) expression of proliferation-related genes, a subtype enriched for ERBB2 (HER2)-amplified tumors, and the " basal " or triple-negative subtype (ER-, PR-, ERBB2-) (Antonelli et al., 2017). In a series of n=1106 samples (with ER (estrogen receptor) status (n=1055) positive 79%/negative 21%, PGR (progesterone receptor ) status (n=1052) positive 67%/negative 33%, ERBB2 (n=1066) positive 13%/negative 87%, ER/PR/ERBB2 (n=1013) Positive for at least one 86%/triple negative 14%, immunohistochemical TP53 positive 21%/negative 79%), ATM protein expression was reduced more frequently among BRCA1 (33%) and BRCA2 (30%) tumors than in non-BRCA1/2 tumors (11%). In a series of 1013 non-BRCA1/2 cases, ATM was more commonly deficient (20%) and TP53 was overabundant (47%). The non-BRCA1/2 tumors with reduced ATM expression were more often ER-, PR- and were of higher grade (Tommiska J et al., 2008).
    Entity name
    Ovarian carcinoma and Fallopian tube carcinoma
    Note
    High grade serous ovarian cancer is the most common epithelial ovarian cancer subtype.
    Germline mutations: Pathogenic germline BRCA1, BRCA2, and several other gene variants predispose women to primary ovarian, fallopian tube, and peritoneal carcinoma, classified as high-grade serous carcinoma. Pathogenic variants of 11 genes were identified in 41 (18%) women: 19 (8%; BRCA1), 8 (4%; BRCA2), 6 (3%; mismatch repair genes), 3 (1%; RAD51D), 2 (1%; ATM) (Hirasawa et al., 2017). Germline mutations in 174 cases of extrauterine high-grade serous carcinomas (located in the fallopian tube, ovary, or peritoneum) were studied. 79% of tumors were high-grade serous ovarian carcinoma (n=138), and the most common mutations in high-grade serous carcinomas were TP53 (94%), BRCA1 (25%), BRCA2 (11%), and ATM (7%). ATM mutations were found in high-grade serous carcinoma (6 of 138 cases), endometrioid carcinoma (2 of 12 cases), and clear cell carcinoma (1 of 10 cases). (Ritterhouse et al., 2016). Germline pathogenic variants: a large study involving more than 600,000 cancer patients with various cancers found that 0.7% patients had an ATM pathogenic variant. Low-to-moderate risks were seen for ovarian cancer (OR= 1.57; 95% CI, 1.35-1.83) (Hall et al., 2021).
    Germline/somatic mutations: 390 ovarian carcinomas were screened. Thirty-one percent of ovarian carcinomas had a deleterious germline (24%) and/or somatic (9%) mutation in one or more of the 13 homologous recombination genes: BRCA1, BRCA2, ATM, BARD1, BRIP1, CHEK1, CHEK2, ABRAXAS1, MRE11, NBN, PALB2, RAD51C, and RAD51D. Germline and somatic mutations in ATM were present respectively in 0 (0%) and 2 (6%) of cases (Pennington et al. 2014).
    Somatic mutations: Among 207 ovarian cancer patients, ATM somatic mutation was more frequently detected in clear cell carcinomas (9%) and endometrioid carcinomas patients (18%) than in high-grade serous carcinomas patients (4%) (Sugino et al., 2019).
    Expression: Wildtype ATM is upregulated in high grade serous ovarian cancer patients compared to normal fallopian tube tissue, as indicated by increased S1981 autophosphorylation (Chen et al., 2020). Seventeen primary serous ovarian cancers were studied; a worse outcome was found in patients with low EZH2 and high-ATM-expressing tumors, compared with patients with low EZH2 and low-ATM-expressing tumors. In the group with low EZH2 expression, the median survival was higher in low-ATM than in high-ATM (20 months vs. 14 months (Naskou et al., 2020).
    Entity name
    Uterine cancers
    Note
    One-hundred forty-one primary uterine cervical lesions and corresponding normal tissues were collected. In cervical intraepithelial neoplasia, comparable frequency of alterations (deletion/methylation) was seen in CADM1 (24%) and ATM (21%). In cervical carcinoma, higher alterations were seen in CADM1 (59%) and ATM (49%). In primary cervical carcinoma, high reduction in expression of CADM1, ATM, and PPP2R1B were found. Low expression and high alterations (55-59%) of ATM, CADM1 or CHEK1 was correlated with poor outcome (Mazumder Indra et al., 2011).
    Somatic mutations in ATM were found in 2 of 7 uterine leiomyosarcomas, 2 of 7 endometrial stromal sarcomas, and 1 of 5 uterine carcinosarcomas (da Costa et al,. 2021.). In another study, 5 mutations were found in 4 of 25 cases of leiomyosarcomas (Lee et al., 2017). ATM deletions were found in 16 of 20 leiomyosarcomas (Ul-Hassan et al., 2009).
    Entity name
    Testicular germ-cell tumors
    Note
    Expression: Among testicular germ-cell tumors, high levels of S1981-phosphorylated ATM (pS-ATM) was observed especially in embryonal carcinomas, less in seminomas, rarely in teratomas or carcinoma in situ. However, it was lower in testicular germ-cell tumors than is known in carcinomas (Bartkova et al., 2005).
    Entity name
    Prostate cancer
    Note
    Germline mutations: Hereditary factors: The proportion of all prostate cancers due to high-risk hereditary factors is about 8-12%. Germline mutations/alterations associated with elevated risk of prostate cancer were: BRCA1 1.8-3.8 fold increased relative risk; BRCA2 2.5-4.6 fold increased relative risk
    Prognosis
    PARP1 inhibitors were found to be cytostatic, but not cytotoxic in ATM-deficient cancer cell lines and combination of PARP1 inhibitors with an ATR inhibitor is needed to induce cell death. Inhibition of ATR potentiates the effects PARP1 inhibition (Jette et al., 2020). Trials using PARP1 inhibitors are reviewed in Nizialek and Antonarakis 2020. The PARP1 inhibitors are now approved for prostate cancer. The combination of anti-hormonal therapy with DNA damage response inhibitors also has a potential enhanced efficacy (Wengner et al., 2020).
    Recommendations for prostate cancer early detection in carriers of high-risk mutations are the following: Begin screening at age 40 yrs: annual PSA dosage and digital rectal examination (men with a high PSA level are at higher risk for prostate cancer and aggressive prostate cancer.); If PSA is low and no other indication for biopsy, repeat screening in 12 months. If PSA is high, recheck PSA; if increased, consider biopsy (Cheng et al., 2019).
    Entity name
    Bladder urothelial carcinoma
    Note
    Germline mutations: In a study of 1,038 patients with urothelial carcinoma, 24% harbored pathogenic germline variants. MLH1 and MSH2 were considered as urothelial carcinoma risk genes, and ATM (germline variant found in 13 of 827 cases) and BRCA2 (18 of 867 cases) were found as potential urothelial carcinoma predisposition genes (Nassar et al., 2020).
    Somatic mutations: Another study included 53 patients with urothelial carcinoma. 11% (6/53) of patients harbored ATM alterations in the tumor. ATM somatic alterations were associated with a significantly shorter overall survival (median survival 18 months vs 39 months) (Joshi et al., 2020). From an immunotherapy cohort (n = 210) and The Cancer Genome Atlas (TCGA)-Bladder cancer cohort, a series of analyses was performed to evaluate the prognostic value of ATM in bladder cancer immunotherapy. It was found that bladder cancer patients with ATM somatic mutation had greater benefit from Immune checkpoint inhibitors. ATM mutation significantly upregulated the number of DNA damage repair pathway gene mutations. ATM mutations resulted in increased bladder cancer sensitivity to 29 drugs, including an IGF1R inhibitor (Yi et al., 2020). Mutations of DNA repair genes, e.g. ATM/RB1, are frequently found in urothelial cancer and have been associated with better response to cisplatin-based chemotherapy. Overall, 31 out of 130 patients (24%) had somatic mutations in either ATM (19/130) or RB1 (ATM/RB1) genes in a TCGA dataset, while 18 out of 81 patients (22%) had mutations in hospitals dataset (with 12/81 ATM mutations). ATM/RB1 mutations may be a biomarker of poor prognosis in unselected UC patients (Yin et al., 2018).
    Entity name
    Renal cell carcinomas
    Note
    Polymorphisms: (germline genetic variants): A total of 2,657 renal cell carcinoma (75% clear cell renal cell carcinoma, 8% papillary renal cell carcinoma, 17% other histology types) cases and 5,315 healthy controls were studied. Two single nucleotide polymorphisms (SNPs) that map to PIK3CG and ATM (rs611646:T, located in an intron of ATM) were significantly associated with renal cell carcinoma risk (Shu et al., 2018).
    Somatic mutations: 229 patients with metastatic clear cell renal cell carcinoma were included. The most frequently altered genes were CHEK2 (n=10; including three somatic and seven germline), ATM (n=8; all somatic), MSH6 (n=4; including three somatic and one germline) and MUTYH (n=4; all germline) (Ged et al., 2020). A total of 110 patients were selected in a study concerning the expression level of ATM. The expression of ATM in clear cell renal cell carcinoma is significantly lower than that in adjacent normal tissues. Further analysis found that the expression of ATM in the clear cell renal cell carcinoma tissues above grade II was lower than that of grade II or below. The survival time of the ATM low expression group was significantly shorter than that of the ATM high expression group (Ren et al., 2019).
    Entity name
    Lung cancers
    Note
    Polymorphisms: ATM gene polymorphisms was evaluated in 616 lung cancer patients ( adenocarcinomas 27%, squamous cell carcinomas 39%, other non-small cell lung carcinoma 18%, small cell lung carcinoma 16%) and 616 controls. Subjects with the A allele at the site (IVS62+60G>A) have significantly higher risk of lung cancer than those with the G allele. Looking at the haplotypes of four ATM single nucleotide polymorphism sites (-4518A>G, IVS21-77C>T, IVS61-55T>C and IVS62+60G>A), the ATTA haplotype showed significantly increased risk of lung cancer (Kim JH et al., 2006). In a meta-analysis showed that the polymorphism rs189037 (G>A) was associated with the risk of lung cancer and breast cancer (Zhao et al., 2019). Moreover, the homozygote AA variant allele was found to be significantly related with the prognosis of lung cancer (Bhowmik et al., 2015). A comprehensive meta-analysis of 14 studies including 4,731 cases of small and non-small cell lung carcinomas and 5,142 controls was conducted to evaluate the association between ATM gene polymorphisms and both susceptibility to lung cancer. ATM rs189037, rs664677 and rs664143 gene polymorphisms are risk factors for lung cancer (Yan et al., 2017). ATM rs189037 (G>A), located at the 5UTR of its promoter, is an important variant. Four eligible studies on lung carcinomas were included for meta-analysis; the histological type is however not defined in one of these studies. ATM rs189037 AA carriers had more risk of lung cancers than wild-type carriers. Specially, the association was more notable in non-smokers whereas no association of this variant with lung cancer risk was found in smokers (He et al., 2019). Homozygous variants rs227060 and rs170548) were associated with elevated risk for non-small cell lung carcinoma (Yang et al., 2007). In a case-control study, 852 "lung cancer" patients and 852 healthy controls, individuals carrying variant AA genotype of rs189037 had higher lung cancer risk (Liu et al. 2014). In a study of 720 non-small cell lung cancer patients (22% with squamous cell carcinoma, and 50% with adenocarcinomas) that have undergone radiation or chemo-radiation therapies: patients with rs664143 GA or AA genotype and patients with rs189037 AG/GG had poorer overall survival (Mou J et al., 2020).
    Smokers: Variant rs652311 may enhance the effect of smoking on lung cancer development and thereby increase lung cancer risk in smokers (Hsia et al., 2013). Polymorphisms rs189037, rs228597, rs228592, rs664677, rs609261, rs599558, rs609429, rs227062, and rs664982) were significantly associated with lung cancer among never-smokers, but not among smokers (Lo et al., 2010).See also the review by Xu et al., 2017).
    Somatic mutations: 188 primary lung adenocarcinomas were studied and 623 candidate genes were screened for somatic mutations. 1,013 non-synonymous somatic mutations were identified, including 14 ATM mutations in 13 tumors. Altogether, the genes most frequently mutated were: TP53 6.6%, KRAS 6.1%, STK11 3.4%, EGFR 3.4%, LRP1B 1.7%, NF1 1.6%, ATM 1.4%, APC 1.3%, EPHA3 1.1%, and PTPRD 1.0. A negative correlation between mutations in ATM and TP53 was detected, suggesting that mutations in ATM and TP53 may be independently sufficient for the loss of cell-cycle checkpoint control (Ding et al., 2008). In a vast multiple-cohort study, co-mutation of both ATM and TP53 has been shown to occur in 3-4% of non-small cell lung cancer. No significant differences in the TP53 and ATM comutation frequency was observed within the histologic subtypes (adenocarcinoma vs squamous cell carcinoma). ATM and TP53 comutation correlated with better response to immune checkpoint inhibitors therapy (Chen et al., 2019). Simple nucleotide variation (SNV), transcriptome profiling, copy number variation (CNV) and clinical data of patients were downloaded using TCGAbiolinks R package. EGFR, MGA, SMARCA4, ATM, RBM10, and KDM5C genes were found to be mutated only in lung adenocarcinoma, but not in lung squamous cell carcinoma. CDKN2A, PTEN, and HRAS genes are mutated only in lung squamous cell carcinoma samples. Both lung adenocarcinoma cases and lung squamous cell carcinoma cases have important gene alterations such as CDKN2B deletions (Zengin and Onal-Suzek, 2021).
    Expression: ATM is highly expressed in cisplatin-resistant non-small cell lung cancer cell lines. ATM enhances epithelial-to-mesenchymal transition (EMT) and metastatic potential via upregulation of CD274 (also called PD-L1), through JAK1,2/STAT3 pathway, in cisplatin-resistant non-small cell lung cancer cells, and inhibition of ATM suppresses tumor metastasis in xenograft mouse models (Shen et al. 2019). Methylation data in TCGA datasets revealed significant negative correlations between ATM promoter methylation and ATM gene expression in lung adenocarcinoma, and colon adenocarcinoma (Jette et al., 2020).
    Pulmonary neuroendocrine tumors: A total of 130 mutations were found in 29 genes and 49 patients (17 typical carcinoids, 17 atypical carcinoids, 19 large-cell neuroendocrine carcinomas, and 17 small-cell lung cancers. Four out of five ATM-mutated patients (1 large-cell neuroendocrine carcinoma) and 4 small-cell lung cancers, no carcinoid) showed an additional alteration in TP53, which was by far the most frequently altered gene (28 out of 130; 22%). Correlations between tumor type and grade for ATM and TP53-mutated patients were found. Both mutated genes were also associated with lymph node invasion and distant metastasis. The mutation frequency of APC and ATM in high-grade neuroendocrine lung cancer patients was associated with progression-free survival (Vollbrecht et al., 2015).
    Prognosis
    One hundred and sixty five non-small cell lung cancer samples (adenocarcinoma 54%, squamous cell carcinoma 29%, bronchoalveolar carcinoma 11%, other 6%) were analyzed for ATM expression. There was a moderate overexpression in tumor tissue and/but ATM loss was identified in 22% of patients. Compared to patients with a high malignant cell-specific ATM expression, patients with a low ATM expression had worse overall survival. Patients with low ATM expression treated with platin based perioperative treatment showed a strong trend toward improved disease free survival. This suggests that low ATM expression may be predictive of benefit from adjuvant platin based treatment (Petersen et al., 2017).
    Entity name
    Brain tumors
    Note
    Polymorphisms: Five hundred and three meningioma cases, and 1,555 controls were analyzed in association with five polymorphisms in ATM (rs228599, rs3092992, rs664143, rs170548, rs3092993). Haplotypes were constructed using the HAPLOSTAT program. The haplotype analysis in ATM revealed an increased frequency of the 1-1-1-2-1 haplotype (34%) (Malmer et al., 2007). Pediatric brain tumors (93 pilocytic astrocytoma, 13 diffuse astrocytoma, 11 anaplastic astrocytoma, 20 "other gliomas", 19 ependymoma, 49 "other specified intracranial neoplasms" such as germinomas or dysembryoplastic neuroepithelial tumors, 26 ganglioglioma, 16 "unspecified intracranial neoplasm") were investigated. An increased risk of the non-astrocytoma subtypes (ependymomas u2026) was associated with the ATM single nucleotide polymorphisms rs170548 as well as polymorphisms in EGFR, EME1, and BICRA (Adel Fahmideh et al., 2016).
    Germline/somatic mutations: Genomic analysis of chordoid meningiomas from 30 patients was performed. Mutations in NF2 was detected in 18 cases (60%), LRP1B in 30%, TRAF7 in 27%, NF2 in 20%, ATM in 7% (1 germline and 1 somatic), in other genes in 47% (Georgescu et al., 2020).
    Somatic mutations: Tumors from 37 patients with pediatric low grade gliomas (31 pilocytic astrocytomas, 4 pleomorphic xanthoastrocytomas and 2 diffuse gliomas) were analyzed. Genetic alterations were found in 97% of cases, The KIAA1549 / BRAF fusion was the most common alteration (57%) followed by AFAP1/NTRK2 (5%) and TBL1XR1/PIK3CA (5%) fusions that were observed at much lower frequencies. The most frequently mutated genes were NOTCH genes (19%), ATM (11%), RAD51C, RNF43, SLX4, and NF1 (Mobark et al., 2020).
    Expression: In a study of 95 gliomas of different grades, the methylation index of a set of genes was tested. RASSF1A, RUNX3, GATA6, and MGMT were most frequently methylated, whereas the CDKN2A-CDKN2B locus, PTEN, RARB, and ATM were methylated to a lesser extent (Majchrzak-Celinska et al., 2015). One of the most important inactivation mechanisms of ATM gene is promoter methylation. 30 cases of different types of brain tumors (14 medulloblastoma, 6 astrocytoma, 7 glioblastoma multiforme (7), 3 others) were studied. ATM promoter was methylated in 73% of patients (Mehdipour et al., 2015). In a sample of 52 brain tumors, ATM, CCND2, TP53 and RB1 had higher expression in astrocytoma than in meningioma tumors. Higher grade astrocytoma tumors had up-regulation for CCND2 and ATM (Kheirollahi et al., 2011).
    Prognosis
    To identify the association between ATM somatic mutations and improved radio-sensitivity, 39 IDH-wildtype high-grade glioma (2 diffuse astrocytoma, 10 anaplastic astrocytoma, and 27 glioblastoma) were studied. ATM mutations were detected in 26% of cases (6 glioblastoma and 4 anaplastic astrocytoma cases). ATM mutations might be involved in the increased radio-sensitivity (Kim et al., 2020). Silencing of ATM expression via siRNA technique was found to improve radiosensitivity of glioma stem cell in vitro and in vivo (Li et al.. 2017).
    Entity name
    Skin Melanoma
    Note
    Germline mutations: A large study involving more than 600,000 cancer patients with various cancers found that 0.7% patients had an ATM pathogenic variant. c.7271T>G was associated with higher melanoma risk (OR= 1.46; 95% CI, 1.18-1.81) (Hall et al., 2021).
    Expression: A total of 366 melanoma patients (230 primary melanoma and 136 metastatic melanoma) and 59 cases of nevi (27 normal nevi and 32 dysplastic nevi) were studied. Phosphorylated ATM at serine-1981 was explored. Both loss of phospho-ATM expression, and high phospho-ATM expression were associated with progression of melanoma from normal nevi to metastatic melanoma (normal nevi: 100% moderate expression, dysplastic nevi 81%, primary melanomas 73%, metastatic melanomas 60%). High phospho-ATM expression was correlated to the worse outcome at 5 and 10yrs, negative phospho-ATM expression was correlated with an intermediate survival at 5yrs, but a poor survival at 10yrs, comparable to high phospho-ATM expression cases. Moderate phospho-ATM expression was correlated with the best outcome (Bhandaru et al., 2015).
    Entity name
    Uveal melanoma
    Note
    Expression of ATM was investigated in 69 choroid melanoma samples. Loss of ATM was observed in 65% of cases. Loss of ATM was associated with a poor prognosis. (Jha et al., 2019).
    Entity name
    Soft tissue sarcomas
    Note
    Somatic mutations: Somatic mutations of ATM were found in 2 of 43 cases of myxofibrosarcoma and in 1 of 15 cases of undifferentiated spindle cell sarcoma, but not in undifferentiated pleomorphic sarcomas (0 of 18 cases) nor in dedifferentiated liposarcoma (0 of 6) (Lewin et al., 2018). Somatic mutations were found in 7 of 86 patients with liposarcoma (Kanojia et al., 2015).
    Expression: 17 cases of rhabdomyosarcoma specimens were studied and 7 of the 17 cases were negative for ATM expression (41%) (2 embryonal rhabdomyosarcoma, 1 alveolar rhabdomyosarcoma, and 4 "unknown subtype") (Zhang et al., 2003).
    Note
    Atm-knockout in p210Bcr/Abl1 transgenic mice resulted in the acceleration of the blast crisis, which supports that the DNA damage-response pathway plays a vital role for determination of susceptibility to blast crisis in CML (Takagi et al., 2013).
    Disease
    Chronic myelogenous leukemia (9975/3) is a myeloproliferative neoplasm
    Entity name
    Acute myeloid leukemias
    Note
    MIR100 was highly expressed in bone marrow of pediatric acute myeloid leukemia (AML) patients and cell lines, and MIR100 depletion inhibits cell viability and induces cell apoptosis. The expression of ATM was downregulated in bone marrow of AML patients and AML cell lines whereas ectopic expression of ATM repressed cell viability while enhanced apoptosis. ATM is inhibited by MIR100 (Sun et al., 2020). MIR181A1 was overexpressed in pediatric AML, which showed an inverse association with ATM expression (Liu et al., 2016). 33 cases of blastic plasmacytoid dendritic cell neoplasm (9727/3) were studied. Somatic point mutations were found in NRAS (27% of cases), ATM (21%), MET, KRAS, IDH2, KIT (9% each), etcu2026
    Disease
    Acute myeloid leukaemia (AML) and related precursor neoplasms
    Entity name
    Acute lymphoblastic leukemia
    Note
    A series of 57 childhood acute lymphoblastic leukemia (ALL) cases (26 B-precursor ALL and 31 T-ALL) were studied. 28 distinct ATM alterations were found in 14 patients (25%). Six alterations of potential biological significance were observed in 5 cases of B-precursor ALL (19%), and 5 were found in 3 cases of T-ALL (10%). In two cases of B-precursor ALL cases, the ATM alterations were germline (Gumy Pause et al., 2003). Chromothripsis was found in ALLs developing in patients with Ataxia Telangiectasia inherited disease (Ratnaparkhe et al., 2017).
    Disease
    Acute lymphoblastic leukemia is a precursor lymphoid neoplasm
    Entity name
    Chronic lymphocytic leukemia
    Note
    Germline mutations: 516 cases plus additional cohorts of 106 exomes and24 genomes were studied. Two genes were significantly associated with chronic lymphocytic leukemia (CLL): CDK1 and ATM (Tiao et al., 2017).
    Germline/somatic mutations: In early studies, six CLL cases were studied. Both germ-line and somatic ATM mutations in CLL were found. The observed AT heterozygotes frequency of 6% in CLL was greater than the 1% estimate in the general population. 41 patients with CLL were studied for loss of heterozygosity. 14% had heterozygosity in ATM and a mutation in the second allele. Patients with ATM deficiency had significantly shorter survival times (Starostik et al. 1998; Bullrich F et al. 1999). 16 of 50 B-CLLs analyzed had ATM mutations (7 biallelic; 3 germline, 11 somatic, 6?), and 6 had mutations in TP53. All 16 ATM mutant B-CLLs showed the absence of somatic variable region heavy chain hypermutation indicating a pregerminal center cell origin (Stankovic T et al. 2002). Fifteen of 56 patients analyzed (27%) showed a pattern compatible with the presence of a somatic mutation (Lahdesmaki et al., 2004). Del(11q) can be found in 10% of patients with CLL in early stage and 20-25% in patients with advanced disease. About 40% of patients with a del(11q) have an inactivating mutation of the second ATM allele and these cases show a poor chemotherapy response. In addition, patients carrying a del(11q) clone typically show rapid progression, and reduced overall survival (review in Knittel et al., 2015). Samples from 54 patients with CLL were used. Twelve somatic mutations and 15 germline mutations were found. In the 12 CLL samples with somatic mutations, 8 were deficient in ATM function, while only one of the 15 CLL samples with only germline mutations had diminished ATM function, indicating that the germline mutations have minimal impact on ATM activity. Seventeen of the 26 samples with mutations retained ATM function. Patients with deficient ATM function had lesser progression-free survival than those with normal ATM function (Jiang et al., 2016). Patients with biallelic ATM alteration had shorter overall survival than patients with isolated del(11q), similar to patients with delTP53 (Lozano-Santos et al., 2017).
    Somatic mutations: 1,043 CLL patients were studied; 42% patients displayed abnormal karyotypes, and 10% had complex karyotypes. The group with complex karyotypes included patients with more advanced disease at diagnosis, including a higher proportion of patients diagnosed at Binet stage B/C, delATM (25% vs 7% in not-complex karyotypes) and delTP53 (40% vs 5%). Median overall survival was 87 months in delATM patients, 79 months in complex karyotypes patients, and 56 months in delTP53 patients. A subgroup of CLL patients with complex karyotypes lacking "high-risk-FISH abnormalities" (namely: delATM/delTP53) showed an equivalent impaired clinical evolution as those with high-risk-FISH abnormalities and no complex karyotypes (Puiggros et al., 2017). A cohort of 249 CLL patients was studied. ATM mutations were found in 19%. Short telomeres were significantly associated with both reduced time-to-first-treatment and overall survival in subset #2 (Note: focusing on rearrangements of IGH genes, subset #2 is defined by: mutational status: mostly mutated, IGHV: V3-21, IGHJ: J6, VH CDR3 length 9, pattern: [AVLI]x[DE]xxxM[DE]x, see Agathangelidis et al, 2012. The highest ATM mutation frequency was observed in subset #2, and was associated with particularly short telomeres (Navrkalova et al., 2016); In 499 CLL cases, ATM mutations were observed in 37 cases (7%), without evidence of any mutational hotspots. BIRC3, POT, BRAF, XPO1 and KRAS were also mutated in 7 to 6% of cases. Biallelic BIRC3 deleted patients had reduced overall survival in comparison to sole del(11q) patients, while ATM abnormalities did not significantly differ in median survival times compared with sole del(11q) cases (Blakemore et al., 2020). Unmutated IGHV genes are strongly associated with poor survival, while Binet stage A patients with mutated IGHV genes have a much better prognosis. 150 samples from patients with CLL were analyzed. Tumor mutational burden (or mutational complexity) can be evaluated from an eight gene estimator that comprises ATM, SF3B1, NOTCH1, BIRC3, XPO1, MYD88, TNFAIP3, and TP53. Median tumor mutational burden was 1.75. ATM, SF3B1, NOTCH1 and BIRC3 were the genes the most frequently mutated. Tumor mutational burden evaluated from the eight gene estimator was significantly higher in high risk (del(11q), del(17p) or complex karyotype) than in low risk (isolated del(13q) or normal karyotype) cytogenetic categories and was strongly associated with cytogenetic complexity. Any mutation in this set of 8 genes is associated with poor prognosis cytogenetics (eg, del(11q), del(17p) and complex karyotype) as well as with unmutated IGHV genes. Tumor mutational burden also predicted shorter treatment-free survival even in Binet stage A patients or patients with a good prognosis karyotype. These results indicate that Binet stage, IGHV mutational status, and tumor mutational burden identify patients at diagnosis who will need to be rapidly treated despite a clinically non-progressive disease. Tumor mutational burden could also help to predict evolution of patients in Binet stage A or with good prognosis cytogenetics (Chauzeix et al., 2020).
    Disease
    Chronic lymphocytic leukemia (9823/3) is a mature B-cell neoplasm.
    Oncogenesis
    Combination Sf3b1 mutation with Atm deletion in mouse B cells leads to the overcoming of cellular senescence and the development of CLL-like disease in elderly mice (Yin et al., 2019).
    Note
    Atm deficiency promotes development of murine B-cell lymphomas that resemble diffuse large B-cell lymphoma (Hathcock et al., 2015).
    Disease
    Diffuse large B-cell lymphoma (9680/3) is a mature B-cell neoplasm.
    Note
    ATM is the most frequently mutated gene in mantle cell lymphoma (MCL) (about 50%) (Swerdlow et al , 2008). ATM mutation in MCL is mostly of somatic origin.
    In an early study, 8 ATM gene mutations were detected in 7 of 20 patients with mantle cell lymphoma. Somatic origin was demonstrated in 3 cases, and one mutation was germline. Chromosomal imbalances were significantly higher in typical MCL with ATM inactivation (Camacho et al., 2002). ATM and TP53 mutations in 72 MCL patients were analyzed. Mutated ATM and TP53 alleles were found in 51% and 22% respectively, with only three patients harboring mutations in both genes. Mutated TP53 gene was associated with a reduced overall survival (Mareckova et al., 2019). A total of 552 patients samples from six studies published in the literature were reviewed. Somatic mutations in ATM (40-50%), CCND1 (14-35%), TP53 (7-31%), KMT2D (12-20%), KMT2C (16%), NOTCH1 (5-14%), NSD2 (7-13%), BIRC3 (5-10%), UBR5 (7-18%) were most frequently encountered in mantle cell lymphoma (Ahmed et al. 2016). Data were extracted from 2,045 MCL patients in another meta-analysis of 32 selected articles. The mutations detected are likely to be somatic, but data is missing concerning the possibility of germline mutations. ATM was the most frequently mutated gene (44%) at diagnosis, followed by IGH (38%), TP53 (27%), CDKN2A (24%), MYC (21%), and CCND1 (20%). During disease progression the level of mutations increases. The highest increases were found in TP53 (+16%), ATM (+14%), KMT2A (+13%), MAP3K14 (+12%), BTK (+12%), TRAF2 (+11%). (Hill et al, 2020).
    Disease
    Mantle cell lymphoma (9673/3) is a mature B-cell neoplasm.
    Prognosis
    Mantle cell lymphoma (MCL) cells deficient in both ATM and TP53 are more sensitive to PARP inhibitors than cells lacking ATM function alone. Combining inhibitors of PARP and ATM may have utility in TP53-deficient MCL (Williamson et al., 2012)
    Note
    Absence of ATM deletions in 16 cases of splenic marginal-zone B-cell lymphoma was noted (Salido et al., 2003).
    Disease
    Splenic marginal-zone B-cell lymphoma (9689/3) is a mature B-cell neoplasm.
    Entity name
    Non-Hodgkin lymphomas
    Note
    Twenty-seven cases of childhood non-Hodgkin lymphoma (NHL) were screened. ATM alterations were detected in 12 cases. there were 13 B-cell NHL NOS (with 2 pathogenic mutations, 2 polymorphisms); 7 Burkitt lymphoma (2 pathogenic mutations, 1 reduced protein expression, 2 polymorphisms); 1 diffuse large B-cell lymphoma (1 pathogenic mutation); 2 T-lymphoblastic lymphoma (1 pathogenic mutation); 2 T-cell non-Hodgkin lymphoma (1 polymorphism), and 1 lymphoblastic lymphoma and 1 NHL-NOS. ATM methylation status was normal (Gumy-Pause et al., 2006). In a population of 1,297 NHL cases and 1,946 controls, ATM polymorphisms were associated with diffuse large B-cell lymphoma (412 cases) and CLL (164 cases), in particular rs227060, but not with follicular lymphomas (301 cases) (Rendleman et al., 2014).
    Disease
    Non-Hodgkin lymphomas are either mature B-cell neoplasms or mature T- and NK-cell neoplasms.
    Entity name
    T-cell acute lymphoblastic leukemia
    Note
    Somatic inactivation of Atm in hematopoietic cells predisposes mice to T-cell acute lymphoblastic leukemia and Ccnd3 was necessary for initiation of T-ALLs (Ehrlich et al., 2015).
    Disease
    T-cell acute lymphoblastic leukemia (9827/3) is a mature T- and NK-cell neoplasm.
    Note
    Mutations of ATM was found in 17 of 37 patients with T prolymphocytic leukemia (T-PLL). Two of seventeen mutated T-PLL samples had a previously reported A-T allele (Vorechovsky et al., 1997). In a panel of eight cases of T-cell prolymphocytic leukemia, there were structural lesions of ATM in all samples, and in four samples both alleles were affected (Yuille et al., 1998). Among 25 T-cell prolymphocytic leukemia cases, gain of MYC (67%), loss of ATM (64%), and gain (75%) and/or rearrangement (78%) of TCL1A were found (Hsi et al., 2014). In another study, 70% of 40 T-PLL samples harbored somatic mutations in the tumor suppressor ATM (Kiel et al., 2014).
    Disease
    T-cell prolymphocytic leukemia (9834/3) is a mature T- and NK-cell neoplasm.
    Note
    Matched peripheral T cell lymphomas and non-malignant samples from 12 patients revealed 104 somatic mutations. Somatic mutations in ATM were found in 5 out of the 12 patients (Simpson et al., 2015). Mutations in ATM were also found in a panel of 125 peripheral T cell lymphomas (Palomero et al., 2014).
    Disease
    Peripheral T cell lymphoma (9702/3) is a mature T- and NK-cell neoplasm.
    Note
    The expression of ATM in nasal extranodal NK/T cell lymphoma was decreased compared with that in nasal benign lymphoid proliferative disease (Ye et al., 2021). Phospho-ATM (pATM) (Ser1981) was distinguishingly expressed in eleven of the 12 cases of extranodal natural killer/T cell lymphoma, while pATM was negative in all normal lymph nodes (Sui et al., 2020).
    Disease
    Extranodal NK/T cell lymphoma is a mature T- and NK-cell neoplasm.
    Entity name
    Note
    Tumors were obtained from 23 children with Hodgkin disease. Two patients (9%) had germline mutation, and both had a more aggressive disease. Four out of 12 informative patients exhibited loss of heterozygosity centromeric to ATM locus. Polymorphisms were also detected (Liberzon et al., 2004).

    Bibliography

    Reference NumberPubmed IDLast YearTitleAuthors

    Other Information

    Locus ID:

    NCBI: 472
    MIM: 607585
    HGNC: 795
    Ensembl: ENSG00000149311

    Variants:

    dbSNP: 472
    ClinVar: 472
    TCGA: ENSG00000149311
    COSMIC: ATM

    RNA/Proteins

    Gene IDTranscript IDUniprot
    ENSG00000149311ENST00000278616Q13315
    ENSG00000149311ENST00000278616A0A024R3C7
    ENSG00000149311ENST00000452508Q13315
    ENSG00000149311ENST00000452508A0A024R3C7
    ENSG00000149311ENST00000526567Q6P7P1
    ENSG00000149311ENST00000527805E9PIN0
    ENSG00000149311ENST00000527891E9PIQ5
    ENSG00000149311ENST00000529588H0YEC6
    ENSG00000149311ENST00000530958A0A087X0E9
    ENSG00000149311ENST00000531525H0YDU7
    ENSG00000149311ENST00000532931E9PRG7
    ENSG00000149311ENST00000601453M0QXY8
    ENSG00000149311ENST00000638443Q6P7P1
    ENSG00000149311ENST00000639240Q6P7P1
    ENSG00000149311ENST00000639953Q6P7P1
    ENSG00000149311ENST00000640388Q6P7P1

    Expression (GTEx)

    0
    5
    10
    15
    20
    25

    Pathways

    PathwaySourceExternal ID
    Homologous recombinationKEGGko03440
    Cell cycleKEGGko04110
    p53 signaling pathwayKEGGko04115
    ApoptosisKEGGko04210
    Homologous recombinationKEGGhsa03440
    Cell cycleKEGGhsa04110
    p53 signaling pathwayKEGGhsa04115
    ApoptosisKEGGhsa04210
    HTLV-I infectionKEGGko05166
    HTLV-I infectionKEGGhsa05166
    BRCA1-associated genome surveillance complex (BASC)KEGGhsa_M00295
    Transcriptional misregulation in cancerKEGGko05202
    Transcriptional misregulation in cancerKEGGhsa05202
    NF-kappa B signaling pathwayKEGGhsa04064
    NF-kappa B signaling pathwayKEGGko04064
    MicroRNAs in cancerKEGGhsa05206
    MicroRNAs in cancerKEGGko05206
    BRCA1-associated genome surveillance complex (BASC)KEGGM00295
    FoxO signaling pathwayKEGGhsa04068
    DNA damage-induced cell cycle checkpointsKEGGhsa_M00691
    DNA damage-induced cell cycle checkpointsKEGGM00691
    Gene ExpressionREACTOMER-HSA-74160
    Generic Transcription PathwayREACTOMER-HSA-212436
    Transcriptional Regulation by TP53REACTOMER-HSA-3700989
    Cell CycleREACTOMER-HSA-1640170
    Cell Cycle CheckpointsREACTOMER-HSA-69620
    G1/S DNA Damage CheckpointsREACTOMER-HSA-69615
    p53-Dependent G1/S DNA damage checkpointREACTOMER-HSA-69580
    p53-Dependent G1 DNA Damage ResponseREACTOMER-HSA-69563
    Stabilization of p53REACTOMER-HSA-69541
    Autodegradation of the E3 ubiquitin ligase COP1REACTOMER-HSA-349425
    p53-Independent G1/S DNA damage checkpointREACTOMER-HSA-69613
    p53-Independent DNA Damage ResponseREACTOMER-HSA-69610
    Ubiquitin Mediated Degradation of Phosphorylated Cdc25AREACTOMER-HSA-69601
    G2/M CheckpointsREACTOMER-HSA-69481
    G2/M DNA damage checkpointREACTOMER-HSA-69473
    MeiosisREACTOMER-HSA-1500620
    Meiotic recombinationREACTOMER-HSA-912446
    DNA RepairREACTOMER-HSA-73894
    Cellular responses to stressREACTOMER-HSA-2262752
    Cellular response to heat stressREACTOMER-HSA-3371556
    Regulation of HSF1-mediated heat shock responseREACTOMER-HSA-3371453
    Cellular SenescenceREACTOMER-HSA-2559583
    DNA Damage/Telomere Stress Induced SenescenceREACTOMER-HSA-2559586
    DNA Double-Strand Break RepairREACTOMER-HSA-5693532
    DNA Double Strand Break ResponseREACTOMER-HSA-5693606
    Sensing of DNA Double Strand BreaksREACTOMER-HSA-5693548
    Recruitment and ATM-mediated phosphorylation of repair and signaling proteins at DNA double strand breaksREACTOMER-HSA-5693565
    Homology Directed RepairREACTOMER-HSA-5693538
    HDR through Homologous Recombination (HR) or Single Strand Annealing (SSA)REACTOMER-HSA-5693567
    Processing of DNA double-strand break endsREACTOMER-HSA-5693607
    HDR through Homologous Recombination (HRR)REACTOMER-HSA-5685942
    Homologous DNA Pairing and Strand ExchangeREACTOMER-HSA-5693579
    Presynaptic phase of homologous DNA pairing and strand exchangeREACTOMER-HSA-5693616
    Resolution of D-Loop StructuresREACTOMER-HSA-5693537
    Resolution of D-loop Structures through Holliday Junction IntermediatesREACTOMER-HSA-5693568
    Resolution of D-loop Structures through Synthesis-Dependent Strand Annealing (SDSA)REACTOMER-HSA-5693554
    HDR through Single Strand Annealing (SSA)REACTOMER-HSA-5685938
    Nonhomologous End-Joining (NHEJ)REACTOMER-HSA-5693571
    TP53 Regulates Transcription of Cell Death GenesREACTOMER-HSA-5633008
    TP53 Regulates Transcription of Genes Involved in Cytochrome C ReleaseREACTOMER-HSA-6803204
    TP53 Regulates Transcription of Caspase Activators and CaspasesREACTOMER-HSA-6803207
    TP53 Regulates Transcription of DNA Repair GenesREACTOMER-HSA-6796648
    Regulation of TP53 ActivityREACTOMER-HSA-5633007
    Regulation of TP53 Expression and DegradationREACTOMER-HSA-6806003
    Regulation of TP53 DegradationREACTOMER-HSA-6804757
    Regulation of TP53 Activity through PhosphorylationREACTOMER-HSA-6804756
    Regulation of TP53 Activity through MethylationREACTOMER-HSA-6804760
    Platinum drug resistanceKEGGko01524
    Platinum drug resistanceKEGGhsa01524

    Protein levels (Protein atlas)

    Not detected
    Low
    Medium
    High

    PharmGKB

    Entity IDNameTypeEvidenceAssociationPKPDPMIDs
    PA110CHEK1GenePathwayassociated
    PA128406956fluorouracilChemicalClinicalAnnotationassociatedPD
    PA33744PRKAA1GenePathwayassociated22722338
    PA33745PRKAA2GenePathwayassociated22722338
    PA33746PRKAB1GenePathwayassociated22722338
    PA33747PRKAB2GenePathwayassociated22722338
    PA33751PRKAG1GenePathwayassociated22722338
    PA33752PRKAG2GenePathwayassociated22722338
    PA33753PRKAG3GenePathwayassociated22722338
    PA404CHEK2GenePathwayassociated
    PA443560Breast NeoplasmsDiseaseClinicalAnnotationassociatedPD
    PA443890Diabetes Mellitus, Type 2DiseaseClinicalAnnotationassociatedPKPD28834135
    PA449165cyclophosphamideChemicalClinicalAnnotationassociatedPD
    PA449412doxorubicinChemicalClinicalAnnotationassociatedPD
    PA450395metforminChemicalClinicalAnnotationassociatedPKPD28834135

    References

    Pubmed IDYearTitleCitations
    125568842003DNA damage activates ATM through intermolecular autophosphorylation and dimer dissociation.1118
    126126512003ATM and related protein kinases: safeguarding genome integrity.795
    195974882009Persistent DNA damage signalling triggers senescence-associated inflammatory cytokine secretion.634
    201773972010Genomic instability--an evolving hallmark of cancer.633
    234862812013The ATM protein kinase: regulating the cellular response to genotoxic stress, and more.519
    157908082005ATM activation by DNA double-strand breaks through the Mre11-Rad50-Nbs1 complex.514
    163277812006ATM- and cell cycle-dependent regulation of ATR in response to DNA double-strand breaks.449
    209662552010ATM activation by oxidative stress.386
    210349662010The ATM-Chk2 and ATR-Chk1 pathways in DNA damage signaling and cancer.381
    186575002008ATM signaling facilitates repair of DNA double-strand breaks associated with heterochromatin.356

    Citation

    Jean Loup Huret

    ATM (ataxia telangiectasia mutated)

    Atlas Genet Cytogenet Oncol Haematol. 2021-04-01

    Online version: http://atlasgeneticsoncology.org/gene/123/tumors-explorer/css/lib/bootstrap.min.css

    Historical Card

    2016-10-01 ATM (ataxia telangiectasia mutated) by  Yossi Shiloh 

    2002-11-01 ATM (ataxia telangiectasia mutated) by  Nancy Uhrhammer,Jacques-Olivier Bay,Richard A Gatti 

    jean-loup.huret@atlasgeneticsoncology.org

    1999-10-01 ATM (ataxia telangiectasia mutated) by  Nancy Uhrhammer,Jacques-Olivier Bay,Richard A Gatti 

    jean-loup.huret@atlasgeneticsoncology.org

    1998-04-01 ATM (ataxia telangiectasia mutated) by  Jean-Loup Huret 

    Genetics, Dept Medical Information, University of Poitiers; CHU Poitiers Hospital, F-86021 Poitiers, France