hSMG-1 (human suppressor with morphogenic effect on genitalia) is the most recent addition to the family of phosphatidyl-inositol-3 kinase related kinases (PIKK). This family includes proteins such as Ataxia Telangiectasia Mutated (ATM), DNA Dependent Protein Kinase (DNA-PK), ATM and Rad3 related kinase (ATR) which are involved in stress induced signal transduction, cell cycle checkpoint control and DNA damage repair. hSMG-1 was first described in Caenorhabditis elegans where it was shown to be essential for Nonsense Mediated mRNA Decay (NMD). More recently hSMG-1 has been implicated in NMD and in the DNA damage response in mammalian cells. Three hSMG-1 isoforms have been described in the literature to date. Isoform 1 is 3 657 amino acids long with isoforms 2 (3 521 amino acids) and 3 (3 031 amino acids) N-terminal truncations of isoform 1. To explore the role of hSMG-1 in the DNA damage response three separate antibodies were generated. Each antibody was raised against a different region of hSMG-1, which allowed detection of specific hSMG-1 isoforms. These hSMG-1 antibodies were found to be suitable for immunoprecipitations, immunoblotting and immunofluorescence. The specificity of the antibodies was further confirmed using mass spectrometric analysis which identified the immunoprecipitated band as hSMG-1. Using these antibodies hSMG-1 was found to be located primarily in the cytoplasm, a novel location for a PIKK protein predicted to be involved in the DNA damage response. In addition, it was found that hSMG-1 isoform 2 was the only isoform present in the cytoplasm and is the major cellular isoform. All three isoforms were detected in the nucleus, with isoforms 1 and 3 only present in this cellular compartment, consistent with a proposed role in the DNA damage response. The generation of hSMG-1 specific antibodies allowed the characterisation of endogenous hSMG-1 kinase activity, which was found to be Mn2+ dependent and was stimulated after DNA damage (ionising radiation, ultraviolet radiation and hydrogen peroxide). Endogenous hSMG-1 was also demonstrated to bind to N-terminal hSMG-1 GST fusion proteins, suggesting that hSMG-1 can from multimers. In addition, hSMG-1 was found to associate with p53, a key component of the DNA damage response pathway, which is also targeted by other members of the PIKK family in response to DNA damage. In response to DNA damage hSMG-1 was observed by immunofluorescence to localise to discrete cytoplasmic granules. These sites were subsequently identified as stress granules (SG). During NMD hSMG-1 targets Upf1, a key step leading to the degradation of aberrant mRNA. Upf1 is recruited to the aberrant mRNA by Upf2 and both proteins were detected in SG. DNA damage induced SG were also demonstrated to be sites of phosphorylated hSMG-1 target motifs [phospho S(/T)Q], suggesting that hSMG-1 has kinase activity within SG. The presence of this phosphorylated site in SG parallels the presence of hSMG-1 at all times observed during SG formation and disassociation. Interestingly, not all treatments with genotoxic agents resulted in hSMG-1 inclusion in SG, indicating that hSMG-1 is not a core component of SG. Indeed, hSMG-1 was excluded from SG induced with the mitochondrial poisons clotrimazole (CZ) and sodium arsenite (NaAs), agents commonly used to induce SG. In addition to hSMG-1, Upf1, Upf2 and phospho S(/T)Q motifs were also excluded from CZ and NaAs induced SG. Reducing the cellular levels of hSMG-1 using small interfering RNA (siRNA) abrogated the formation of SG in response to DNA damage, indicating a crucial role for hSMG-1 in the formation of these SG. Surprisingly, a role for ATM in SG formation after DNA damage was suggested. Abrogation of ATM activity with small molecule inhibitors resulted in decreased SG formation after DNA damage. This novel role of ATM was not required for the induction of SG after treatment with CZ or NaAs. This was confirmed by observing SG induction with the same agents in ATM deficient (A-T) cells. Given that hSMG-1 is excluded from CZ or NaAs induced stress granules this indicates that not only is ATM signalling required for SG formation after DNA damage but this ATM-dependent pathway for SG formation involves the recruitment of hSMG-1, as well as the previously described proteins. This suggests that there are at least two distinct signaling pathways responsible for SG formation, one pathway which is both hSMG-1 and ATM dependent and the other which is ATM and hSMG-1 independent. Here I describe a novel and essential role for hSMG-1 and ATM in stress granule formation after exposure to agents that damage DNA and produce physiological stress.