Encyclopedia of Signaling Molecules

2018 Edition
| Editors: Sangdun Choi

Apoptosis-Inducing Factor 1, Mitochondrial

  • Patricia Ferreira
  • Raquel Villanueva
  • Marta Martínez-Júlvez
  • Milagros Medina
Reference work entry
DOI: https://doi.org/10.1007/978-3-319-67199-4_101522


Historical Background

Mitochondria play a central role in both cellular redox metabolism and programmed cell death (PCD) induction, since they contain certain agents relevant in these two key cellular functions. Among them is the apoptosis-inducing factor (AIF), initially described as the first caspase-independent apoptotic inducer provoking PCD after liberation from mitochondria (Susin et al. 1999). In healthy cells, AIF is found in the mitochondrial intermembrane space (IMS), where it also has a vital role in the redox metabolism of this organelle that is not well described yet. AIF is released from the IMS to the cytosol in response to apoptotic stimuli and then translocated into the nucleus, where it acts as a proapoptotic factor.

AIFM1 Processing and Localization in the Cell

The canonical human AIF mitochondrion associated 1 (AIFM1) gene is located on chromosome X and codifies for a protein that is synthesized in the cytosol as a 613 amino acids precursor that contains an N-terminal mitochondrial localization signal (MLS) (Susin et al. 1999) (Fig. 1). After translocation into the IMS, the N-terminal 54-residues segment of AIFM1 is proteolytically cleaved, yielding the mature mitochondrial form, AIFΔ1–54. AIFΔ1–54 anchors to the inner mitochondrial membrane through the specific inner-membrane-sequence (IMSS) contained in its N-terminal helix, while the main protein portion folds in the IMS and incorporates the flavin adenine dinucleotide (FAD) cofactor (Otera et al. 2005; Modjtahedi et al. 2006). In response to specific death signals (See also encyclopedia section “ Apoptosis-Inducing Factor (AIF)”), AIF is further processed between residues 101–102 yielding the soluble apoptotic form, AIFΔ1–101, that is released to the cytoplasm, and then translocated into the nucleus assisted by its two nuclear localization sequences (Fig. 1). Once in the nucleus, AIFΔ1–101 induces chromatin condensation and large-scale DNA fragmentation into 50 kb pieces in a caspase-independent manner (Otera et al. 2005).
Apoptosis-Inducing Factor 1, Mitochondrial, Fig. 1

Human AIFM1 processing, subcellular localization, and partners. AIFM1 is synthesized in cytoplasmic ribosomes and imported into the IMS. Once in mitochondria the MLS is cleaved, the protein is tethered to the inner-membrane (by its IMSS), and its soluble portion folds incorporating the FAD cofactor. Upon mitochondrial outer membrane permeabilization by an apoptotic insult, AIF is cleaved at amino acid 101, released from mitochondria, and translocated into the nucleus. The schematic representation of the AIF processing is shown in colored boxes related to its structural domains. The figure also indicates the main interactions of AIF (shown in yellow) with other proteins in the different cellular compartments

AIFM1 NADH-Dependent Redox Activity

AIFM1 plays an essential function in the mitochondrial bioenergetics metabolism through its NADH-dependent redox activity. It indirectly regulates the biogenesis of the mitochondrial respiratory chain, through the assembly and maintenance of respiratory complexes I, III, and IV (Miramar et al. 2001; Modjtahedi et al. 2006; Sevrioukova 2011). The in vivo AIF NADH-dependent activity has been scarcely studied, and consequently it is less understood than the apoptotic one. Under physiological conditions in healthy mitochondria AIF is present in a monomer-dimer equilibrium displaced towards the monomer, while NADH binding and flavin reduction considerably increase the proportion of dimers (Ferreira et al. 2014). The in vitro reaction of the AIFΔ1–101 recombinant enzyme with NADH yields a highly oxygen-stable FADH-NAD+ charge transfer complex (CTC) (Fig. 2) coupled to an increase in the proportion of protein dimers. Both the CTC and the dimer are proposed to be physiologically relevant in a model where AIF would act as a sensor of the mitochondrial redox state. In this model FAD reduction would be sufficient to promote AIF dimerization, while coenzyme binding might initiate the conformational changes associated with CTC formation (see below) (Sorrentino et al. 2015; Ferreira et al. 2014). The low efficiency of AIF as a reductase when using artificial electron acceptors (kcat/KmNADH 5.6 ± 0.2, 2.3 ± 0.4, and 6.4 ± 0.7 s−1 mM−1 for dichlorophenolindophenol, ferricyanide, and cytochrome c, respectively, electron acceptors (Ferreira et al. 2014)) together with the high stability of the CTC dimer suggests that NAD+ dissociation limits the overall reaction rate. In vivo, the suitable physiological acceptor might decrease the enzyme affinity for NAD+ (Ferreira et al. 2014). In addition, mutational disruption of the dimerization interface decreases the enzyme capacity to stabilize the CTC, suggesting a close relationship between the formation of the CTC and the dimer.
Apoptosis-Inducing Factor 1, Mitochondrial, Fig. 2

Spectral evolution and CTC formation upon reduction of hAIF Δ1–101 by NADH. Spectra recorded for the reaction of hAIFΔ1–101 with NADH at several times after mixing. The dashed line corresponds to the spectrum of oxidized protein before mixing. The inset shows time evolution at 451 nm (flavin reduction) and 750 nm (formation of CTC)

Main AIFM1 Structural Characteristics

AIFM1 is organized into three domains: a FAD binding domain (residues 128–262 and 401–480), a NADH binding domain (residues 263–400), and a C-terminal proapoptotic domain (residues 481–613) (Ye et al. 2002; Mate et al. 2002). The FAD and NADH binding domains have a typical Rossmann fold, while the C-terminal domain folds in five β-strands and two α-helices (Mate et al. 2002) (Fig. 3a). The isoalloxazine ring of the FAD cofactor is stabilized at the interface of the three modules, being partially accessible from the solvent. The crystal structure of reduced hAIF in complex with NAD(H) exhibits a dimer and two NAD(H) molecules per protomer (Fig. 3b) (Ferreira et al. 2014). The NAD(H)A molecule shows an extended conformation with its nicotinamide stacking between the re-face of the FAD flavin and the F310 rings. Its binding is also stabilized through a H-bond network involving several residues (G308, F310, L311, E314, E336, G399, E453, H454, and W483). The side chains of K177, F310, and H454 displaced upon coenzyme binding, when compared with the free enzyme, to allocate the NADH nicotinamide portion (Fig. 3c). Mutational analysis indicated that K177 and E314 set up the active site conformation; P173 determines the flavin properties and modulates the enzyme affinity for NADH; and F310 and H454 contribute to the compact active site essential for NADH binding, CTC stabilization, and strong NAD+ affinity for the reduced state of hAIF. All these features are key determinants of the particular behavior of hAIF as a low efficiency NADH-dependent oxidoreductase (Villanueva et al. 2015).
Apoptosis-Inducing Factor 1, Mitochondrial, Fig. 3

Human AIFM1 structure. (a) Cartoon representation of the structure of hAIFΔ1–121ox (PDB ID 4BV6). The NADH binding, the FAD binding, and the apoptotic domains are drawn in yellow, raspberry red, and blue, respectively. A dash line accounts for the 545–559 disordered segment. (b) Surface representation of the dimeric hAIFΔ1–101rd:2NAD(H) complex (PDB ID 4BUR). (c) Network of H-bonds and hydrophobic stacking interactions at the FAD and NAD(H) binding sites in the hAIFΔ1–101rd:2NAD(H) complex. Relevant residues are shown in CPK-colored sticks. (d) Detail of conformational changes suffered by hAIF upon NADH induced dimerization in cartoon representation of hAIFΔ1–121ox (blue, upper layer) and hAIFΔ1–101rd:2NAD(H) (yellow, down layer). The 508–560 segment of hAIFΔ1–121ox is highlighted in deep blue while it is not observed in hAIFΔ1–101rd:2NAD(H). The 190–202 β-hairpin is shown in green for hAIFΔ1–121ox and in raspberry red for hAIFΔ1–101rd:2NAD(H). Key residues are shown in sticks. In all structures, FAD, NAD(H)A, and NAD(H)B are shown as sticks with its C atoms in green, blue, and pink, respectively

The second NAD(H) molecule, NAD(H)B, binds at the si-face of the flavin, with the side chain of W483 stacking at one side with its nicotinamide ring and at the other with the flavin ring (Fig. 3c) (Ferreira et al. 2014). F582 and particularly W196 are displaced to accommodate the adenine of NAD(H)B through stacking interactions, also contributing to stabilize its pyrophosphate, ribose, and nicotinamide moieties. Comparison of the oxidized and reduced structures shows that binding of NAD(H)B induces conformational changes in the 190–202 β-hairpin and in the 509–560 segment (Fig. 3d), as well as in the molecular dimerization interface (439–453 segment). Noticeably, the 509–560 insertion in mammalian enzymes is a part of the apoptotic C-terminal domain that in the oxidized enzyme shows its 517–533 moiety folded into two short helices which decrease the solvent accessibility to the flavin ring active site. In the complex of the reduced protein with the coenzyme these helices become disordered, and their positions are occupied by NAD(H)B.

To date, six pathological hAIF mutations have been associated with neurological disorders. Five of them (ΔR201, V243L, G262S, G308E, and G338E) cause severe mitochondriopathies associated to reduced expression of respiratory chain complexes and OXPHOS failure, while the sixth mutation (E493V) results in increased cell death via apoptosis causing the Cowchock syndrome but without affecting the OXPHOS activity (Rinaldi et al. 2012; Ardissone et al. 2015; Kettwig et al. 2015; Ghezzi et al. 2010). Mutations E493V and ΔR201 locate in the NAD(H)B site. The E493V mutation probably produces a negative impact in the allocation of NAD(H)B. The variant ΔR201 is structurally unstable and prone to FAD release, facts probably related with R201 stabilizing the two short helices in the oxidized structure (Ghezzi et al. 2010). These two mutations support the functional importance of the folding conservation in the NAD(H)B binding site in both redox states of the protein. The rest of mutations are distributed all over the hAIF structure, but in general contribute to structural elements involved in the allocation of the adenine moiety portions of either NAD(H)A or FAD. Thus, mutations at residues V243 and G262 led to a decreased level of protein expression probably related to a defective folding (Ardissone et al. 2015; Kettwig et al. 2015) or FAD incorporation, while those at G308 and G338 were critical for NAD(H)A stabilization. Structural alterations in ligand binding sites in the pathogenic mutations highlight the importance of the proper folding of the protein as well as the correct cofactor and coenzyme binding for its suitable functioning.

AIFM1 Partners

AIFM1 has been hypothesized to act as a sensor of the NADH/NAD+ intracellular levels, with coenzyme binding and flavin reduction modulating its monomer-dimer equilibrium and the interaction with other molecules that could be determining its biological activity. So far, it has been described in the interaction of AIF with a variety of proteins in the different cellular compartments where it can be located. Nevertheless, other partners might still be unknown, and the exact role of AIF in mitochondria remains a conundrum. In mitochondria the interaction of the AIFred−NAD+ complex with CHCHD4, a protein that participates in the import of several mitochondrial proteins and their oxidative folding by disulfide bond formation, has linked AIF to the biogenesis of respiratory chain complexes (Hangen et al. 2015; Lui and Kong 2007). The interaction of AIFΔ1–101 with Hsp70 in the cytosol induces cellular survival (Lui and Kong 2007), while its interaction with cyclophilin A (CypA) in the same compartment promotes the optimal nuclear translocation of both proteins to induce PCD (Cande et al. 2004). Once in the nucleus, AIF forms a degradosome with CypA and the phosphorylated histone H2AX that provokes chromatin condensation and DNA degradation (Artus et al. 2010; Cande et al. 2004). In addition, the cytosolic and nuclear interactions of AIF with reduced thioredoxin 1 (Trx1) are proposed to provide a mechanism to attenuate the AIF lethal action (Shelar et al. 2015). Under proapoptotic oxidative stress conditions, Trx1 becomes oxidized in the cytosol and the Trx1-AIF complex dissociates, which likely promote nuclear translocation of AIF. In the nucleus, the complexation of AIF with reduced Trx1 would hinder its interaction with DNA and, as a consequence, apoptosis (Shelar et al. 2015).


AIFM1 is a ubiquitously distributed protein in mammals that regulates caspase-independent programmed cell death, but it is also a key factor in the biogenesis of the respiratory complexes through its NADH-oxidoreductase activity. In each particular cell the main AIFM1 function is determined by its subcellular localization, with the protein in the mitochondrial intermembrane space in healthy cells being translocated to the nucleus upon apoptotic stimuli. In addition, current research has provided compelling evidences supporting the critical role of the AIFM1 redox state and monomer-dimer equilibrium in both its apoptotic and respiratory chain maintenance functions. At the molecular level AIFM1 folds in three domains, two of them related with its oxidoreductase activity and the C-terminal with its apoptotic function. Upon reduction by NADH it forms a FADH-NAD+ CTC highly stable versus oxygen reoxidation and dimerizes, being in addition able to allocate two coenzyme molecules, one in the oxidoreductase portion and the other one in the apoptotic domain, whose nicotinamide portions get placed in the flavin ring environment. In addition, AIFM1 is able to interact with several protein partners, being such interactions modulated by the enzyme subcellular localization, redox state, and oligomeric state. Further elucidations of the mechanisms of processes involving AIFM1 are at this point required, since they will surely provide a better understanding of the molecular mechanisms underlying the diseases in which it is involved and will help to develop novel pharmacological therapies.



We thank Luis Martinez Lostao (University of Zaragoza) for helping us with the graphical illustration in Figure 1.


  1. Ardissone A, Piscosquito G, Legati A, Langella T, Lamantea E, Garavaglia B, et al. A slowly progressive mitochondrial encephalomyopathy widens the spectrum of AIFM1 disorders. Neurology. 2015;84:2193–5. doi: 10.1212/WNL.0000000000001613.PubMedCentralCrossRefPubMedGoogle Scholar
  2. Artus C, Boujrad H, Bouharrour A, Brunelle MN, Hoos S, Yuste VJ, et al. AIF promotes chromatinolysis and caspase-independent programmed necrosis by interacting with histone H2AX. EMBO J. 2010;29:1585–99. doi: 10.1038/emboj.2010.43.PubMedCentralCrossRefPubMedGoogle Scholar
  3. Cande C, Vahsen N, Kouranti I, Schmitt E, Daugas E, Spahr C, et al. AIF and cyclophilin A cooperate in apoptosis-associated chromatinolysis. Oncogene. 2004;23:1514–21. doi: 10.1038/sj.onc.1207279.CrossRefPubMedGoogle Scholar
  4. Ferreira P, Villanueva R, Martínez-Júlvez M, Herguedas B, Marcuello C, Fernandez-Silva P, et al. Structural insights into the coenzyme mediated monomer-dimer transition of the pro-apoptotic apoptosis inducing factor. Biochemistry. 2014;53:4204–15. doi: 10.1021/bi500343r.CrossRefPubMedGoogle Scholar
  5. Ghezzi D, Sevrioukova I, Invernizzi F, Lamperti C, Mora M, D’Adamo P, et al. Severe X-linked mitochondrial encephalomyopathy associated with a mutation in apoptosis-inducing factor. Am J Hum Genet. 2010;86:639–49. doi: 10.1016/j.ajhg.2010.03.002.PubMedCentralCrossRefPubMedGoogle Scholar
  6. Hangen E, Féraud O, Lachkar S, Mou H, Doti N, Fimia GM, et al. Interaction between AIF and CHCHD4 regulates respiratory chain biogenesis. Mol Cell. 2015;58:1001–14. doi: 10.1016/j.molcel.2015.04.020.CrossRefPubMedGoogle Scholar
  7. Kettwig M, Schubach M, Zimmermann FA, Klinge L, Mayr JA, Biskup S, et al. From ventriculomegaly to severe muscular atrophy: expansion of the clinical spectrum related to mutations in AIFM1. Mitochondrion. 2015;21C:12–8. doi: 10.1016/j.mito.2015.01.001.CrossRefGoogle Scholar
  8. Lui JC, Kong SK. Heat shock protein 70 inhibits the nuclear import of apoptosis-inducing factor to avoid DNA fragmentation in TF-1 cells during erythropoiesis. FEBS Lett. 2007;581:109–17. doi: 10.1016/j.febslet.2006.11.082.CrossRefPubMedGoogle Scholar
  9. Mate MJ, Ortiz-Lombardia M, Boitel B, Haouz A, Tello D, Susin SA, et al. The crystal structure of the mouse apoptosis-inducing factor AIF. Nat Struct Biol. 2002;9:442–6. doi: 10.1038/nsb793.CrossRefPubMedGoogle Scholar
  10. Miramar MD, Costantini P, Ravagnan L, Saraiva LM, Haouzi D, Brothers G, et al. NADH oxidase activity of mitochondrial apoptosis-inducing factor. J Biol Chem. 2001;276:16391–8. doi: 10.1074/jbc.M010498200.CrossRefPubMedGoogle Scholar
  11. Modjtahedi N, Giordanetto F, Madeo F, Kroemer G. Apoptosis-inducing factor: vital and lethal. Trends Cell Biol. 2006;16:264–72. doi: 10.1016/j.tcb.2006.03.008.CrossRefPubMedGoogle Scholar
  12. Otera H, Ohsakaya S, Nagaura Z, Ishihara N, Mihara K. Export of mitochondrial AIF in response to proapoptotic stimuli depends on processing at the intermembrane space. EMBO J. 2005;24:1375–86. doi: 10.1038/sj.emboj.7600614.PubMedCentralCrossRefPubMedGoogle Scholar
  13. Rinaldi C, Grunseich C, Sevrioukova IF, Schindler A, Horkayne-Szakaly I, Lamperti C, et al. Cowchock syndrome is associated with a mutation in apoptosis-inducing factor. Am J Hum Genet. 2012;91:1095–102. doi: 10.1016/j.ajhg.2012.10.008.PubMedCentralCrossRefPubMedGoogle Scholar
  14. Sevrioukova IF. Apoptosis-inducing factor: structure, function, and redox regulation. Antioxid Redox Signal. 2011;14:2545–79. doi: 10.1089/ars.2010.3445.PubMedCentralCrossRefPubMedGoogle Scholar
  15. Shelar SB, Kaminska KK, Reddy SA, Kumar D, Tan CT, Yu VC, et al. Thioredoxin-dependent regulation of AIF-mediated DNA damage. Free Radic Biol Med. 2015;87:FRBMD1500111. doi: 10.1016/j.freeradbiomed.2015.06.029.CrossRefGoogle Scholar
  16. Sorrentino L, Calogero AM, Pandini V, Vanoni MA, Sevrioukova IF, Aliverti A. Key role of the adenylate moiety and integrity of the adenylate-binding site for the NAD(+)/H binding to mitochondrial apoptosis-inducing factor. Biochemistry. 2015;54:6996–7009. doi: 10.1021/acs.biochem.5b00898.CrossRefPubMedGoogle Scholar
  17. Susin SA, Lorenzo HK, Zamzami N, Marzo I, Snow BE, Brothers GM, et al. Molecular characterization of mitochondrial apoptosis-inducing factor. Nature. 1999;397:441–6. doi: 10.1038/17135.CrossRefPubMedGoogle Scholar
  18. Villanueva R, Ferreira P, Marcuello C, Usón A, Miramar MD, Peleato ML, et al. Key residues regulating the reductase activity of the human mitochondrial apoptosis inducing factor. Biochemistry. 2015;54:5175–84. doi: 10.1021/acs.biochem.5b00696.CrossRefPubMedGoogle Scholar
  19. Ye H, Cande C, Stephanou NC, Jiang S, Gurbuxani S, Larochette N, et al. DNA binding is required for the apoptogenic action of apoptosis inducing factor. Nat Struct Biol. 2002;9:680–4. doi: 10.1038/nsb836.CrossRefPubMedGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  • Patricia Ferreira
    • 1
  • Raquel Villanueva
    • 1
  • Marta Martínez-Júlvez
    • 1
  • Milagros Medina
    • 1
  1. 1.Department of Biochemistry and Molecular and Cellular BiologyInstitute for Biocomputation and Physics of Complex Systems, University of ZaragozaZaragozaSpain