An alternative, nonapoptotic form of programmed cell death Sabina Sperandio*, Ian de Belle†, and Dale E. Bredesen*‡§¶ *Buck Institute for Age Research, 8001 Redwood Boulevard, Novato, CA 94945; †Program on Oncogenes and Tumor Suppressor Genes and ‡Program on Aging and Cancer, The Burnham Institute, 10901 North Torrey Pines Road, La Jolla, CA 92037; and §Neuroscience Department, University of California, 9500 Gilman Drive, San Diego, CA 92093 Communicated by Leonard A. Herzenberg, Stanford University School of Medicine, Stanford, CA, September 19, 2000 (received for review May 3, 2000) The term apoptosis often has been used interchangeably with the FBS and 1% penicillin͞streptomycin. Transient transfections term programmed cell death. Here we describe a form of pro- used Lipofectamine (GIBCO͞BRL) according to the manufac- grammed cell death that is distinct from apoptosis by the criteria turer’s instructions. A total of 1 ϫ 106 293T cells or 2 ϫ 105 of morphology, biochemistry, and response to apoptosis inhibi- Apaf-1 null cells were seeded in 6-cm dishes and transfected the tors. Morphologically, this alternative form of programmed cell next day by using a ratio of DNA͞Lipofectamine of 1 ␮g͞5 ␮l. death appears during development and in some cases of neuro- Transfection efficiency was 60–80% for 293T and 10–20% for degeneration. Despite its lack of response to caspase inhibitors and the Apaf-1 KO cells, determined by 5-bromo-4-chloro-3-indolyl Bcl-xL, we show that this form of cell death is driven by an ␤-D-galactoside staining after transfection of a ␤-galactosidase alternative caspase-9 activity that is Apaf-1-independent. Charac- construct. terization of this alternative form of programmed cell death should T-butyloxycarbonyl-Asp(O-methyl)-fluoromethyl ketone lead to new insight into cell death programs and their roles in (BAF) and benzyloxycarbonyl-valyl-alanyl-aspartyl fluoro- development and degeneration. methyl ketone (zVAD.fmk) were obtained from Enzyme Systems Products (Livermore, CA). Actinomycin D and cyclo- he original definition of apoptosis as a form of cell death heximide were from Sigma. Tdistinct from necrosis was based on morphological criteria (1). Genetic studies in Caenorhabditis elegans led to the Constructs. The human insulin-like growth factor I receptor ␤ identification of key components of the apoptotic cell death (IGFIR) subunit intracellular domain was subcloned from a pathway, including the protease Ced-3 (2–5). However, it has NT2 (human teratocarcinoma) cDNA library by PCR amplifi- become clear that the morphological descriptions of apoptosis cation and inserted into pcDNA3 (Invitrogen). Fifteen amino and necrosis are not adequate to describe all cell deaths; acids containing the Src myristylation signal sequence were examples that do not conform to the criteria for either of incorporated into the construct. these forms of cell death include: (i) certain developmental Mutants of IGFIR intracellular domain (IGFIR-IC) were cell deaths, such as autophagic cell death (6–10) and cyto- obtained by QuikChange mutagenesis (Stratagene) from the plasmic cell death (7, 9, 11–14); (ii) some neurodegenerative wild-type construct. The sequences of all constructs were con- cell deaths, such as those in a transgenic mouse model of firmed, and Western blot analyses were performed to verify amyotrophic lateral sclerosis (15); and (iii) some ischemia- protein expression. related cell deaths featuring cell swelling, referred to as oncosis (16). Assessment of Cell Death and Apoptosis. Cell death was assessed by The biochemical basis for these alternative morphological trypan blue exclusion. Floating cells were collected at the forms of cell death remains unknown. Understanding the mech- indicated times after transfection, by centrifugation for 10 min at 1,500 rpm. The cell pellet was resuspended in 50 ␮l of PBS, to anisms for these forms has implications for the understanding of ␮ evolutionary aspects of cell death programs, developmental cell which 50 l of trypan blue was added. Dead cells were counted death, neurodegeneration, and cancer therapeutics and for the by four independent hemocytometer counts. design of novel therapeutic agents for diseases featuring these For the Apaf-1 KO cells, both floating and adherent cells alternative forms of cell death. were collected by trypsinization, and the cell death was Here we describe a form of nonapoptotic programmed cell determined as the percentage of dead cells over the total death (paraptosis, from para ϭ next to or related to, and number of cells. apoptosis) that fails to fulfill the criteria for apoptosis (Table 1). For the assessment of cell death after treatment with actino- We also show that this form of cell death is programmatic, mycin D and cycloheximide, 293T cells were transfected as because it requires gene expression. Despite its lack of response described above; 8.5 h after the beginning of transfection the drugs were added at the indicated concentrations and subse- to caspase inhibitors and Bcl-xL, however, this alternative form of cell death nonetheless is inhibited by a catalytic mutant of quently removed after 12 h, at which time the cells were fed with caspase-9 zymogen. Furthermore, in complementary studies we fresh medium. Floating cells were collected 24 h after the show that the expression of caspase-9 zymogen leads to both removal of the drugs, and dead cells were counted in the apoptotic and nonapoptotic programmed cell death; block of the presence of trypan blue. apoptotic arm of this response by caspase inhibitors (17) converts the cell death to the nonapoptotic form. Finally, the alternative, Abbreviations: zVAD.fmk, benzyloxycarbonyl-valyl-alanyl-aspartyl fluoromethyl ketone; nonapoptotic pathway mediated by caspase-9 is Apaf-1- BAF, t-butyloxycarbonyl-Asp(O-methyl)-fluoromethyl ketone; IGFIR, insulin-like growth independent and is not inhibited by mutation of the sites of factor I receptor; IGFIR-IC, IGFIR intracellular domain; TUNEL, terminal deoxynucleotidyl- zymogen processing to the apoptotically active forms (D315 and transferase-mediated dUTP nick-end labeling; xiap, X-chromosome-linked inhibitor of apoptosis. D330). ¶To whom reprint requests should be addressed at: Buck Institute for Age Research, 8001 Methods Redwood Boulevard, Novato, CA 94945. E-mail: [email protected]. The publication costs of this article were defrayed in part by page charge payment. This Cells and Culture Conditions. 293T cells and Apaf-1 null mouse article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. embryonic fibroblasts were cultured in DMEM containing 10% §1734 solely to indicate this fact. 14376–14381 ͉ PNAS ͉ Demcember 19, 2000 ͉ vol. 97 ͉ no. 26 Downloaded by guest on September 29, 2021 Table 1. Comparison of apoptosis, necrosis and paraptosis Apoptosis Necrosis Paraptosis Morphology Nuclear ϩϪϪ fragmentation Chromatin ϩϪϮ condensation Apoptotic bodies ϩϪϪ Cytoplasmic Ϫϩϩ vacuolation Mitochondrial Sometimes ϩ Late swelling Genomic effect TUNEL ϩ Usually ϪϪ Internucleosomal ϩϪϪ DNA fragmentation Caspase activity DEVD-cleaving ϩϪϪ activity Caspase-3 ϩϪϪ processing PARP cleavage ϩ (85-kDa ϩ (50- to Ϫ fragment) 62-kDa fragments) Inhibition by: zVAD.fmk ϩϪϪ BAF ϩϪϪ p35 ϩϪϪ xiap ϩϪϪ Bcl-xL ϩ Usually ϪϪ Actinomycin D Sometimes Ϫϩ Cycloheximide Sometimes Ϫϩ DEVD, Asp-Glu-Val-Asp; PARP, poly(ADP-ribose) polymerase. CELL BIOLOGY CPP32 (caspase-3)-like activity was determined by using the ApoAlert Caspase-3 Assay Kit (CLONTECH), following the manufacturer’s instructions. Terminal deoxynucleotidyltransferase-mediated dUTP nick- end labeling (TUNEL) staining was performed by using the Apoptosis Detection System (Fluorescein) from Promega, fol- lowing the manufacturer’s instructions. Oligonucleosomal DNA fragmentation was assessed by ex- traction of soluble DNA from 3 ϫ 106 transfected cells, by the method of Frisch and Francis (18). Fig. 1. The IGFIR-IC induces cell death. (A) Comparison between the cell death Coimmunoprecipitation Assay. Cell lysis and immunoprecipitation induced by pIGFIR-IC and nonmyristylated form of the IC domain and deletion were performed as described in Ye et al. (19). mutants. Asterisks indicate a highly significant difference from the control as determined by one-way ANOVA (F ϭ 63.3; P Ͻ 0.0001); P Ͻ 0.0001 for pIGFIR-IC Electron Microscopy. Samples for electron microscopy were pre- and pIGFIR-IC⌬1264–1276, Bonferroni͞Dunn posthoc test. Error bars represent pared as described (20) and examined on a Hitachi 600 electron the SEM from three independent experiments. (B) Light microscopy, using Hoff- microscope. man optics, demonstrating morphological changes induced by IGFIR-IC. IGFIR-IC expression caused a rounding of the cells and intracellular vacuoles. (C) Ultra- Results structural characteristics of IGFIR-induced cell death. 293T cells transfected with control empty vector pcDNA3 (a), pC9 (b), or pIGFIR-IC (c and d). Although The IGFIR Induces Cell Death. During studies of dependence re- caspase-9-transfected cells displayed characteristic features of apoptosis, includ- ceptors (21) [i.e., receptors such as the common neurotrophin ing chromatin condensation, IGFIR-IC-transfected cells did not. In contrast, ex- receptor (22), the androgen receptor (23), and DCC (deleted in tensive cytoplasmic vacuolization was observed in the absence of nuclear frag- colorectal cancer) (24) that induce cell death unless blocked by mentation, cellular blebbing, or apoptotic body formation. Note that the cell