Special Feature

Immunology and Cell Biology (2006) 84, 66–71; doi:10.1111/j.1440-1711.2005.01409.x

Addressing the mysteries of perforin function

Ilia Voskoboinik1 and Joseph A Trapani1

1 Cancer Immunology Program, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia

Correspondence: Ilia Voskoboinik, Cancer Immunology Program, Peter MacCallum Cancer Centre, St Andrew's Place, East Melbourne, Vic. 3002, Australia. Email: ilia.voskoboinik@petermac.org

Received 19 October 2005; Accepted 20 October 2005.



Perforin is a cytolytic protein stored in secretory granules of CTL and NK cells. It synergizes with proapoptotic serine proteases, granzymes, to deliver the lethal hit to virus-infected or transformed target cells. The mechanism of perforin action has not been described beyond its original characterization in the 1980s, and its role in human disease has remained elusive. This article addresses recent key advances in genetic, clinical and biochemical studies that have reignited the current interest in perforin biology.


cytotoxic lymphocyte, FHL, haemophagocytosis



Perforin is a soluble, pore-forming cytolytic protein synthesized in CTL and NK cells (collectively cytolytic lymphocytes [CL]) and sequestered into secretory cytotoxic granules. Upon the formation of the immunological synapse between a CL and a virus-infected or a transformed target cell, cytotoxic granules fuse with the plasma membrane of the CL and release their contents (which also include granzyme serine proteases) into the synapse.1, 2, 3 In the synapse, perforin synergizes with proapoptotic granzymes by allowing their unfettered entry into the target cell. Granzymes, predominantly granzymes A and B, then initiate caspase-dependent and caspase-independent apoptotic pathways, which rapidly lead to target cell death. The mechanism of synergy between perforin and granzymes is not fully understood. Although the loss of individual granzymes has no universally detrimental effect on granule-mediated death pathways,4 functional perforin is essential for the function of CTL and NK cells.5, 6, 7, 8, 9

Perforin was first described in 1985 and characterized as a calcium-dependent, pore-forming cytolytic protein.10, 11, 12 Perforin protein was first identified through its antigenic cross-reactivity with several components of the complement membrane-attack complex (MAC), particularly C9. However, subsequent cloning of perforin (PRF1) cDNA and deduction of its primary amino acid sequence showed very little similarity between perforin and MAC components. In fact, perforins from various species have shown very little amino acid similarity to any other known proteins and yet are highly conserved in species as divergent as humans and fish.13 Such a high level of conservation over at least 300 million years of evolution and the lack of any known isoforms of perforin suggested that it has some universally essential biological properties. In contrast, other granule components such as the granzymes are represented by a variety of isoforms and there are also considerable structural and functional differences between the same isoforms of closely related species, suggesting that granzymes might have evolved in response to epigenetic immune challenges.

Over a number of years, all attempts to produce functional recombinant perforin proved to be futile either because of its inherent cytotoxicity or because of its folding and stability problems. As a result, studies on perforin structure and biochemical function could not proceed beyond its original characterization and demonstration of its synergy with granzymes, and progress essentially ceased by the early 1990s.2, 14 In 2004, fully functional recombinant perforin was purified for the first time in our laboratory, and a new reliable recombinant system for the analysis of perforin function at the cellular level was described.15 These advances reignited interest in perforin biology and its role in immune homeostasis and also allowed us to revisit some of the original, often speculative, ideas and paradigms concerning its molecular function.


Novel method for studying perforin

In the early 1990s, Henkart and co-worker reported the expression of functional perforin in rat basophilic leukaemia (RBL) cells.16 However, certain technical problems limited these studies to the analysis of wild-type perforin. More than a decade later, we revisited that approach, but rather than stably expressing perforin, we used a transient transfection system (Figure 1).15 Perforin cDNA was cloned into an internal ribosome entry site (IRES)-based expression vector (pIRES2-EGFP), and following transfection by electroporation, the cells were sorted using FACS on the basis of their green fluorescent protein (GFP) fluorescence. Because a single transcript is produced for both perforin cDNA and GFP, it is expected that these proteins will be produced at molar equivalence. To ensure consistent sorting, as indicated by GFP fluorescence, it was imperative to calibrate the FACS using FITC-conjugated beads before each experiment. The second condition for achieving reproducible and valid results was to collect the various transfected cell population with identical mean fluorescence intensities. This allows a valid comparison of the activity of the wild-type perforin with those of its mutant forms that have altered stability as determined by western immunoblotting.17 Fcalt epsilon receptors on the sorted RBL cells were bound to IgE antibodies raised against trinitrophenol, which was used to label the surface of target cells preloaded with 51Cr. Cross-linking with the antigen (trinitrophenol) at the target cell surface then triggered degranulation, providing a reliable surrogate model of perforin-dependent, CTL-induced cell death. As a result, perforin was released from RBL cells and triggered lysis of targets was measured by 51Cr release. Importantly, the rigorous standardization of cell selection processes has provided consistent and reproducible levels of cytotoxic activity of transfected RBL cells with respect to targets as measured by the 51Cr release assay.17 Concurrent with the study of perforin mutants in the cellular setting, recombinant perforins were produced and purified from baculovirus-infected insect cells. These were assayed in vitro to identify the nature of defects in perforin function mapping to the level of the target cell. These included, for example, limited binding to the target cell or the inability to polymerize at the cell surface causing the loss of or reduction in cytolytic activity.18 In conclusion, the design of a reliable expression and purification system using baculovirus and insect cells, and a high throughput RBL cell-based assay, provided a tremendous methodological advance in the analysis of molecular mechanism of perforin-induced cell death. These systems have been recently applied to analyse the structure/function of one of the key functional domains of perforin and to directly investigate functional defects of perforin mutants linked to familial haemophagocytic lymphohistiocytosis (FHL) syndrome.17

Figure 1.
Figure 1 - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author

Expression of perforin (PRF) in rat basophilic leukaemia (RBL) cells. RBL cells degranulate upon cross-linking of their Fcalt epsilon receptor by IgE detecting an appropriate antigen on the target cell (trinitrophenol [TNP] in the case of the described experimental system) (a). As a result, perforin is released from the effector cell (RBL) granules and lyses the cross-linked target cell. RBL cells are transiently transfected (by electroporation) with perforin cDNA cloned into pIRES2-EGFP vector (Clontech) (b), and cells are FACS sorted 20–24 h later on the basis of enhanced green fluorescent protein (EGFP) fluorescence. The FACS sorter is calibrated before each experiment using FITC-labelled beads (c), thus ensuring consistent set-up of the instrument. Cells expressing mutated and wild-type perforin are sorted on the basis of identical mean relative fluorescence intensity so that perforin function can be validly compared (c). IRES, internal ribosome entry site; FU, fluorescence units.

Full figure and legend (60K)


Perforin and human disease

The importance of perforin in maintaining immune homeostasis and immune surveillance has become a virtual paradigm of immunology as a result of extensive studies using perforin knockout mice.5, 6, 7, 8, 9 These animals are healthy when raised in pathogen-free conditions, but develop spontaneous aggressive B-cell lymphomas as a result of either external pathogen challenge or endogenous retrovirus activation.19, 20, 21 When transplanted, these spontaneous lymphomas are avidly rejected by wild-type mice but not by perforin-deficient mice. These knockout animals were also more susceptible to chemical-induced, non-haematological sarcomas, and a small number of mice (10–15%) developed adenocarcinoma as they aged.21, 22 Finally, the perforin-mediated cytotoxic pathway has been shown to play a key antimetastatic role in NK cells in experimental tumours as diverse as melanoma and breast carcinoma.19, 23, 24 Interestingly, unless the animals were immune challenged,7 perforin deficiency did not cause such an aggressive and fatal manifestation as observed in humans suffering from the only proven perforin-related disorder, FHL.

Familial haemophagocytic lymphohistiocytosis was first observed and characterized in the 1950s,25, 26 but the genetic basis of the syndrome was discovered only recently. In 1999, G. de Saint Basile's group reported the association of PRF1 mutations with several cases of FHL (FHL2).27 During the next few years, dozens of mutations in PRF1 were identified in FHL patients and were considered to be the primary cause of the disease in up to 60% of cases.28, 29, 30, 31, 32, 33 FHL patients fall into one of several categories based on the types of mutation in PRF1: (i) both alleles of PRF1 with nonsense or frame-shift mutations, which result in premature termination of the protein; (ii) compound heterozygotes with one allele carrying a nonsense or frame-shift mutation and the other having a missense mutation; (iii) homozygotes with both PRF1 alleles having the same missense mutation; and (iv) compound heterozygotes with both alleles bearing missense mutations. Although nonsense and frame-shift mutations can clearly result in loss of function/expression of perforin, the effect of missense mutations on perforin function was harder to predict. In these cases, the loss of or reduction in perforin expression concurrent with the loss of cytotoxic activity of NK cells and, to a lesser extent, of CTL was the only marker of perforin malfunction. However, it is important to note that primary cells were commonly obtained from patients during the acute stage of the disease, when CL activity had already been depleted. Indeed, the in vitro activation of 'nonfunctional', alloreactive primary CTL cells from FHL patients with missense mutations in PRF1 can result in significant restoration of cytotoxic activity.34 Also, at least 40% of FHL patients had no mutations in PRF1, thus indicating the heterogenous genetic nature of the disorder. Consistent with this notion was the discovery of a different genetic defect in at least 20% FHL patients, who had detrimental mutations in the granule trafficking protein Munc13-4.35, 36

To elucidate the role of PRF1 mutations in FHL, we recently analysed the majority of suspected disease-causing perforin alleles using our RBL cell-based expression system.15, 17 Consistent with clinical observations, the majority of mutations appeared to be detrimental at the presynaptic level as indicated by the dramatic loss of perforin expression compared with the expression of wild-type protein. However, three suspected disease-causing mutations surprisingly had wild-type levels of expression and activity and are therefore unlikely to play a causative role in FHL. Most intriguingly, four mutations that we studied resulted in post-synaptic defects as the protein appeared to be expressed at the wild-type level, yet showed significantly reduced or no cytolytic activity. With no prior findings on structure/function of perforin, the nature of defects in these mutant proteins was expected to give some indication of the mechanism of action of wild-type perforin. Indeed, the analysis of one of these mutations, Gly429Glu, has led to functional characterization and structural prediction of the C-terminal C2 domain.

Ala91Val is a common human perforin variant

One of the most contentious issues in the role of perforin in human disease is the effect of a common polymorphism, Ala91Val, on the function of perforin and immune homeostasis and surveillance in general.37, 38, 39 The Ala91Val allele is observed in the general population with a frequency of 4–17%.33, 40 Such a level of occurrence suggests that the allele encodes a neutral polymorphism, because the predicted frequency of Ala91Val homozygosity, up to 3%, greatly exceeds the frequency of perforin-related FHL, 0.002% (1:50 000 live births).41 Nevertheless, several cases of late-presentation FHL (early teens to middle twenties) have been linked to Ala91Val. However, close examination showed that in these cases, the second allele of PRF1 had either a frame-shift or an inactivating missense mutation so that Ala91Val was the only functional allele remaining.39, 42, 43 We investigated Ala91Val perforin using RBL expression and found that the mutation resulted in approximately 50% reduction in perforin activity, largely because of loss of protein stability as shown by western immunoblotting.17 These experiments proved beyond doubt that Ala91Val is in fact a functionally impaired mutant allele, not a neutral polymorphism. The loss of stability was, in turn, probably attributed to incorrect folding of Ala91Val perforin, as shown by the rapid decay of cytolytic function of purified Ala91Val perforin compared with wild type. Taken together, our findings suggest that wild-type/Ala91Val heterozygous individuals are predicted to have approximately 75% and Ala91Val homozygous individuals approximately 50% of perforin activity compared with wild-type perforin homozygotes. However, an Ala91Val/null heterozygote is predicted to show as little as 25% of normal perforin activity in CTL or NK cells. As shown by clinical observations, such a reduction in the level of activity could predispose an individual to late or atypical FHL.39, 42, 43 Furthermore, significant reduction in cytolytic activity of perforin could potentially impair immune surveillance leading to, under some unfavourable conditions of immune challenge, other immunodeficiency-related disorders. Indeed, several reports have recently indicated the link between mutations in PRF1, including the Ala91Val polymorphism, and various types of cancer, such as B-cell and T-cell lymphomas and acute childhood lymphoblastic leukaemia.37, 38, 39 Although the evidence of the link between malignancy and perforin mutations that cause only a partial loss of function is purely circumstantial and preliminary, it does suggest that chronic perforin deficiency is associated with human immunopathology.

Another important aspect of perforin deficiency, as shown in knockout mice, is that in the absence of immune challenge, for example, in the form of bacterial or viral infection, the animals have no obvious pathophysiological manifestation. Similarly, FHL patients with an identical perforin genotype may succumb to the disease at widely different ages, varying from a few months to several years.17 These observations suggest that additional genetic and/or environmental factors might play a major role in triggering disease. It will be important to identify and understand these factors in future studies.


Regulation of perforin cytotoxicity

Perforin is a multidomain protein44 comprising a cleavable leader peptide that directs perforin into the secretory pathway, followed by a positively charged N-terminal sequence of approximately 30–35 amino acids, with predicted lytic capabilities. The next region of 150 odd residues has no significant homology to any characterized protein and has no function assigned to it. The midregion of perforin (residues approximately 200–240) is predicted to form an amphipathic alpha-helix, one side of which consists of polar or positively charged amino acids, whereas the other half is almost invariably hydrophobic. This region is believed to regulate membrane insertion and transmembrane stabilization of perforin at the time of pore formation, and it might also participate in perforin polymerization. The next region, approximately 130 amino acids long, has no designated function, and the following putative epidermal growth factor-like domain has a characteristic disulfide bridge signature. Yet this domain has no predicted functional role in perforin, but by analogy with other proteins, it might participate in intraprotein interactions. The C-terminal, calcium-dependent C2 domain is essential for membrane binding of perforin.18 The final 20 amino acids of perforin contain an N-glycosylation site and a putative cleavage site, which are thought to be important for perforin activation.45 Overall, perforin can be described as a complex protein consisting of several discrete domains, which act in concert to provide the protein with its distinct biological properties.

Protection of 'self' from perforin

A unique feature of perforin is its potent cytotoxicity to the target cell, with half-maximum lysis achieved with as few as 750 molecules per cell.18 Yet, perforin is completely innocuous to the host cell that secrets it. How can such a duality be achieved? At what level is the effector cell protected from perforin: is it at the level of synthesis and post-translational modification, in the endoplasmic reticulum or Golgi, or at the level of cytotoxic granules?

Recently, while analysing FHL-related mutations, we found that one mutation that caused a post-synaptic defect, Gly429Glu, was mapped to the putative C2 domain.15, 18 We used that mutation as a lead to analyse the functional importance of the C-terminus of perforin and showed that indeed it encodes a functional, calcium-dependent, membrane-binding C2 domain as had been previously proposed.46 C2 domains have been found in a number of unrelated proteins (e.g. synaptotagmin, protein kinase C-beta and phospholipase C), which are often involved in signal transduction and membrane trafficking, and operate under restricted (low micromolar or submicromolar) calcium levels in the intracellular environment. C2 domains usually coordinate calcium ions through the exposed anionic side chains of multiple acidic aspartate or glutamate residues protruding from loops formed by the beta-sheet scaffold of the C2 motif. Calcium coordination then facilitates binding of the C2 domain to the phospholipid membrane through electrostatic or hydrophobic interaction.46, 47 Perforin is a unique member of C2 domain proteins in that it is destined to function in an extracellular milieu, where calcium concentrations are typically much higher than in the cytosol (approximately 1.3 mmol/L in the plasma vs 50–200 nmol/L in resting cells). Moreover, cytosolic free calcium concentration in activated degranulating CL increases does not exceed 2–5 mumol/L.48 Several groups have clearly shown that perforin is unable to bind to phospholipid membrane at such restricted levels of ionic calcium, thus preventing the first essential step in perforin pore formation, namely, calcium-dependent membrane binding. In fact, at least 200 mumol/L free calcium is required to maximize membrane binding capacity of perforin and to provide it with full cytolytic activity.18

So what makes perforin so insensitive to physiological intracellular calcium concentrations? Our mapping and functional characterization of the C2 domain of perforin showed its unusual properties.18 Thus, conserved aspartate residues in positions 428, 435, 483 and 485 were clearly essential for perforin function. Surprisingly, the residue in position 491, which according to the 3-D models was expected to play an important role in calcium binding, appeared to be dispensable at >800 mumol/L extracellular calcium but was essential at <500 mumol/L. These observations, together with overall low affinity of perforin for calcium-dependent membrane binding, suggested that the affinity of Asp-491 for calcium is somewhat limited. On this basis, it has been proposed that calcium binding to Asp-491 is a crucial step that differentially controls perforin activity intracellularly and extracellularly, ensuring that perforin binds to phospholipid membranes only after it is released from the effector cell into a high-calcium milieu. Importantly, perforin has been shown to bind to the lipid bilayer only at pH > 6.2. This is consistent with protonation of key aspartate residues of the C2 domain under acidic conditions that makes them unable to coordinate calcium ions. Therefore, the strongly acidic environment of cytolytic granules (pH 4.9–5.2) becomes a safeguard for cytolytic activity of perforin, which is unable to bind to the luminal membrane in C2 domain-dependent manner, which appears to be the prerequisite for its cytolytic activity.

The properties of the C2 domain protect the cell from perforin at the level of the cytosol and cytolytic granules. However, a functional C2 domain might be a disadvantage in the endoplasmic reticulum, where neutral pH and high calcium concentrations might enable perforin to bind to these membranes in the same manner it binds to the target cell to induce cytolysis. Does perforin have any other built-in mechanism of host cell protection? The answer appears to be 'yes'. Asparagine residue Asn-548, only seven residues from the C-terminus, appears to be heavily glycosylated, as the glycosylated and unglycosylated forms differ by approximately 10 kDa.45 Amazingly, the glycosylation appears to be sufficient either to make perforin completely inactive or to dramatically reduce its cytolytic activity to the extent that can be well tolerated by the cell. Consistent with this hypothesis, we have found that the expression of deglycosylated perforin was detrimental to the host cell (I. Voskoboinik and J. A. Trapani, unpubl. obs.). Deglycosylation appeared to result in a C2 domain-dependent cytolytic function, as the simultaneous disruption of a calcium-binding site in the C2 domain restored the expression of the glycosylation mutant to the wild-type levels. Others and we have postulated that once the perforin is delivered to the cytolytic granules in a mature (glycosylated) form, it undergoes either direct deglycosylation or proteolytic cleavage of the last 12–20 amino acids, which makes perforin fully active for killing the target cell.45 In conclusion, finely tuned processing and peculiarities of structure ensure protection of the host cell and, at the same time, allow perforin to exhibit its cytolytic activity in an extracellular environment.


Concluding remarks

Until very recently, perforin has been one of the most mysterious molecules in the immune system: it is absolutely essential, yet its structure, mechanisms of lysis and synergy with granzymes are not understood. Recent identification of perforin mutations in FHL patients has, for the first time, shed light on the role of perforin in human disease, which until now has only been found in experimental animal models. Given enormous interest in perforin as the key player of CL-regulated immunity and progress in experimental methods and clinical studies, we are confident that concerted efforts in biochemical, structural and epidemiological areas will resolve more perforin mysteries in the near future.



  1. Barry M, Bleackley RC. Cytotoxic T lymphocytes: all roads lead to death. Nat. Rev. Immunol. 2002; 2: 401–9. | Article | PubMed | ISI | ChemPort |
  2. Trapani JA, Smyth MJ. Functional significance of the perforin/granzyme cell death pathway. Nat. Rev. Immunol. 2002; 2: 735–47. | Article | PubMed | ISI | ChemPort |
  3. Trapani JA, Sutton VR. Granzyme B: pro-apoptotic, antiviral and antitumor functions. Curr. Opin. Immunol. 2003; 15: 533–43. | Article | PubMed | ChemPort |
  4. Heusel JW, Wesselschmidt RL, Shresta S et al. Cytotoxic lymphocytes require granzyme B for the rapid induction of DNA fragmentation and apoptosis in allogeneic target cells. Cell 1994; 76: 977–87. | Article | PubMed | ISI | ChemPort |
  5. Clark WR. Immunology. The hole truth about perforin. Nature 1994; 369: 16–17. | Article | PubMed | ChemPort |
  6. Kagi D, Hengartner H. Different roles for cytotoxic T cells in the control of infections with cytopathic versus noncytopathic viruses. Curr. Opin. Immunol. 1996; 8: 472–7. | Article | PubMed | ISI | ChemPort |
  7. Kagi D, Ledermann B, Burki K et al. Cytotoxicity mediated by T cells and natural killer cells is greatly impaired in perforin-deficient mice. Nature 1994; 369: 31–7. | Article | PubMed | ISI | ChemPort |
  8. van den Broek ME, Kagi D, Ossendorp F et al. Decreased tumor surveillance in perforin-deficient mice. J. Exp. Med. 1996; 184: 1781–90. | Article | PubMed | ChemPort |
  9. Walsh CM, Matloubian M, Liu CC et al. Immune function in mice lacking the perforin gene. Proc. Natl Acad. Sci. USA 1994; 91: 10 854–8.
  10. Masson D, Tschopp J. Isolation of a lytic, pore-forming protein (perforin) from cytolytic T-lymphocytes. J. Biol. Chem. 1985; 260: 9069–72. | PubMed | ISI | ChemPort |
  11. Podack ER, Young JD, Cohn ZA. Isolation and biochemical and functional characterization of perforin 1 from cytolytic T-cell granules. Proc. Natl Acad. Sci. USA 1985; 82: 8629–33. | Article | PubMed | ChemPort |
  12. Liu CC, Perussia B, Cohn ZA et al. Identification and characterization of a pore-forming protein of human peripheral blood natural killer cells. J. Exp. Med. 1986; 164: 2061–76. | Article | PubMed | ChemPort |
  13. Hwang JY, Ohira T, Hirono I et al. A pore-forming protein, perforin, from a non-mammalian organism, Japanese flounder, Paralichthys olivaceus. Immunogenetics 2004; 56: 360–67. | PubMed | ChemPort |
  14. Trapani JA. Target cell apoptosis induced by cytotoxic T cells and natural killer cells involves synergy between the pore-forming protein, perforin, and the serine protease, granzyme B. Aust. NZ J Med. 1995; 25: 793–9. | ChemPort |
  15. Voskoboinik I, Thia MC, De Bono A et al. The functional basis for hemophagocytic lymphohistiocytosis in a patient with co-inherited missense mutations in the perforin (PFN1) gene. J. Exp. Med. 2004; 200: 811–16. | Article | PubMed | ChemPort |
  16. Shiver JW, Henkart PA. A noncytotoxic mast cell tumor line exhibits potent IgE-dependent cytotoxicity after transfection with the cytolysin/perforin gene. Cell 1991; 64: 1175–81. | Article | PubMed | ISI | ChemPort |
  17. Voskoboinik I, Thia MC, Trapani JA. A functional analysis of the putative polymorphisms A91V and N252S and 22 missense perforin mutations associated with familial hemophagocytic lymphohistiocytosis. Blood 2005; 105: 4700–706. | Article | PubMed | ChemPort |
  18. Voskoboinik I, Thia MC, Fletcher J et al. Calcium-dependent plasma membrane binding and cell lysis by perforin are mediated through its C2 domain: a critical role for aspartate residues 429, 435, 483, and 485 but not 491. J. Biol. Chem. 2005; 280: 8426–34. | PubMed | ChemPort |
  19. Smyth MJ, Thia KY, Street SE et al. Perforin-mediated cytotoxicity is critical for surveillance of spontaneous lymphoma. J. Exp. Med. 2000; 192: 755–60. | Article | PubMed | ISI | ChemPort |
  20. Street SE, Hayakawa Y, Zhan Y et al. Innate immune surveillance of spontaneous B cell lymphomas by natural killer cells and gammadelta T cells. J. Exp. Med. 2004; 199: 879–84. | Article | PubMed | ChemPort |
  21. Street SE, Trapani JA, MacGregor D et al. Suppression of lymphoma and epithelial malignancies effected by interferon gamma. J. Exp. Med. 2002; 196: 129–34. | Article | PubMed | ISI | ChemPort |
  22. Street SE, Cretney E, Smyth MJ. Perforin and interferon-gamma activities independently control tumor initiation, growth, and metastasis. Blood 2001; 97: 192–7. | Article | PubMed | ISI | ChemPort |
  23. Kodama T, Takeda K, Shimozato O et al. Perforin-dependent NK cell cytotoxicity is sufficient for anti-metastatic effect of IL-12. Eur. J. Immunol. 1999; 29: 1390–96. | Article | PubMed | ISI | ChemPort |
  24. Smyth MJ, Thia KY, Cretney E et al. Perforin is a major contributor to NK cell control of tumor metastasis. J. Immunol. 1999; 162: 6658–62. | PubMed | ISI | ChemPort |
  25. Farquhar JW, Macgregor AR, Richmond J. Familial haemophagocytic reticulosis. Br. Med. J. 1958; 30: 1561–4.
  26. Farquhar JW, Claireaux AE. Familial haemophagocytic reticulosis. Arch. Dis. Child. 1952; 27: 519–25. | PubMed | ChemPort |
  27. Stepp SE, Dufourcq-Lagelouse R, Le Deist F et al. Perforin gene defects in familial hemophagocytic lymphohistiocytosis. Science 1999; 286: 1957–9. | Article | PubMed | ISI | ChemPort |
  28. Clementi R, zur Stadt U, Savoldi G et al. Six novel mutations in the PRF1 gene in children with haemophagocytic lymphohistiocytosis. J. Med. Genet. 2001; 38: 643–6. | Article | PubMed | ISI | ChemPort |
  29. Ericson KG, Fadeel B, Nilsson-Ardnor S et al. Spectrum of perforin gene mutations in familial hemophagocytic lymphohistiocytosis. Am. J. Hum. Genet. 2001; 68: 590–97. | Article | PubMed | ChemPort |
  30. Feldmann J, Le Deist F, Ouachee-Chardin M et al. Functional consequences of perforin gene mutations in 22 patients with familial haemophagocytic lymphohistiocytosis. Br. J. Haematol. 2002; 117: 965–72. | Article | PubMed | ChemPort |
  31. Kogawa K, Lee SM, Villanueva J et al. Perforin expression in cytotoxic lymphocytes from patients with hemophagocytic lymphohistiocytosis and their family members. Blood 2002; 99: 61–6. | Article | PubMed | ISI | ChemPort |
  32. Suga N, Takada H, Nomura A et al. Perforin defects of primary haemophagocytic lymphohistiocytosis in Japan. Br. J. Haematol. 2002; 116: 346–9. | Article | PubMed | ISI | ChemPort |
  33. Molleran Lee S, Villanueva J, Sumegi J et al. Characterisation of diverse PRF1 mutations leading to decreased natural killer cell activity in North American families with haemophagocytic lymphohistiocytosis. J. Med. Genet. 2004; 41: 137–44. | Article | PubMed | ChemPort |
  34. Ishii E, Ueda I, Shirakawa R et al. Genetic subtypes of familial hemophagocytic lymphohistiocytosis: correlations with clinical features and cytotoxic T lymphocyte/natural killer cell functions. Blood 2005; 105: 3442–8. | Article | PubMed | ChemPort |
  35. Feldmann J, Callebaut I, Raposo G et al. Munc13-4 is essential for cytolytic granules fusion and is mutated in a form of familial hemophagocytic lymphohistiocytosis (FHL3). Cell 2003; 115: 461–73. | Article | PubMed | ISI | ChemPort |
  36. Ishii E, Ohga S, Imashuku S et al. Review of hemophagocytic lymphohistiocytosis (HLH) in children with focus on Japanese experiences. Crit. Rev. Oncol. Hematol. 2005; 53: 209–23. | PubMed |
  37. Santoro A, Cannella S, Trizzino A et al. A single amino acid change A91V in perforin: a novel, frequent predisposing factor to childhood acute lymphoblastic leukemia? Haematologica 2005; 90: 697–8. | PubMed |
  38. Clementi R, Locatelli F, Dupre L et al. A proportion of patients with lymphoma may harbor mutations of the perforin gene. Blood 2005; 105: 4424–8. | Article | PubMed | ISI | ChemPort |
  39. Busiello R, Galgani M, De Fusco C et al. Role of A91V mutation in perforin gene in hemophagocytic lymphohistiocytosis. Blood 2004; 104: 1910. | ChemPort |
  40. Zur Stadt U, Beutel K, Weber B et al. A91V is a polymorphism in the perforin gene not causative of an FHLH phenotype. Blood 2004; 104: 1909–10. | Article | PubMed | ChemPort |
  41. Henter JI, Elinder G, Soder O et al. Incidence in Sweden and clinical features of familial hemophagocytic lymphohistiocytosis. Acta Paediatr. Scand. 1991; 80: 428–35. | PubMed | ChemPort |
  42. Clementi R, Emmi L, Maccario R et al. Adult onset and atypical presentation of hemophagocytic lymphohistiocytosis in siblings carrying PRF1 mutations. Blood 2002; 100: 2266–7. | Article | PubMed | ISI | ChemPort |
  43. Busiello R, Adriani M, Locatelli F et al. Atypical features of familial hemophagocytic lymphohistiocytosis. Blood 2004; 103: 4610–12. | Article | PubMed | ChemPort |
  44. Trapani JA. Dual mechanisms of apoptosis induction by cytotoxic lymphocytes. Int. Rev. Cytol. 1998; 182: 111–92. | PubMed | ChemPort |
  45. Uellner R, Zvelebil MJ, Hopkins J et al. Perforin is activated by a proteolytic cleavage during biosynthesis which reveals a phospholipid-binding C2 domain. EMBO J 1997; 16: 7287–96. | Article | PubMed | ChemPort |
  46. Nalefski EA, Falke JJ. The C2 domain calcium-binding motif: structural and functional diversity. Protein Sci. 1996; 5: 2375–90. | PubMed | ISI | ChemPort |
  47. Murray D, Honig B. Electrostatic control of the membrane targeting of C2 domains. Mol. Cell 2002; 9: 145–54. | Article | PubMed | ISI | ChemPort |
  48. Esser MT, Haverstick DM, Fuller CL et al. Ca2+ signaling modulates cytolytic T lymphocyte effector functions. J. Exp. Med. 1998; 187: 1057–67. | Article | PubMed | ISI | ChemPort |


J. A. T. is supported by senior Fellowship and a Program Grant from the National Health and Medical Research Council of Australia. J. A. T. and I. V. also received Program Grant support from the Juvenile Diabetes Research Foundation of Australia. We also thank the many members of the Cancer Immunology Program and our collaborators for contributions over many years to many of the findings referred to in this article.


These links to content published by NPG are automatically generated.


Granzyme A: the road less traveled

Nature Immunology News and Views (01 Feb 2003)

The hole truth about perforin

Nature News and Views (05 May 1994)

Cytolytic T-cell melodrama

Nature News and Views (07 May 1987)