Battling brain-eating amoeba: Enigmas surrounding immunity to Naegleria fowleri

Abstract

Yet, because distinguishing PAM from other types of meningitis and encephalitis can be difficult, it’s likely that many cases of PAM are simply never identified, especially in areas of the world with under-resourced healthcare systems. Because N. fowleri is a thermophilic organism, rising global temperatures will prolong growth seasons and expand compatible habitats. [...]although N. fowleri infection and PAM are not a nationally notifiable disease, there is evidence that the latitude of reported infections has broadened over the past 10 to 15 years [1]. [...]a critical component of N. fowleri’s lethal opportunism likely lies in the barrier being breached within nasal turbinates. [...]pathogenic variants of N. fowleri are resistant to downstream complement mediated lysis [16], and there is scant in vivo evidence that complement is critical to N. fowleri containment. Because neutrophils show no intrinsic chemotactic response toward N. fowleri [15], how is the alarm bell rung?

About the Authors:

E. Ashley Moseman

* E-mail: [email protected]

Affiliation: Department of Immunology, Duke University School of Medicine, Durham, North Carolina, United States of America

ORCID logo http://orcid.org/0000-0002-6064-6116

Introduction

Free-living amoebas (FLA) are remarkable single-cell engines, foraging their way through a range of environments. However, occasionally these amoebas find themselves within a human host, and an unusual and deadly opportunistic infection unfolds. The FLA Naegleri fowleri is the causative agent for an invasive and lethal form of meningoencephalitis known as primary amoebic meningoencephalitis (PAM). According to the CDC, 147 patients have been diagnosed with PAM in the United States since 1962. Yet, because distinguishing PAM from other types of meningitis and encephalitis can be difficult, it’s likely that many cases of PAM are simply never identified, especially in areas of the world with under-resourced healthcare systems. Because N. fowleri is a thermophilic organism, rising global temperatures will prolong growth seasons and expand compatible habitats. In fact, although N. fowleri infection and PAM are not a nationally notifiable disease, there is evidence that the latitude of reported infections has broadened over the past 10 to 15 years [1]. When paired with potentially increased water recreation, a warming climate may facilitate a collision course of amoebic growth and human activity.

Collision course: N. fowleri encounter with mammalian hosts

As a free-living amoeba, N. fowleri is fully capable of reproducing without a host, and mammals are certainly not a requisite step in the N. fowleri life cycle. Indeed, N. fowleri are found in warm fresh water across the globe, making human contact commonplace and typically benign. N. fowleri can exist in 3 forms: a dormant cyst form, a migratory flagellate, and the pathogenic trophozoite that feeds and divides. PAM occurs when trophozoites access the nasal turbinates and cross the olfactory epithelium (OE) to enter olfactory nerve bundles and migrate into the brain, where they provoke an intense inflammatory reaction and lethal increases in intracranial pressure. Even though N. fowleri infection is purely opportunistic, unlike many other opportunistic infections, it is not associated with immunocompromised individuals; on the contrary, PAM patients are typically young and seemingly healthy at the time of exposure [2]. The sudden infection and death of otherwise healthy young people underlies the 2 biggest mysteries surrounding N. fowleri infections: Why are some people infected, while others are not, when exposed to seemingly similar conditions? Why is nasal N. fowleri exposure the only route with dire consequences? These 2 unknowns are probably linked because although there are animal models of visceral/peripheral naegleriosis [3], human peripheral infection is virtually unknown, even when people undoubtedly swallow parasites or have exposed open wounds.

Therefore, a critical component of N. fowleri’s lethal opportunism likely lies in the barrier being breached within nasal turbinates. There is evidence that N. fowleri can penetrate the respiratory epithelium [4], but our experiments suggest this is rather uncommon. However, penetration of adjacent OE provides N. fowleri with immediate access to olfactory sensory axon bundles [5] that serve as de facto “tunnels” for amoebas to migrate directly into the brain (Fig 1D), bypassing conventional central nervous system (CNS) barrier protections. Although it has been suggested that amoeba actively chemotax toward brain tissue [6], it may be that the anatomical structures of the OE simply provide a path of least resistance that lead to the brain. In any event, anatomy alone cannot explain why immune mechanisms sufficient elsewhere in the periphery fail within the OE. Might differences in the immune response partially explain why certain individuals develop PAM? Studies of human serum and mucosal antibody titers have found widespread evidence of anti-Naegleria immune responses resulting from subclinical N. fowleri exposure [7–10]. These immune responses may arise after a nonolfactory exposure or olfactory clearance of less pathogenic strains of N. fowleri. Although there is no evidence that overt immunodeficiency predisposes toward N. fowleri infection, the presence of detectable but variable immune responses suggests that differences in innate and adaptive immunity contribute to developing PAM.

[Figure omitted. See PDF.]

Fig 1. Model of early olfactory immune response to N. fowleri infection.

(A) Water entry into the olfactory turbinates delivers N. fowleri to the olfactory epithelial surface. IgA, IgM, and IgG secreted within the airway potentially interfere with attachment to the epithelial surface, and the parasite is washed harmlessly away. (B) If this antibody response is insufficient, N. fowleri interaction with the olfactory surface results in an early (approximately 10 hours) and robust innate response. This early cellular response is characterized by inflammatory cell entry, particularly neutrophils, into the airway space, where they reduce the number of amoebas through effector mechanisms or mechanically inhibiting amoeba interaction with the epithelial surface. (C) When initial mechanisms fail to prevent parasite olfactory invasion, innate responses within the OE can be triggered. These responses involve macrophage and monocyte lineage cells but are again dominated by neutrophils. However, some parasites evade immune detection at this point, allowing them to continue invading deeper into the lamina propria. N. fowleri invasion of the lamina propria results in the parasite gaining entry into the olfactory nerve bundles. (D) These structures serve as conduits for OSNs axons to reach the brain, yet they now become a direct parasite pathway into the brain. Occasionally, innate cells will recognize amoeba within the axon bundles, but numerous amoebas make their way into the brain. (E) Once in the brain, N. fowleri proliferate and eventually provoke a massive inflammatory infiltration consisting of neutrophils, monocytes, and eosinophils that drives lethality. IgA, immunoglobulin A; IgG, immunoglobulin G; IgM, immunoglobulin M; IgM; N. fowleri, Naegleria fowleri; OE, olfactory epithelium; OSN, olfactory sensory neurons.

https://doi.org/10.1371/journal.ppat.1008406.g001

Hurdles to understanding immunity to Naegleria

Because there remain no effective clinical treatments for PAM, defining the mechanisms underlying the early immune failure and the factors that precipitate the subsequent fulminant inflammation may suggest improvements in clinical care. Deciphering these immune mechanisms and retrospectively understanding the human immune response is particularly challenging because of the swiftness of PAM and the rarity of surviving patients. Luckily, animal models of PAM appear remarkably similar to human infections and offer a powerful tool for characterizing how the immune system perceives and responds to N. fowleri. While in vitro experiments have revealed many pathogenic mechanisms employed by N. fowleri [11], a shortage of mechanistic in vivo studies on the immune response to N. fowleri has left many basic questions unanswered. Does breach of the olfactory barrier unequivocally result in death, or must there also be a combined failure of adaptive and innate mechanisms to result in PAM? What protective immune responses could prevent individuals from being infected in the first place? Which cells and mechanisms are critical for killing N. fowleri in vivo? Is the immune response beneficial to the host at all or simply causing further damage? We cannot fully answer these questions, but this review highlights our current understanding as well as what remains unclear.

Innate immune response to Naegleria: Intense and incomplete

Immune cell infiltration into the CNS is closely associated with the edema that drives herniation and death in PAM patients. However, depletion of CD11b-expressing cells hastened death in animal models, suggesting that neutrophils and infiltrating hematopoietic cells provide an important source of antiamoebic pressure [12], even if overexuberance may contribute to lethality. In contrast to the intense inflammation of end-stage PAM, the initial invasive process of the amoeba is remarkably uninflammatory. Rojas-Hernández and colleagues characterized a very early cellular exudate within the nasal turbinates hours after infection [4], yet parasites then invaded the OE and followed the olfactory axon tracts toward the brain without eliciting significant numbers of innate inflammatory cells [5,13] (Fig 1B–1D). Amoeba are eventually detected, and the subsequent infiltration of neutrophils, eosinophils, monocytes, and macrophages [13] ignites a cascade of hemorrhage and lytic necrosis within the brain 3 to 4 days after infection. Although inflammation ultimately characterizes this disease, the early failure to detect parasites and employ effective antiamoebic mechanisms is particularly noteworthy. How phagocytic parasites could migrate undetected within the CNS is difficult to reconcile with conventional viewpoints on innate cell recruitment to sites of cellular injury. This capacity of individual parasites to enter the CNS without immune detection likely plays a crucial role in the failure to control amoeba and prevent lethality.

How then does the immune system recognize amoebic invasion? Unlike bacterial or viral pathogens, N. fowleri is eukaryotic, and most mammalian pattern recognition receptors will not recognize it as foreign. Complement activation, particularly that mediated by antibody, can drive enhanced neutrophil activity against the amoeba [14,15]. Additionally, complement cleavage products are known to serve as a chemotactic impetus for immune cell recruitment. However, pathogenic variants of N. fowleri are resistant to downstream complement mediated lysis [16], and there is scant in vivo evidence that complement is critical to N. fowleri containment. Because neutrophils show no intrinsic chemotactic response toward N. fowleri [15], how is the alarm bell rung? Complement activation may contribute, but beyond that, it’s unclear which cell types can recognize the presence of the amoeba within the airway, OE, or brain. Amoebas likely damage host cells as they break down intercellular matrices or tear off host cell membranes to feed via trogocytosis [17]. This “bull in a china shop” approach should result in extracellular ATP release from host cells, yet there are few data to support purinergic receptor based recruitment [18] of innate cells to N. fowleri within the OE or CNS. Several studies have indicated that antiamoebic neutrophil responses are regulated by cytokines produced by other cells [14,19].Tumor necrosis factor alpha (TNFα), in particular, has been shown to “license” or “awaken” neutrophils to engage and target amoebas for destruction, likely through mechanisms that are dependent upon myeloperoxidase, superoxide formation, or neutrophil extracellular trap (NET) release [19–23]. Individually, immune cells stand little chance against N. fowleri; however, neutrophils can employ a group attack technique in which numerous neutrophils encircle an individual amoeba to centrally target effector activity [13,14] (Fig 1C and 1E). How neutrophils are able to locate and target the amoeba is unclear, but the failure of sentinel cells to detect N. fowleri and elicit a licensing stimulus such as TNFα may be partially responsible for N. fowleri’s ultimate immune evasion. Indeed, bypassing sentinel recognition through injection of a potent TNFα inducer (muramyl dipeptide) protects animals from PAM even after disease manifestations have begun [24]—a truly remarkable finding that suggests it may be possible to augment or attenuate specific immune functions to tailor a less pathologic and more protective immune response [25].

Adaptive immunity to Naegleria—A critical mediator of prevention?

Studying the adaptive immune response to N. fowleri infection in PAM patients is especially difficult because of the rapidly lethal disease course. In vitro studies have shown that N. fowleri can rapidly internalize surface binding antibodies [26,27], a behavior that has been suggested to undermine the utility of the humoral response. Nonetheless, in vivo, amoeba may initially internalize antibodies, but serum will continuously replenish local antibody concentrations and drive effector activity. Several immunization strategies utilizing amoebic lysate, cell culture supernatant, live amoeba, fixed amoeba, and specific protein via different inoculation routes have resulted in measurable antibody titers and varying degrees of protection [2]. Immune serum transfer experiments have confirmed that circulating antibodies are the dominant protective adaptive immune mechanism [28,29] (Fig 1A). In addition, intrathecal therapeutic monoclonal antibody administration has been shown to prolong survival in animals [30]. Antibodies potentially impinge on amoebic lifestyle in several ways depending upon their isotype: they can opsonize to facilitate fragment crystallizable (Fc) receptor–mediated phagocytosis or effector activity, activate complement to target immune cells to the amoeba, as well as promote direct lysis. However, analysis of human sera has found the majority of anti-Naegleria antibodies are directed toward internal structures [31], rather than protective surface antigens. And of those, the primary antibody isotype generated upon N. fowleri infection in humans is immunoglobulin M (IgM) [2]. Although IgM can drive N. fowleri agglutination [32] and complement activation, IgM’s high molecular weight (approximately 900 kD) may impede crossing the blood–brain barrier to access the infected brain.

An IgM bias in the N. fowleri humoral response could be the result of several factors. Amoebic surface antigens may favor T-independent responses (as is this case with bacterial polysaccharides [33]), or there may be defects in the Naegleria-specific CD4+ repertoire, priming, and functions that facilitate antibody class switch. Cell-mediated immunity against N. fowleri has been observed in the form of delayed-type hypersensitivity [34], but there has not been a careful dissection of amoeba-specific CD4+ T-cell functionality after infection or vaccination. In immunized animals, IL-4 levels correlated with survival—an effect dependent upon STAT6, suggesting a role for Th2 cells in facilitating an antiamoebic vaccine response [35]. Tissue-specific exposure is known to guide immune bias, and indeed, intranasal vaccinations have yielded substantial increases in N. fowleri-binding IgG and IgA antibody titers, which correlated with protection from lethal challenge [35,36] (Fig 1A). And yet, we fundamentally don’t understand how N. fowleri–specific adaptive immune responses are generated, specifically, how and where antigen-presenting cells acquire amoebic antigen, as well as how these cells could coordinate an adaptive immune response within the relevant tissues.

There are many open questions surrounding fundamental immunological processes during N. fowleri infection (Table 1). The protective potential of antibody responses is clear; however, dissection of the relevant antibody isotypes and Fc receptors that provide protective immunity is still needed to guide vaccine design or immunotherapeutic approaches. And while antibody titers are easy to measure, addressing the roles for other lymphocytes such as natural killer (NK) cells, NKT cells, gamma/delta T cells, or even potentially CD8+ T cells requires comprehensive mechanistic in vivo studies. This will require reversing a historical lack of funding for basic research into the host response to N. fowleri and other free-living amoebic pathogens but will catalyze transformative changes in the prophylactic and therapeutic clinical options for a devastating disease.

[Figure omitted. See PDF.]

Table 1.

https://doi.org/10.1371/journal.ppat.1008406.t001

Concluding remarks

It is impossible to know how frequent “subclinical” Naegleria infection is, though there is evidence that individuals from warmer southern states with more presumed Naegleria exposure have more serum antibody activity [32]. It’s possible that asymptomatic Naegleria infections occur with regularity but are simply contained and cleared [37]. However, this is little solace for those few individuals at the confluence of exposure, insufficient immunity, and luck. Understanding the role of immunity in preventing PAM should shed light on a frustratingly fundamental question asked by the families of PAM victims: Why?

Citation: Moseman EA (2020) Battling brain-eating amoeba: Enigmas surrounding immunity to Naegleria fowleri. PLoS Pathog 16(4): e1008406. https://doi.org/10.1371/journal.ppat.1008406

References

1. Kemble SK, Lynfield R, DeVries AS, Drehner DM, Pomputius WF, Beach MJ, et al. Fatal Naegleria fowleri infection acquired in Minnesota: possible expanded range of a deadly thermophilic organism. Clin Infect Dis. 2012 Mar;54(6):805–809. pmid:22238170

2. Visvesvara GS, Moura H, Schuster FL. Pathogenic and opportunistic free-living amoebae: Acanthamoebaspp., Balamuthia mandrillaris, Naegleria fowleri, and Sappinia diploidea. FEMS Immunol Med Microbiol. 2007 Jun;50(1):1–26. pmid:17428307

3. Culbertson CG, Ensminger PW, Overton WM. Amebic cellulocutaneous invasion by Naegleria aerobia with generalized visceral lesions after subcutaneous inoculation: an experimental study in guinea pigs. Am J Clin Pathol. 1972 Mar;57(3):375–86. pmid:5018608

4. Rojas-Hernández S, Jarillo-Luna A, Rodríguez-Monroy M, Moreno-Fierros L, Campos-Rodríguez R. Immunohistochemical characterization of the initial stages of Naegleria fowleri meningoencephalitis in mice. Parasitol Res. 2004 Sep;94(1):31–36. pmid:15338289

5. Jarolim KL, McCosh JK, Howard MJ, John DT. A light microscopy study of the migration of Naegleria fowleri from the nasal submucosa to the central nervous system during the early stage of primary amebic meningoencephalitis in mice. J Parasitol. 2000 Feb 1;86(1):50–55. pmid:10701563

6. Baig AM. Primary Amoebic Meningoencephalitis: Neurochemotaxis and Neurotropic Preferences of Naegleria fowleri. ACS Chem Neurosci. 2016 Apr 13;7(8):1026–1029. pmid:27447543

7. DUBRAY BL, WILHELM WE, JENNINGS BR. Serology of Naegleria fowleri and Naegleria lovaniensis in a Hospital Survey1. J Eukaryot Microbiol. 1987 Aug 1;34(3):322–327.

8. Marciano-Cabral F, Cline ML, Bradley SG. Specificity of antibodies from human sera for Naegleria species. J Clin Microbiol. 1987 Apr;25(4):692–627. pmid:2437151

9. Rivera V, Hernández D, Rojas S, Oliver G, Serrano J, Shibayama M, et al. IgA and IgM anti-Naegleria fowleri antibodies in human serum and saliva. Can J Microbiol. 2001 May;47(5):464–466. pmid:11400739

10. Lares-Jiménez LF, Borquez-Román MA, Alfaro-Sifuentes R, Meza-Montenegro MM, Casillas-Hernández R, Lares-Villa F. Detection of serum antibodies in children and adolescents against Balamuthia mandrillaris, Naegleria fowleri and Acanthamoeba T4. Exp Parasitol. 2018 May 31;189:28–33. pmid:29673623

11. Marciano-Cabral F, Cabral GA. The immune response to Naegleria fowleri amebae and pathogenesis of infection. FEMS Immunol Med Microbiol. 2007 Nov;51(2):243–259. pmid:17894804

12. Ferrante A, Carter RF, Lopez AF, Rowan-Kelly B, Hill NL, Vadas MA. Depression of immunity to Naegleria fowleri in mice by selective depletion of neutrophils with a monoclonal antibody. Infect Immun. 1988 Sep;56(9):2286–2291. pmid:3410537

13. Cervantes-Sandoval I, Serrano-Luna J de J, García-Latorre E, Tsutsumi V, Shibayama M. Characterization of brain inflammation during primary amoebic meningoencephalitis. Parasitol Int. 2008 Sep;57(3):307–313. pmid:18374627

14. Ferrante A, Thong YH. Unique Phagocytic Process in Neutrophil-Mediated Killing of Naeglaria-Fowleri. Immunol Lett. 1980;2(1):37–41.

15. Rowan-Kelly B, Ferrante A, Thong YH. Activation of complement by Naegleria. Trans R Soc Trop Med Hyg. 1980;74(3):333–336. pmid:7434428

16. Whiteman LY, Marciano-Cabral F. Resistance of highly pathogenic Naegleria fowleri amoebae to complement-mediated lysis. Infect Immun. 1989 Dec;57(12):3869–3875. pmid:2807551

17. John DT, Cole TB, Marciano-Cabral FM. Sucker-like structures on the pathogenic amoeba Naegleria fowleri. Appl Environ Microbiol. 1984 Jan;47(1):12–14. pmid:6696410

18. Jassam YN, Izzy S, Whalen M, Mcgavern DB, Khoury El J. Neuroimmunology of Traumatic Brain Injury: Time for a Paradigm Shift. Neuron. 2017 Sep 13;95(6):1246–1265. pmid:28910616

19. Ferrante A, Mocatta TJ. Human neutrophils require activation by mononuclear leucocyte conditioned medium to kill the pathogenic free-living amoeba, Naegleria fowleri. Clin Exp Immunol. 1984 Jun;56(3):559–566. pmid:6378454

20. Michelson MK, Henderson WR, Chi EY, Fritsche TR, Klebanoff SJ. Ultrastructural studies on the effect of tumor necrosis factor on the interaction of neutrophils and Naegleria fowleri. Am J Trop Med Hyg. 1990 Mar;42(3):225–233. pmid:2316792

21. Ferrante A, Hill NL, Abell TJ, Pruul H. Role of myeloperoxidase in the killing of Naegleria fowleri by lymphokine-altered human neutrophils. Infect Immun. 1987 May;55(5):1047–1050. pmid:3032797

22. Ferrante A. Augmentation of the neutrophil response to Naegleria fowleri by tumor necrosis factor alpha. Infect Immun. 1989 Oct;57(10):3110–3115. pmid:2777375

23. Contis-Montes de Oca A, Carrasco-Yepez M, Campos-Rodriguez R, Pacheco-Yépez J, Bonilla-Lemus P, Pérez-López J, et al. Neutrophils extracellular traps damage Naegleria fowleri trophozoites opsonized with human IgG. Parasite Immunology. 2016 Aug;38(8):481–495. pmid:27189133

24. Ferrante A, Lederer E. Curative properties of muramyl dipeptide in experimental Naegleria meningoencephalitis. Trans R Soc Trop Med Hyg. 1986 Jan;80(2):323–326. pmid:3538525

25. Adrover JM, Aroca-Crevillén A, Crainiciuc G, Ostos F, Rojas-Vega Y, Rubio-Ponce A, et al. Programmed “disarming” of the neutrophil proteome reduces the magnitude of inflammation. Nat Immunol. 2020 Feb 1;21(2):135–144. pmid:31932813

26. Shibayama M, Serrano-Luna J de J, Rojas-Hernández S, Campos-Rodríguez R, Tsutsumi V. Interaction of secretory immunoglobulin A antibodies with Naegleria fowleri trophozoites and collagen type I. Can J Microbiol. 2003 Mar;49(3):164–170. pmid:12795402

27. Ferrante A, Thong YH. Antibody induced capping and endocytosis of surface antigens in Naegleria fowleri. Int. J. Parasitol.1979 Dec 1;9(6):599–601. pmid:541172

28. Reilly MF, White KL, Bradley SG. Host resistance of mice to Naegleria fowleri infections. Infect Immun. 1983 Nov;42(2):645–652. pmid:6642646

29. Thong YH, Ferrante A, Shepherd C, Rowan-Kelly B. Resistance of mice to Naegleria meningoencephalitis transferred by immune serum. Trans R Soc Trop Med Hyg.1978 Jan 1;72(6):650–652. pmid:734724

30. Lallinger GJ, Reiner SL, Cooke DW, Toffaletti DL, Perfect JR, Granger DL, et al. Efficacy of immune therapy in early experimental Naegleria fowleri meningitis. Infect Immun. 1987 May;55(5):1289–1293. pmid:3570464

31. Tew JG, Burmeister J, Greene EJ, Pflaumer SK, Goldstein J. A radioimmunoassay for human antibody specific for microbial antigens. J Immunol Methods. 1977;14(3–4):231–241. pmid:320268

32. Reilly MF, Marciano-Cabral F, Bradley DW, Bradley SG. Agglutination of Naegleria fowleri and Naegleria gruberi by antibodies in human serum. J Clin Microbiol. 1983 Apr;17(4):576–581. pmid:6853686

33. Cerutti A, Cols M, Puga I. Marginal zone B cells: virtues of innate-like antibody-producing lymphocytes. Nat Rev Immunol. 2013 Jan 25;13(2):118–132. pmid:23348416

34. Cursons RT, Brown TJ, Keys EA, Moriarty KM, Till D. Immunity to pathogenic free-living amoebae: role of cell-mediated immunity. Infect Immun. 1980 Aug;29(2):408–410. pmid:7011975

35. Carrasco-Yepez M, Rojas-Hernandez S, Rodriguez-Monroy Ma, Terrazas Li, Moreno-Fierros L. Protection against Naegleria fowleri infection in mice immunized with Cry1Ac plus amoebic lysates is dependent on the STAT6 Th2 response. Parasite Immunol. 2010 Sep;32(9–10):664–670. pmid:20691018

36. Lee J, Yoo J-K, Sohn H-J, Kang H-K, Kim D, Shin H-J, et al. Protective immunity against Naegleria fowleri infection on mice immunized with the rNfa1 protein using mucosal adjuvants. Parasitol Res. 2015 Apr;114(4):1377–1385. pmid:25604672

37. Apley J, Clarke SKR, Roome A. P. C. H., Sandry SA, Saygi G, Silk B, et al. Primary Amoebic Meningoencephalitis in Britain. BMJ. 1970 Mar 7;1(5696):596–569. pmid:5440234