Bacterial toxins and small molecules elucidate endosomal trafficking

Spotlights

Bacterial toxins and small molecules elucidate endosomal trafficking Louise H. Slater1, Anne E. Clatworthy2,3, and Deborah T. Hung2,3,4 1

MedImmune, Granta Park, Cambridge, UK Infectious Disease Program, Broad Institute of Harvard and MIT, Cambridge, MA, USA 3 Department of Molecular Biology and Center for Computational and Integrative Biology, Massachusetts General Hospital, Boston, MA, USA 4 Department of Microbiology and Immunobiology, Harvard Medical School, Boston, MA, USA 2

Bacterial toxins and small molecules are useful tools for studying eukaryotic cell biology. In a recent issue of PNAS, Gillespie and colleagues describe a novel small molecule inhibitor of bacterial toxins and virus trafficking through the endocytic pathway, 4-bromobenzaldehyde N-(2,6-dimethylphenyl)semicarbazone (EGA), that prevents transport from early to late endosomes. Bacterial virulence factors have historically been useful tools with which to probe eukaryotic cell biology. For example, Listeria monocytogenes has been instrumental in understanding eukaryotic cytoskeletal rearrangement, and analysis of bacterial toxin entry into host cells has shed valuable insight into vesicular trafficking. Classical genetic approaches utilizing bacterial toxins, such as complementation and RNAi screens, have been successful in the identification of host proteins (e.g., the anthrax toxin receptors CMG-1 and TEM-8, ARAP3 and the chaperones GRP78 and CCT) that are involved in toxin uptake and translocation [1–4]. However, genetic approaches such as RNAi have limitations, for example incomplete knockdown, slow kinetics and indirect effects on host cell physiology. This is particularly problematic for studying essential pathways such as endocytic trafficking, which is required for normal cell function. Alternatively, chemical genetics offers a number of advantages over classical genetics such as rapid kinetics, potent inhibition and the ability to inhibit related, redundant targets. More recently, a combination of both bacterial toxins and small molecule tools is proving to be a powerful approach to understand eukaryotic cell biology, in particular vesicle trafficking. In a recent report, Gillespie et al. conducted a high throughput screen of 30 000 commercially available small molecules to find inhibitors of anthrax lethal toxin (LT)-induced macrophage cell death [5]. LT is a binary AB toxin, consisting of protective antigen (PA), the receptor binding B moiety, and lethal factor (LF), a zinc metalloprotease which cleaves MAP kinase kinases and NLRP1 upon delivery to the cytosol, leading to caspase-1 mediated Corresponding authors: Hung, D.T. ([email protected]). 0966-842X/$ – see front matter ß 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tim.2013.12.002

pyroptosis – an inflammatory form of cell death [6]. Trafficking of LT involves binding of PA to the host cell receptor, proteolytic activation and oligomerization, and binding of LF, followed by endocytosis and trafficking to the endosome (Figure 1) [7]. Similar to other pH-dependent bacterial toxins, PA undergoes a conformational change upon exposure to low pH in the endosome, leading to insertion into the membrane and formation of a pore through which LF translocates into the cytosol in an unfolded conformation. One of the protective compounds identified in the screen, 4-bromobenzaldehyde N-(2,6-dimethylphenyl)semicarbazone (EGA), protected up to 100% of cells when viability was measured 3 hours after toxin addition, with an IC50 of 1.7 mM. The authors showed that EGA inhibited caspase-1 activation, as well as MEK2 cleavage, indicating that it inhibits a step prior to LF activity. Further assays revealed that EGA also inhibits formation of the SDSresistant PA oligomer, which results from the acid-induced conformational change. Notably, EGA did not appear to function by neutralizing the pH of acidic organelles, suggesting that it may have a role in the initial toxin uptake and trafficking events. Interestingly, EGA also inhibited trafficking of EGFR, which was trapped in early endosomes positive for EEA1. This indicates that EGA targets a host factor that is common in both the EGFR and anthrax toxin pathways, and is likely involved in trafficking from early to late endosomes. Furthermore, EGA did not appear to inhibit events involved in phagocytosis or phagosome maturation, suggesting that its target is specifically involved in endosomal trafficking. Consistent with this proposed mode of inhibition, EGA also inhibited entry of related pH-dependent bacterial toxins such as diphtheria and Pseudomonas ExoA, but not ricin, which enters the cytosol via retrograde trafficking through the Golgi to the endoplasmic reticulum (ER), and retrotranslocation from the ER to the cytosol. Furthermore, EGA inhibited infection of cells from pH dependent viruses, such as vesicular stomatitis virus (VSV-G), influenza and lymphocytic choriomeningitis virus (LCMV), which enter the cytosol via acid-induced membrane fusion events in the late endosome. EGA did not inhibit infection of cells by pH-independent viruses such as amphotropic murine leukemia virus (A-MLV) and Nipah virus (NiV), unless pseudotyped by the envelope protein VSV-G, which redirects virus trafficking through the endocytic, pH-dependent route. Further demonstrating the utility of EGA, Trends in Microbiology, February 2014, Vol. 22, No. 2

53

Spotlights

Trends in Microbiology February 2014, Vol. 22, No. 2

PA20

LF

PA83 Furin ANTXR1

ANTXR2

Clathrin coated pit

H+

Early endosome

Cytochalasin D

Mulvesicular body

H+

EGA

vATPase

Late endosome

H+

Bafilomycin A1

LF MAPKK

Cell death

NLRP1

Ammonium chloride TRENDS in Microbiology

Figure 1. Cytoplasmic delivery of anthrax lethal toxin. PA83 binds to one of two anthrax toxin receptors on the cell surface. PA is then processed by a host furin protease into PA20 and PA63. PA63 remains bound to the receptor and oligomerizes into a heptamer or octamer, which can then be bound by LF. Toxin receptor complexes are then internalized by clathrin-mediated endocytosis and trafficked to the early endosome, a process that can be inhibited by cytochalasin D. Upon endosome acidification, a process inhibited by bafilomycin A1 or ammonium chloride, PA undergoes a conformational change and forms a pore in the endosomal membrane. LF is then translocated through the PA pore into the lumen of intraluminal vesicles, or directly from late endosomes into the cytosol. Back-fusion of intraluminal vesicles with the limiting membrane of late endosomes releases LF into the cytoplasm. EGA appears to have a molecular target that is important for early to late endosomal trafficking. Upon release into the cytoplasm, LF exerts its enzymatic effects. Abbreviations: EGA, 4-bromobenzalde-hyde N-(2,6-dimethylphenyl)semicarbazone; LF, lethal factor; PA, protective antigen.

the authors applied it to investigate the mechanism of entry of an uncharacterized cytolethal distending toxin (Ec-CDT) from Escherichia coli. EGA was unable to inhibit intoxication by Ec-CDT, suggesting that this toxin likely does not enter the cytosol through late endosomes. The pH dependence of PA pore formation is reported to be dependent on the specific receptor to which PA is bound. When bound to TEM8 (ANTXR1), PA undergoes pore formation in the mildly acidic environment of the early endosome (pH 6.5); when bound to CMG2 (ANTXR2), the more acidic environment of the late endosome (pH 5) is required [8]. The data presented here suggest that at least in RAW264.7 cells, the majority of PA pores do not form until the toxin reaches the late endosome, which could suggest that CMG2 is the predominant receptor in this cell type. Furthermore, this report is consistent with a model of LT trafficking proposed by Abrami et al., which was based on studies using BHK cells that are reported to express both anthrax toxin receptors [9]. The model proposes that toxin–receptor complexes in early endosomes are sorted into multivesicular endosomes, followed by PA pore formation and LF translocation into the lumen of intraluminal vesicles, and delivery of LF to the cytosol when the vesicles 54

undergo back-fusion with the limiting membrane of late endosomes. Despite the significance of endosomal trafficking in cell biology, few specific chemical probes have actually been identified which serve to elucidate key steps in this process. Small molecule inhibitors currently used to study membrane and vesicular trafficking events include the Arf GTPase inhibitor brefeldin A, which inhibits membrane trafficking; the vacuolar ATPase inhibitor bafilomycin A1, which inhibits organelle acidification and trafficking between organelles; the actin polymerization inhibitor cytochalasin D and the recently discovered dynamin inhibitor, dynasore [10], which both inhibit endocytosis. The majority of other inhibitors that have been identified to disrupt trafficking do so in a relatively non-specific manner. EGA is a valuable addition to this toolbox, with the ability to inhibit a specific step in this complex pathway. Thus, elucidating the exact mechanism of action of EGA will be valuable to furthering our understanding of vesicle trafficking. EGA will therefore have utility in studying endosomal trafficking pathways in both the context of host–pathogen interactions, as demonstrated in this report, as well as normal cellular functions.

Spotlights The authors additionally propose that small molecules, like EGA, that target vesicle trafficking and inhibit both bacterial toxin function and viral entry, might be developed into a broad-spectrum therapeutic agent to treat bacterial and viral diseases. Although targeting host proteins therapeutically to treat infectious diseases is attractive, targeting an essential process such as endosomal trafficking ultimately may be problematic; small molecules that target this step may be intrinsically toxic to a whole organism. Thus, while the jury is still out on such therapeutic applications, the value of such chemical probes has been clearly demonstrated historically and will continue to be well into the future. References 1 Bradley, K.A. et al. (2001) Identification of the cellular receptor for anthrax toxin. Nature 414, 225–229 2 Lu, Q. et al. (2004) EST-based genome-wide gene inactivation identifies ARAP3 as a host protein affecting cellular susceptibility to anthrax toxin. Proc. Natl. Acad. Sci. U.S.A. 101, 17246–17251

Trends in Microbiology February 2014, Vol. 22, No. 2

3 Tamayo, A.G. et al. (2011) GRP78(BiP) facilitates the cytosolic delivery of anthrax lethal factor (LF) in vivo and functions as an unfoldase in vitro. Mol. Microbiol. 81, 1390–1401 4 Slater, L.H. et al. (2013) CCT chaperonin complex is required for efficient delivery of anthrax toxin into the cytosol of host cells. Proc. Natl. Acad. Sci. U.S.A. 110, 9932–9937 5 Gillespie, E.J. et al. (2013) Selective inhibitor of endosomal trafficking pathways exploited by multiple toxins and viruses. Proc. Natl. Acad. Sci. U.S.A. http://dx.doi.org/10.1073/pnas.1302334110 6 Fink, S.L. et al. (2008) Anthrax lethal toxin and Salmonella elicit the common cell death pathway of caspase-1-dependent pyroptosis via distinct mechanisms. Proc. Natl. Acad. Sci. U.S.A. 105, 4312–4317 7 Young, J.A. and Collier, R.J. (2007) Anthrax toxin: receptor binding, internalization, pore formation, and translocation. Annu. Rev. Biochem. 76, 243–265 8 Rainey, G.J. et al. (2005) Receptor-specific requirements for anthrax toxin delivery into cells. Proc. Natl. Acad. Sci. U.S.A. 102, 13278–13283 9 Abrami, L. et al. (2004) Membrane insertion of anthrax protective antigen and cytoplasmic delivery of lethal factor occur at different stages of the endocytic pathway. J. Cell Biol. 166, 645–651 10 Macia, E. et al. (2006) Dynasore, a cell-permeable inhibitor of dynamin. Dev. Cell 10, 839–850

Entrapment exploited Jos A.G. van Strijp and Suzan H.M. Rooijakkers Medical Microbiology, UMC Utrecht, Heidelberglaan 100, 3584CX Utrecht, The Netherlands

In their recent paper in Science, Thammavongsa et al. demonstrate how Staphylococcus aureus degrades the DNA of neutrophil extracellular traps (NETs) into 20 -deoxy-adenosine, which causes incoming macrophages to go into apoptosis, thereby increasing the chance for the bacterium to survive in an abscess. Staphylococcus aureus is a leading cause of severe bacterial infections in humans in both hospital and community settings. Due to its increasing resistance to antibiotics, development of additional therapeutic strategies is required to control this pathogen. S. aureus often enters the human body via surgical wounds, trauma, or prosthetic devices. From there, it can enter the bloodstream and spread to different organs causing abscesses, which are a hallmark of S. aureus infections. The bacteria persist in these abscesses without being cleared by the host immune response. Bacteria can multiply within an abscess and after the abscess ruptures from the organ surface they can easily spread to other organs. The authors of this paper previously developed a mouse model for studying abscess lesions caused by S. aureus [1]. It is now clear that staphylococci grow in the center of these abscesses and actively keep away neutrophils, important immune cells for clearance of staphylococci. Although neutrophils are present in Corresponding author: van Strijp, J.A.G. ([email protected]). 0966-842X/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tim.2013.12.010

the abscess lesion surrounding the bacteria, they are unable to kill them. The present study shows that S. aureus also actively kill macrophages and prevent them from entering the abscess. In the past decades it has become clear that Staphylococcus aureus is equipped with a large variety of proteins that actively downmodulate almost every single step of the innate immune system in order to survive within the human host. Because complement and neutrophils are indispensible in the clearance of Gram-positive bacteria such as S. aureus many of these proteins are directed against elements of the human complement system and degrade, bind, and/or block complement proteins or convertases [2]. Others specifically inhibit neutrophil functions by blocking or cleaving neutrophil receptors, including Fc receptors, Toll-like receptors, and chemokine receptors [3,4]. Other factors that were previously known as general toxins, but now, with the cellular receptors known we understand that these molecules target and kill specific subpopulations of white blood cells without affecting the rest of the host [5]. These proteins are all relatively small, secreted, and present in the majority of all clinically relevant human S. aureus isolates. The manuscript ‘S. aureus degrades neutrophil extracellular traps to promote immune cell death’ by Thammavongsa et al. [6] now demonstrates that S. aureus also modulates macrophages specifically. In vivo sections show that macrophages are actively averted from migration towards the bacterial community in an abscess. This is achieved by a concerted action of two proteins [staphylococcal nuclease (Nuc) and adenosine synthase A (AdsA)] that together convert DNA from neutrophil extracellular 55