Plasmodium yoelii S4/CelTOS is important for sporozoite gliding motility and cell traversal Ryan W. J. Steel1*, Ying Pei1*±, Nelly Camargo1, Alexis Kaushansky1, 2, Dorender A. Dankwa1, Thomas Martinson1, Thao Nguyen1, Will Betz1, Hayley Cardamone1, Vladimir Vigdorovich1, Nicholas Dambrauskas1, Sara Carbonetti1, Ashley M. Vaughan1, Noah D. Sather1 and Stefan H. I. Kappe1, 2, #
1
Center for Infectious Disease Research, formerly Seattle Biomedical Research
Institute, Seattle, Washington, USA 2
Department of Global Health, University of Washington, Seattle, Washington, USA
* Equal
contribution to the work
± Present address: XCell Science Inc, Novarto, California, USA # To
whom correspondence should be addressed. Tel.: +1-206-256-7205; fax: +1-
206-256-7229; E-mail address:
[email protected] Running title: S4/CelTOS for sporozoite gliding motility and cell traversal
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/cmi.12817 This article is protected by copyright. All rights reserved.
Summary Gliding motility and cell traversal by the Plasmodium ookinete and sporozoite invasive stages allow penetration of cellular barriers to establish infection of the mosquito vector and mammalian host, respectively. Motility and traversal are not observed in red cell infectious merozoites, and we have previously classified genes that are expressed in sporozoites but not merozoites (S genes) in order to identify proteins involved in these processes. The S4 gene has been described as critical for Cell Traversal of Ookinetes and Sporozoites (CelTOS), yet knockout parasites (s4/celtos¯) do not generate robust salivary gland sporozoite numbers, precluding a thorough analysis of S4/CelTOS function during host infection. We show here that a failure of oocysts to develop or survive in the midgut contributes to the poor mosquito infection by P. yoelii (Py) s4/celtos¯ rodent malaria parasites. We rescued this phenotype by expressing S4/CelTOS under the ookinete-specific CTRP promoter (S4/CelTOSCTRP), generating robust numbers of salivary gland sporozoites lacking S4/CelTOS that were suitable for phenotypic analysis. Py S4/CelTOSCTRP sporozoites showed reduced infectivity in BALB/c mice when compared to wild-type sporozoites, although they appeared more infectious than sporozoites deficient in the related traversal protein PLP1/SPECT2 (Py plp1/spect2¯). Using in vitro assays, we substantiate the role of S4/CelTOS in sporozoite cell traversal but also uncover a previously unappreciated role for this protein for sporozoite gliding motility.
This article is protected by copyright. All rights reserved.
1. Introduction Plasmodium parasites have distinct extracellular infectious stages for transmission; the ookinete that infects the mosquito vector and the sporozoite that infects the vertebrate host. These stages possess a shared actin/myosin motor that confers active motility, allowing their forward propulsion on or within substrates (gliding motility), traversal of cellular barriers (cell traversal) and active invasion of host cells (Tardieux & Baum, 2016).
Following a mosquitoes’ infectious blood meal, motile ookinetes leave the blood bolus by active motility, then traverse the epithelial layer of the mosquito midgut that requires disruption of epithelial cell membranes (Siden-Kiamos & Louis, 2004; Vinetz, 2005). Ookinetes become sessile beneath the midgut basal lamina where they develop into oocysts and undergo sporulation, forming thousands of sporozoites that will invade the mosquito salivary glands in preparation for transmission to the vertebrate host (Douglas, Amino, Sinnis, & Frischknecht, 2015; Matuschewski, 2006; Vaughan, Aly, & Kappe, 2008). Sporozoites are not delivered directly into blood vessels during an infectious mosquito bite (Sidjanski & Vanderberg, 1997; Yamauchi, Coppi, Snounou, & Sinnis, 2007), but are instead inoculated into the dermis and use active motility and cell traversal to overcome cellular barriers to successful infection. These include traversing skin and endothelial cells to enter the circulation, and once sporozoites are retained in the liver, they traverse liver sinusoidal endothelial cells and Kupffer cells to access the parenchyma (Amino et al., 2008; Amino et al., 2006; Hopp et al., 2015). Here sporozoites traverse a number of hepatocytes before selecting a suitable host hepatocyte for infection with formation of a parasitophorous vacuole membrane (PVM), committing them to become a tropic liver stage form that
This article is protected by copyright. All rights reserved.
initiates intracellular development (Akhouri, Bhattacharyya, Pattnaik, Malhotra, & Sharma, 2004; Amino et al., 2008; Coppi et al., 2007; Frevert et al., 2005; Frischknecht & Matuschewski, 2017; Vaughan et al., 2008; Vaughan & Kappe, 2017).
The function of the actin/myosin motor components are conserved among Plasmodium infectious stages (Tardieux & Baum, 2016). However, the extracellular adhesins and secreted effector proteins that are employed by the ookinete to penetrate the mosquito midgut (Bennink, Kiesow, & Pradel, 2016), and by the sporozoite to invade the salivary gland and passage through the skin and infect the liver (Frischknecht & Matuschewski, 2017), are mostly stage-specific and allow for unique interactions with their respective vector and host environments. Adhesins and secreted effector proteins are stored in specialized organelles known as micronemes, that contain numerous proteins that are secreted apically to mediate motility and cell traversal (Carruthers & Tomley, 2008).
In ookinetes secreted chitinase mediates escape of the ookinete from the peritrophic matrix surrounding the blood meal, the CSP and TRAP-related protein (CTRP) drives ookinete motility, while the secreted ookinete adhesive protein (SOAP), vonWillebrand factor A domain-related protein (WARP) and membrane attack ookinete protein (MAOP) are involved in ookinete traversal of midgut epithelial cells (Bennink et al., 2016; Dessens et al., 1999; Kadota, Ishino, Matsuyama, Chinzei, & Yuda, 2004; Ramakrishnan et al., 2011; Yuda, Yano, Tsuboi, Torii, & Chinzei, 2001). In sporozoites, the secreted protein thrombospondin-related anonymous protein (TRAP) is required for sporozoite motility (Bergman et al., 2003; Buscaglia,
This article is protected by copyright. All rights reserved.
Coppens, Hol, & Nussenzweig, 2003; Ejigiri et al., 2012). The sporozoite protein essential for cell traversal (SPECT) (T. Ishino, Yano, Chinzei, & Yuda, 2004), perforin-like protein 1 (PLP1, also referred to as SPECT2) (Tomoko Ishino, Chinzei, & Yuda, 2005; K. Kaiser et al., 2004), phospholipase (Bhanot, Schauer, Coppens, & Nussenzweig, 2005) and the gamete egress and sporozoite traversal protein (GEST) (Talman et al., 2011) are secreted to disrupt membranes and enable cell traversal by the sporozoite.
Intriguingly, ookinetes and sporozoites share the secreted micronemal protein S4 which was originally identified in the rodent malaria parasite Plasmodium yoelii (Py) as a sporozoite-expressed gene (Kaiser, Matuschewski, Camargo, Ross, & Kappe, 2004). A subsequent study showed that S4 is also expressed in ookinetes, and gene deletion of S4 in the rodent malaria parasite Plasmodium berghei (Pb) indicated its involvement in cell traversal by ookinetes and sporozoites. As such, S4 was renamed CelTOS (Kariu, Ishino, Yano, Chinzei, & Yuda, 2006). Knockout of S4/CelTOS in Pb caused a dramatic defect in mosquito midgut infection by ookinetes that was attributed to failure of the ookinetes to complete traversal of the midgut epithelial cells. This resulted in extreme reduction of oocysts and in consequence sporozoite formation, thereby greatly encumbering an accurate analysis of S4/CelTOS function during sporozoite infection of the host. An accurate account of the role(s) of S4/CelTOS during sporozoite infection is however critical, since S4/CelTOS has been identified as a promising pre-erythrocytic (PE) and transmission blocking vaccine candidate that is conserved across parasite species (Alves et al., 2017; E. S. Bergmann-Leitner, Legler, Savranskaya, Ockenhouse, & Angov, 2011; Elke S. Bergmann-Leitner et al., 2010; Espinosa et al., 2017), and is
This article is protected by copyright. All rights reserved.
immunogenic in attenuated sporozoite vaccination and natural infection settings (Aguiar et al., 2015; Anum et al., 2015; Doolan et al., 2003; Kanoi et al., 2017).
To enable the robust characterization of S4/CelTOS function in Py sporozoites we expressed S4/CelTOS under the promoter of the ookinete micronemal protein CTRP (Yuda, Sakaida, & Chinzei, 1999; Yuda, Sawai, & Chinzei, 1999). The resultant Py S4/CelTOSCTRP parasites rescued the severe mosquito infection defect and generated robust numbers of sporozoites deficient in S4/CelTOS. This tool allowed us to provide the first direct evidence of the role of S4/CelTOS during PE infection. We substantiate the role for S4/CelTOS during sporozoite cell traversal, and reveal a previously unappreciated role for S4/CelTOS in sporozoite gliding motility as well as oocyst survival/development.
2. Results 2.1 Generation and characterization of transgenic parasites to study S4/CelTOS function in mosquito- and mammalian host infection To better understand the role of S4/CelTOS in PE infection, we used a double crossover recombination strategy to generate transgenic Py parasites either lacking S4/CelTOS (Py s4/celtos¯), or expressing S4/CelTOS under the ookinete-specific CTRP promoter (Py S4/CelTOSCTRP; Fig. S1) (Dessens et al., 1999; Menard & Janse, 1997; Yuda, Sakaida, et al., 1999).
We first asked whether mosquito midgut infection was rescued by the promoter swap strategy in our Py S4/CelTOSCTRP parasites (Fig. 1). A partial but incomplete rescue of the number of oocysts per midgut was seen in Py S4/CelTOSCTRP parasites, which This article is protected by copyright. All rights reserved.
in turn showed significantly more oocysts per midgut than Py s4/celtos¯. Surprisingly Py s4/celtos¯ parasites also established a low-level mosquito midgut infection. Whereas the number of oocysts per midgut stayed stable between days 7 – 11 for both Py WT and Py S4/CelTOSCTRP parasites, oocyst numbers for Py s4/celtos¯ parasites declined throughout this period (Fig. 1A). While Py S4/CelTOSCTRP parasites had fewer oocysts per midgut than Py WT, the percent of Py S4/CelTOSCTRP infected midguts was not significantly lower than Py WT and was greater than for Py s4/celtos¯ parasites at all time points analyzed. The fraction of midguts infected with Py s4/celtos¯ declined markedly from day 7 - 11, by which time oocysts were only occasionally observed (Fig. 1B) and these exhibited defects in sporulation compared to Py WT oocysts (Fig. S2). Combined, the data suggest that Py S4/CelTOS has an important role during oocyst survival and/or sporogenesis in addition to the previously reported role in ookinete traversal of the midgut epithelium.
We next analyzed the salivary gland sporozoite loads of mosquitoes infected with the mutant parasites (Fig. 1C). Py s4/celtos¯ parasites did not show sporozoites in the salivary glands of infected mosquitoes, consistent with the dramatic reduction in midgut infection. The rescue of midgut infection by the Py S4/CelTOSCTRP parasite translated into robust salivary gland sporozoite loads, although the number of sporozoites was lower than in Py WT. Expression analysis showed a greater than 25-fold reduction in S4/CelTOS mRNA (Fig. S3A). Furthermore, we were unable to detect S4/CelTOS protein in Py S4/CelTOSCTRP salivary gland sporozoites by either indirect immunoflourescence assay (IFA) or western blot, while expression of other adhesion and motility motor proteins such as CSP, TRAP and MTIP were unaffected (Fig. 1D, Fig. S3B-D). Collectively these data demonstrate that the CTRP promoter
This article is protected by copyright. All rights reserved.
swap strategy substantially rescued the deficit in midgut infection associated with the complete S4/CelTOS knockout, yielding robust salivary gland sporozoite numbers that were deficient in S4/CelTOS.
2.2 S4/CelTOS is important for mammalian host infection. The generation of Py S4/CelTOSCTRP achieved robust salivary gland infection levels and as such enabled an unencumbered study of the function of S4/CelTOS during PE infection. Other studies have previously suggested that S4/CelTOS is important in host cell traversal by sporozoites, and may function in a traversal pathway that includes another micronemal protein, PLP1/SPECT2 (Jimah et al., 2016; Kariu et al., 2006; Risco-Castillo et al., 2015). Thus, we created a parasite deficient in PLP1/ SPECT2 (Py plp1/spect2¯) (Fig. S4), which allowed for direct comparisons with Py S4/CelTOSCTRP parasites.
BALB/cJ mice were infected with Py WT, Py S4/CelTOSCTRP, or Py plp1/spect2¯ salivary gland sporozoites by both intravenous (i.v.) injection and mosquito bite (Fig. 2, Table 1). Injection of 100 sporozoites of each parasite line revealed that Py S4/CelTOSCTRP and Py plp1/spect2¯ sporozoites were significantly less infectious than Py WT sporozoites (p
