**1. Introduction**

In 2020 alone, more than 240 million cases of malaria were reported leading to 627,000 deaths. These values represent a substantial increase in the number of malaria case incidence and deaths estimated globally, fueled by the disruptions caused by the COVID-19 pandemic [1]. In 2021, the World Health Organization recommended for the first time a malaria vaccine, RTS,S/AS01, for use in children living in endemic areas with moderate to high transmission [1]. However, this vaccine only confers moderate protection against clinical disease by *Plasmodium falciparum*, the most dangerous human malaria parasite [2]. RTS,S/AS01 targets the circumsporozoite protein (CSP), the protein that densely coats the surface of sporozoites, the parasite stage deposited in the skin of the mammalian host

**Citation:** Sá, M.; Costa, D.M.; Teixeira, A.R.; Pérez-Cabezas, B.; Formaglio, P.; Golba, S.; Sefiane-Djemaoune, H.; Amino, R.; Tavares, J. MAEBL Contributes to *Plasmodium* Sporozoite Adhesiveness. *Int. J. Mol. Sci.* **2022**, *23*, 5711. https://doi.org/ 10.3390/ijms23105711

Academic Editor: Michail Kotsyfakis

Received: 15 April 2022 Accepted: 13 May 2022 Published: 20 May 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

by infected mosquitoes. Sporozoites actively migrate in the skin and invade blood vessels to complete their development in the liver. Inside hepatocytes, a single sporozoite will transform and multiply into thousands of merozoites, the red blood cells infective forms. Sporozoites and ensuing liver stages, called the pre-erythrocytic phase, represent an attractive target for immune interventions [3].

Sera from individuals immunized with radiation-attenuated *P. falciparum* sporozoites, the gold standard malaria vaccine, contain antibodies against multiple pre-erythrocytic antigens highly associated with sporozoite-induced protection [4]. In an attempt to find novel pre-erythrocytic antigens, Peng and colleagues screened a library of *P. falciparum* antigens with sera from volunteers immunized by mosquito bite under chemoprophylaxis with chloroquine [5]. One of the antigens recognized by the sera from most of the individuals was the membrane-associated erythrocyte binding-like protein (MAEBL) [5].

MAEBL is a large type I transmembrane protein composed of two N-terminal cysteinerich adhesion domains homologous to the apical membrane antigen 1 (AMA-1), named M1 and M2, and a C-terminal cysteine-rich region (C-cys) structurally related with *Plasmodium* Duffy binding-like family of erythrocyte binding proteins [6]. Conserved among *Plasmodium* species [7], MAEBL was initially reported as an erythrocytic-binding protein present in blood-stage parasites [6,8], but was later found to be expressed in sporozoites and late liver stages [9–12]. Although dispensable for asexual blood-stage growth [13–16], immunization with MAEBL M2 domain protects animals from dying of a challenge with the lethal *Plasmodium yoelii* YM strain infected red blood cells [17].

MAEBL is required for the colonization of the mosquito salivary glands by sporozoites [13,14,16]. Two main *maebl* transcripts are expressed in sporozoites as a result of the alternative splicing in 30 exons, encoding a canonical transmembrane and a putative soluble MAEBL isoform [12]. However, only the transmembrane isoform is essential for *P. falciparum* sporozoite infection of salivary glands [16].

In sporozoites, MAEBL is found associated with the micronemes [13,14]. However, immunolabelling studies indicate that its subcellular localization is developmentally regulated during parasite maturation, as it changes from being restricted to the apical pole in immature sporozoites, to covering the surface of mature parasites colocalizing with CSP [11]. In salivary gland sporozoites, the protein was detected both internally and on the parasites surface [11,18]. Nevertheless, antibodies generated against MAEBL domains often recognize multiple bands on western blot analysis of parasite extracts that might hinder conclusions on the localization, particularly when sera reactivity is not evaluated also in a knockout line [18].

While MAEBL was suggested to be dispensable for liver infection by *P. berghei* sporozoites collected from the midgut of mosquitoes [13], MAEBL-deficient *P. falciparum* sporozoites from the hemolymph have been shown to exhibit impaired hepatocyte wounding and invasion capacities along with reduced liver infection of humanized chimeric mice [14]. Indeed, antibodies against MAEBL partially inhibit hepatocyte invasion by sporozoites and/or liver-stage development [5,18], supporting a role for MAEBL in sporozoite infectivity in the mammalian host. In this study, and using the rodent malaria model *P. berghei*, we aimed at understanding the contribution of this protein in the sequence of events that lead to a successful establishment of liver infection by sporozoites.

#### **2. Results**

#### *2.1. Genetic Complementation Reverts the Phenotype of maebl- Parasites in the Mosquito*

A *maebl* knockout (*maebl*-) line was generated in a bioluminescent background of *P. berghei*, by replacing the open reading frame (ORF) of *maebl* with the selectable marker *Toxoplasma gondii* dihydrofolate reductase-thymidylate synthase by double-crossover homologous recombination (Figure S1A). Three *maebl*- isogenic lines (*maebl*- B2, B3, and G3) were generated and their genotype was verified by PCR and Southern blot analysis (Figure S1B,C). The absence of *maebl* transcripts was confirmed for the *maebl*- lines by RT-PCR (Figure S1D) and the data presented throughout this work refers to *maebl*- G3 clone.

A genetic complementation approach was simultaneously adopted to directly link the defective phenotypes of *maebl*- parasites to the absence of MAEBL. As the *P. falciparum maebl* is transcribed along with the upstream gene as a bicistronic transcript [19], the full-length gene was re-introduced into the original locus of *maebl* together with the human dihydrofolate reductase selectable marker cassette, by a single-crossover homologous recombination event (Figure S1E). A *maebl* complemented isogenic line (*maebl\_comp* V3) with the expected genotype was isolated and used in further studies (Figure S1F). defective phenotypes of *maebl*- parasites to the absence of MAEBL. As the *P. falciparum maebl* is transcribed along with the upstream gene as a bicistronic transcript [19], the fulllength gene was re-introduced into the original locus of *maebl* together with the human dihydrofolate reductase selectable marker cassette, by a single-crossover homologous recombination event (Figure S1E). A *maebl* complemented isogenic line (*maebl\_comp* V3) with the expected genotype was isolated and used in further studies (Figure S1F). To investigate the development of *maebl*- and *maebl\_comp* mutant lines in the vector,

G3) were generated and their genotype was verified by PCR and Southern blot analysis (Figure S1B,C). The absence of *maebl* transcripts was confirmed for the *maebl*- lines by RT-PCR (Figure S1D) and the data presented throughout this work refers to *maebl*- G3 clone. A genetic complementation approach was simultaneously adopted to directly link the

*Int. J. Mol. Sci.* **2022**, *23*, x FOR PEER REVIEW 3 of 24

To investigate the development of *maebl*- and *maebl\_comp* mutant lines in the vector, mosquitoes were fed on mice infected with control, *maebl*-, or *maebl\_comp* parasites. Between days 18 and 26 post-infection, mosquitoes were dissected and the numbers of sporozoites collected from their midguts, hemolymph, and salivary glands were determined. While there were no significant differences between the numbers of midgut sporozoites among all lines, we frequently found higher numbers of *maebl*- sporozoites in the hemolymph, an observation consistent with the inability of these parasites to colonize the salivary glands (Figure 1A) [13]. Transmission electron microscopy (TEM) analysis showed no *maebl*sporozoites inside salivary glands even when these are collected at a late time point postinfection such as day 27 (Figure 1B), suggesting that the few sporozoites recovered most likely result from contamination with hemolymph. Importantly, we found no differences between the number of control and *maebl\_comp* salivary gland sporozoites (Figure 1A), which indicates the genetic complementation rescued sporozoite infectivity to the vector. mosquitoes were fed on mice infected with control, *maebl*-, or *maebl\_comp* parasites. Between days 18 and 26 post-infection, mosquitoes were dissected and the numbers of sporozoites collected from their midguts, hemolymph, and salivary glands were determined. While there were no significant differences between the numbers of midgut sporozoites among all lines, we frequently found higher numbers of *maebl*- sporozoites in the hemolymph, an observation consistent with the inability of these parasites to colonize the salivary glands (Figure 1A) [13]. Transmission electron microscopy (TEM) analysis showed no *maebl*- sporozoites inside salivary glands even when these are collected at a late time point post-infection such as day 27 (Figure 1B), suggesting that the few sporozoites recovered most likely result from contamination with hemolymph. Importantly, we found no differences between the number of control and *maebl\_comp* salivary gland sporozoites (Figure 1A), which indicates the genetic complementation rescued sporozoite infectivity to the vector.

**Figure 1.** Development of *maebl*- and *maebl\_comp* parasites in the mosquito. (**A**) Number of sporozoites in the midgut (MG), hemolymph (Hemo), and salivary glands (SG) of mosquitoes infected with Control, *maebl\_comp*, or *maebl*- parasites, on days 18 to 26 post-infection. Symbols represent the counts in independent experiments and bars indicate the mean + SD. Statistical significance was determined using one-way ANOVA (Kruskal–Wallis test with Dunn's multiple comparisons test). (**B**) Transmission electron micrographs of salivary glands of Control- or *maebl*- - **Figure 1.** Development of *maebl*- and *maebl\_comp* parasites in the mosquito. (**A**) Number of sporozoites in the midgut (MG), hemolymph (Hemo), and salivary glands (SG) of mosquitoes infected with Control, *maebl\_comp*, or *maebl*- parasites, on days 18 to 26 post-infection. Symbols represent the counts in independent experiments and bars indicate the mean + SD. Statistical significance was determined using one-way ANOVA (Kruskal–Wallis test with Dunn's multiple comparisons test). (**B**) Transmission electron micrographs of salivary glands of Control- or *maebl*- -infected mosquitoes, dissected on day 27 post-infection. SC, secretory cavity; SD, salivary duct; red arrows, sporozoites. Scale bar, 1 µm (**left** panel) or 2 µm (**right** panel). (**C**) Ratio of hemolymph (Hemo) to midgut (MG) sporozoites in Control- or *maebl*- infected mosquitoes, between days 18 and 21 post-infection. Symbols

represent values of independent experiments and bars indicate the mean + SD. Statistical significance was determined using the Mann–Whitney test. (**D**) Viability of sporozoites. Hemolymph sporozoites were collected from Control- or *maebl*- -infected mosquitoes, on days 18 and 19 post-infection. Sporozoites were activated with DMEM supplemented with 5% FBS at room temperature (activated RT) or incubated with saline phosphate buffer on ice or at room temperature (non-activated Ice and RT, respectively). Propidium iodide (PI) was then added to the parasite suspensions and sporozoites were immediately imaged. **Left**: sporozoites were manually classified as PI+ or PI− sporozoites (dead or viable, respectively). The graphic shows the mean of two independent experiments performed in duplicated + SD. At least 150 sporozoites were analyzed per replicate. Statistical analysis was performed using the unpaired *t*-test. Right: representative images of PI+ or PI− sporozoites. Scale bar, 5 µm. ns, non-significant. \* *p* < 0.05; \*\* *p* < 0.01; \*\*\*\* *p* < 0.0001.

In agreement with what was previously reported [13], our data show that *maebl*sporozoites accumulate in the vector circulatory system (Figure 1A,C). To test whether the accumulation of sporozoites in the hemolymph of infected mosquitoes led to a reduction in parasite viability, we performed a standard propidium iodide (PI) exclusion assay. The percentage of viable *maebl*- hemolymph sporozoites, collected from mosquitoes at day 18/19 post-infection, was not significantly different from that of the control even following activation in the presence of serum (Figure 1D). These results validate the use of *maebl*hemolymph sporozoites in subsequent experiments.

#### *2.2. maebl- Sporozoites Exhibit Decreased Infectivity to Mice*

It has been previously suggested that MAEBL is dispensable for the infection of rat livers by *P. berghei* midgut sporozoites [13]. In contrast, *maebl*- *P. falciparum* sporozoites collected from the mosquito hemolymph showed reduced infectivity to chimeric mice with humanized livers [14]. Therefore, to evaluate whether this phenotype is speciesspecific [13,14], we assessed the infectivity of *maebl*- and *maebl\_comp* hemolymph sporozoites to C57BL/6 mice using in vivo bioluminescence imaging. Mice were inoculated intravenously (i.v.) with control, *maebl*- or *maebl\_comp* hemolymph sporozoites, and the bioluminescent signal in the liver was quantified 1- and 2-days post-infection (D1 and D2, respectively). Animals infected with *maebl*- sporozoites showed a reduced liver burden compared to both control and *maebl\_comp* at D1 and D2 (Figure 2A). *maebl*- parasites only emerged in the blood of 3 out of 4 mice and after a delay of 2 days comparing with the other lines (Figure 2B). These observations were consistently reproducible, as we frequently observed 1 to 2 days of delay in the prepatent period of mice inoculated i.v. with *maebl*sporozoites, in several independent experiments (data not shown). Once in the blood, *maebl*parasites exhibited normal asexual growth kinetics as determined by counting the percentage of infected red blood cells (Figure 2B). Although mice inoculated with *maebl\_comp* hemolymph sporozoites displayed lower parasite loads in the liver at D1 compared to control-infected animals (~3.0-fold reduction), the reduction was no longer observed at a later time-point (Figure 2A). In agreement with this observation, no differences were seen in the prepatent periods or in the blood-stage growth of *maebl\_comp* and control parasites (Figure 2B).

Genetically complemented *maebl*- sporozoites successfully enter the mosquito salivary glands (Figure 1A). To assess whether *maebl\_comp* sporozoites have completed their maturation in the vector and efficiently infect the mammalian host, mice were inoculated i.v. with *maebl\_comp* or control sporozoites collected from the mosquito salivary glands. As expected, no differences were observed in the bioluminescent signal between experimental groups at D1 and D2 (Figure 2C), as well as in the parasitemia of animals (Figure 2D). The *maebl*- line was not used in these experiments due to its reduced number of salivary glands-associated sporozoites. Therefore, mice were inoculated i.v. with the few parasites we could collect but no animal ever became blood-stage positive, contrarily to mice infected with similar numbers of control salivary gland sporozoites (Table S1). Altogether, our data

demonstrate that in the absence of MAEBL, *P. berghei* hemolymph sporozoites exhibit an impaired ability to infect the liver of mice. *Int. J. Mol. Sci.* **2022**, *23*, x FOR PEER REVIEW 5 of 24

> **Figure 2.** Infectivity of *maebl*- and *maebl\_comp* sporozoites to mice. (**A–D**) Infectivity of *maebl*- and *maebl\_comp* sporozoites to C57BL/6 mice. Mice were injected intravenously with 3.5 × 104 Control, *maebl-* and *maebl\_comp* hemolymph sporozoites (panels A-B) or with 2.5 × 104 Control and *maebl\_comp* salivary gland sporozoites (panels C-D), collected from mosquitoes on days 20 or 21 post-infection. (**A**,**C**) *Left:* parasite burdens in the liver were quantified as average radiance (photons/s/cm2/steradian) one (D1) and two (D2) days post-infection. Symbols represent values for individual animals and bars indicate the mean + SD (*n* = 3–4). Dotted line: background level, calculated using non-infected mice. Statistical analysis was performed using one-way ANOVA (Tukey's multiple comparisons test; panel (**A**) or unpaired *t*-test (panel C). *Right*: representative images of infected mice, on D1 and D2 post-infection. (**B**,**D**) Parasitemia of infected mice, determined daily by a Giemsa-stained blood smear. Symbols represent values for individual animals and bars indicate the mean + SD. Statistical analysis was performed using unpaired *t*-test (**B**,**D**) or one-way ANOVA (Tukey's multiple comparisons test) (**B**)]. ns, non-significant; \* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001; NI, non-infected. Genetically complemented *maebl*- sporozoites successfully enter the mosquito **Figure 2.** Infectivity of *maebl*- and *maebl\_comp* sporozoites to mice. (**A**–**D**) Infectivity of *maebl*and *maebl\_comp* sporozoites to C57BL/6 mice. Mice were injected intravenously with 3.5 <sup>×</sup> <sup>10</sup><sup>4</sup> Control, *maebl*- and *maebl\_comp* hemolymph sporozoites (panels **<sup>A</sup>**,**B**) or with 2.5 <sup>×</sup> <sup>10</sup><sup>4</sup> Control and *maebl\_comp* salivary gland sporozoites (panels **C**,**D**), collected from mosquitoes on days 20 or 21 post-infection. (**A**,**C**) **Left**: parasite burdens in the liver were quantified as average radiance (photons/s/cm2/steradian) one (D1) and two (D2) days post-infection. Symbols represent values for individual animals and bars indicate the mean + SD (*n* = 3–4). Dotted line: background level, calculated using non-infected mice. Statistical analysis was performed using one-way ANOVA (Tukey's multiple comparisons test; panel (**A**) or unpaired *t*-test (panel **C**). **Right**: representative images of infected mice, on D1 and D2 post-infection. (**B**,**D**) Parasitemia of infected mice, determined daily by a Giemsa-stained blood smear. Symbols represent values for individual animals and bars indicate the mean + SD. Statistical analysis was performed using unpaired *t*-test (**B**,**D**) or one-way ANOVA (Tukey's multiple comparisons test) (**B**). ns, non-significant; \* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001; NI, non-infected.

> salivary glands (Figure 1A). To assess whether *maebl\_comp* sporozoites have completed their maturation in the vector and efficiently infect the mammalian host, mice were inoculated i.v. with *maebl\_comp* or control sporozoites collected from the mosquito salivary glands. As expected, no differences were observed in the bioluminescent signal between experimental groups at D1 and D2 (Figure 2C), as well as in the parasitemia of animals (Figure 2D). The *maebl-* line was not used in these experiments due to its reduced

*Vitro* 

#### *2.3. maebl- Hemolymph Sporozoites Present Hampered Invasion and Wounding of Host Cells In Vitro* hemolymph sporozoites exhibit an impaired ability to infect the liver of mice. *2.3. maebl- Hemolymph Sporozoites Present Hampered Invasion and Wounding of Host Cells In*

contrarily to mice infected with similar numbers of control salivary gland sporozoites (Table S1). Altogether, our data demonstrate that in the absence of MAEBL, *P. berghei* 

*Int. J. Mol. Sci.* **2022**, *23*, x FOR PEER REVIEW 6 of 24

Next, we conducted several in vitro experiments to further explore the infectivity of *maebl*- sporozoites to the mammalian host. We started by evaluating sporozoite invasion and liver stage development inside the hepatoma cell line HepG2, using immunofluorescence microscopy. To that end, sporozoites were collected from the hemolymph or salivary glands and incubated with cells for 2 h to evaluate the invasion of host cells. Parasite development was analyzed at 48 h after infection. The percentage of cells with intracellular sporozoites was significantly reduced for the *maebl*- line compared to the control and *maebl\_comp* line (Figure 3A), leading to the formation of a lower number of exoerythrocytic forms (EEFs) (Figure 3B). No differences in the size of EEFs were observed among all lines (Figure 3C), suggesting that MAEBL is not required for liver stage development. As expected, *maebl\_comp* sporozoites did not show impaired hepatocyte invasion nor exoerythrocytic development (Figure 3A–C). Next, we conducted several in vitro experiments to further explore the infectivity of *maebl-* sporozoites to the mammalian host. We started by evaluating sporozoite invasion and liver stage development inside the hepatoma cell line HepG2, using immunofluorescence microscopy. To that end, sporozoites were collected from the hemolymph or salivary glands and incubated with cells for 2 h to evaluate the invasion of host cells. Parasite development was analyzed at 48 h after infection. The percentage of cells with intracellular sporozoites was significantly reduced for the *maebl-* line compared to the control and *maebl\_comp* line (Figure 3A), leading to the formation of a lower number of exoerythrocytic forms (EEFs) (Figure 3B). No differences in the size of EEFs were observed among all lines (Figure 3C), suggesting that MAEBL is not required for liver stage development. As expected, *maebl\_comp* sporozoites did not show impaired hepatocyte invasion nor exoerythrocytic development (Figure 3A–C).

**Figure 3.** Evaluation of HepG2 cells invasion and cell wounding activity of *maebl*- sporozoites. (**A– C**) Invasion of cells by *maebl-* and *maebl\_comp* sporozoites and exoerythrocytic forms (EEFs) development. Control, *maebl-* or *maebl\_comp* hemolymph (Hemo) or salivary glands (SG) sporozoites, collected from mosquitoes on days 19 to 21 post-infection, were incubated with cells for 2 h (panel A) or 48 h (panel B-C). (**A**) *Left:* percentage of cells containing intracellular sporozoites. Bars represent the mean + SD of experimental replicates. Statistical analysis was performed using unpaired *t*-test (SG) or one-way ANOVA (Tukey's multiple comparisons test; Hemo). *Right:*  **Figure 3.** Evaluation of HepG2 cells invasion and cell wounding activity of *maebl*- sporozoites. (**A**–**C**) Invasion of cells by *maebl*- and *maebl\_comp* sporozoites and exoerythrocytic forms (EEFs) development. Control, *maebl*- or *maebl\_comp* hemolymph (Hemo) or salivary glands (SG) sporozoites, collected from mosquitoes on days 19 to 21 post-infection, were incubated with cells for 2 h (panel **A**) or 48 h (panel **B**,**C**). (**A**) **Left**: percentage of cells containing intracellular sporozoites. Bars represent the mean + SD of experimental replicates. Statistical analysis was performed using unpaired *t*-test (SG) or one-way ANOVA (Tukey's multiple comparisons test; Hemo). **Right**: representative immunofluorescence images of intracellular (white arrow) and extracellular sporozoites (red arrow). DAPI-stained cell nuclei, cyan. Scale bar, 10 µm. (**B**) Number of EEFs in cells per well. Bars represent the mean + SD of experimental replicates. Values are representative of at least two independent experiments (panel **A**,**B**). Statistical analysis was performed using unpaired *t*-test (SG) or one-way ANOVA (Tukey's multiple comparisons test; Hemo). (**C**) **Left**: area of EEFs. Box plots showing the median, maximum, minimum, and the 25th and 75th percentiles of the area of individual EEFs. At least 45 EEFs were analyzed per

condition. Statistical significance was determined using one-way ANOVA (Kruskal–Wallis test with Dunn's multiple comparisons test). **Right**: representative immunofluorescence images of EEFs. CSP, red; GFP, green; DAPI-stained nuclei, cyan. Scale bar, 3 µm. (**D**) Cell wounding capacity of Control, *maebl*- and *maebl\_comp* Hemo or SG sporozoites collected from mosquitoes on day 19 post-infection. Sporozoites were incubated with cells for 60 min in the presence of propidium iodide (PI). The graph shows the percentage of wounded cells (PI+) assessed by flow cytometry analysis. Bars represent the mean + SD of experimental replicates; values are representative of 3 independent experiments. Horizontal dotted line: percentage of PI+ cells following incubation with medium only. Statistical analysis was performed using unpaired *t*-test (SG) or one-way ANOVA (Tukey's multiple comparisons test; Hemo). ns, non-significant. \* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001.

It has also been demonstrated that cell traversal activity is disrupted in the MAEBLdeficient *P. falciparum* sporozoites [14]. To assess whether this process is also affected in the *P. berghei* knockout line, we performed a standard in vitro cell wounding assay using PI [20]. During traversal, the plasma membrane of host cells is breached, allowing the incorporation of cell-impermeant dyes, such as PI. Thus, sporozoites were allowed to traverse HepG2 cells in the presence of PI for 1 h before quantification of the percentage of wounded cells by flow cytometry analysis. Whereas the percentage of PI+ cells obtained upon incubation with control and *maebl\_comp* hemolymph sporozoites was 13.3 ± 0.8% and 16.4 ± 1.8%, respectively, *maebl*- sporozoites induced PI-incorporation levels on host cells close to those of cells incubated with medium alone (4.1 ± 0.9%, Figure 3D and Figure S2). No differences were observed in the percentage of wounded cells by control and *maebl\_comp* sporozoites collected from either the hemolymph or the salivary glands (Figure 3D). Altogether, our data indicate that the absence of MAEBL results in a decrease of host cell invasion and wounding by *P. berghei* sporozoites in vitro.
