Overall Introduction and Aims
Mosquito-borne pathogens such as parasites and arthropod-borne viruses (arboviruses) pose a significant risk to public health [
1]. Malaria is one of the most common parasitic diseases globally with an estimated 249 million cases and 608,000 deaths reported in 2022, mostly in children under five in sub-Saharan Africa [
2]. Over half the world’s population are at risk of infection by arboviruses, including Rift Valley fever virus (RVFV), yellow fever virus (YFV), Zika virus (ZIKV), Japanese encephalitis virus (JEV), West Nile virus (WNV), Dengue virus (DENV), and Chikungunya virus (CHIKV) [
3].
Due to climate change and other anthropogenic factors, the burden of mosquito-borne diseases is intensifying [
4,
5,
6]. There is no single solution for the control of mosquito-borne disease and multiple strategies are required. This multi-pronged approach will require location-specific strategies influenced by environmental and economic factors, governing bodies and disease prevalence [
7]. In this regard, novel strategies and tools are urgently required to develop an integrated control strategy.
Paratransgenesis, the genetic engineering of symbionts with anti-pathogenic effectors to control disease transmission, represents a potentially promising strategy. The technique was originally developed by Beard et al. to control
Rhodnius prolixus (triatomine/kissing bug) from spreading the causal parasite of Chagas disease (
Trypanosoma cruzi). A gram-positive bacteria,
Rhodococcus rhodnii, that occurs at high concentrations within the hindgut of
R. prolixus was genetically engineered to express a trypanocidal immune peptide, Cecropin A. This resulted in a decreased
T. cruzi infection rate in
R. prolixus and was approved as an integrated pest management program in South and Central America [
8].
Paratransgenesis offers several advantages. It is scalable because transgenic microorganisms can be grown to large quantities at low cost. The technique is not limited to single mosquito species because symbiotics can potentially colonise multiple important vector species. Moreover, the symbiont can be maintained within the ecosystem by vertical, horizontal, and trans-stadial transmission, mitigating the need for re-introduction [
9,
10]. Finally, and perhaps most importantly, it is a manipulable system that can be altered to target different pathogens or keep pace with resistance by exploiting different effectors. As such, the discovery and development of novel anti-pathogen molecules is critical for paratransgenesis implementation.
Venom toxins are excellent candidates for effectors in paratransgenesis. Venoms are complex mixtures of toxic proteins, peptides, and small molecules that are delivered through the infliction of a wound [
11,
12]. Venoms of hymenopteran insects such as bees and wasps are diverse, consisting of peptides, enzymes, and neurotransmitters [
13], whilst scorpion and spider venoms largely consist of neurotoxins, which modulate a variety of channels including voltage-gated potassium, sodium, and calcium ion channels, acid sensing ion channels, calcium-activated potassium channels, glutamate receptors, and glutamate transporters [
14,
15]. Snake venom consists of haemotoxins, cytotoxins, and neurotoxins that can be grouped into superfamilies by structure, with phospholipase A2s (PLA
2s), snake venom metalloproteinases, snake venom serine proteinases, and three-finger toxins being the most abundant [
16]. Venom toxins have high specificity, potency, and stability [
11], and are less susceptible to bioaccumulation than chemical insecticides [
17]. The venoms of many hymenopteran insects, scorpions, spiders, and snakes have been studied for their potential antiparasitic [
18,
19,
20] and antiviral [
21,
22,
23,
24] properties. This review highlights the untapped potential of venom toxins as effectors in paratransgenesis. We discuss successful paratransgenesis studies that have been undertaken with antimalarial venom toxins as proof of principle and the need for specific screening of venom toxins to identify effectors is highlighted. Regarding mosquito-borne diseases, paratransgenesis strategies have focused on targeting
Plasmodium, the causal agent of malaria. However, we suggest that paratransgenesis could be applicable to target arboviruses through the use of antiviral venom toxins.
This review highlights the untapped potential of venom toxins as effectors in paratransgenesis. We discuss successful paratransgenesis studies that have been undertaken with antimalarial venom toxins as proof of principle and the need for specific screening of venom toxins to identify effectors is highlighted. Regarding mosquito-borne diseases, paratransgenesis strategies have focused on targeting Plasmodium, the causal agent of malaria. However, we suggest that paratransgenesis could be applicable to target arboviruses through the use of antiviral venom toxins.
Antiparasitic Venom Toxins as Effectors to Target Plasmodium
Previous mosquito paratransgenesis strategies have focussed on targeting
Plasmodium, the casual parasite of malaria. The species of
Plasmodium responsible for causing malaria in humans are
P. falciparum,
P. vivax,
P. ovale,
P. knowlesi, and
P. malariae, with the former being responsible for more than 90% of the malaria deaths. Mosquitoes from the Anopheles genus are responsible for the transmission of malaria. Paratransgenesis targeting
Plasmodium must use effectors that inhibit the parasite stages within the mosquito: gametes, ookinetes, oocysts or sporozoites (
Figure 1) [
25]. Two venom toxins have effectively been utilised as effectors (
Table 1): Scorpine, an excitatory neurotoxin from
Pandinus imperator with antibacterial and antiparasitic properties, and mPLA
2, a PLA
2 from bee venom with a point mutation (H67N) to prevent enzyme activity and toxicity to bacteria. mPLA
2 expressed in
Escherichia coli induced a moderate reduction of oocyst numbers from
P. berghei, a rodent malaria model, when fed to
Anopheles stephensi, however, the bacterium survived poorly in the mosquito [
26]. mPLA
2 and Scorpine expressed in
P. agglomerans [
27] and
Serratia [
28] were able to effectively colonise the midgut of
An. gambiae and decreased the number of
P. falciparum oocytes in infected mosquitoes. Scorpine has also been expressed in
Asaia [
29], a bacteria found in
Anopheles sp.,
Aedes aegypti, and
Aedes albopictus [
30,
31,
32,
33,
34,
35,
36] that is transmitted vertically, horizontally and transstadially [
31]. Transgenic
Asaia expressing Scorpine significantly reduced the number of
P. berghei oocytes in the mosquito midgut, however, constitutive expression of the toxin compromised bacterial fitness. To improve bacterial fitness, blood meal inducible promoters within the mosquito microbiome were identified and used to conditionally express Scorpine. This enabled
Asaia to maintain fitness and compete with wild type
Asaia, whilst oocyst midgut number in
A. stephensi decreased by approximately 90% and prevalence decreased by up to 20%, indicating a decrease in infection potential [
37]. Future work should assess transgenic bacteria in semi-field trials to assess suitability for practical use within the field.
Despite promising preliminary mosquito paratransgenesis data, only a limited number of effector molecules have been assessed with mPLA
2 and Scorpine being the only venom toxins effectors that have been experimentally tested in mosquito paratransgenesis. An expanded arsenal of molecules is required to allow a multi-faceted and adaptable approach to paratransgenesis. Importantly, expression of multiple effectors has been shown to enhance efficacy [
27,
28] and can enable several stages of the pathogen life cycle to be targeted, increasing robustness. The risk of resistance development can be reduced through identification and use of multiple effectors with different mechanisms of actions and/or broad-spectrum actions. There is also a potential to target multiple pathogens through co-expression of effectors or use of effectors with multiple mechanism of actions. Finally, it is important to have a diverse effector library available to mitigate resistance and enable new paratransgenesis replacement strategies.
Venom toxins have the potential to act as effectors due to their antiparasitic activity (
Supplementary Table S1,
Figure 1). However, the majority of these studies have focussed on the intraerythrocytic asexual stages of
Plasmodium within the mammalian host [
19], in line with research more applicable to the identification of antimalaria therapeutics. Few studies have screened toxins to identify effectors for paratransgenesis, but for effectors to be useful they must target the
Plasmodium stages occurring in the mosquito [
25].
Several α-helical linear peptides such as Anoplin and Mastoparan X from wasp venom [
38], Melittin from European honeybee venom [
38], and MeuTXKβ [
39] from
Mesobuthus scorpion venom inhibit ookinete development. Whilst, another linear helical peptide, specifically scorpion toxin VmCT1 from
Vaejovis mexicanus is effective
in vitro against
P. gallinaceum sporozoites, a poultry model of the last stage of
Plasmodium development within the mosquito [
40]. Antimicrobial peptides from scorpions including Scorpine and synthetic peptides based on Vejovine and Hadrurin also inhibit ookinete development
in vitro [
41,
42,
43].
In vivo studies have found PLA
2 derived from the venom of the rattlesnake
Crotalus adamanteus reduced the number of oocysts by 99% when mixed with cultured
P. falciparum gametocytes and fed to
An. gambiae or
An. stephensi mosquitos [
44]. A similar reduction in
P. gallinaceum oocyst number in
Ae. aegypti was achieved. Interestingly, the PLA
2 toxin did not affect ookinete viability, but acted on the midgut surface, preventing ookinete maturation to oocytes. A similar effect was observed for a PLA
2 from bee venom in
Ae. fluviatilis [
45] and Melittin [
38].
Ookinete to oocyst development in the midgut, the bottleneck of malaria transmission [
46], is the best target for effector screening. The emphasis should be on
P. falciparum, the species primarily infecting humans and causing most malaria deaths. Although experiments with
P. falciparum are challenging and limited to
in vitro culture of the intraerythrocytic stages of the parasite, ookinetes can be generated
in vitro by gametocyte differentiation in specialised medium enabling toxins to be rapidly screened [
47]. However, studies have mainly been performed with
P. berghei and
P. gallineceum, species that infect rodents and poultry, respectively, as described above. This is because they can be maintained as intraerythrocytic stages in mice or chickens to generate a high density of gametocytes, the sexual stage that is required for oocyst development within the mosquito. An ideal high-throughput pipeline would be the screening of toxins on ookinete development
in vitro with successful candidates taken forward to
in vivo development of
Plasmodium in the mosquito, as described by Carter et al., 2013 [
38].
In addition, downstream assays must be undertaken to ensure the toxin is specific for the pathogen. This should involve assessing any potential effect of mosquito fitness, as performed in a previous study to identify effectors [
38]. Assays included basic
in vitro cell viability assays on mosquito cell lines and feeding toxins to mosquitoes. However, the compatibility of the toxins with the
mosquito symbiont must also be assessed by performing minimum inhibitory concentration assays. From an ecological perspective it is important to assess effects on other insects, especially pollinator species, using similar assays. From a safety perspective, mammalian toxicity should also be reviewed using
in vivo and
in vitro assays. These types of studies have largely been neglected to date.
Could Paratransgenesis Be Used to Target Arboviruses?
Paratransgenesis to target arboviruses has not been attempted thus far but the antiviral properties of venom toxins are encouraging for this strategy. Antiviral compounds can target various stages of virus infection including pre-entry and/or post-entry stages [
48]. Compounds can inactivate the virus pre-entry by inactivating the virus before it attaches to the cell, a process known as neutralisation; inhibiting surface proteins required for attachment; inhibiting virus endocytosis or inhibiting fusion of the viral envelope and host cell membrane. Alternatively, compounds can act at the post-entry stages, by inhibiting viral uncoating, replication, transcription, translation, virus assembly, and virus release. Antivirals can also induce the host immune response, through stimulating the production of interferons, other cytokines, and chemokines, affecting both pre- and post-entry stages. Targeting any of these stages within the mosquito midgut, as the location of arbovirus infection after the mosquito takes a blood meal from an infected host, has the potential to prevent viral dissemination into salivary glands. Blocking this step, as with
Plasmodium, would prevent the mosquito becoming infectious and transmitting the arbovirus and has been suggested previously as a strategy to control arbovirus transmission [
49].
Venom toxins have shown antiviral activity against ZIKV, DENV, YFV, JEV, and CHIKV. However, there is limited research on the antiviral properties of venom toxins against RVFV. Many antiviral venom toxins have been shown to target the pre-entry stages, the most studied of which being group I and II snake venom PLA
2 toxins (
Figure 2,
Supplementary Table S2). Group I PLA
2, consisting of PLA
2 produced by
Elapidae (cobras, mambas, coral snakes) and
Hydrophidae (sea snakes) whilst group II PLA
2 are produced by
Viperidae (rattlesnakes) [
12]. Group II PLA
2s derived from
Bothrops alteratus [
50],
B. leucurus [
51], and
B. asper venom [
52] can neutralise several strains of DENV, whilst group II PLA
2s from
B. jararacussu [
53,
54] and
Crotalus durissus terrificus venom have shown inhibition activity against YFV, CHIKV, DENV, and ZIKV [
55,
56,
57,
58,
59]. LaPLA
2-1, a group III PLA
2 from the scorpion
Liocheles australasiae, can neutralise DENV and JEV [
60]. Interestingly, DENV propagated in mosquito cell lines was more sensitive to Mt-I, a catalytically inactive PLA
2 from
B. asper venom, than viruses propagated in mammalian cells [
52]. Neutralisation by group I, II, and III PLA
2 likely occurs by hydrolysis of the virus lipid bilayer [
55,
59,
60]. Viral neutralisation has also been shown with to occur with ZY13, a peptide analogue of cathelicidin from
Bungarus fasciatus venom [
61] and the Scorpine-like peptide Smp76 from
Scorpio maurus palmatus venom [
62,
63].
Venom toxins can also inhibit virus post-entry stages (
Figure 2). Ev37, a Scorpine-like peptide from scorpion
Euscorpiops validus venom that selectively inhibit K
v1.3 potassium channel, prevents viral genome release into the cytoplasm by acidifying viral genome-containing vesicles preventing membrane fusion [
64]. The host defense peptide Av-LCTX-An1a from
Alopecosa nagpag spider venom can inhibit viral protease activity preventing virus maturation [
62]. Studies assessing the host immune response have found that Scorpine-like peptide rSmp76 from scorpion
Scorpio maurus palmatus venom and ZY13 have antiviral effects by activating interferon signaling [
61,
62]. However, it is important to stress that most of these studies have been undertaken with mammalian cell lines, and their translatability into mosquito cells is unknown. Promisingly, recombinant Scorpine generated in
Anopheles gambie cells can inhibit DENV serotype 2 replication in mosquito cells [
42] showing the potential of venom toxins to have antiviral activity within mosquitoes. Similar studies with the aforementioned venom toxins are necessary to determine if the antiviral activity seen within mammalian cells is transferable to mosquitoes.
In vivo studies assessing viral load, for example by RT-qPCR and plaque assays, in mosquitoes fed with toxins and virus must also be conducted to confirm
in vitro findings.
Screening to identify new potential antiviral effectors should involve assessing the ability of toxins to inhibit virus at all stages of infections. Viral neutralization should be assessed by incubating the test compound with the virus then assessing virus titre. Effect of the toxins at pre-entry stages should be evaluated by simultaneously adding the toxin and virus to mosquito cells at 4 °C (to prevent virus internalisation) and quantifying the levels of bound virus, as well as simultaneously adding the compounds and virus at 37 °C to determine effects on virus internalization and entry. Toxins should also be added after viral infection to evaluate post-entry antiviral activity. Finally, the ability of the toxin to induce a cellular antiviral response can be determined by addition of the toxin to the host cells pre-viral infection. Similarly, with antiparasitic effectors, any potential candidates should be further tested to ensure the toxin does not affect the fitness of the mosquito and symbiont. This is a vital step before moving forward with genetically engineering the symbiont.