Bactericidal Properties of Proline-Rich Aedes aegypti Trypsin Modulating Oostatic Factor (AeaTMOF)

The antimicrobial properties of proline-rich Aedes aegypti decapeptide TMOF (AeaTMOF) and oncocin112 (1–13) were compared. Incubations with multidrug-resistant Escherichia coli cells showed that AeaTMOF (5 mM) was able to completely inhibit bacterial cell growth, whereas oncocin112 (1–13) (20 mM) partially inhibited bacterial growth as compared with bacterial cells that were not multidrug-resistant cells. AeaTMOF (5 mM) was very effective against Acinetobacter baumannii and Pseudomonas aeruginosa, completely inhibiting cell growth during 15 h incubations. AeaTMOF (5 mM) completely inhibited the Gram-positive bacteria Staphylococcus aureus and Bacillus thurengiensis sups. Israelensis cell growth, whereas oncocin112 (1–13) (10 and 20 mM) failed to affect bacterial cell growth. E. coli cells that lack the SbmA transporter were inhibited by AeaTMOF (5 mM) and not by oncocin112 (1–13) (10 to 20 mM), indicating that AeaTMOF can use other bacterial transporters than SbmA that is mainly used by proline-rich antimicrobial peptides. Incubation of E. coli cells with NaAzide showed that AeaTMOF does not use ABC-like transporters that use ATP hydrolysis to import molecules into bacterial cells. Three-dimensional modeling and docking of AeaTMOF to SbmA and MdtM transporters showed that AeaTMOF can bind these proteins, and the binding location of AeaTMOF inside these protein transporters allows AeaTMOF to be transported into the bacterial cytosol. These results show that AeaTMOF can be used as a future antibacterial agent against both multidrug-resistant Gram-positive and -negative bacteria.


Molecular Modeling and Docking of AeaTMOF 2.3.1. SbmA Transporter
Homology modeling of the ABC transporter corresponding to the sbmA gene of Escherichia coli K-12 (accession number WP_001304845.1) was conducted with the YASARA Structure program [27]. The SbmA model was built from the X-ray coordinates of the Staphylococcus aureus ABC transporter (PDB code 2HYD) [28] as a template. The geometric quality of the model was assessed using PROCHECK [29]. The residues of SbmA were correctly assigned to the allowed regions in the Ramachandran plot except for H229 and A290, which occupy a non-allowed region of the plot. Using ANOLEA [30] to evaluate the model, only 10 residues of SbmA out of 402 exhibited an energy over the threshold value. These residues are located in the loop region connecting α-helices and β-sheets. The calculated QMEAN score for the SbmA model is −3.92 [31,32]. Atomic coordinates of the E. coli ABC transporter MalFGK 2 (PDB code 4KHZ) [33], and E. coli ABC transporter McjD (PDB code 5OFP) [34], were obtained from the PDB [35].
The hydrophilic and hydrophobic regions distributed on the surface of SbmA were identified using Chimera [36]. The coulombic charges on the surface of SbmA were calculated using the classic values for the inner and outer dielectric constants obtained from to the proteins and the solvent (4.0 and 80.0, respectively).
The decapeptide AeaTMOF (YDPAPPPPPP) was built as a left-handed α-helix [37] using Chimera [36], and the structure was minimized by 1000 steps of steepest descent and 100 steps of conjugated gradient. Docking of AeaTMOF to SbmA was performed by using YASARA [27]. Additional docking experiments were also submitted to the GRAMM_X [38,39] web server. Molecular cartoons were drawn and rendered using YASARA Structure, Chimera, and ChimeraX programs [36].

MdtM Transporter
The E. coli membrane transporter MdtM was modeled using YASARA [27], and the following templates: E. coli multidrug transporter MtdA in complex with deoxycholate (PDB code 4ZP0) [40], the E. coli multidrug transporter Emrd (PDB code 2GFP) [41], the YajR transporter from E. coli (PDB code 3WDO) [42], and the E. coli POT transporter (PDB code 6EI3) [43]. A hybrid model was built and used as the final model for the E. coli MdtM transporter. PROCHECK [28] was used to assess the geometric quality of the model. All the residues of the modeled MdtM were correctly assigned in the allowed regions in the Ramachandran plot except for two residues, H8 and A195, which occur in a non-allowed  [31,32]. Atomic coordinates of the E. coli transporters were retrieved from the PDB [35]. The decapeptide AeaTMOF (YDPAPPPPPP) was built as a left-handed α-helix as was described above (Section 2.3.1). Docking of AeaTMOF to MdtM and POT (proton coupled) transporters was performed with YASARA. Additional docking experiments were submitted to the GRAMM_X [38,39] web server. Molecular cartoons were drawn and rendered as was described above for the SbmA transporter (Section 2.3.1) [44].

Statistical Analysis
Statistical analyses were conducted using GraphPad Prism and two-tailed unpaired student's t-test. Results were considered significant when p < 0.05. Each experimental point is an average of 3 determinations ± SEM.

Gram Negative Bacteria
E. coli cells (10 6 cpu/mL) that exhibit multi drug resistance were grown in 100 µL of LB medium containing ampicillin (100 µg/mL), kanamycin, and tetracycline (50 µg/mL/each) in a 96 well-plate at 37 • C for 15 h. Incubation of the bacterial cells with different concentrations of TMOF (0.1 to 5 mM) showed that AeaTMOF (5 mM) completely stopped bacterial growth and caused 99.9% cell death; 1-2 cpu were detected at 15 h as compared with 10 6 cpu/mL that were inoculated into each well. Lower concentrations of TMOF (0.1 and 1 mM, respectively) were not effective ( Figure 1A). On the other hand, oncocin112 (1-13) concentrations of 5, 10, and 20 mM were not effective, although 20 mM oncocin112 (1-13) stopped bacterial growth for 8 h; however, after 15 h the cells stated to rapidly grow. Cell growth, however, was 1.9-, 1.4-, and 1.2-fold lower for concentrations of 20 mM, 10 mM, and 5 mM oncocin112 (1-13), respectively, when compared with cells that were grown without oncocin112 (1-13) ( Figure 1B). Oncocin112 (1-13) was much more effective when incubated with E. coli cells that were not multi-drug resistant. It completely stopped bacterial growth at 20 mM, causing 90% inhibition at 10 mM and 50% inhibition at 5 mM after 15 h incubations as compared with control cells that were incubated without onconcin112 (1-13) ( Figure 1C).

Effect on A. baumannii
To find out if AeaTMOF inhibits the growth of A. baumanni, a bacterial pathogen that is primarily associated with hospital-acquired infection exhibiting multidrug resistance. A. baumanni cells (10 6 cpu/mL) were incubated with different concentrations of AeaTMOF (0.2, 1 and 5 mM). Control cells were incubated without AeaTMOF. Bacterial cells that were incubated with AeaTMOF did not grow and 99.9% of the cells were dead at 15 h. AeaTMOF concentrations (0.2 mM and 1 mM) were not effective, and bacterial cells that were incubated with these AeaTMOF concentrations grew at a similar rate as control cells that were not treated with AeaTMOF ( Figure 2A).

Effect on P. aeruginosa
To find out if AeaTMOF inhibits the growth of P. aeruginosa, a multidrug-resistant pathogen associated with serious hospital-acquired infections. P. aeruginosa cells (10 6 cpu/mL) were incubated with AeaTMOF (0.1 mM, 1 mM, and 5 mM) for 15 h. AeaTMOF (5 mM) completely stopped bacterial growth, and at the end of the incubation period (15 h), 99.9% of the bacterial cells were dead. Lower concentrations, however, of 0.1 mM and 1 mM did not inhibit bacterial growth and cells grew at a similar rate to control cells that were not incubated with AeaTMOF ( Figure 2B).

Effect on A. baumannii
To find out if AeaTMOF inhibits the growth of A. baumanni, a bacterial pathogen that is primarily associated with hospital-acquired infection exhibiting multidrug resistance. A. baumanni cells (10 6 cpu/mL) were incubated with different concentrations of AeaTMOF (0.2, 1 and 5 mM). Control cells were incubated without AeaTMOF. Bacterial cells that were incubated with AeaTMOF did not grow and 99.9% of the cells were dead at 15 h. AeaTMOF concentrations (0.2 mM and 1 mM) were not effective, and bacterial cells that  were incubated with these AeaTMOF concentrations grew at a similar rate as control cells that were not treated with AeaTMOF (Figure 2A). .

Effect on P. aeruginosa
To find out if AeaTMOF inhibits the growth of P. aeruginosa, a multidrug-resistant pathogen associated with serious hospital-acquired infections. P. aeruginosa cells (10 6 cpu/mL) were incubated with AeaTMOF (0.1 mM, 1 mM, and 5 mM) for 15 h. AeaTMOF (5 mM) completely stopped bacterial growth, and at the end of the incubation period (15 h), 99.9% of the bacterial cells were dead. Lower concentrations, however, of 0.1 mM and 1 mM did not inhibit bacterial growth and cells grew at a similar rate to control cells that were not incubated with AeaTMOF ( Figure 2B).

Effect on B. thuringiensis subsp. Israelensis
To find out whether AeaTMOF can inhibit the growth of B. thurengiensis subsp. Israelenis, a bacterium that produces Cry and Cyt toxins for the control of mosquito larvae, different concentrations of AeaTMOF (1 and 5 mM) were incubated with the bacterial cells (10 6 cpu/mL). A control was incubated without AeaTMOF for 15 h. Bacterial cells that were incubated in the presence of AeaTMOF (5 mM) did not grow, and 99.9% of the initial bacterial cells were dead. On the other hand, cells that were incubated in the presence of AeaTMOF (1 mM) or without AeaTMOF grew at a similar growth-rate, and at 15 h the number of bacterial cells grown with AeaTMOF and control cells were not significantly different ( Figure 4A). To find out if oncocin112 (1-13) also affects the bacterial cell growth, B. thuringiensis subsp. israelensis (10 6 cpu/mL) were incubated with different concentrations of oncocin112 (1-13) (10 and 20 mM) and bacterial growth was monitored at 1 h intervals as above. Bacterial cells that were incubated without oncocin112 (1-13) served as a control group. Although there was a slight growth inhibition initially when onconcin112 (1-13) (20 mM) was incubated with the bacterial cells, at 15 h all the cells grew similarly as compared with the control group that was incubated without oncocin112 (1-13) ( Figure 4B). These results indicate that oncocin112 (1-13) does not affect this bacterium's growth, as was shown for S. aureus (see above Section 3.2.1).
AeaTMOF (1 mM) or without AeaTMOF grew at a similar growth-rate, and at 15 h the number of bacterial cells grown with AeaTMOF and control cells were not significantly different ( Figure 4A). To find out if oncocin112 (1-13) also affects the bacterial cell growth, B. thuringiensis subsp. israelensis (10 6 cpu/mL) were incubated with different concentrations of oncocin112 (1-13) (10 and 20 mM) and bacterial growth was monitored at 1 h intervals as above. Bacterial cells that were incubated without oncocin112 (1-13) served as a control group. Although there was a slight growth inhibition initially when oncon-cin112 (1-13) (20 mM) was incubated with the bacterial cells, at 15 h all the cells grew similarly as compared with the control group that was incubated without oncocin112 (1-13) ( Figure 4B). These results indicate that oncocin112 (1-13) does not affect this bacterium's growth, as was shown for S .aureus (see above Section 3.2.1). .

The Role of the E. coli SbmA Transporter
SbmA is a bacterial inner membrane protein that forms a dimer and imports into the cytoplasm of E. coli peptides, proline-rich peptides, nucleic acids, antisense peptides, and several oligomers [45]. To find out if E. coli sbmA − cells lacking the transporter gene are affected by AeaTMOF and oncocin112 (1-13), E. coli CGSC strain 8547 sbmA − (10 6 cpu/mL) was incubated with different concentrations of AeaTMOF (1, 3, and 5 mM) or without TMOF (control) for 15 h, and at 1 h intervals cell growth was determined at 630 nm. AeaTMOF (5 mM) completely inhibited bacterial growth and partially inhibited cells growth (23%) at AeaTMOF (3 mM) as compared with control cells that were not incubated with AeaTMOF, indicating that other importers aside from SbmA transport AeaTMOF into the bacterium cell ( Figure 5A). On the other hand, incubations with higher concentrations of oncocin112 (1-13) (20 mM) were needed to achieve 43% inhibition in bacterial cell growth after 15 h incubation, and no inhibition was noted at a lower concentration of 10 mM ( Figure 5B), indicating that oncocin112 (1-13) is mainly imported into bacterial cells by the SbmA transporter. AeaTMOF, indicating that other importers aside from SbmA transport AeaTMOF into the bacterium cell ( Figure 5A). On the other hand, incubations with higher concentrations of oncocin112 (1-13) (20 mM) were needed to achieve 43% inhibition in bacterial cell growth after 15 h incubation, and no inhibition was noted at a lower concentration of 10 mM (Figure 5B), indicating that oncocin112 (1-13) is mainly imported into bacterial cells by the SbmA transporter. .

E. coli sbmA − Cell Growth in the Presence of AeaTMOF and NaAzide
Our results indicate that AeaTMOF is imported into E. coli cells that lack an SbmA importer ( Figure 5A). To find out if AeaTMOF uses an ABC importer like the AeaTMOF receptor/importer in Ae. aegypti [46], E. coli CGSC strain 8547 sbmA − (10 6 cpu/mL) was incubated for 18 h with AeaTMOF in the presence of NaAzide (25 and 250 mM) in order to inhibit ATPase activity that is associated with ABC importers. Controls were incubated without AeaTMOF and with NaAzide (50 and 250 mM) or without NaAzide. E. coli sbmA − cells that were incubated in the presence of AeaTMOF (5 mM) and in the presence of NaAzide (50 and 250 mM) or without NaAzide did not grow and were significantly

E. coli sbmA − Cell Growth in the Presence of AeaTMOF and NaAzide
Our results indicate that AeaTMOF is imported into E. coli cells that lack an SbmA importer ( Figure 5A). To find out if AeaTMOF uses an ABC importer like the AeaTMOF receptor/importer in Ae. aegypti [46], E. coli CGSC strain 8547 sbmA − (10 6 cpu/mL) was incubated for 18 h with AeaTMOF in the presence of NaAzide (25 and 250 mM) in order to inhibit ATPase activity that is associated with ABC importers. Controls were incubated without AeaTMOF and with NaAzide (50 and 250 mM) or without NaAzide. E. coli sbmA − cells that were incubated in the presence of AeaTMOF (5 mM) and in the presence of NaAzide (50 and 250 mM) or without NaAzide did not grow and were significantly inhibited as compared with controls (p < 0.0001), whereas controls that were incubated without AeaTMOF in the presence of NaAzide (50 and 250 mM) or in the absence of NaAzide grew normally and no significant difference was found when the three controls were compared ( Figure 6). These results suggest that in the absence of SbmA, AeaTMOF is using another transporter to enter the bacterial cells.
inhibited as compared with controls (p < 0.0001), whereas controls that were incubated without AeaTMOF in the presence of NaAzide (50 and 250 mM) or in the absence of NaAzide grew normally and no significant difference was found when the three controls were compared ( Figure 6). These results suggest that in the absence of SbmA, AeaTMOF is using another transporter to enter the bacterial cells.

Docking of AeaTMOF to SbmA and MdtM Transporters
Our results indicate that AeaTMOF is imported into bacterial cells using transporters that are not ABC importers. To show that AeaTMOF uses SbmA and MdtM importers, both do not use ATP to import proline rich molecules into bacterial cells, we docked AeaT-MOF to these importers.

SbmA
The modeled α-helical domain (α-Hd) of the proline-rich peptide transporter (SbmA) of E. coli exhibits a typical V-shaped structure (Figure 7(Aa)), embedded in the membrane lipid bilayer (MLB) via a highly hydrophobic region (Figure 7(Ab)), which extends the α-Hd into an intracellular domain, and linked with a smaller extracellular domain ( Figure  7(A(a,b)). Both domains are located on the sides of a hydrophobic α-helical region and are mainly hydrophilic (Figure 7(Ab)), displaying electropositive and electronegative charged patches on their surfaces (Figure 7(Ac)). Docking experiments performed with the AeaT-MOF decapeptide as a ligand resulted in optimal position of the decapeptide in the substrate-binding cavity located at the fork of the V-shaped α-Hd structure (Figure 7(Ad-f)). The binding-site accommodating the AeaTMOF decapeptide exhibits electrostatic, hydrophilic, hydrophobic, and stacking interactions, including a few hydrogen bonds that participate in anchoring AeaTMOF to the binding site (Figure 7(Af)). The α-helices and loops at the top of the extracellular α-helical domain are sufficiently open to allow flexibility that allows access to AeaTMOF and subsequent anchoring of the molecule to the binding site located in the intracellular part of E. coli SbmA, at the fork of α-Hd (Figure 7(Ad-f)).

Docking of AeaTMOF to SbmA and MdtM Transporters
Our results indicate that AeaTMOF is imported into bacterial cells using transporters that are not ABC importers. To show that AeaTMOF uses SbmA and MdtM importers, both do not use ATP to import proline rich molecules into bacterial cells, we docked AeaTMOF to these importers.

SbmA
The modeled α-helical domain (α-Hd) of the proline-rich peptide transporter (SbmA) of E. coli exhibits a typical V-shaped structure (Figure 7(Aa)), embedded in the membrane lipid bilayer (MLB) via a highly hydrophobic region (Figure 7(Ab)), which extends the α-Hd into an intracellular domain, and linked with a smaller extracellular domain (Figure 7(A(a,b)). Both domains are located on the sides of a hydrophobic α-helical region and are mainly hydrophilic (Figure 7(Ab)), displaying electropositive and electronegative charged patches on their surfaces (Figure 7(Ac)). Docking experiments performed with the AeaTMOF decapeptide as a ligand resulted in optimal position of the decapeptide in the substrate-binding cavity located at the fork of the V-shaped α-Hd structure (Figure 7(Ad-f)). The binding-site accommodating the AeaTMOF decapeptide exhibits electrostatic, hydrophilic, hydrophobic, and stacking interactions, including a few hydrogen bonds that participate in anchoring AeaTMOF to the binding site (Figure 7(Af)). The α-helices and loops at the top of the extracellular αhelical domain are sufficiently open to allow flexibility that allows access to AeaTMOF and subsequent anchoring of the molecule to the binding site located in the intracellular part of E. coli SbmA, at the fork of α-Hd (Figure 7(Ad-f)).

MdtM
The modeled membrane transporter of E. coli MdtM exhibits a canonical organization of membrane transporters and consists of 12 α-helices spanning the lipid bilayer (Figure 7(Bg)). The α-helices delineate a central channel that allows transport of peptides across the bacterial membrane (Figure 7(Bh)). Docking experiments show that AeaTMOF readily interacts with the transporter via hydrogen bonds and stacking interactions with aromatic residues surrounding the central channel of MdtM (Figure 7(Bj)). Anchoring of AeaTMOF to MdtM involves a network of hydrogen bonds to Y26, A118, T119, Y122, Y227, M230, and M231. Additional stacking interactions between the proline and tyrosine rings of AeaTMOF and aromatic residues Y26, Y122, Y227, F259, F326, and F330 of MdtM strengthen the anchoring of TMOF to MdtM (Figure 7(Bj)). The modeling results show that AeaTMOF has high affinity for both SbmA and MdtM indicating that they could be used to transport AeaTMOF into bacterial cells. The positions of MdtM and SbmA in the bilayer lipid inner membrane of E. coli and the binding of AeaTMOF to the transporters and oncocin112 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13) to SbmA indicate that both peptides bind the importers inside the membrane at positions that are dictated by charge, hydrophobic, and stacking interactions with the transporters (Figure 8(b,e,f)). Although AeaTMOF and oncocin112 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13) are bound in the same active site, the binding is different (Figure 8(Be,f)): oncocin112 (1-13) extends towards the top of the bacterial bilayer membrane because of a different 3D conformation, whereas TMOF exhibiting a left-handed helix is more compact binding in the middle of the active pocket at the end of the bacterial bilayer membrane, which is also observed in binding MdtM (Figure 8(Ab,Be,f)).

MdtM
The modeled membrane transporter of E. coli MdtM exhibits a canonical organization of membrane transporters and consists of 12 α-helices spanning the lipid bilayer ( Figure  7(Bg)). The α-helices delineate a central channel that allows transport of peptides across the bacterial membrane (Figure 7(Bh)). Docking experiments show that AeaTMOF readily interacts with the transporter via hydrogen bonds and stacking interactions with aromatic residues surrounding the central channel of MdtM (Figure 7(Bj)). Anchoring of AeaTMOF to MdtM involves a network of hydrogen bonds to Y26, A118, T119, Y122, Y227, M230, and M231. Additional stacking interactions between the proline and tyrosine rings of AeaTMOF and aromatic residues Y26, Y122, Y227, F259, F326, and F330 of MdtM strengthen the anchoring of TMOF to MdtM (Figure 7(Bj)). The modeling results show that AeaTMOF has high affinity for both SbmA and MdtM indicating that they could be used to transport AeaTMOF into bacterial cells.
Because antibiotic resistance is a problem in many hospitals in the USA and beyond, AeaTMOF could perhaps be used in the future to control bacterial-resistant strains [1,2]. We also tested the effect of AeaTMOF on two medically important Gram-negative bacteria A. baumannii and P. aeruginosa. The first causes outbreaks infection in hospital intensive care units (ICUs) and healthcare settings affecting very ill or disabled patients causing pneumonia symptoms, bloodstream infection, wound infection, and urinary tract infection. The second bacterial species causes infections in the blood, lungs (pneumonia), or other parts of the body after surgery. Both bacterial species are known to acquire antibiotic resistance and can become multi-drug resistant [47,48].
A. baumannii and P. aeruginosa growth was completely inhibited in the presence of AeaTMOF (5 mM), and lower concentrations (0.1, 0.2, and 1 mM) were not effective. At 15 h only one or two colonies were found, indicating that AeaTMOF is very effective in stopping the growth of these bacterial cells (Figure 2). One explanation why P. aeruginosa was not inhibited by oncocin112 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13) is because this bacterial species is known to lack the SbmA transporter that is required to import PrAMP including oncocin112 (1-13) [49], whereas AeaTMOF apparently can use other transporters and was not affected. S. aureus, a Grampositive bacterium, is a major human pathogen causing a wide range of clinical infections. It plays a major role in bacteremia and infective endocarditis, as well as skin and soft tissue and pleuropulmonary infections that are caused by virulent strains that exhibit β-lactam antibiotics resistance [50]. AeaTMOF incubation with these bacterial cells caused 99.9% inhibition of bacterial growth during a 15 h incubation period when AeaTMOF (5 mM) was used, lower concentrations (0.2 and 1 mM) were not effective ( Figure 3A). Although we did not test the bacterial cells for antibiotic resistance, most of the S. aureus strains are currently β-lactam resistant. On the other hand, onconcin112 (1-13) did not affect the bacterial cells' growth even at high concentration (10 mM) ( Figure 3B). Because Gram-positive bacteria generally lack an SbmA transporter, PrAMP peptides such as oncocin112 (1-13) do not enter the bacterial cells and do not affect their growth; on the other hand, AeaTMOF, a proline-rich decapeptide, can use other transporters to enter the bacterial cells and affect their growth [51]. Similar results were obtained when Gram-positive B. thuringiensis subsp. israelensis were grown in the presence of AeaTMOF and onconcin112 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13). Only AeaTMOF (5 mM) was able to completely inhibit bacterial cell growth, whereas onconcin112 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13) failed to inhibit bacterial cell growth even at high concentration (20 mM) ( Figure 4A,B), indicating that this bacterium lacks a SbmA transporter and thus preventing oncocin112 (1-13) from entering the bacterial cell, whereas AeaTMOF uses other transporter(s) to enter the bacterial cells [51]. Indeed, E. coli cells, lacking the SbmA transporter, that were incubated with AeaTMOF and oncocin112 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13) were inhibited only in the presence of AeaTMOF indicating that oncocin112 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13) uses SbmA exclusively to enter the E. coli cells, whereas AeaTMOF in the absence of SbmA can use other transporters such as MdtM, YjiL, or YgdD to enter the bacterial cells [20]. Our earlier reports [25,46] show that AeaTMOF is transported into the midgut epithelial cells of Ae. Aegypti by an ABC importer that uses ATP hydrolysis to transport AeaTMOF into the mosquito gut epithelial cells before it binds the mosquito ribosome, stopping trypsin biosynthesis [25,46]. To eliminate the possibility that AeaTMOF uses a similar transporter to enter E. coli cells, we incubated E. coli cells lacking the SbmA transporter in the presence of NaAzide (25 and 250 mM) to inhibit ATPase activity, thus inhibiting AeaTMOF-like ABC transporters in E. coli. Our results show that AeaTMOF (5 mM) significantly (p < 0.0001) inhibited E. coli cell growth in the presence of NaAzide (50 and 250 mM), whereas the growth of bacterial cells (control) that were incubated without AeaTMOF and in the presence of NaAzide were not inhibited ( Figure 6). These results indicate that AeaTMOF in the absence of SbmA uses different transporter(s) [20], although several genes are alleged to be alternative transporters for PrAMP, including the yjiL-mdtM gene cluster, which aside from its role as an efflux pump can also be used to import PrAMP into bacterial cells [20]. YjiL is a putative ATPase activator of (R)-hydroxyglutaryl-CoA dehydratase (Gene ID 948837) (www.ncbi.nlm.nih.gov/gene/ (accessed on 10 November 2022)), and its function, although unknown, could possibly facilitate ABC transporters [51]. On the other hand, SbmA and MdtM can each form a channel through the bacterial cytoplasmic membrane to transport proline-rich peptides such as AeaTMOF. SbmA 3D modeling shows that it has high similarity to the ABC transporters except for nucleotide binding sites (Figure 7(Aa)) and uses electrochemical gradient and not ATP hydrolysis to import PrAMP into the bacterial cells [5]. Our docking results show that AeaTMOF binds SbmA at a specific binding cavity using electrostatic interactions, hydrophilic and hydrophobic interactions, stacking interactions, and a few hydrogen bonds (Figure 7(Ad-f)). The α-helices and the loops at the top of the extracellular part of the α-Hd that extend outside the inner membrane are sufficiently loose and flexible and allow AeaTMOF to enter and anchor at the binding site in the fork of the a-HD (Figure 7(Ad,e)), indicating that AeaTMOF uses SbmA to enter E. coli cells. Similarly, a 3D modeling of MdtM and docking of AeaTMOF show that the decapeptide binds the central channel of MdtM using network of hydrogen bonds and stacking interactions between Pro and Tyr rings of AeaTMOF and the aromatic residues of MdtM ( Figure 7B). AeaTMOF exhibited low hemolytic activity when different concentrations of AeaTMOF (0-6.5 mM) were tested by Professor Robert Hodges (university of Colorado Anschutz School of Medicine, department of Biochemistry and Molecular Genetics) using females' (ages 20 to 40 years old) blood samples that were obtained at the University of Colorado Anschutz Medical School Hospital (results not shown). Therefore, in future work we will clone and express SbmA and MdtM and determine the AeaTMOF K D for both transporters as was performed for the AeaTMOF ABC receptor [45]. To make AeaTMOF more amenable for animal and clinical work, it will be necessary in the future to alter the peptide as was achieved for oncocin112 [21], reducing the effective bactericidal concentrations from millimolar to micromolar range. For future clinical work, the peptide should be encapsulated to allow a slow release into the circulation of test animals.

Conclusions
AeaTMOF is a mosquito decapeptide hormone that is synthesized by female mosquito ovaries after the blood meal and secreted into the hemolymph binds a AeaTMOF ABC gut receptor and is imported into the gut epithelial cells regulating trypsin biosynthesis by binding the gut ribosomes stopping the translation of the trypsin transcript after the blood meal. Because of this unique property of AeaTMOF, we explored the possibility that bacterial transporters SbmA and MdtM can also import AeaTMOF into the bacterial cytosol, allowing the peptide to stop protein translation by the bacterial ribosome and cause bacterial death. We first used 3D molecular modeling to show that SbmA and MdtM importers bind and could transport AeaTMOF by comparing it with oncocin112 (1-13), a proline-rich peptide that is transported by SbmA into the bacterial cytosol. Oncocin112 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13) was shown by X-ray crystallography to bind the bacterial ribosome at the protein exit tunnel blocking mRNA translation. Indeed, incubation of both Gram-negative and -positive bacterial cells in the presence of AeaTMOF caused bacterial death. Because AeaTMOF exhibits low hemolytic activity, it could be used to control bacterial infections in future.