*2.5. A. aegypti Piwi4 Expresses a Nuclear Localization Signal in the N-Terminal Region of the Protein*

To further explore AePiwi4 nuclear localization, we identified a putative NLS in the N-terminal region of the protein (Supplemental Figure S8A). In *Drosophila melanogaster* Piwi, the NLS is expressed in the intrinsically disordered domain in a similar region of the N-terminal. We therefore generated a phylogenetic tree of the intrinsically disordered regions of the *A. aegypti* and *D. melanogaster* Piwi proteins to see if Piwis with known or putative NLS signals would cluster together (Supplemental Figure S8B). We found that the intrinsically disordered domains of *A. aegypti* Ago3 and *D. melanogaster* Ago3 clustered together, that *Drosophila* Aub and *A. aegypti* Piwi7 clustered together, and that *A. aegypti* Piwis2-6 clustered with *Drosophila* Piwi, the only *Drosophila* Piwi with an NLS. These results suggested that *A. aegypti* Piwi2-6 may also harbor nuclear localization signals in their intrinsically disordered domains in the N-terminal regions.

To confirm that the putative AePiwi4 NLS was responsible for protein nuclear localization, we cloned the putative AePiwi4 NLS (amino acid residues 42–83, Supplemental Figure S8A), as well as the entire N-terminal region containing the NLS (amino acid residues 1–83), fused to an eGFP; we henceforth named these constructs AePiwi4NLS-eGFP and AePiwi4Nterminal-eGFP, respectively. We used the same backbone containing either a known SV40 NLS fused to the eGFP [52] as a positive control or an eGFP alone as a negative control; we henceforth named these constructs SV40NLS-eGFP and eGFP, respectively. We then transfected HEK293 cells with these constructs and visualized eGFP and DAPI colocalization 24 h post-transfection. As expected, we found that the known SV40NLSeGFP localized in the nucleus while the eGFP alone appeared diffused throughout the cells (Figure 6A,B). Plasmids harboring either AePiwi4NLS-eGFP or AePiwi4Nterminal-eGFP migrated into the nucleus, as evidenced by eGFP expression colocalized with DAPI staining (Figure 6C,D). The nuclear staining appeared punctated, perhaps indicative of nucleolar staining. We also observed that both AePiwi4NLS-eGFP and AePiwi4Nterminal-eGFP displayed cytoplasmic eGFP expression as well, which was not observed in cells transfected with the SV40 NLS plasmid (Figure 6C,D).

To quantitively compare eGFP fluorescent intensities across sample types, we subtracted total eGFP fluorescent intensity sums from eGFP fluorescent intensity sums in nuclear surfaces, normalized by number of cells, in three independent views across slides (Figure 6E). We found that the resulting eGFP nuclear intensity sums outside of the nuclear surfaces was significantly higher for cells transfected with the eGFP construct as compared to cells transfected with SV40NLS-eGFP (*p* = 0.006), AePiwi4NLS-EGFP (*p* = 0.01), or AePiwi4Nterminal-eGFP (*p* = 0.05). There were no significant differences in eGFP fluorescent intensity sums outside of nuclear surfaces between cells transfected with AePiwi4NLS-EGFP (*p* = 0.34) or AePiwi4Nterminal-eGFP (*p* = 0.23) compared to those transfected with the SV40NLS-eGFP positive control. Taken together, these results suggested that *A. aegypti* Piwi4 expresses an NLS in the intrinsically disordered domain in the N-terminal region of the protein.

**Figure 6.** *Cont*.

**Figure 6.** AePiwi4 harbors an NLS in the N-terminal region of the protein. Representative slices from image stacks of HEK293 cells transfected with (**A**) SV40NLS-eGFP, (**B**) eGFP, (**C**) AePiwi4NLSeGFP, or (**D**) AePiwi4Nterminal-eGFP. Cells were stained with DAPI (blue) 24 h post-transfection. DAPI (left), eGFP (middle), and merged (right) channels are shown separately. Orthogonal views presented in merged channel. Scale bar = 15 μM. (**E**) Quantification of total eGFP fluorescence intensity sums subtracted from eGFP intensity sums in nuclear surfaces for each sample type, normalized by number of cells. Each black dot represents an individual picture. Red bars indicate SEM. \*\* = *p* ≤ 0.01, \* = *p* ≤ 0.05, ns = non-significant. **3***.* **Discussion Figure 6.** AePiwi4 harbors an NLS in the N-terminal region of the protein. Representative slices from image stacks of HEK293 cells transfected with (**A**) SV40NLS-eGFP, (**B**) eGFP, (**C**) AePiwi4NLS-eGFP, or (**D**) AePiwi4Nterminal-eGFP. Cells were stained with DAPI (blue) 24 h post-transfection. DAPI (left), eGFP (middle), and merged (right) channels are shown separately. Orthogonal views presented in merged channel. Scale bar = 15 µM. (**E**) Quantification of total eGFP fluorescence intensity sums subtracted from eGFP intensity sums in nuclear surfaces for each sample type, normalized by number of cells. Each black dot represents an individual picture. Red bars indicate SEM. \*\* = *p* ≤ 0.01, \* = *p* ≤ 0.05, ns = non-significant.

#### In this study, we characterized *A. aegypti* Piwi4 structural features involved in RNA **3. Discussion**

binding and nuclear localization to gain insights into the protein's function. AePiwi4 had previously been associated with various 28–30 nt 3′ 2′ O-methylated piRNAs, so we focused our efforts on the PAZ domain that binds the 3′ ends of piRNAs. We assessed AePiwi4 PAZ RNA-binding dynamics by SPR and found that AePiwi4 PAZ bound to both mature and unmethylated piRNAs with micromolar affinities in a sequence independent manner. We identified key residues in AePiwi4 PAZ involved in RNA binding and found that they were highly conserved across organisms. We also highlighted a unique arginine amino acid flanking a tyrosine residue necessary for 3′nm RNA binding that was present in most other organisms' Piwi PAZ but was only present in a single *A. aegypti* Piwi PAZ (Ago3). Mutating this residue in AePiwi4 PAZ to match that of Ago3 improved both 3′m and 3′nm RNA binding. Through circular dichroism, we showed that single amino acid changes in Piwi PAZ changes the secondary structure of the protein. Finally, we found that AePiwi4 was both cytoplasmic and nuclear in mosquito tissues, and that signals in the intrinsically disordered region drove nuclear localization. We report herein Piwi-RNA-binding affinities for a Piwi protein of an arthropod vec-In this study, we characterized *A. aegypti* Piwi4 structural features involved in RNA binding and nuclear localization to gain insights into the protein's function. AePiwi4 had previously been associated with various 28–30 nt 30 2 0 O-methylated piRNAs, so we focused our efforts on the PAZ domain that binds the 30 ends of piRNAs. We assessed AePiwi4 PAZ RNA-binding dynamics by SPR and found that AePiwi4 PAZ bound to both mature and unmethylated piRNAs with micromolar affinities in a sequence independent manner. We identified key residues in AePiwi4 PAZ involved in RNA binding and found that they were highly conserved across organisms. We also highlighted a unique arginine amino acid flanking a tyrosine residue necessary for 30nm RNA binding that was present in most other organisms' Piwi PAZ but was only present in a single *A. aegypti* Piwi PAZ (Ago3). Mutating this residue in AePiwi4 PAZ to match that of Ago3 improved both 30m and 30nm RNA binding. Through circular dichroism, we showed that single amino acid changes in Piwi PAZ changes the secondary structure of the protein. Finally, we found that AePiwi4 was both cytoplasmic and nuclear in mosquito tissues, and that signals in the intrinsically disordered region drove nuclear localization.

tor, which complements studies performed with human and *Drosophila melanogaster* Piwi PAZ. We found that *A. aegypti* Piwi4 PAZ bound 3′ 2′ O-methylated and non-methylated piRNAs with KDs of 1.7 ± 0.8 μM and 5.0 ± 2.2 μM, respectively. The preference of AePiwi4 PAZ for 3′m piRNAs over 3′nm piRNAs was less pronounced than what has been reported for other Piwi PAZ (Supplemental Table S2). For example, Hiwi1, Hiwi2, and Hili bound 3′m piRNAs with KDs of 6.5 μM, 2 μM, or 10 μM, respectively, but they bound nonmethylated piRNAs with weaker affinities—KDs of 16 μM, 12 μM, or 34 μM, respectively [44]. Immunoprecipitations of AePiwi4 from uninfected or infected Aag2 cells followed by sRNA sequencing of associated RNAs have revealed the protein associates with bona fide piRNAs resistant to beta-elimination, a method that selects for 3′m piRNAs and depletes 3′nm miRNAs [32,39]. Our results suggest, however, that AePiwi4 is able to bind to both 3′m and 3′nm sRNAs with only a marginally higher affinity for the former over the We report herein Piwi-RNA-binding affinities for a Piwi protein of an arthropod vector, which complements studies performed with human and *Drosophila melanogaster* Piwi PAZ. We found that *A. aegypti* Piwi4 PAZ bound 30 2 0 O-methylated and non-methylated piRNAs with KDs of 1.7 ± 0.8 µM and 5.0 ± 2.2 µM, respectively. The preference of AePiwi4 PAZ for 30m piRNAs over 30nm piRNAs was less pronounced than what has been reported for other Piwi PAZ (Supplemental Table S2). For example, Hiwi1, Hiwi2, and Hili bound 30m piRNAs with KDs of 6.5 µM, 2 µM, or 10 µM, respectively, but they bound non-methylated piRNAs with weaker affinities—KDs of 16 µM, 12 µM, or 34 µM, respectively [44]. Immunoprecipitations of AePiwi4 from uninfected or infected Aag2 cells followed by sRNA sequencing of associated RNAs have revealed the protein associates with bona fide piRNAs resistant to beta-elimination, a method that selects for 30m piRNAs and depletes 30nm miRNAs [32,39]. Our results suggest, however, that AePiwi4 is able to bind to both 30m and 30nm sRNAs with only a marginally higher affinity for the former

over the latter. Further investigations on AePiwi4-associated sRNAs from different cellular compartments may provide new insights on protein-RNA trafficking and the range of sRNAs with which AePiwi4 interacts. For example, the role(s) of *A. aegypti* Piwis may function with both pre-processed non-methylated RNAs and mature piRNAs across cellular compartments or with sRNA populations outside the piRNA pathway. Halbach et al. [39] compared AePiwi4-mediated silencing of a satellite repeat-derived target by way of a piRNA to that of miRNA silencing [39]. In that study, the authors found that the 30 end of a satellite repeat-derived piRNA (tapR1) was not absolutely required for silencing, while the seed region was not sufficient for silencing, a pattern they compared to miRNA-mediated silencing [39]. In another study, Tassetto and colleagues found that silencing *AePiwi4* impacted both 30m piRNA and siRNA production and argued that AePiwi4 links the siRNA and piRNA pathways [32]. Our AePiwi4 RNA-binding studies indicate that the PAZ domain of AePiwi4 is indeed able to interact with diverse populations of sRNAs with similar affinities, perhaps suggesting AePiwi4 has broad functions or unique roles in RNA binding that may differ from model Piwis. Future studies comparing RNA-binding dynamics across the *A. aegypti* Piwis will elucidate the roles they play in RNAi.

In this study, we studied 30m and 30nm RNA binding with protein partners by SPR. Other studies have characterized PAZ RNA binding by isothermal calorimetry (ITC) using small eight nucleotide RNAs, and we note that caution should be taken when comparing hard dissociation constant values across these different techniques. While ITC provides valuable information on number of binding sites and heat released from a binding reaction, we found that immobilizing the 50 end of longer, more physiologically relevant RNAs by SPR enabled us to efficiently calculate dissociation constants for many protein-binding partners against stabilized ligands in a single experiment. This method may be useful for other studies aimed at understanding protein-RNA binding at specific motifs.

Our data suggest that small differences in Piwi PAZ amino acid composition across Piwi proteins alter protein secondary structure, which thereby impact the protein's affinity for certain RNAs. Given that preferential binding of 30m RNA over 30nm RNA was mostly dictated by backbone confirmation of the protein in Hiwi1, it is likely that subtle differences in Piwi protein structure may have profound impacts on preferential RNAbinding behaviors that are important for defining the functions of different Piwis. We found that although the amino acids that directly form hydrogen bonds with RNA were highly conserved, residues that flank these sites tended to be more variable. Perhaps those residues that impact the stability of the PAZ structure, rather than those that directly bind the RNA itself, drive Piwi functional divergence.

The number of Piwi proteins has expanded in culicine mosquitoes as compared to anophelines and drosophilids [53], and understanding their evolutionary relationships with other Piwis, their shared or unique structural features, and their interactions with diverse RNA populations may provide insights into how their functions have diverged. We found that the PAZ domains of *A. aegypti* Piwis 2-7 are more evolutionarily related to that of *D. melanogaster* Aubergine than to *D. melanogaster* Piwi or Ago3 PAZ. Aubergine is a cytoplasmic, germline-specific protein that participates in ping-pong amplification by binding antisense primary piRNAs and producing secondary sense piRNAs that fuel the cycle [54], a role similar to that of *A. aegypti* Piwi5 [25]. A recent study showed that piRNA binding to Aub PAZ induces a protein confirmational change that triggers symmetric dimethylarginine (sDMA) methylation of the Aub intrinsically disordered domain in the N-terminal region [55]. The sDMA modification then serves as a binding site for Krimper, which simultaneously binds unmethylated Ago3 to bring the proteins in close proximity for RNA transfer during ping-pong amplification [55]. Indeed, Joosten and colleagues recently characterized a Krimper ortholog in *Aedes*, Atari, which bound Ago3 without sDMA modifications [26]. Perhaps a similar mode of RNA binding-dependent autoregulation and sDMA signaling also govern the *Aedes aegypti* Piwis Piwi2-6.

*Drosophila* Piwi, the only nuclear Piwi in the fly, expresses a bipartite NLS in its intrinsically disordered domain [56]. *Drosophila* Piwi nuclear localization is autoregulated

by confirmational changes that occur once the protein binds piRNAs; the NLS remains buried within the protein structure until RNA binding triggers a conformational change and exposes the NLS [56]. Once the protein is imported into the nucleus and releases the piRNA, the protein is trafficked back into the cytoplasm. We found that *A. aegypti* Piwi4 also expresses signals in the intrinsically disordered region that drive proteins to the nucleus (Figure 6C,D), which, if similar to *Drosophila* Piwi, could autoregulate protein trafficking based on protein confirmational changes that dictate signal exposure. We observed that the AePiwi4 NLS did not drive complete expression of eGFP into the nucleus, as evidenced by diffused cytoplasmic fluorescence in addition to punctated nuclear staining (Figure 6C,D). It is possible that the intrinsically disordered region of *AePiwi4* contains both an NLS and nuclear export signals (NES) that drive protein trafficking in and out of the nucleus. Investigations on how AePiwi4 regulates its trafficking into different cellular compartments require future studies and could be useful in understanding its role in different RNAimediated processes. It is also possible that several *A. aegypti* Piwis autoregulate their subcellular localization in similar manners as AePiwi4. Our phylogenetic analyses revealed that the regions of the *AePiwi4* and *Drosophila melanogaster* Piwi proteins that harbored nuclear localization signals, the intrinsically disordered domains, also clustered with *A. aegypti* Piwi2, Piwi3, Piwi5, and Piwi6 (Supplemental Figure S8B). Different Piwis likely have sophisticated and diverse regulation mechanisms that control their expression patterns in different compartments of the cell.

Growing evidence reveals that the piRNA pathway is involved in gene regulation in somatic tissues and contributes to diverse human diseases including cancer [23,57] and neurodegenerative disorders [58]. Because somatic piRNAs and Piwi expression are common in arthropods [24], they could be valuable models for understanding the molecular mechanisms underlying the lesser understood Piwi or piRNA functions. Future studies aimed at understanding how Piwi-RNA binding impacts protein structure and function will be useful for learning more about how this pathway is involved in immunity, gene regulation, and disease in arthropod vectors as well as in other organisms.

#### **4. Materials and Methods**

#### *4.1. A. aegypti Piwi4 Structure Model Prediction*

A model of the *A. aegypti* Piwi4 structure was generated using the I-TASSER software (version 5.1 Zhang Lab, University of Michigan, Ann Arbor, MI, USA [46,59,60]) and visualized using Chimera (University of California, San Francisco, CA, USA). This software predicts secondary and tertiary structures based on the similarity of other proteins whose structures have been solved. The AePiwi4 amino acid sequence was queried against the *Drosophila* Piwi structure that was crystalized in Yamaguchi et al., 2020 [45], which allowed us to determine the predicted *A. aegypti* Piwi4 PAZ domain. AePiwi4 PAZ was then superimposed to other crystalized PAZ proteins, including Hili PAZ [44], Hiwi1 PAZ [44], Miwi PAZ [47], and Siwi PAZ [48].

#### *4.2. Cloning*

*Aedes aegypti Piwi4* (*AAEL007698*), including a 6xHis-tag, was synthesized by BioBasic Inc. (Markham, ON, Canada). Both *AePiwi4* full length (FL) and PAZ domain (residues 270–380) nucleotide sequences were sub-cloned into pCR-Blunt II-TOPO vector using the Zero Blunt TOPO PCR Cloning Kit (Thermo Fisher Scientific, Waltham, MA, USA) following the manufacturer's instructions. pCR-Blunt II-TOPO vectors containing either *AePiwi4* FL or *AePiwi4* PAZ were transformed in OneShot Top10 chemically competent *E. coli* (Invitrogen, Waltham, MA, USA). Using standard restriction enzyme-mediated cloning and the pCR-Blunt II-TOPO vectors described above as PCR templates, *AePiwi4* FL or *AePiwi4* PAZ were then cloned into pET-17b vectors. pET-17b vectors containing either *AePiwi4* FL or *AePiwi4* PAZ were transformed in OneShot Top10 chemically competent *E. coli* (Invitrogen, Waltham, MA, USA). Inserts were confirmed by Sanger sequencing. Primers used in this study are displayed in Supplemental Table S3.

The putative AePiwi4 NLS, as well as the entire N-terminal region of AePiwi4 containing the putative NLS, were cloned from the *AePiwi4* FL-containing pCR-Blunt II-TOPO vector into a backbone containing an eGFP by In-Fusion cloning (TakaRa, San Jose, CA, USA) following the manufacturer's instructions. The parent SV40NLS-eGFP backbone was a gift from Rob Parton (Addgene plasmid # 67652; http://n2t.net/addgene:67652 (accessed on 28 April 2021); RRID: Addgene\_67652) [52]. Briefly, an eGFP-alone plasmid was generated by NcoI digestion and religation with the T4 DNA ligation Mighty Mix (TakaRa, San Jose, CA, USA) of the SV40NLS-eGFP plasmid. AePiwi4NLS-eGFP or AePiwi4Nterminal-eGFP were then cloned into the eGFP alone plasmid using In-Fusion primers listed in Supplemental Table S3. All constructs were transformed into Stellar Competent Cells (TakaRa, San Jose, CA, USA), and inserts were confirmed by Sanger sequencing.

#### *4.3. Recombinant Protein Expression*

pET-17b vectors containing either *AePiwi4* FL or PAZ were transformed into BL21(DE3) pLysS chemically competent *E. coli* cells (Thermo Fisher Scientific, Waltham, MA, USA). Transformed *E. coli* were plated onto Luria–Bertani (LB) agar plates with 100 µg/mL ampicillin and 34 µg/mL chloramphenicol that were left O/N at 37 ◦C. Individual colonies were picked into starter cultures of 4 mL LB broth (supplemented with 100 µg/mL ampicillin and 34 µg/mL chloramphenicol) that were left shaking at 220 RPM O/N at 37 ◦C. Starter cultures were then added to 150 mL LB broth (supplemented with 100 µg/mL ampicillin and 34 µg/mL chloramphenicol) that were left shaking at 220 RPM 37 ◦C until OD<sup>600</sup> = 0.4 (~1 h). Protein expression was then induced with 0.1 mM isopropyl β- d-1-thiogalactopyranoside (IPTG) for 4 h shaking at 160 RPM at 25 ◦C. Bacteria was then pelleted and stored at −30 ◦C until protein purification.

For larger scale expression, 150 mL LB broth (supplemented with 100 µg/mL ampicillin and 34 µg/mL chloramphenicol) starter cultures that had been inoculated with glycerol scrapings of BL21(DE3) pLysS *E. coli* containing either pET-17b-*AePiwi4* FL or pET-17b-*AePiwi4* PAZ were left shaking at 220 RPM O/N at 37 ◦C. Starter cultures were then added to 1 L LB broth (supplemented with 100 µg/mL ampicillin and 34 µg/mL chloramphenicol) and expression was induced following the above protocol.

Expression was confirmed by SDS-PAGE separation and anti-6xHis-tag Western blot in both the soluble and inclusion body fractions for both proteins.

#### *4.4. Recombinant Protein Purification*

The soluble AePiwi4 PAZ protein was purified by affinity chromatography followed by size-exclusion chromatography using Nickel-charged HiTrap Chelating HP (GE Healthcare, Chicago, IL, USA) and Superdex 200 10/300 GL columns (GE Healthcare, Chicago, IL, USA), respectively. Frozen *E. coli* pellets were resuspended with Buffer A (10 mM Tris, 500 mM NaCl, 5 mM imidazole, pH 8), left on ice for 10–15 min, and pulse sonicated 4× for 30 s—2 min. The lysates were then spun at 15,000× *g* for 30 min at 4 ◦C. The resulting supernatants were filtered with a 0.8 µM filter (MilliporeSigma, Burlington, MA, USA) and loaded onto a pre-equilibrated Nickel-charged HiTrap Chelating HP column using a peristaltic pump. The column was then pre-washed with 3 column volumes (CV) of Buffer A, followed by a 3 CV wash with Wash Buffer 1 (10 mM Tris, 500 mM NaCl, 20 mM imidazole, pH 8) and 3 CV wash with Wash Buffer 2 (10 mM Tris, 500 mM NaCl, 100 mM imidazole, pH 8). The protein was eluted from the column using 3 CV of Elution Buffer (10 mM Tris, 500 mM NaCl, 300 mM imidazole, and pH 8) and visualized by SDS-PAGE gel electrophoresis.

Eluted protein was concentrated down to ~500 µL using an Amicon stirred cell with a cellulose membrane of 3 kDa nominal molecular weight (MilliporeSigma, Burlington, MA, USA). The resulting concentrated protein was spun down at 4000× *g* for 10 min to remove large debris and loaded onto a Superdex 200 10/300 GL column that had been equilibrated with 20 mM Tris-HCl, 150 mM NaCl, pH 7.4. Peak elutions that corresponded to the correct

size of Piwi4 PAZ (14 kDa) were confirmed by SDS-PAGE gel electrophoresis. N-terminal protein sequence was also confirmed by Edman degradation.

## *4.5. SDS-PAGE*

All proteins were heated to 95 ◦C for 5 min under reducing conditions in 1X LDS (Thermo Fisher Scientific, Waltham, MA, USA) and were separated using 4–12% Bis-Tris protein gels (Thermo Fisher Scientific, Waltham, MA, USA). Gels were stained with Coomassie Brilliant Blue (GenScript, Piscataway, NJ, USA). Protein concentrations were determined using the Nanodrop ND-1000 spectrophotometer adjusted by the molar extinction coefficient.

#### *4.6. Western Blot*

*Aedes aegypti* mosquito midguts and ovaries, as well as recombinant proteins, were separated by SDS-PAGE gel electrophoresis for Western blots. *A. aegypti* that had been fed defibrinated sheep blood (Denver Serum Company, Denver, CO, USA) were collected 48 h post-bloodmeal, and their midguts (cleaned of blood in 1× PBS) and ovaries were dissected and flash frozen on dry ice. 15 midguts or ovaries/tube were resuspended in 100 µL cold hypotonic lysis buffer (10 mM Hepes pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.2 mM PMSF) and left on ice for 15 min. Samples were vortexed vigorously for 30 s and then pelleted at 1000× *g* for 15 min. The supernatant was collected as the cytoplasmic fraction. The remaining pellets were then resuspended in 100 µL solubilization buffer (15 mM Tris, 150 mM NaCl, 5 mM EDTA, 0.5% Triton X-100, 10% glycerol, 0.2 mM PMSF) and spun down at 100,000× *g*. The supernatant was collected as the nuclear fraction.

30 µg of protein was processed for SDS/PAGE separation, as described previously, and run alongside 10 µM Piwi4 PAZ or Piwi4 FL inclusion bodies. Proteins were transferred to a PVDF membrane (iBlot, Invitrogen, Waltham, MA, USA) that was blocked for 2 h at RT in blocking buffer (5% powdered milk (Carnation), 50 mM Tris-HCl pH 7.4, 150 mM NaCl, 1% Tween 20 (TBST)). Membranes were incubated O/N at 4 ◦C with anti-Piwi4 PAZ mouse serum (1:500 in blocking buffer), a 6xHis-tag monoclonal antibody (ThermoFisher Scientific, Waltham, MA, USA, diluted 1:5000 in blocking buffer), or anti-Histone H3 as a nuclear marker (Novus Biologicals, Littleton, CO, USA; generated in rabbit, diluted 1:1000 in blocking buffer).

Membranes were washed with TBST (3× for 10 min) and with TBS (1× 10 min) and incubated at RT for 1–2 h with goat anti-mouse or anti-rabbit antibodies conjugated to alkaline phosphatase (1 mg/mL, diluted 1:10,000). Membranes were again washed with TBST and TBS, and proteins were detected for 5–10 min using Western Blue Stabilized alkaline phosphatase substrate (Promega, Madison, WI, USA).

#### *4.7. Mosquito Rearing*

*A. aegypti* mosquitoes (Liverpool (LVP) strain) were reared in standard insectary conditions at the Laboratory of Malaria and Vector Research, NIAID, NIH (28 ◦C, 60–70% humidity, 14:10 h light/dark cycle) under the expert supervision of Karina Sewell, Andre Laughinghouse, Kevin Lee, and Yonas Gebremicale. Mosquitoes had a solution of 10% sucrose ad libitum and were offered defibrinated sheep blood (Denver Serum Company, Denver, CO, USA) in an artificial feeding system. Larva were fed Tetramin.

#### *4.8. Sequence Alignment*

Nucleotide and amino acid sequences were retrieved from the NCBI databases. Multiple alignments and phylogenetic trees were obtained by Clustal Omega [61] and visualized on Jalview [62].

## *4.9. Site Directed Mutagenesis*

AePiwi4 PAZ protein mutants were generated using the QuikChange II Site-Directed mutagenesis kit (Agilent, Santa, Clara, CA, USA) following the manufacturer's instructions.

Primers were designed using PrimerX (https://www.bioinformatics.org/primerx/cgi-bin/ DNA\_1.cgi, (accessed on 14 March 2021)) and are displayed in Supplemental Table S3. The pET-17b vector containing *AePiwi4* FL was used as template for the reactions, and XL1-Blue supercompetent cells (Agilent, Santa, Clara, CA, USA) were transformed with the mutant plasmids. Mutation nucleotide sequences were confirmed by Sanger sequencing, and protein mutant sequences were confirmed by mass spectrometry.

The pET-17b vectors containing the AePiwi4 PAZ mutations were transformed into BL21(DE3) pLysS chemically competent *E. coli* cells (Thermo Fisher Scientific, Waltham, MA, USA), and all proteins were expressed and purified, as described previously.

#### *4.10. Surface Plasmon Resonance (SPR)*

All SPR experiments were carried out in a T100 instrument (GE Healthcare, Chicago, IL, USA) following the manufacturer's instructions. Sensor CM5, amine coupling reagents, and HBS-P buffers were also purchased from GE Healthcare (Chicago, IL, USA). HBS-P was supplemented with EDTA (HBS-PE, 10 mM Hepes pH 7.4, 150 mM NaCl, 3 mM EDTA, and 0.005% (*v*/*v*) P20 surfactant) and was used as the running buffer while Conditioning Solution 2 (50 mM NaOH, 1 M NaCl) was used as the regeneration and conditioning solution for all experiments. Briefly, the CM5 sensor was coated 40 µg/mL neutravidin and pre-conditioned with 3 × 60 s injections of Conditioning Solution 2. ~500–1000 RUs of biotinylated RNAs were then captured to flow cells 2 or 4, which were then conditioned with 3 × 60 s injections of Conditioning Solution 2. Protein analyte was introduced unto the surface with 180 s injections (30 µL/s). Results were analyzed using the Biacore T200 Evaluation software v2.0.3 provided by GE Healthcare (Chicago, IL, USA). Equilibrium dissociation constants were calculated from steady-state binding levels (Req) against molar concentration of the analyte (C). The fitted equation was Req = ((CRmax)/(K<sup>D</sup> + C)) + offset, where Rmax = analyte-binding capacity of the surface in response units (RU) and offset = response at zero analyte concentration, which accounts for buffer-mediated effects on the refractive index. SPR experiments were carried out 2–4×.

## *4.11. Circular Dichroism*

0.1 mg/mL of purified AePiwi4 PAZ WT, T39A, Y40A, T41A, or T41R in 20 mM Tris 75 mM, NaCl pH 7.4 were used for CD analyses. Continuous measurements with a pitch of 0.2 nm were recorded from 200–260 nm wavelengths with a bandwidth of 1 nm. Mean residue ellipticity was calculated with the following equation: (molecular weight of each protein in daltons/((number of amino acids − 1) × θλ))/(10 × pathlength in cm × protein concentration in g/mL). All readings were normalized by subtracting with blank (buffer) mean residue ellipticity. Data were analyzed using CAPITO [51].

#### *4.12. RNA Synthesis*

The 30 2 <sup>0</sup> O-CH<sup>3</sup> and 3<sup>0</sup> 2 0 OH 28 nt RNAs were synthesized by Eurofins Genomics (Louisville, KY, USA). Sequences are listed in Supplemental Table S4. RNA was resuspended at 1–2 mM in DEPC-treated water and stored at −80 ◦C.

#### *4.13. Mouse Polyclonal Antibody Production*

Polyclonal antibodies against *A. aegypti* Piwi4 PAZ were raised in mice. Mice (Balb/c; Charles River, Frederick, MD, USA) were IM immunized with 10 µg of AePiwi4 PAZ in combination with Magic Mouse Adjuvant (CD Creative Diagnostics, Shirley, NY, USA). Negative control mice were immunized with Magic Mouse Adjuvant alone. At 21 d post-immunization, mice received a 2nd booster immunization with 10 µg of AePiwi4 PAZ in combination with Magic Mouse Adjuvant (or adjuvant alone for negative control group). Blood was collected 35 d post-immunization. The antibody levels were confirmed by ELISA.

#### *4.14. Mass Spectrometry*

Mosquito tissue samples and recombinant proteins were prepared and separated by an SDS-PAGE gel as previously described, which was then stained with Coomassie blue. Bands of interest were excised from the gel and submitted for liquid chromatography coupled with mass spectrometry at the Research and Technology Branch (NIAID, NIH, Rockville, MD, USA). Briefly, the gel slices from the SDS-PAGE gel were cut into small pieces and subjected to in-gel trypsin digestion. The gel slices were destained to remove Coomassie blue staining and were then reduced and alkylated. After dehydration with acetonitrile and air-drying, a sequencing grade trypsin (Promega, Madison, WI, USA) solution was added onto the gel slices and was allowed to be absorbed into the gel slice. The gel slices were then incubated overnight at 30 ◦C for in-gel digestion. The peptides released from in-gel digestion were extracted by acetonitrile and then applied for LC-MS/MS analysis. Proteomic analyses were performed, as previously described [63].

#### *4.15. IFA*

Mosquito midguts or ovaries were dissected 48 h post-bloodmeal, flash fixed for 30 s in cold 4% paraformaldehyde (PFA) in PBS and cleaned of blood in cold PBS. The midguts or ovaries were then left shaking in 4% PFA in PBS O/N at 4 ◦C. The next day, midguts or ovaries were washed 3X in PBS and blocked O/N in blocking buffer (2% BSA, 0.5% Triton-X-100, PBS). The midguts or ovaries were then incubated with either serum from mice immunized with AePiwi4 PAZ or with Magic Mouse adjuvant alone (1:500, diluted in blocking buffer) O/N at 4 ◦C. The midguts or ovaries were washed with blocking buffer a minimum of 3× for 30 min and were then incubated with secondary goat anti-mouse antibodies conjugated to Alexa Fluor 594 (Thermo Fisher Scientific, Waltham, MA, USA; diluted 1:1000 in blocking buffer) for 1 h at RT. The midguts or ovaries were again washed with blocking buffer a minimum of 3× for 30 min, followed by incubation with 1 µg/mL DAPI (diluted 1:1000 in blocking buffer) and phalloidin conjugated to Alexa Fluor 488 (Thermo Fisher Scientific, Waltham, MA, USA; diluted 1:250 in blocking buffer) for 20 min at RT. The midguts or ovaries were washed 2× for 20 min with blocking buffer and 1× with 0.5% Triton-X-100 in PBS and were then mounted onto slides with ProLong Gold antifade mountant with DAPI (Thermo Fisher Scientific, Waltham, MA, USA).

#### *4.16. HEK293 Cell Culture and Transfection*

HEK293 cells were cultured in 35 mm dishes with a No. 15 coverslip pre-coated with Poly-D-Lysine (MatTek Life Sciences, Ashland, MA, USA). Briefly 300,000 cells were plated on individual dishes in DMEM media. The next day, the cells were transfected with transfection complex containing 500 ng of (1) SV40NLS-eGFP, (2) eGFP alone, (3) AePiwi4NLSeGFP, or (4) AePiwi4Nterminal-eGFP in 0.5 µL Lipofectamine 3000 (Invitrogen, Waltham, MA, USA) in serum free Opti-MEM media, according to the manufacturer's protocol. Twenty-four hours post-transfection, the cells were washed 3X with PBS and fixed with 4% PFA in PBS for 30 min at RT. The cells were then washed 3X with PBS and permeabilized with 0.5% Triton-X-100 in PBS for 30 min at RT. The cells were then stained with DAPI (1 µg/mL in 2% BSA, 0.5% Triton-X-100, PBS). The cells were visualized using a Leica Confocal SP8 microscope. Images were processed with Imaris software version 9.2.1 and post-processing was carried out in Fiji ImageJ version 1.52n for representative purposes.

#### *4.17. Statistics*

Surface Plasmon Resonance equilibrium curves were fitted with a non-linear regression generated by the Biacore Evaluation software v2.0.3 provided by GE Healthcare (described in "Surface Plasmon Resonance" Methods section), which were then visualized with GraphPad Prism. The equilibrium dissociation constants, calculated based on steady state, were generated by that same software. Differences between dissociation constants were compared using an unpaired two-tailed t-test with GraphPad Prism.

Quantifications of eGFP fluorescent intensities were calculated by subtracting eGFP pixel total intensity sums by average nuclear intensity sums, normalized by the number of cells, in three independent views across a slide. Nuclear surfaces were determined by DAPI display and eGFP pixel intensity values were extracted using Imaris software version 9.2.1. Differences between eGFP intensities outside of nuclear surfaces were compared using an unpaired two-tailed T-test with GraphPad Prism.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/ 10.3390/ijms222312733/s1.

**Author Contributions:** Conceptualization, A.E.W., K.E.O. and E.C.; methodology, A.E.W., G.S., A.G.G., S.G., I.M.-M., P.C.V.L., K.E.O. and E.C.; software, A.E.W., A.G.G., S.G., I.M.-M. and E.C.; validation, A.E.W., G.S., A.G.G., S.G. and E.C.; formal analysis, A.E.W., A.G.G., S.G. and E.C.; investigation, A.E.W., G.S., A.G.G., S.G., I.M.-M. and P.C.V.L.; resources, A.G.G., S.G. and E.C.; data curation, A.E.W., G.S., A.G.G., S.G., I.M.-M., P.C.V.L., K.E.O. and E.C.; writing—original draft preparation, A.E.W., K.E.O. and E.C.; writing—review and editing, A.E.W., G.S., A.G.G., S.G., I.M.-M., P.C.V.L., K.E.O. and E.C.; visualization, A.E.W., A.G.G., S.G., I.M.-M., K.E.O. and E.C.; supervision, A.E.W., A.G.G., S.G., I.M.-M., K.E.O. and E.C.; project administration, A.E.W., K.E.O. and E.C.; funding acquisition, K.E.O. and E.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Intramural Research Program of the NIH/NIAID (AI001246). The research was also supported by NIH/NIAID, grant number R01 AI130085-02.

**Institutional Review Board Statement:** Public Health Service Animal Welfare Assurance #A4149-01 guidelines were followed according to the National Institute of Allergy and Infectious Diseases (NIAID) and the National Institutes of Health (NIH) Animal Office of Animal Care and Use (OACU). These studies were carried out according to the NIAID-NIH animal study protocols (ASP) approved by the NIH Office of Animal Care and Use Committee (OACUC), with the approval ID ASP-LMVR3. Mice used in this study were housed in one of the animal facilities from the NIAID/NIH and were humanly treated according to OACU regulations.

**Informed Consent Statement:** Not applicable.

**Acknowledgments:** The authors thank Karina Sewell, Kevin Lee, Andre Laughinghouse, and Yonas Gebremicale for excellent and essential mosquito rearing. The authors also thank Dave Garboczi and Jose Ribeiro for stimulating scientific discussion as well as John Andersen for technical assistance with SPR studies. The authors thank Glenn Nardone, Lisa (Renne) Olano, and Ming Zhao, Research Technology Branch, NIH, for mass spectrometry analysis, and Brian Martin, Research Technology Branch, NIH, for N-terminal sequencing. Finally, the authors thank Carol Blair for manuscript feedback and review.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **References**

