Embryo lysate (0–2 h) of D. melanogaster was prepared and diluted to 2 mg/mL. The lysate was mixed with an equal volume of adjuvant (GERBU). An alpaca was immunized with 1 mL of mixture by 4 subcutaneous injections at fortnightly intervals performed by Shenzhen Kangti Life Technology in Shenzhen, China, as previously described (Fang et al., 2020). Isolation of lymphocytes from the immunized alpaca, RNA extraction, cDNA synthesis, and VHH library construction were conducted as described in our previous study (Fang et al., 2020). Specifically, 50 mL of peripheral blood was collected from the alpaca, and lymphocytes were isolated by Ficoll-Hypaque density gradient centrifugation. Total RNA was extracted by TRIzol Reagent, and cDNA was synthesized. Sequences encoding various domains of all heavy chains (vh) were first amplified by PCR. The vhh sequences were then amplified using the vh sequences as templates, which were then ligated to a phage display vector pADL-10b. Subsequently, the ligation products were transduced into SS320 competent cells to generate the VHH library.

CG7544 contains 305 amino acids (aa), and the Drosophila Myc and CyclinE proteins consist of 717 and 709 aa, respectively. Considering the difficult expression and purification of large proteins in E. coli, truncated Myc and CyclinE were used in our study to screen corresponding nanobodies. Full-length CG7544 and amino acids 1–400 of Myc were PCR-amplified from Drosophila cDNA with the respective primer pairs 7544F: 5′-CGGGATCCATGGTAAAAACTAAGGGAAACCAAAAG-3′/7544R: 5′-CCC​AAG​CTT​TTA​TGT​TGT​CCC​CTG​CTG​TAG-3′ and mycF: 5′-CGGAATTCGCCCTTTACCGCTCTGATCCGTAT-3′/mycR: 5′-CCC​AAG​CTT​TTA​ACC​ATC​GTC​CAC​CAT​ATC​GTT​GCA​G-3′ and were cloned into pET-28a individually. Amino acids 1–361 of CyclinE were PCR-amplified from cDNA with the primers cycEF: 5′-CGGAATTCAGTGTTTGTTCTACCAGCAGCACTGAG-3′/cycER: 5′-CCC​AAG​CTT​TTA​CAA​GAG​AAT​GGC​ACG​CAT​ACG​TG-3′ and cloned into pET-28a-GFP-TEV (a plasmid constructed in our laboratory with gfp and tev sequences inserted into pET-28a). These expression vectors were transformed into E. coli BL21 (DE3), and CG7544 and Myc were purified by affinity chromatography on nickel columns in a denatured form and then refolded. GFP-TEV-CyclinE was purified in a soluble form. CyclinE was obtained by cleaving GFP-TEV-CyclinE with TEV protease and purified by affinity chromatography on a nickel column.

To pan for VHH binders of CG7544, Myc, and CyclinE, a phage display library was first obtained from the vhh library, and its titre was determined by preparing serial tenfold dilutions according to the methods described previously (Pardon et al., 2014). The panning process was conducted as described in our previous study (Fang et al., 2020) with some modifications. Specifically, 60 μg of CG7544, Myc, or CyclinE was incubated with 30 μL of Dynabeads His-Tag Isolation and Pulldown (Invitrogen) for 1 h at room temperature (RT). The protein Dynabeads were then washed three times with PBST (containing 0.05% Tween-20). Meanwhile, the phage display library containing 5 × 10¹² phages in 500 μL of incubation buffer (PBST with 0.05% BSA) was pretreated by incubating with 30 μL of Dynabeads for 30 min at RT. Then, the pretreated phage display library was added to the protein Dynabeads, which were incubated with rotation at RT for 2 h. After that, the beads were washed 25 times with PBST to remove free and weakly bound phages. Finally, 500 μL of trypsin (0.25 mg/mL) was added to the beads and incubated at RT for 30 min to disassociate the binding phages. The eluted phages were neutralized with 10 μL of protease inhibitor cocktail (Roche). The first sublibrary and corresponding subphage library were obtained as previously described (Fang et al., 2020). This panning process was repeated three times to enrich phages displaying antigen-specific binders, and the number of input phages in the second and third rounds was reduced to 1 × 10¹². Some colonies from the third sublibrary were randomly selected and sequenced.

Subsequently, ELISA was applied to select positive clones expressing antigen-specific binders from the corresponding third sub-vhh library. Specifically, a 96-well microtiter plate was coated with CG7544, Myc, or CyclinE and blocked with 3% BSA. ELISA was performed as previously described (Fang et al., 2020). Wells with an absorbance value at 405 nm two fold higher than those in negative control wells were defined as positive wells, and the corresponding colonies were defined as positive colonies.

Eight sequences encoding the potential VHHs against Myc (a8, b10), CyclinE (n3, n4, and n6), and CG7544 (c3, c4, and c8) were identified and named as indicated in the brackets. To express and purify these VHHs, a prokaryotic expression vector, pADL-10b-Flag-His, was constructed by inserting Flag- and 6 × His-encoding sequences into pADL-10b, which contains pelB sequences to secrete VHHs to the periphery. Then, the eight vhh sequences were PCR-amplified individually from corresponding phagemids using the primer pair nbF: CCG​GAA​TTC​ATG​GCA​GAT​GTG​CAG​CTG​CAG/nbR: GGC​CTC​GAG​ACC​AGA​ACC​ACC​GCT​GGA​GAC​GGT​GAC​CTG​GGT​C, digested with EcoRI/Xho I (indicated with underlines), and cloned into pADL-10b-Flag-His individually to generate pADL-10b-Flag-Nb-His. A GGSG linker (indicated with wavy lines) was inserted between the nanobody and 6× His sequences. pADL-10b-Flag-Nb-His was then transformed individually into competent SS320 for expression. VHHs were induced with IPTG at a final concentration of 0.4 mM when SS320 reached exponential phase at 37°C, and SS320 was then cultured at 30°C overnight. Crude extracts of VHHs were obtained by osmometry and then purified by affinity chromatography on a nickel column.

ELISA was first performed to confirm the binding ability between purified VHHs and corresponding antigens. Specifically, a 96-well microtiter plate was coated with 1 μg/well purified CG7544, Myc, or CyclinE and blocked with 3% BSA. This plate was then incubated sequentially with 1 μg of purified VHHs for 2 h at RT, anti-Flag antibody for 1 h at RT, HRP-conjugated goat anti-mouse antibody for 1 h at RT, and TMB (200 μL/well) (Thermo Scientific). TMB was incubated for no more than 30 min in the dark. The reaction was terminated with H₂SO_4, and the absorbance at 450 nm was determined. As negative controls, no Flag antibody or VHHs were used. Each VHH assay was conducted in triplicate, and the results represent three independent assays.

SPR was applied to further confirm the binding ability between purified VHHs and corresponding antigens. The SPR experiment was performed using the bScreen LB 991 Label-free Microarray System (BERTHOLD TECHNOLOGIES, Germany). The chemically modified label-free photo-cross-linker sensor chips were provided by Betterways Inc., (China). Purified CG7544, Myc, and CyclinE (in PBS) were individually immobilized onto the surface of the optical cross-linked chip. Purified C4, A8, and N4 (in PBS) were diluted separately with running buffer at concentrations of 200 nM, 400 nM, 800 nM, 1,600 nM, and 3,200 nM and were injected as flow fluid. The solvent for proteins (PBS, pH = 7.4) was crosswise tested as blank controls and background noise controls. The captured signals were monitored in real time using the PlexArray^® HT system (Plexera^® Bioscience, Beijing, China). The processing and analysis of the association and dissociation rate constants (Ka and Kd, respectively) and the equilibrium dissociation constant (KD, kd/ka) were performed using the data analysis software of the bScreen LB 991 unlabelled microarray system according to a single-site binding model (1:1 Langmuir binding) with mass transfer limitations for determination of the binding kinetics.

First, an immune VHH library was constructed as described in the Materials and Methods. PCR amplification of vh sequences resulted in two products with lengths of nearly 800 bp and 600 bp (Figure 1A). vhh sequences were then amplified using the 600 bp fragment as a template, and an 400-bp product was obtained (Figure 1B). Finally, a bacterial library containing the vhh fragment was constructed, and its capacity was approximately 3 × 10⁷. The quality and diversity of the library were determined by PCR and DNA sequencing. As shown in Figure 1C, 15 of 16 ransom-selected colonies contained vhh fragments, suggesting a nearly 94% insertion rate of the library. These vhh fragments were further sequenced, and their amino acid sequences showed different CDRs among these colonies, suggesting a high diversity of the library (Figure 1D).

Myc encodes a transcription factor that is homologous to vertebrate Myc proto-oncogenes. It contributes to cell growth, cell competition and regenerative proliferation. CyclinE is essential for the control of the cell cycle at the G1/S (start) transition. CG7544 is the RNA N6-methyltransferase that mediates N6-methylation of adenine of U6 small nuclear RNA in Drosophila. They all have critical roles in Drosophila development. The three Drosophila proteins were selected as antigens to screen corresponding nanobodies from the immune library. Prokaryotic expression vectors for full-length CG7544 and truncated Myc and CyclinE were constructed, and the three proteins were purified from E. coli (Figure 2A). VHHs against these proteins were enriched individually for three rounds by phage display. The number of input phages was 5 × 10¹² in the first round and 1 × 10¹² in the second and third rounds. The number of output phages in each round is shown in Figure 2B, and those in the third round were 5.5 × 10⁸, 3 × 10⁸, and 7 × 10⁸ for Myc, CyclinE, and CG7544, respectively, showing an increasing number of output phages. These results indicated the successful enrichment of VHH binders. To estimate the efficiency of enrichment, colonies were randomly selected from the third round sublibrary of Myc, CyclinE, or CG7544, and the inserted vhhs were sequenced. As shown in Supplementary Figure S1, enriched sequences with the same CDR3 could be found in the VHH libraries against Myc, CyclinE, or CG7544, also suggesting the successful enrichment of VHH binders.

Subsequently, positive colonies containing VHHs against Myc, CyclinE, or CG7544 were screened from the third-round sublibrary by ELISA. Among randomly selected colonies, colonies with positive reactions were selected and sequenced. Amino acid sequences of positive colonies against Myc or CG7544 are shown in Supplementary Figure S2. Consistently, sequences of the colonies with strong positive reactions in Myc and CG7544 were the same as the most enriched sequences in Supplementary Figure S1. Randomly selected colonies (N1, N2, N3, N4, and N6 shown in Supplementary Figure S1C) from the third-round sublibrary of CyclinE were directly tested by ELISA, and N1, N3, N4, and N6 showed positive reactions. In summary, two different VHHs against Myc (A8 and B10), three different VHHs against CyclinE (N3, N4, and N6), and three different VHHs against CG7544 (C3, C4, and C8) were obtained, and their sequences are shown in Figure 2C. The DNA sequences encoding the eight VHHs were deposited in the GenBank of the National Center for Biotechnology Information, and their accession numbers are as follows: A8 [OQ822238], B10 [OQ822239], N3 [OQ822240], N4 [OQ822241], N6 [OQ822242], C3 [OQ822243], C4 [OQ822244], and C8 [OQ822245]. Moreover, the most enriched VHHs against Myc, CyclinE, and CG7544 were A8, N4, and C4, respectively. According to the results of our previous study and other studies, the most enriched VHH has a strong binding ability with its antigen (Fridy et al., 2014; Fang et al., 2020). Therefore, A8, N4, and C4 were expected to bind Myc, CyclinE, or CG7544 with high affinity.

To confirm the interaction between selected nanobodies and their corresponding antigens, a8, b10, n3, n4, n6, c3, c4, and c8 were first cloned into the prokaryotic expression vector pADL-10b-Flag-His (Figure 3A) individually and purified from bacteria as described in Materials and methods. The purity and molecular sizes of these nanobodies are shown in Figures 3B, C.

ELISA was then performed to determine the binding ability between purified nanobodies and corresponding purified antigens. As shown in Figures 3D–F, all nanobodies showed positive interactions with antigens. For the most enriched VHHs, the absorbance values of N3 (Figure 3E) and C4 (Figure 3F) at OD₄₅₀ were more than 4-fold and 3-fold higher than those of the negative controls, respectively. In particular, the absorbance value at OD₄₅₀ of A8 was more than 11-fold higher than that of the negative controls (Figure 3D), suggesting the strong binding between A8 and Myc.

Subsequently, SPR was applied to further determine the binding affinity between nanobodies and antigens. As shown in Figure 4A, binding signals were hardly detected in the negative groups of PBS-A8, PBS-N3, and PBS-C4, while Myc-A8, CyclinE-N3, and CG7544-C4 showed strong binding signals. The maximum values of the binding signal are shown in Figure 4B. The equilibrium constants of Myc-A8, CyclinE-N3, and CG7454-C4 were 2.85E-06, 4.83E-06, and 1.05E-05, respectively, indicating strong, strong, and medium binding affinities (Table 1).