The genotypic characteristics of the S. pombe and S. cerevisiae strains used in this study are detailed in Table 1. Schizosaccharomyces pombe strains were cultured on yeast extract with supplements (YES) medium under non-selective growth conditions, following established protocols (Sabatinos and Forsburg, 2010). For S. pombe strains requiring plasmid integration or transformation, synthetic Edinburgh minimal medium (EMM) lacking a specific ribonucleotide base or specific amino acids was used to allow the selection a particular plasmid in transformed yeast cells. For the selection of S. pombe transformants via geneticin (G418) resistance, 200 mg/L of G418 was added to YES medium. Yeast extract (1%), bactopeptone (2%), and dextrose (2%) (YPD) medium was used for the routine growth of S. cerevisiae strains. Synthetic dropout (SC) medium was utilized when DNA plasmid transformation was required in S. cerevisiae cells (Yun et al., 2000; Sherman, 2002). A modified synthetic dextrose minimal medium depleted for copper and iron (SD^-Cu-Fe) was used for cross-feeding experiments (Brault et al., 2022). To perform cross-feeding assays, aliquots of the indicated cultures of S. pombe (1 × 10⁷ cells/10 μL) and S. cerevisiae (3 × 10³ cells/10 μL) strains were spotted in the vicinity of one another as previously described (Brault et al., 2022).

Plasmids pJKsib3⁺-GFP, pJKsib3⁺-TAP, and pSP1sib3⁺-TAP have been previously described (Brault et al., 2022). Plasmid pJKsib3⁺-GFP served as the template for introducing site-specific mutations using a PCR overlap extension method (Ho et al., 1989). In the case of pJKsib3H248A-GFP, the sib3⁺ codon corresponding to His²⁴⁸ was substituted with an alanine codon. Similarly, the codon corresponding to Glu²⁸⁶ was mutated either using pJKsib3⁺-GFP or pJKsib3H248A-GFP as the template. The codon was replaced with a nucleotide triplet encoding an alanine residue, and the resulting plasmids were denoted as pJKsib3E286A-GFP and pJKsib3H248A/E286A-GFP, respectively. An identical PCR site-directed mutagenesis procedure was used to generate a series of mutants, this time employing either pJKsib3⁺-TAP or pSP1sib3⁺-TAP as templates. The resulting plasmids were designated as pJKsib3H248A-TAP, pJKsib3E286A-TAP, pJKsib3H248A/E286A-TAP, pSP1sib3H248A-TAP, pSP1sib3E286A-TAP, and pSP1sib3H248A/E286A-TAP.

To generate a deletion from the C-terminal end of Sib3, the sib3⁺ gene was isolated by PCR using primers amplifying the sib3⁺ locus starting at −501 bp from the start codon up to the codon 210 of the gene. This 1,143-bp ApaI-XmaI PCR-amplified DNA segment was inserted into the corresponding sites of pSP1 (Cottarel et al., 1993), creating plasmid pSP1-501sib3codons1–210nostop. The TAP sequence was isolated by PCR, digested with XmaI and SacI, and then insert in-frame into the sib3 coding region (residues 1–210). This new plasmid was denoted as pSP1-501sib3codons1-210-TAP.

The sib3⁺ promoter, containing 501-bp, was inserted into the ApaI and EcoRI sites of pSP1, resulting in plasmid pSP1-501sib3prom. To create truncated versions of Sib3 from the N-terminal end, four DNA fragments containing C-terminal codons 135–334, 165–334, 211–334, and 290–334 were PCR-amplified with an initiation codon specifying methionine and cloned into the EcoRI and XmaI sites of pSP1-501sib3prom. Following the aforementioned procedure, the XmaI-SacI TAP-encoded DNA fragment was inserted in-frame with the 3′-terminal coding sequence of truncated versions of Sib3. The resulting plasmids were denoted as pSP1-501sib3codons135-334-TAP, pSP1-501sib3codons165-334-TAP, pSP1-501sib3codons211-334-TAP, and pSP1-501sib3codons290-334-TAP, respectively.

The plasmid pJB1tpsib2⁺-GFP has been previously described (Brault et al., 2022). Three DNA fragments containing distinct N-terminal regions of Sib2 were generated by PCR amplification. After purification, these PCR products were digested with EcoRV and XmaI, and then used to replace the corresponding wild-type sib2⁺ DNA segment in pJB1tpsib2⁺-GFP. The resulting plasmids were designated as pJB1tpsib2codons1-349-GFP, pJB1tpsib2codons1-279-GFP, and pJB1tpsib2codons1-135-GFP, respectively. Three additional PCR-amplified fragments were generated, each containing different C-terminal regions of Sib2. An identical molecular cloning strategy, as described for the C-terminal truncated versions of Sib2, was employed, with the exception that the EcoRV-XmaI fragments of the sib2⁺ gene possessed an initiator codon methionine in-frame with different 5′-terminal coding sequences of sib2⁺. The resulting plasmids were denoted as pJB1tpsib2codons82-431-GFP, pJB1tpsib2codons174-431-GFP, and pJB1tpsib2codons359-431-GFP, respectively. Two additional plasmid constructs, encompassing a central region of Sib2, were designated as pJB1tpsib2codons149-345-GFP and pJB1tpsib2codons281-358-GFP. These central regions were amplified by PCR using primers designed to generate EcoRV and XmaI restriction sites at the upstream and the downstream termini of a sequence spanning 196 or 77 codons within the middle region of sib2⁺. Moreover, the sense primer contained an initiator codon in-frame with the coding sequences of the middle region (codons 149–345 or 281–358) of sib2⁺. Subsequently, the EcoRV-XmaI DNA fragments were swapped with the corresponding sib2⁺ EcoRV-XmaI fragment in the pJB1tpsib2codons82-431-GFP plasmid. Plasmids p415GPDARN1-GFP and pBPnmt41x-GFP, used for cross-feeding assays and coimmunoprecipitation experiments, respectively, have been previously described (Brault and Labbé, 2020; Brault et al., 2022).

To assess protein steady-state levels, cell lysates were prepared by glass bead disruption using an extraction buffer containing 50 mM Tris–HCl (pH 7.5), 150 mM NaCl, 5 mM MgCl₂, 0.1% Nonidet P-40, and a complete protease inhibitor mixture (Sigma-Aldrich, P8340). Cell lysis was achieved using a FastPrep-24 instrument (MP Biomedicals, Solon, OH). Equal amounts of each protein sample were resolved by electrophoresis on 8% sodium dodecyl sulfate (SDS)-polyacrylamide gels. Protein transfer from gel to membrane in Western blot assays was carried out as described previously (Ioannoni et al., 2016). The immunodetection of Sib3-TAP, Sib3-GFP, Sib2-GFP, and α-tubulin was performed using polyclonal anti-mouse IgG antibody (ICN Biomedicals), monoclonal anti-GFP antibody B-2 (Santa Cruz Biotechnology), and monoclonal anti-α-tubulin antibody (clone B-5-1-2, Sigma-Aldrich). Subsequently, the membranes were washed and incubated with the appropriate horseradish peroxidase-conjugated secondary antibodies (Amersham Biosciences), developed using enhanced chemiluminescence (ECL) reagents, and visualized via chemiluminescence using an ImageQuant LAS 4000 instrument (GE Healthcare) equipped with a Fujifilm High Sensitivity F0.85 43 mm camera.

Protein–protein interaction assays were performed to further delineate the regions of Sib2 and Sib3 necessary for their mutual interaction. Full-length or truncated versions of Sib2-GFP were coexpressed with Sib3-TAP in sib2Δ sib3Δ cells. Similarly, sib2Δ sib3Δ cells were used to coexpress full-length Sib2-GFP with wild-type Sib3-TAP or its mutant derivatives. For each indicated pair of coexpressed proteins, cotransformed sib2Δ sib3Δ cells were grown in EMM medium until mid-logarithmic phase. At this point, cells were incubated with Dip (250 μM) and L-ornithine (15 mM) (Sigma-Aldrich) for 2 h. Subsequently, cells were washed and transferred to YES medium with the same concentrations of Dip and L-ornithine for an additional 3 h. Pull down assays were performed as previously described (Jacques et al., 2014), with the following modifications. The cells were disrupted with glass beads in lysis buffer A containing 50 mM HEPES (pH 7.5), 140 mM NaCl, 5 mM MgCl₂, 1 mM EDTA, 1% Triton X-100, 0.1% sodium deoxycholate, 1 mM phenylmethylsulfonyl fluoride, 50 mM sodium fluoride, 0.2 mM sodium orthovanadate, and a mixture of protease inhibitors (Sigma-Aldrich, P8340). Cell lysates (~2.0 mg of total proteins per sample) were incubated with IgG-Sepharose 6 Fast-Flow beads (GE Healthcare), and the suspensions were mixed end-over-end on a rotating wheel for 2 h at 4°C. The beads underwent four washes. The first wash used a modified buffer A containing 500 mM NaCl instead of 140 mM. The second and third washes were performed with buffer B containing 10 mM Tris–HCl (pH 8.0), 250 mM lithium chloride, 0.5% NP-40, 0.5% sodium deoxycholate, and 1 mM EDTA. Prior to the final wash, samples were transferred to fresh microtubes, and the last wash was carried out using buffer C containing 10 mM Tris–HCl (pH 8.0) and 1 mM EDTA. Subsequently, the attached complexes were dissociated from the beads using an elution buffer consisting of 50 mM Tris–HCl (pH 8.0), 10 mM EDTA, and 1% SDS. Further dissociation occurred at 65°C in a ThermoMixerC (Eppendorf) with a rotating speed of 1,200 rpm for 12 min. The resulting immunoprecipitated fractions were mixed with an SDS loading buffer containing urea (4.0 M) and thiourea (1.0 M), heated at 95°C for 5 min, and then resolved by electrophoresis on 8% SDS-polyacrylamide gels. Proteins were electroblotted onto nitrocellulose membranes and immunodetection assays were performed as described previously (Brault et al., 2022).

The extraction method for isolating Fc from S. pombe cells has been previously described (Plante and Labbé, 2019; Brault et al., 2022). Following their isolation and lyophilization, Fc from the indicated strains was resuspended in water and applied onto preheated silica gel 60 F₂₅₄ thin-layer chromatography (TLC) plastic sheets (EMD Millipore). TLC assays were conducted using a solvent containing 80% aqueous methanol. Commercially purified holo-Fc (Sigma-Aldrich) served as the positive control for signal detection.

Fluorescence and differential interference contrast images (Nomarski) of cells were captured using a Nikon Eclipse E800 epifluorescence microscope (Nikon, Melville, NY) outfitted with a Hamamatsu ORCA-ER digital cooled camera (Hamamatsu, Bridgewater, NJ). The cells were observed at 1,000× magnification using an excitation wavelength of 465–495 nm for detecting GFP signal. The representative cell fields depicted in this study are derived from a minimum of five independent experiments. Furthermore, displayed cell fields represent protein localization in 200 cells examined from GFP-expressing strains.

The UniProt knowledgebase (UniProtKB) amino acid sequences for S. pombe Sib3, U. maydis Fer5, M. tuberculosis MbtK/Rv1347c, A. fumigatus SidL and SidF are Q9UUE3, Q4PEN1, P9WK15, Q4WJX7, and Q4WF55, respectively.

Our previous findings have shown that Sib3 is required for Fc production in S. pombe (Brault et al., 2022). Through comparisons of amino acid sequences, Sib3 was predicted to contain a Gcn5-related N-acetyltransferase (GNAT) domain (Rutherford et al., 2024). The presence of this domain suggested that Sib3 has the capability to catalyze the transfer of an acetyl group to N⁵-hydroxyornithine, resulting in the production of N⁵-acetyl-N⁵-hydroxyornithine, which is involved in Fc biosynthesis. Alignment of an amino acid region (residues 167–291), encompassing the putative GNAT-like domain of Sib3, with four other predicted or known siderophore-biosynthetic acetyltransferases, revealed a high degree of conservation for His²⁴⁸ and Glu²⁸⁶ (Figure 1). In the case of Glu²⁸⁶, its conservation was greater with the three other fungal acetyltransferases from A. fumigatus (SidL and SidF) and U. maydis (Fer5) compared to that of the bacterium M. tuberculosis Rv1347c (MbtK). In the case of His²⁴⁸, it corresponds to residue His¹³⁰ in the M. tuberculosis Rv1347c protein (Figure 1) (Frankel and Blanchard, 2008). In M. tuberculosis, mutagenesis of His¹³⁰ blocks the acetyltransferase reaction by Rv1347c, disrupting its activity and consequently siderophore biosynthesis. As for S. pombe Glu²⁸⁶, the corresponding residue is Asp¹⁶⁸ in M. tuberculosis. According to the predicted positioning of amino acid side chains within the Rv1347c active site, it has been suggested that the Asp¹⁶⁸ residue of Rv1347c may enhance the basicity of His¹³⁰ or stabilize the proximal protonated imidazolium ion to optimize proper coordination of the substrate for its acetylation (Frankel and Blanchard, 2008).

Our previous findings have shown that S. pombe cells lacking Sib3 (sib3Δ) are unable to grow on YES medium in the presence of the iron chelator Dip (Brault et al., 2022). Given the requirement of M. tuberculosis His¹³⁰ and Asp¹⁶⁸ residues for the catalytic activity of the bacterial acetyltransferase, we initiated experiments to examine whether mutations in the His²⁴⁸ and Glu²⁸⁶ residues of Sib3 affected the S. pombe’s ability to grow under iron-limiting conditions. Spot assays were performed using sib3Δ cells expressing either TAP-tagged sib3⁺ or TAP-tagged mutant versions of sib3, where either His²⁴⁸, Glu²⁸⁶ or both His²⁴⁸ and Glu²⁸⁶ had been substituted by Ala residues. As shown in Figure 2A, sib3Δ cells expressing the sib3H248A-TAP or sib3H248A/E286A-TAP allele exhibited a severe growth defect when spotted on iron-depleted medium compared to the isogenic wild-type strain or sib3Δ cells expressing sib3⁺-TAP. In contrast, sib3Δ cells expressing the sib3E286A-TAP allele exhibited a robust growth equivalent to the sib3Δ mutant strain expressing sib3⁺-TAP (Figure 2A). As negative controls, a sib1Δ sib2Δ strain or sib3Δ cells containing an empty integrative vector (v. alone) failed to grow on YES medium supplemented with Dip (Figure 2A).

Consistent with the involvement of Sib3 in Fc biosynthesis, the results showed that growth of sib3Δ cells expressing TAP-tagged Sib3H248A and Sib3H248A/E286A mutants or an empty vector was rescued when exogenous Fc (10 μM) was added in YES medium (Figure 2A). To ensure that all TAP epitope-tagged sib3 alleles were expressed in sib3Δ cells, steady-state protein levels of Sib3 and its mutant derivatives were analyzed by immunoblotting. The results showed that all these proteins were expressed in the sib3Δ strain (Figure 2B). Taken together, these results indicate that Sib3 His²⁴⁸ is required for S. pombe growth in iron-poor media without Fc supplementation.

Given the findings that His²⁴⁸ of Sib3 was required to promote cell growth under low-iron conditions, we sought to determine whether Fc was produced in sib3Δ mutant cells expressing the sib3H248A-TAP or sib3H248A/E286A-TAP allele. Whole cell extracts were prepared from the wild-type, sib1Δ sib2Δ, fep1Δ, sib3Δ, and sib2Δ sib3Δ strains. Similarly, extracts were isolated from sib3Δ cells expressing sib3⁺-TAP and its mutant derivatives or sib2Δ sib3Δ cells co-expressing sib2⁺-GFP and sib3⁺-TAP or sib3-TAP mutated alleles. When Fc was isolated from extracts of the wild-type strain, a weak but reproducible Fc signal was detected by thin-layer chromatography (TLC) (Figures 3A,B). The detection of the Fc signal was stronger in extracts isolated from a fep1Δ strain due to the absence of Fep1 and, therefore, the lack of Fep1-mediated transcriptional repression of sib1⁺ and sib2⁺ genes (Figures 3A,B) (Mercier and Labbé, 2010; Brault et al., 2022). In contrast, extract preparations from sib3Δ, sib1Δ sib2Δ, and sib2Δ sib3Δ mutant strains exhibited no detectable Fc signal (Figures 3A,B). In the case of sib3Δ and sib2Δ sib3Δ strains expressing sib3H248A-TAP and sib3H248A/E286A-TAP mutant alleles, their extract preparations were devoid of detectable Fc (Figures 3A,B). Conversely, Fc signals were detected in extract preparations from sib3Δ cells expressing sib3⁺-TAP and sib3E286A-TAP or from sib2Δ sib3Δ cells co-expressing sib2⁺-GFP and sib3⁺-TAP or sib3E286A-TAP alleles (Figures 3A,B). Together, these results indicated that the His²⁴⁸ residue of Sib3 is essential for its activity in Fc production.

Our previous studies have shown that wild-type S. pombe cells can support the growth of S. cerevisiae cells by secreting Fc-iron as the sole source of iron (Brault et al., 2022). This Fc-dependent cross-feeding between S. pombe and S. cerevisiae requires S. pombe cells competent in Fc biosynthesis, thus expressing functional Sib1, Sib2, and Sib3 proteins. For S. cerevisiae, its ability to acquire exogenous Fc is revealed by co-culturing a S. cerevisiae fet3Δ arn1-4Δ mutant, in which the Fc-iron transporter ARN1 gene is reintroduced to mediate the uptake of Fc (Brault et al., 2022).

Studies have shown that a functional Sib3-GFP protein is constitutively expressed and distributed throughout the cytoplasm and nucleus of cells (Matsuyama et al., 2006; Brault et al., 2022). To verify that mutant derivatives of Sib3 exhibited the same cellular localization as the wild-type protein, microscopic analyses of GFP-tagged Sib3H248A, Sib3E286A, and Sib3H248A/E286A proteins were performed alongside Sib3-GFP. Results showed that Sib3-GFP and its mutant derivatives exhibited similar fluorescence signal patterns, being present in both the cytoplasm and nucleus (Figure 4A). Whole-cell extracts from aliquots of the cultures used for microscopic analyses were subjected to immunoblot assays. The results showed that Sib3-GFP and its mutant variants were expressed, as detected through immunoblot analysis (Figure 4B).

We next assessed the impact of Sib3 mutants on the ability of S. pombe to promote the growth of ARN1-expressing S. cerevisiae fet3Δ arn1-4Δ cells in co-culture experiments. An S. pombe sib3Δ strain expressing Sib3-GFP, Sib3H248A-GFP, Sib3E286A-GFP, or Sib3H248A/E286A-GFP was grown to logarithmic phase under iron-replete conditions (10 μM FeCl₃). At this point, cell cultures were spotted (1 × 10⁷ cells/10 μL) onto a copper-and iron-poor medium, designated SD^-Cu-Fe. Saccharomyces cerevisiae fet3Δ arn1-4Δ cells expressing a functional Arn1-GFP protein were cultured in SD^-Cu-Fe medium until logarithmic phase and then diluted 10,000-fold before being point-inoculated in the vicinity of the indicated S. pombe strains. The results showed that growth of the S. cerevisiae fet3Δ arn1-4Δ ARN1-GFP strain was severely limited when co-cultured with an S. pombe sib3Δ strain expressing sib3H248A-GFP or sib3H248A/E286A-GFP allele, due to the lack of Fc production (from S. pombe) necessary to fuel the growth of S. cerevisiae cells (Figure 4C). In contrast, growth of S. cerevisiae fet3Δ arn1-4Δ ARN1-GFP cells was supported when point-inoculated near the wild-type S. pombe strain or S. pombe sib3Δ cells expressing sib3⁺-GFP or sib3E286A-GFP allele (Figure 4C). As negative controls, S. pombe sib1Δ sib2Δ and sib3Δ cells containing an empty vector were unable to promote the growth of S. cerevisiae fet3Δ arn1-4Δ cells expressing Arn1-GFP (Figure 4C). Similarly, an S. cerevisiae fet3Δ arn1-4Δ mutant strain lacking Arn1-GFP was unable to grow in the vicinity of the wild-type S. pombe and S. pombe sib1Δ sib2Δ strains (Figure 4C). Collectively, these results indicated that the mutation of Sib3 His²⁴⁸ leads to cross-feeding inhibition between S. pombe and S. cerevisiae fet3Δ arn1-4Δ cells expressing Arn1-GFP.

Our previous studies have shown that functional Sib2-GFP and Sib3-TAP fusion proteins interact to form a protein complex (Brault et al., 2022). Considering the necessity of the His²⁴⁸ residue for Sib3 function in Fc biosynthesis, we investigated whether this amino acid is required for the association between Sib2 and Sib3. Furthermore, we examined the impact of substituting the Glu²⁸⁶ residue with an alanine on the Sib2-Sib3 association in parallel experiments (Figure 5A). We first created a sib2Δ sib3Δ mutant strain, wherein the sib3⁺-TAP, sib3H248A-TAP, sib3E286A-TAP, or sib3H248A/E286A-TAP allele was co-expressed with the sib2⁺-GFP allele. Cells co-expressing each combination of protein pairs were cultured to the mid-logarithmic phase and then incubated in the presence of Dip for 90 min. Subsequently, whole-cell extracts were prepared, and immunoprecipitation was performed using IgG-Sepharose beads, which specifically bound to TAP-tagged Sib3 and its mutant derivatives. Immunoblot analysis of proteins retained by the beads indicated the presence of both Sib3-TAP and Sib2-GFP in the immunoprecipitated fraction (IP) (Figure 5B). In the case of the IP fractions from cells co-expressing Sib3H248A-TAP and Sib2-GFP, Sib3E286A-TAP and Sib2-GFP, or Sib3H248A/E286A-TAP and Sib2-GFP, the results showed that Sib2-GFP remained an interacting partner of these Sib3 mutant derivatives (Figure 5B). The specificity of coimmunoprecipitation assays was validated by the absence of α-tubulin in the IP fractions (Figure 5B). Together, these results revealed that substituting the His²⁴⁸ and Glu²⁸⁶ residues with alanines in Sib3 does not hinder the ability of Sib3 to associate with Sib2.

To gain insight into the regions of Sib2 that interact with Sib3, we created a series of sib2 alleles encoding truncated versions of Sib2, either from its C-terminal or N-terminal end (Figure 6A). C-terminal deletions of Sib2-GFP were generated by removing 82, 152, and 296 amino acid residues, resulting in the creation of ¹Sib2³⁴⁹-GFP, ¹Sib2²⁷⁹-GFP, and ¹Sib2¹³⁵-GFP mutants, respectively. Each of these mutants was co-expressed with full-length Sib3-TAP in sib2Δ sib3Δ cells, which were then treated with Dip for 3 h. Moreover, sib2Δ sib3Δ cells were used to co-express sib3⁺-TAP and GFP alleles under the same growth conditions. Cell lysates were extracted from all these cultures and then incubated with IgG-Sepharose beads. Results showed that ¹Sib2³⁴⁹-GFP, ¹Sib2²⁷⁹-GFP, and ¹Sib2¹³⁵-GFP truncated proteins co-immunoprecipitated with TAP-tagged Sib3, whereas GFP alone was consistently absent in the IP fraction (Figures 6B–E). For N-terminal deletions of Sib2-GFP, they were generated by deleting 81, 173, and 358 amino acid residues, resulting in the production of ⁸²Sib2⁴³¹-GFP, ¹⁷⁴Sib2⁴³¹-GFP, and ³⁵⁹Sib2⁴³¹-GFP mutants, respectively. Furthermore, two additional truncated versions of Sib2 were created by combining N-and C-terminal deletions of the protein. These mutant versions of Sib2 were designated as ¹⁴⁹Sib2³⁴⁵-GFP, and ²⁸¹Sib2³⁵⁸-GFP, respectively (Figure 6A). sib2Δ sib3Δ cells co-expressing the aforementioned mutated forms of Sib2-GFP with full-length Sib3-TAP were incubated in the presence of Dip (250 μM). After 3 h, whole-cell extracts were prepared and subjected to co-immunoprecipitation assays using IgG-Sepharose beads. Results showed that the IP fractions contained ⁸²Sib2⁴³¹-GFP, ¹⁷⁴Sib2⁴³¹-GFP, ¹⁴⁹Sib2³⁴⁵-GFP, and ²⁸¹Sib2³⁵⁸-GFP proteins, which were identified as interacting partners with Sib3-TAP (Figures 6F–J). In contrast, co-immunoprecipitation assays failed to reveal an interaction between ³⁵⁹Sib2⁴³¹-GFP and Sib3-TAP (Figure 6H). Notably, α-tubulin was not detected in the IP fractions (Figures 6B–J), indicating the specificity of the pulldown assays. Taken together, these results highlight two minimal regions of Sib2 implicated in its association with Sib3: one encompassing residues 1–135 and the other spanning residues 281–358.

Our experiments on Sib2 prompted us to determine the region on Sib3 required for interacting with Sib2. Truncations were created from both the C-and N-terminal ends of Sib3-TAP (Figure 7A). The first construct removed the last 124 amino acids of Sib3 (¹Sib3²¹⁰-TAP). Subsequent constructs removed the first 134, 164, 210, and 289 amino acids of Sib3, designated as ¹³⁵Sib3³³⁴-TAP, ¹⁶⁵Sib3³³⁴-TAP, ²¹¹Sib3³³⁴-TAP, and ²⁹⁰Sib3³³⁴-TAP, respectively. Following this, TAP pull-down experiments were performed using protein lysates from cells co-expressing full-length Sib2-GFP with either wild-type Sib3-TAP, ¹Sib3²¹⁰-TAP, ¹³⁵Sib3³³⁴-TAP, ¹⁶⁵Sib3³³⁴-TAP, ²¹¹Sib3³³⁴-TAP, or ²⁹⁰Sib3³³⁴-TAP. Results consistently showed the presence of Sib2-GFP as an interacting partner with wild-type Sib3-TAP in the IP fraction (Figure 7B). In contrast, the interaction between ¹Sib3²¹⁰-TAP and Sib2-GFP ceased when the C-terminal region of Sib3, encompassing residues 211 to 334 was deleted, as Sib2-GFP was absent in the IP fraction. On the other hand, Sib2-GFP was detected in the IP fraction containing the other truncated forms of Sib3-TAP from the N-terminal end, including ¹³⁵Sib3³³⁴-TAP, ¹⁶⁵Sib3³³⁴-TAP, ²¹¹Sib3³³⁴-TAP, and ²⁹⁰Sib3³³⁴-TAP. Taken together, these results indicated that the minimal C-terminal region of Sib3 encompassing amino acids 290–334 is sufficient for Sib3-Sib2 association.