Re-assessment of the subcellular localization of Bazooka/Par-3 in Drosophila: no evidence for localization to the nucleus and the neuromuscular junction

Baz/Par-3 is a multi-domain scaffold protein with well-characterized roles in the establishment and maintenance of cell polarity. Baz is a core member of the polarity regulating Par complex, which it forms together with its evolutionarily conserved binding partners aPKC and Par-6 (Goldstein and Macara, 2007; Johnson and Wodarz, 2003; St Johnston and Ahringer, 2010; Tepass, 2012). Baz/Par-3 was first identified in C. elegans (Etemad-Moghadam et al., 1995) with the subsequent identification of homologues in Drosophila (Kuchinke et al., 1998), and mammals (Izumi et al., 1998). In C. elegans, Baz/Par-3 functions as a key regulator of spindle orientation and polarity in the early embryo (Etemad-Moghadam et al., 1995) and the assembly of adherens junctions in the intestine (Achilleos et al., 2010). In Drosophila, Baz has been shown to function as a key regulator of apico-basal polarity in epithelial cells (Harris, 2012; Tepass, 2012). Baz functions at the top of a genetic hierarchy in the establishment of adherens junctions during cellularization (Harris and Peifer, 2004, 2005) and in the maintenance of apico-basal polarity during embryonic epithelial development (Bilder et al., 2003; Müller and Wieschaus, 1996; Tanentzapf and Tepass, 2003). Baz is also required for oocyte differentiation (Cox et al., 2001) and polarization of the developing oocyte along the anterior/posterior axis (Doerflinger et al., 2010). In the developing nervous system, Baz is essential for the maintenance of stem cell fate in neuroblasts through the establishment of apico-basal polarity, spindle orientation and asymmetric cell division (Kuchinke et al., 1998; Loyer and Januschke, 2020; Schober et al., 1999; Wodarz et al., 1999). In mammalian epithelial cells, Par-3 has been shown to play a crucial role in the assembly and maintenance of tight junctions and adherens junctions as well as epithelial spindle orientation (Chen and Macara, 2005; Hao et al., 2010; Ooshio et al., 2007).

Aside from their cortical localization and roles in regulating cell polarity, several studies have reported that members of the Par complex also localize to the nucleus (Cline and Nelson, 2007; Fang et al., 2007; Perander et al., 2000; Seidl et al., 2012; Speese et al., 2012). Studies in mammalian cell culture have revealed that in response to DNA damage induced by γ-irradiation, Par3 translocates to the nucleus where it associates with DNA-dependent protein kinase (DNA-PK) to mediate DNA double strand break repair (Fang et al., 2007). A study by Speese et al., 2012 revealed that Par proteins also show nuclear localization in Drosophila body wall muscles where they promote neuro-muscular junction (NMJ) formation. They reported that Baz localizes to the nuclear envelope, including nuclear envelope foci containing a C-terminal cleavage product of the Drosophila Wingless/Wnt1 receptor DFrizzled 2 (DFz2C). At these foci, Baz is required for phosphorylation of the nuclear envelope component LaminC (LamC) by aPKC to promote nuclear envelope budding, facilitating the export of ribonucleoprotein particles containing DFz2C and mRNAs encoding post-synaptic proteins (Speese et al., 2012).

In this study, we sought to further analyze Baz nuclear envelope localization in Drosophila to gain insight into any additional functions it may have in the nucleus. By immunostaining analysis and two independent Baz-GFP fusion protein lines (Besson et al., 2015; Buszczak et al., 2007) under the endogenous promotor, we assessed whether Baz nuclear envelope localization was ubiquitous or restricted to specific cell types or developmental stages. Immunostaining analysis using multiple antibodies raised against different domains of Baz revealed that immunostaining for Baz at the nuclear envelope was detectable in some but not all tissues, in particular in polyploid cells. However, to our surprise mutational analysis revealed that Baz nuclear envelope immunostaining persisted in baz mutant clones although cortical staining was completely abandoned. Furthermore, analysis of Baz-GFP fusion protein lines showed no Baz-GFP nuclear envelope localization. The same was true for localization of Baz to the NMJ. Thus, our results provide strong evidence that Baz does neither localize to the nuclear envelope nor to the NMJ, requiring a reassessment of its reported function at these subcellular sites.

We assessed by indirect immunofluorescence whether Baz also localizes to the nucleus, in addition to its well-described localization to the cytocortex and intercellular junctions. Several antibodies were used to stain fat body tissue dissected from wild-type 3rd instar larvae. Antibodies raised in rabbit and rat against the Baz N-terminal region (amino acids 1-297, numbering refers to isoform PA of Baz; Wodarz et al., 1999, 2000), an antibody raised in guinea pig against the Baz PDZ domains (amino acids 309-747; Shahab et al., 2015) and an antibody raised in guinea pig against a region in the C-terminal half of Baz (amino acids 905-1221, this work) all showed nuclear staining (Fig. S1). For all further analyses we made use of the antibody against the N-terminal region of Baz raised in rabbit, as in the Drosophila community it is the most commonly used antibody to detect Baz. Also, it showed the least background staining in comparison to the other antibodies tested. Closer inspection of the immunofluorescence pattern detected by this antibody in fat body showed a localization to the periphery of the nucleus colocalizing with the nuclear envelope component Lamin C (Fig. 1A-A″), indicating that Baz localizes to the nuclear envelope.

We next sought to determine whether Baz nuclear envelope localization occurred in a tissue-specific manner. In addition to larval fat body cells (Fig. 1A-A″; Fig. S1), we observed Baz nuclear envelope localization in larval body wall muscle (Fig. 1B-B″), adult ovary nurse cells (Fig. 1C-C″), follicular epithelial cells (Fig. 1D-D″; Fig. S3B) and oocytes (Fig. 1E-E″). Baz nuclear envelope immunostaining signal was absent in embryonic tissue, e.g. epidermis (Fig. S2A-A″) and neuroblasts (Fig. S2A-A″), in larval midgut imaginal islands (Fig. S2B-B″) and in larval wing imaginal disc cells (Fig. S2C-C″).

While strong nuclear envelope immunostaining was observed using several independently raised anti Baz antibodies (Fig. 1; Fig. S1), no nuclear envelope localization was detected in follicular epithelial cells and in larval body wall muscles using a Baz-GFP BAC line (Besson et al., 2015) (Figs S3C-D″, S4A,A') nor in a GFP-Baz protein-trap line (Buszczak et al., 2007) (Figs S3E-F″, S4C,C'). In the GFP-Baz protein-trap line an engineered exon encoding for GFP is inserted into the second untranslated exon (Fig. S5). This exon encoding for GFP is predicted to be spliced in frame into the mRNAs RA and RC encoding for isoforms PA and PC whose translation starts in exon 1 (Fig. S5), resulting in insertion of GFP between amino acid residues K40 and P41 of isoforms PA and PC. The transcripts RB and RD encoding Baz isoforms PB and PD have their translation start within exon 3 and thus cannot form fusion proteins with GFP inserted in exon 2 (Fig. S5). However, GFP-Baz protein trap flies are homozygous viable and are phenotypically indistinguishable from wild-type flies, indicating that the corresponding GFP fusion protein is fully functional and faithfully reflects the expression pattern and subcellular localization of Baz isoforms PA and PC. The BAC line integrates the GFP within exon 10 between amino acid residues L1424 and Q1425 of isoform PA, giving rise to GFP fusion proteins for all four isoforms (Fig. S6) (Besson et al., 2015). Like the protein-trap GFP-Baz fusion protein, the Baz-GFP fusion protein in the BAC line is fully functional as it completely rescued lethality and fertility of the baz^EH747 (Fig. S7D-D″) and baz^815-8 alleles (Besson et al., 2015). For both lines GFP expression was detected at adherens junctions, colocalizing with the signal from the Baz N-terminal antibody (Fig. 3E,G; Figs S3C,C',E,E', S7D',D″,E',E″). However, even after signal enhancement using an anti-GFP antibody we did not detect a signal at the nuclear envelope in follicular epithelial cells like seen when using the anti-Baz antibodies (Fig. S3D',D″,F',F″), nor at the nuclear envelope of the oocyte (Fig. S7D',D″,E',E″).

In the follicular epithelium, Baz immunostaining detected Baz at the adherens junctions (Fig. S3A) and in a different focal plane at the nuclear envelope (Fig. S3B). To test whether both junctional and nuclear envelope staining of Baz was specific, we eliminated baz expression in the follicular epithelium using two different methods. First, we used RNAi against baz driven by traffic jam (tj)::Gal4, which is expressed in all follicle cells of the developing egg chamber until stage 12 (Fig. 2A') (Li et al., 2003). During early stages of egg chamber development the expression of tj::Gal4 in the follicle cells is heterogeneous (Fig. 2A'), while it becomes homogeneous in older egg chambers (inset in Fig. 2A'). Whereas the junctional staining for Baz close to the apical surface of the follicular epithelium was lost upon RNAi against baz (Fig. 2B), the staining at the nuclear envelope persisted in the same egg chamber imaged at the optical plane containing the nuclei (Fig. 2C). The staining for Armadillo/beta-catenin (Arm) (Fig. 2B',C') shows junctions that are still intact although Baz is downregulated (Shahab et al., 2015).

For clonal analysis the strong loss-of-function allele baz^EH747 was used, where a point mutation in exon 4 results in a premature stop close to the N-terminus of all four isoforms (the codon for amino acid residue Q51 is mutated to a stop in isoform A) (Krahn et al., 2010). In follicular epithelial cells loss of Baz has no relevance for the integrity of adherens junctions and development of the egg chamber is unaffected (Shahab et al., 2015). FLP-FRT and MARCM clones were generated in follicle cells and in the germline. Both small and large clones in the follicle epithelium showed the loss of Baz staining at the junctions (Fig. 2D,D') but a persistent immunostaining signal for Baz around the nucleus (Fig. 2E,E'). Large follicle cell clones encompassing all follicle cells of an egg chamber marked by the loss of nuclear GFP (Fig. S7B') showed the loss of junctional Baz in the baz^EH747 mutant follicle cells (Fig. S7B″). The loss of Baz in the follicle cells did not result in any apparent junctional defect and the junctional marker Arm was still localized as in wild type (Fig. S7B‴). In the germ line Baz was detectable at the junctions between the germ line cells, in particular between nurse cells and the oocyte. In addition, a strong signal was consistently detected at the nuclear envelope of the oocyte (Fig. S7A″). While in baz^EH747 mutant germ line clones marked by the loss of nuclear GFP (Fig. S7C′), the junctional staining within the germline was lost, the staining around the oocyte nucleus persisted (Fig. S7C″). The staining for Arm revealed that the junctions between the germline cells were still intact (Fig. S7C‴). The staining around the oocyte nucleus was not detectable with the anti GFP antibody in the Baz-GFP BAC line (Fig. S7D′) nor in the GFP-Baz protein-trap line (Fig. S7E′). Staining with rabbit anti Baz antibody again showed the oocyte nuclear envelope marked in these egg chambers (Fig. S7D″,E″).

In parallel to our observations in the ovary we looked for Baz nuclear localization in the larval body wall muscle. Baz immunostaining was detectable at the nuclear envelope and at the neuromuscular junction (NMJ) as published before (Ruiz-Canada et al., 2004; Speese et al., 2012) (Fig. 3A-A″). To test if this signal was specific for the antibody, we stained with the pre-immune serum, revealing a striped pattern in the muscle, but no signal at the NMJ or the nucleus (Fig. 3B-B″). Staining with the secondary antibody only without any primary antibody resulted in no signal at the NMJ or the nucleus (Fig. 3C-C″). Anti Discs large antibody and Hoechst were used in the same stainings as control to detect the NMJs and the nuclei (Fig. 3A-C″). Together, these experiments showed that the Baz antibody was indeed responsible for the signal at the NMJ and the nuclear envelope of somatic body wall muscles.

By contrast, we did not detect GFP at the NMJ or at the nuclear envelope in larval body wall muscles in the Baz-GFP BAC and the GFP-Baz trap lines stained with anti GFP antibody (Fig. 3D′,F′), whereas we observed a strong signal with the anti Baz antibody in these lines (Fig. 3D,F). In the wing imaginal discs of the same L3 larvae we observed colocalization of Baz and GFP immunostaining at intercellular junctions as expected (Fig. 3E,G). These findings were confirmed by analysis of fixed larval tissues that were imaged for GFP fluorescence without anti GFP antibody staining (Fig. S4). Neither in the Baz-GFP BAC line (Fig. S4A,A′), nor in the GFP-Baz trap line (Fig. S4C, C′) any nuclear envelope or NMJ signal was detectable in somatic muscles, whereas junctional signal in wing imaginal discs was readily detectable in both lines (Fig. S4B,D).

It has been published that heterozygous baz⁴ mutant larvae show a significant decrease in immunofluorescence signal of Baz and also of Spectrin at the NMJ (Ruiz-Canada et al., 2004). Another publication showed a significant decrease in Baz and Spectrin immunostaining at the NMJ of larvae heterozygous for the baz^815-8 allele (Ramachandran et al., 2009). We did not attempt to reproduce these findings. However, in our hands mitotic clones generated with FRT chromosomes carrying these latter two baz alleles showed polarity phenotypes in the follicular epithelium, whereas clones of the clean baz^EH747 null allele did not show any polarity defect (Shahab et al., 2015), raising the possibility that the NMJ phenotypes observed by Ruiz-Canada et al. (2004) and Ramachandran et al. (2009) were caused by second site mutations on these chromosomes rather than by reduced Baz activity.

To reassess a potential localization of Baz at NMJs, we downregulated Baz by RNAi. Clonal analysis using a null allele of baz is not feasible in the muscle due to the syncytial nature of this tissue. To test whether RNAi works in body wall muscles we conducted β-spectrin-RNAi using the muscle specific (M12) driver line 5053-Gal4. We observed a strong decrease of α-spectrin staining in the postsynaptic membrane within the M12 muscle as published before (Pielage et al., 2006) (Fig. 4A-A″). We pursued the same approach to downregulate Baz in muscle M12. The functionality of baz-RNAi was demonstrated in the follicular epithelium, where expression under the control of tj::Gal4 led to strong downregulation of Baz at epithelial junctions (Fig. 2B). While it has been proposed that Baz is localized at the postsynaptic membrane within the NMJ (Ruiz-Canada et al., 2004), we did not detect any decrease in immunoreactivity for Baz or α-Spectrin upon baz-RNAi in M12 compared to other muscle segments not expressing baz-RNAi (Fig. 4B-C″; compare NMJs marked by red arrowhead and red arrow). Baz nuclear envelope localization remained unaffected as well. We also did not see any downregulation of Baz or α-spectrin upon baz-RNAi in M12 at 29°C, when the UAS-Gal4 system is maximally active (Fig. S8). Taken together, our data strongly indicate that the nuclear envelope and NMJ immunostaining pattern observed using anti-Baz antibodies does not reflect the true Baz localization.

In line with results from (Speese et al., 2012), our initial immunostaining analysis revealed strong localization of Baz to the nuclear envelope and colocalization with the nuclear envelope protein Lamin C. The fact that four different anti Baz antibodies raised against three different regions of the Baz protein showed nuclear localization provided strong evidence to believe that the observed immunostaining pattern was unlikely to be an immunostaining artifact. Further analysis revealed that localization of Baz to the nuclear envelope was detectable only in specific cell types. We found Baz nuclear envelope immunostaining mostly in polyploid cells undergoing endoreduplication such as larval fat body cells, larval body wall muscle, nurse cells and late-stage follicular epithelial cells, but also in diploid oocytes.

However, analysis of Baz immunostaining in mutant clones for the strong loss-of-function allele baz^EH747 lacking all but the first 51 amino acid residues of Baz indicated that the observed nuclear envelope localization pattern was not specific to Baz. We initially speculated whether the persistence of Baz nuclear envelope immunostaining in the baz mutant follicular epithelial cells and oocyte could be due to increased Baz protein stability and low turnover of Baz associated with the nuclear envelope. However, this is unlikely given that no change in the intensity of Baz nuclear envelope staining was observed when large late-stage follicular epithelial clones were compared to small early-stage clones. If perdurance would be the explanation for the persistent nuclear envelope staining, then the signal should become weaker as cells divide and the tissue expands. Similarly, in baz mutant nurse cells, Baz staining at the junctions was completely lost whereas no change in nuclear envelope staining intensity was observed.

We also considered the possibility that the short N-terminal peptide of 51 amino acids that may still be expressed in the baz^EH747 allele could give rise to the observed staining at the nuclear envelope of follicular epithelial cells. However, we consider this possibility very unlikely for two reasons: 1) RNAi affects the baz mRNA and thus should knock down all epitopes to the same degree. However, we see a complete loss of junctional Baz signal but no reduction of the signal at the nuclear envelope or the NMJ upon RNAi targeting baz. 2) The GFP-Baz fusion proteins do not show any signal at the nuclear envelope upon imaging of the native GFP fluorescence or upon antibody staining with an anti GFP antibody, although both the Baz-GFP BAC line and the GFP-Baz protein trap line express full-length Baz including the N-terminal epitope that is potentially still expressed in the baz^EH747 allele.

Thus, our data strongly indicate that Baz does not localize to the nuclear envelope and that nuclear envelope immunostaining observed with anti-Baz antibodies is due to cross-reactivity with an unidentified epitope. We can only speculate about the reason for the nuclear envelope signal detected with the different anti Baz antisera. All four sera were raised against GST fusion proteins, so it could be that antibodies against GST that are present in the antisera cross-react with a component of the nuclear envelope. However, as we were interested in a potential function of Baz at the nuclear envelope, we did not further investigate this after we found that signal to be unrelated to Baz.

Finally, our data also revealed that the published localization of Baz to the postsynaptic membrane of the NMJ (Ramachandran et al., 2009; Ruiz-Canada et al., 2004; Speese et al., 2012) is an artifact, as the staining is unaffected by RNAi against Baz and neither the GFP-Baz trap line nor the Baz-GFP BAC line showed any GFP signal at the NMJ. Work from the Budnik lab reported NMJ phenotypes such as reduced staining intensity for α-Spectrin and reduced number of synaptic boutons in animals heterozygous mutant for the baz alleles baz⁴ and baz^815-8 as well as upon RNAi against baz (Ramachandran et al., 2009; Ruiz-Canada et al., 2004). We did not attempt to reproduce these findings because our data do not provide evidence for localization of Baz at the NMJ. The reported effects on the NMJ may be explained by the existence of second site mutations on the chromosomes carrying the baz⁴ and baz^815-8 alleles used for the analyses. These second site mutations apparently also cause defects in epithelial apical-basal polarity that are not observed using clean null alleles of baz (Shahab et al., 2015).

Altogether, the reported findings for a function of Baz at the nuclear envelope (Speese et al., 2012) and at the NMJ (Ramachandran et al., 2009; Ruiz-Canada et al., 2004) should be regarded with great caution as we did not find any evidence for localization of Baz to these subcellular structures.

The following fly stocks were used in this study: w¹¹¹⁸ (was used as wild-type control, BL 3605), Baz-GFP BAC (Besson et al., 2015), GFP-Baz gene trap CC01941 (BL 51572) (Buszczak et al., 2007) UAS::baz-RNAi57 (VDRC v2915, on II.), UAS::baz-RNAi58 (VDRC v2914, on II.), UAS::baz-RNAi73 HMS01412 (BL 35002), UAS::β-spectrin-RNAi (Pielage et al., 2005), UAS::CD8-GFP (BL 32184), traffic jam::Gal4 (Li et al., 2003), 5053::Gal4 M12 (BL 2702), FRT19A (BL 1709), baz^EH747 FRT19A, (Shahab et al., 2015), hsFlp¹²² FRT19A H2AvD-GFP (BL 32045), FRT19A tubP::Gal80LL1 hsFLP; tubP::Gal4 UAS::mCD8-GFP (for generation of MARCM clones, gift from Heinrich Reichert). Stock numbers from the Bloomington Drosophila Stock Center (BL #) and from the Vienna Drosophila Research Center (VDRC #) are given in parentheses. Crossings for RNAi experiments were set up at 25°C if not indicated otherwise. For generating follicle cell clones in ovaries by Flipase-mediated mitotic recombination of the FRT sites flies were heat shocked for 1 h at 37°C 5-7 days prior to preparation of the ovaries. For generation of germ line clones by Flipase-mediated mitotic recombination of the FRT sites flies were heat shocked twice for 2 h at 37°C on two consecutive days in late 2nd, early 3rd instar larval stages.

One day before preparation the females were fed with fresh yeast. Ovaries were removed and ovariole tubules separated by pipetting the ovaries prior to fixation two times with a cut 1000 µl tip to enlarge tip opening. Ovaries were fixed in 3.7% formaldehyde, washed three times in PBS and blocked in PTX (0,1% Trixon X-100 in PBS) for at least 3 h.

L3 larvae were washed in PBS to remove residual food and placed on a plate with PBS. Larvae were fixed with needles and cut open between the lateral trunks of the tracheae along the a/p-centerline. Larval cuticle was nicked at the anterior and posterior end to the lateral side and the cuticle was unfolded to fix it with needles onto the plate to flatten it. Inner organs were removed, imaginal discs remained. Torsos were washed with PBS to remove residual tissue and then shortly once with 3.7% formaldehyde in PBS and then fixed for 15 min in 3.7% FA. Fixative was removed and the torso washed three times with PBS. After the needles were removed, the torsos were transferred to a tube and washed three times with PTX (0.1% Triton X-100 in PBS) for 10 min.

The following antibodies were used for immunostainings: rabbit anti Baz-N-term (aa 1-297) (Wodarz et al., 2000) 1:1000, rabbit anti Baz-N-term (aa1-297) pre-immune 1:1000, rat anti Baz-N-term (aa 1-297) (Wodarz et al., 1999) guinea pig anti Baz-PDZ (aa 309-747) (Shahab et al., 2015) 1:1000, guinea pig anti Baz-C-term (aa 905-1221) 1:1000 (this work), mouse anti α-Spectrin (3A9, DSHB) 1:10, mouse anti Dlg (4F3, DSHB) 1:20, mouse anti Armadillo (N2-7A1, DSHB) 1:20, mouse anti Lamin C (ADL 76.10, DSHB) 1:100, rat anti DE-Cadherin (DCAD2, DSHB) 1:5, goat anti HRP-Alexa647 (Jackson ImmunoResearch), mouse anti GFP (A11120 Molecular Probes) 1:1000, rabbit anti GFP (A11122 Molecular Probes) 1:1000. Secondary antibodies conjugated to Alexa-Fluor-488/555/647 (Invitrogen) were used at 1:400. DNA was stained with HOECHST 33258 (Sigma). Immunostainings for the 1st and 2nd antibodies were performed in 5% NHS in PTX. Tissues were imaged on a Zeiss LSM880 Airyscan confocal microscope using 25x LCI Plan Neofluar NA 0.8 and 63x Plan Apochromat NA 1.4 oil immersion objectives. If not stated otherwise in the figure legend, all confocal images are single optical sections taken at a pinhole setting of 1 Airy unit. Images were processed with Zen black software (Zeiss) without contrast enhancement. Figures were assembled with Inkscape 1.2 (Inkscape.org) and Powerpoint (Microsoft).

Images were analyzed for the presence or absence of a fluorescence signal at the nuclear envelope or the NMJ compared to negative or positive controls, either in the same tissue (mutant clones in the follicular epithelium, RNAi in a specific body wall muscle, junctional versus nuclear signal, anti-Baz staining versus Baz-GFP signal) or in samples processed in parallel (ovaries with follicle cell and germ line clones). Fluorescence intensities were not quantified because the results were obvious and fully penetrant. Therefore, no statistical analysis of the results was required.

We thank Uli Thomas for showing us how to prepare and stain larval muscles, Jan Pielage, Trudi Schüpbach, Dorothea Godt, Francois Schweisguth, Heinrich Reichert, the Bloomington Drosophila stock center at the University of Indiana, the Vienna Drosophila Research Center (VDRC) and the Developmental Studies Hybridoma Bank (DSHB) at the University of Iowa for sending fly stocks and reagents. We thank Mona Honemann-Capito, Ferdi Grawe and Monique Ulepic for technical assistance and members of the Wodarz lab for discussion.