The AP-2 complex has a specialized clathrin-independent role in apical endocytosis and polar growth in fungi

Abstract

Filamentous fungi provide excellent systems for investigating the role of the AP-2 complex in polar growth. Using Aspergillus nidulans, we show that AP-2 has a clathrin-independent essential role in polarity maintenance and growth. This is in line with a sequence analysis showing that the AP-2 β subunit (β2) of higher fungi lacks a clathrin-binding domain, and experiments showing that AP-2 does not co-localize with clathrin. We provide genetic and cellular evidence that AP-2 interacts with endocytic markers SlaB^End4 and SagA^End3 and the lipid flippases DnfA and DnfB in the sub-apical collar region of hyphae. The role of AP-2 in the maintenance of proper apical membrane lipid and cell wall composition is further supported by its functional interaction with BasA (sphingolipid biosynthesis) and StoA (apical sterol-rich membrane domains), and its essentiality in polar deposition of chitin. Our findings support that the AP-2 complex of dikarya has acquired, in the course of evolution, a specialized clathrin-independent function necessary for fungal polar growth.

Keywords: Aspergillus nidulans; cell biology; endocytosis; flippase; polarity maintenance; transporters.

Figure 1.. The β subunit of fungal AP complexes lacks clathrin-binding domains.

(A) Cartoon depicting the absence in the β subunits of AP complexes of higher fungi (A. nidulans, S. cerevisiae or S. pombe) of a C-terminal region that includes putative clathrin binding domains. The cartoon includes the human AP1-3 β subunits as examples of canonical AP complexes, as well as, selected primitive fungi (Rozella allomycis and Spizellomycetales punctatus) and F. alba, as examples of lower eukaryotes conserving degenerate versions of putative clathrin binding domains. One signifies the N-terminal adaptin domain (pfam01602, ~534 amino acids) common in β subunits of all AP complexes. Two is the α-adaptin C2 domain (pfam02883, ~111 amino acids) present in the β subunits of AP-2 and AP-1. Three is the β2-adaptin appendage (pfam09066, ~112 amino acid residues) present in the β subunits of AP-2 and AP-1, required for binding to clathrin. Four is the so-called clathrin adaptor protein complex β1 subunit domain, found in AP-3 complexes, probably required for Golgi association (pfam14796, ~148 amino acids). (B) Phylogenetic relationships of the C-terminal region of the β subunit of AP-2 that includes clathrin-binding domains. The tree includes selected organisms representing major taxonomic groups, from the amoeba A. castellanii to metazoa such as C. elegans and H. sapiens. The tree was reconstructed with the maximum-likelihood method and bootstrap-method testing (shown in red). The branch scale used was 0.2 and the branch lengths (shown in black) reflect the expected number of substitutions per site. DOI: http://dx.doi.org/10.7554/eLife.20083.002

Figure 1—figure supplement 1.. Phylogenetics of the four subunits of AP1, AP-2 and AP-3 of Aspergilli with three model organisms used as out-groups.

The sequences were retrieved from AspGD (RRID: SCR_002047) database using BLASTp with A.nidulans sequences as query respectively. Picture (A) illustrates the evolutionary relationship between the alpha subunits of AP-1, AP-2 and AP-3 of Aspergilli with the model organisms; S.cerevisiae, S.pombe and U.maydis. Picture (B) illustrates the relationship between the beta subunits, picture (C) between the mu subunits and picture (D) between the sigma subunits, accordingly. Red colour is used for the AP-1 complexes, light blue for the AP-2 and green for the AP-3. A. nidulans sequences are highlighted with red colour in all pictures whereas the model organisms are highlighted with blue. Moreover, the alpha subunits of AP-1 complexes are close to the alpha subunits of AP-3 complexes whereas the other three subunits (beta, mu, and sigma) of AP-1 are closer to the AP-2 complexes. In picture (C), it is important to note that the S.cerevisiae AP-3 mu subunit is found to be closer to the AP-1 and AP-2 mu subunits than the other AP-3 mu subunits. The branch length used is 0.2. All trees were reconstructed using the phylogeny.fr website on advance mode (Dereeper et al., 2008), analysed using MEGA6 software and subsequently manipulated on Figtree (RRID: SCR_008515) and Adobe Photoshop software (RRID: SCR_002078). DOI: http://dx.doi.org/10.7554/eLife.20083.003

Figure 1—figure supplement 2.. Phylogenetics of the four subunits of AP-1, AP-2 and AP-3 of major fungal groups with out-groups from model organisms.

The sequences were retrieved from JGI database BLASTp with A. nidulans sequences as query respectively. From each major fungal group individual BLASTp searches were performed and the top hits were used as representatives. Picture (A) illustrates the evolutionary relationship between the alpha subunits of AP-1, AP-2 and AP-3 of fungal groups with nine model organisms; S. cerevisiae, S. pombe, U. maydis, A. thaliana, C. elegans, D. melanogaster, D. rerio, M. musculus, and H. sapiens. Picture (B) illustrates the relationship between the beta subunits, picture (C) between the mu subunits and picture (D) between the sigma subunits, accordingly. Red colour is used for the AP-1 complexes, light blue for the AP-2 and green for the AP-3. A. nidulans sequences are highlighted with red colour and the model organisms with blue. On picture (B), the yellow highlighted area on the tree is indicating the existence of beta AP-1 subunits that are very closely related to the beta AP-2 (C. elegans, D. rerio, M. musculus, and H. sapiens) and the AP-1/2 (D. melanogaster) subunits. In picture (C), it is important to note that the AP-3 mu subunit of N. fluitans of the Pucciniomycotina group is closerrelated to the AP-1 mu subunits than the other AP-3. In addition, similarly to Figure 1—figure supplement 1, the alpha subunits of AP-1 complexes are closer to the alpha subunits of AP-3 complexes instead of the AP-2. The branch length used is 0.2. All trees were reconstructed as in Figure 1—figure supplement 1. DOI: http://dx.doi.org/10.7554/eLife.20083.004

Figure 2.. AP-1 and AP-2, but not AP-3, complexes are critical for A. nidulans growth.

(A) Colony growth (bottom left inserts) and microscopic morphology (20 hr hyphal cells stained with calcofluor) of isogenic thiA_p-ap1^σ, ap2^σΔ, ap3^σΔ, ap2^σΔ thiA_p-ap1^σand ap2^σΔ ap3^σΔ mutant strains, compared to wild-type (wt). Biological/Technical replicates: 4/25, 3/12, 3/15, 3/10, 2/10, 2/10 for wild-type and mutant strains, respectively. For the definition of the two categories of replicates see Materials and methods. (B) Representative types of morphological phenotypes related to unipolar or multipolar germination (upper panel) and relative quantitative analysis of n = 100 hyphae of wild-type and n = 84, 58, 60, 48 and 31 hyphae of thiA_p-ap1^σ, ap2^σΔ, ap3^σΔ, ap2^σΔ thiA_p-ap1^σ and ap2^σΔ ap3^σΔ, respectively (lower panel). Replicates as in (A). (C) Representative types of septa formation (upper panel) and relative quantitative analysis of n = 32 hyphae of wild-type and n = 32, 23, 32, 32 and 27 hyphae of thiA_p-ap1^σ, ap2^σΔ, ap3^σΔ, ap2^σΔ thiA_p-ap1^σ and ap2^σΔ ap3^σΔ respectively (lower panel). Replicates as in (A). p<0.001 for thiA_p-ap1^σ, ap2^σΔ, ap2^σΔ thiA_p-ap1^σ and ap2^σΔ ap3^σΔ, compared to wt. See Materials and methods for statistical analysis methods and statistical tests used. (D) Calcofluor deposition at the tip of growing hyphae in AP mutants and wild-type. A standard endocytic mutant, sagAΔ, showing reduced calcofluor staining at the tip is included for comparison. For experimental details see Materials and methods. Replicates as in (A). DOI: http://dx.doi.org/10.7554/eLife.20083.005

Figure 3.. AP complexes are dispensable for transporter membrane traffic and endocytosis.

All panels show Epifluorescence microscopy analyses of 18–20 hr growing hyphal cells in supplemented Minimal Media. (A) Subcellular localization of UapA-GFP under standard growing conditions (NO₃⁻ as sole N source) and in response to endocytic signals (standard conditions followed by 2 hr addition of either NH₄⁺ or uric acid, UA) in AP (thiA_p-ap1^σ, ap2^σΔ, ap3^σΔ) or endocytic mutants (sagAΔ and thiA_p-slaB), compared to isogenic wild-type. Biological/Technical replicates: 2/10. (B) Time course (min) of NH₄⁺-elicited endocytosis of UapA-GFP in a ap2^σΔ genetic background. An identical picture is obtained in a wild-type (ap2^σ+) background (Gournas et al., 2010). Biological/Technical replicates: 2/10. (C) Subcellular localization of six GFP-tagged transporters belonging to distinct protein families (PrnB, AgtA, FurA, FurD, FurE and FcyB; for details see text) in response to endocytic trigger (2 hr NH₄⁺). Staining with FM4-64 is included to show that in the presence of NH₄⁺ all transporters are eventually sorted for degradation in the vacuoles, similarly to what was observed with UapA. Biological/Technical replicates: 2/20, 2/20, 2/15, 2/15, 2/15 and 3/10. (D) Relative subcellular localization of UapA-GFP in response to endocytic trigger (2 hr NH₄⁺) in wild-type and ap2^μΔ genetic backgrounds. CMAC staining highlights terminal sorting in the vacuoles in the presence of NH₄⁺. Notice that UapA-GFP is normally endocytosed in ap2^μΔ, similarly to ap2^σΔ or the wild-type strain. Biological/Technical replicates: 2/12. See also Figure 3—figure supplement 2. (E–F) Quantitative analysis of transporter endocytosis presented in (C) as depicted by measurements of vacuolar surface or vacuolar GFP fluorescence. n = 5 hyphae per condition (Control or Endocytosis). For the method of measurements, statistical analysis and other experimental details see Materials and methods. Replicates as in (C). DOI: http://dx.doi.org/10.7554/eLife.20083.006

Figure 3—figure supplement 1.. Phenotypic characterization of a conditional null mutant of SlaB constructed using the thiA_p repressible promoter.

(A) Microscopic morphology of thiA_p-slaB hyphal cells grown at 25°C under repressing conditions (addition of thiamine), stained with calcofluor. (B) Colony growth of thiA_p-slaB and isogenic wild-type control strain in the presence (repressing conditions) of absence (non-repressing conditions) of thiamine, at 25 or 37°C. DOI: http://dx.doi.org/10.7554/eLife.20083.007

Figure 3—figure supplement 2.. Phenotypic characterization of conditional ap2^μΔ null mutants constructed using standard reverse genetics and genetic crossing.

(A, B) Colony growth and microscopic morphology of isogenic wild-type, ap2^σΔ, ap2^μΔ and ap2^σΔ ap2^μΔ strains at 25°C. In (B) cells are stained with calcofluor. DOI: http://dx.doi.org/10.7554/eLife.20083.008

Figure 4.. Clathrin is essential for growth and transporter secretion or/and endocytosis.

(A, B) Colony growth phenotypes and microscopic morphology (hyphal cells stained with calcofluor) of ClaL knock-out (claLΔ) or conditionally knocked-down (thiA_p-claL and thiA_p-claH) mutants. Representative types of morphological phenotypes (a-d) of thiA_p-claH are shown (B, right panel). For details see text. Biological/Technical replicates in (B): 4/25, 2/20 and 2/100, for the three strains respectively. Unless otherwise stated, thiamine was added ab initio (16 hr) at a final concentration of 10 μg ml⁻¹. (C–E) Quantitative analysis of growth phenotypes (categorized as in Figure 2 in A,B or C), tip morphology and number of septa, in thiA_p-claL (thiamine-repressed) and an isogenic wild-type control (wt). (C) Analysis of n = 100 hyphae of wild-type and n = 45, 94 hyphae of thiA_p-claL at 25°C and 37°C respectively. (D) Analysis of n = 95, 69, 95 hyphal tips of wt, thiA_p-claL at 25°C and 37°C respectively. (E) Analysis of n = 32 hyphae of wt and knock-down strains. Replicates as in (B). (F) Relative quantitative analysis of growth types shown in (B), of n = 200 hyphae of thiA_p-claH (thiamine-repressed). Replicates as in (B). (G) Epifluorescence microscopy showing the relative subcellular localization of UapA-GFP under control or endocytic conditions (2 hr NH₄⁺) in isogenic wild-type and thiA_p-claL or thiA_p-claH genetic backgrounds. Notice that repression of claL expression (o/n thiamine) blocks UapA-GFP endocytic turnover, ab initio repression of claH expression (o/n thiamine) severely blocks UapA-secretion to the PM, whereas claH repression (10 hr thiamine) after pre-secretion of UapA-GFP into the PM (14 hr) leads to an apparent block in secretion, but a fraction of UapA-GFP still remains in the PM. For more explanations see the text. Biological/Technical replicates: 4/10, 3/15 for thiA_p-claL UapA-GFP and thiA_p-claH UapA-GFP, respectively. DOI: http://dx.doi.org/10.7554/eLife.20083.009

Figure 4—figure supplement 1.. Western blot analysis of thiA_p-claH-GFP.

Comparison of protein levels of ClaH-GFP in the presence or absence of thiamine, added ab initio (o/n), for 5 or for 10 hr. Quantification of the signal intensity ratio (GFP/actin) is given below. DOI: http://dx.doi.org/10.7554/eLife.20083.010

Figure 4—figure supplement 2.. Time course of FM4-64 internalization in wild-type and mutant strains.

In all cases FM4-64 internalization becomes evident after 5–10 min leading to prominent vacuolar staining at later times (Peñalva, 2010). DOI: http://dx.doi.org/10.7554/eLife.20083.011

Figure 5.. AP-2 is essential for the polar localization of lipid flippases DnfA and DnfB.

(A) Epifluorescence microscopy showing the subcellular localization of apical markers DnfA, DnfB, SagA, SlaB, SynA and AbpA in wild-type and ap2^σΔ genetic backgrounds. DnfA and DnfB are lipid flippases, SagA and SlaB are factors involved in the formation of endocytic vesicles, AbpA is an actin-polymerization marker, and SynA is a v-SNARE marking the apical tip (for more details see the text). Notice that lack of a functional AP-2 complex leads to detectable depolarization of solely DnfA and DnfB. Representative phenotypes selected from 30–40 hyphae for wt and mutant strains. Biological replicates: 4. (B) Quantitative analysis of protein (apical marker) distribution (polarized versus non-polarized) of n = 58, 56, 51, 40, 32, 43 and 48, 50, 37, 60, 45, 61 hyphal tips of DnfA-GFP, DnfB-GFP, SagA-GFP, SlaB-GFP, mCherry-SynA, AbpA-mRFP in wild-type and ap2^σΔ genetic backgrounds, respectively. Replicates as in (A). (C) Epifluorescence microscopy of the subcellular localization of DnfA, SagA and SlaB in a thiA_p-claL genetic background. Notice that ClaL repression (o/n thiamine) does not affect DnfA, SagA or SlaB polarization. Representative phenotypes selected from 20–30 hyphae for each strain. Biological replicates: 3. (D) Epifluorescence microscopy of the subcellular localization of DnfA or DnfB in a thiA_p-claH genetic background. Notice that ab initio ClaH repression (o/n thiamine) affects DnfA or DnfB polarization (upper and lower left panels), apparently due to Golgi collapse (see text), while in samples repressed (10 hr thiamine) after a period of pre-growth (16 hr), a degree of polarization is retained (upper and lower right panels). The middle panel depicts DnfA localization in the presence of the Golgi inhibitor Brefeldin A (BFA), under ClaH repressed (o/n thiamine, 150 min BFA) or de-repressed conditions (25 min BFA). Notice the apparent block in DnfA-GFP secretion (also refer to Figure 5—figure supplement 1 and the text for more details). Biological/Technical replicates: 3/50, 2/50 for thiA_p-claH DnfA-GFP and thiA_p-claH DnfB-GFP respectively. (E) Quantitative analysis of fluorescence intensity of DnfA-GFP or DnfB-GFP in wt, ap2^σΔ or thiA_p-claL (thiamine-repressed), along 4 μm of hyphal tips. For details of fluorescence intensity measurements see Materials and methods. DOI: http://dx.doi.org/10.7554/eLife.20083.012

Figure 5—figure supplement 1.. Subcellular localization of DnfA-GFP or DnfB-GFP in wild-type, ap2^σΔ, thiA_p-claL or thiA_p-claH isogenic backgrounds.

(A) Representative images of epifluorescence microscopy showing the subcellular localization of DnfA in wild-type, ap2^σΔ, thiA_p-claL genetic backgrounds (upper panel) and DnfB in wild-type and ap2^σΔ genetic backgrounds (lower panel). Notice the apparent loss of flippase polarization in ap2^σΔ. (B) Epifluorescence microscopy of DnfA-GFP (left panel) and DnfB-GFP (right panel) in a thiA_p-claH genetic background. Notice that the secretion of flippases is lost when thiA_p-claH transcription is repressed from the beginning of growth (o/n), apparently reflecting the block in flippase secretion. Instead, in samples repressed for thiA_p-claH transcription for 10 hr, after a period of pre-growth (14–16 hr), a degree of polarization is retained. Notice also a fluorescent ring at the apex, where the Spitzenkörper is usually localized (white arrow). o/n: overnight ab initio addition of thiamine; 10 hr: addition of thiamine for 10 hr after 14 hr of pre-growth. Scale bars represent 5 μm in all cases. DOI: http://dx.doi.org/10.7554/eLife.20083.013

Figure 5—figure supplement 2.. Time course of Brefeldin A effect on DnfA-GFP subcellular localization in a thiA_p-claH mutant.

Epifluorescence microscopy following the subcellular localization of DnfA-GFP in a thiA_p-claH strain under repressing conditions (10 μM thiamine added ab initio, upper panel) compared to non-repressing conditions (absence of thiamine, lower panel), in the presence of Brefeldin A (BFA, 100 μg ml^-1). Notice that under conditions of claH repression (upper panel), addition of BFA for periods > 2 hr leads to the formation of bright cortical foci, a picture compatible with apparent Golgi collapse and progressive hyphae death. When claH is de-repressed (lower panel), BFA addition leads to rapid depolarization of DnfA-GFP within the first 15 min, apparently due to a block in secretion. Notice also the changes in hyphae morphology that are characteristic of Golgi disorganization and arrest in normal cargo secretion, visible in the period of 15–60 min of BFA addition (Pinar et al., 2013). Finally, notice that the BFA effect on Golgi functioning and secretion in the presence of ClaH (lower panel), but not when ClaH expression is blocked (upper panel), is temporary, as polarity of DnfA-GFP and normal hyphae morphology are recovered after 60 min (Pantazopoulou and Penalva, 2009). Scale bars represent 2 μm. DOI: http://dx.doi.org/10.7554/eLife.20083.014

Figure 6.. AP-2 shows polar co-localization with DnfA, DnfB, SagA and SlaB, but not with clathrin or UapA.

(A) Subcellular localization of functional Ap2^σ-mRFP or Ap2^σ-GFP in wild-type background, or of Ap2^σ-mRFP in ap2^σΔ or thiA_p-claL backgrounds. Notice that the absence of a functional μ subunit leads to non-polar and non-cortical fluorescent signal of Ap2^σ-mRFP, whereas Ap2^σ-mRFP remains apically localized in the absence of clathrin (left panel). Biological/Technical replicates: 5/10, 5/10, 3/6 and 3/12, respectively. Cartoon depicting the hyphal tip of A. nidulans (middle panel). Kymograph analysis showing the rather static localization of Ap2^σ-GFP at the hyphal tip (right panel). Biological/Technical replicates: 2/3. (B) Subcellular localization experiments related the possible co-localization of Ap2^σ-mRFP or Ap2^σ-GFP with GFP- or mRFP-tagged SagA, SlaB, DnfA, DnfB, SynA, AbpA, ClaL and UapA. Notice the apparent cortical co-localization, especially at the collar region, of AP-2 with SagA, SlaB, DnfA, DnfB, SynA and AbpA, but not with ClaL or UapA. Biological replicates: 2, Technical replicates: 5–7. (C) Quantification of co-localization by calculating Pearson's Correlation Coefficient (PCC) for n = 5 hyphae, confirming significant co-localization of AP-2 with SagA, SlaB, DnfA, DnfB, SynA and AbpA. P-values are p<0.0001 for co-localization of AP-2 with SagA, SlaB, DnfA, DnfB, SynA, AbpA, p=0.0002 and p=0.0007 for ClaL and UapA respectively (upper panel). Quantification of co-localization by calculating Pearson's Correlation Coefficient (PCC) specifically at the apical region of tips for n = 7, 10, 6 tip regions of strains co-expressing fluorescent-tagged AP-2 and DnfA, SynA, ClaL respectively, showing that AP-2 does not co-localize with SynA, ClaL or DnfA (lower panel). P-values are 0.0026, 0.0250 and 0.0001 respectively. See Materials and methods for statistical analysis methods and statistical tests used. (D) Subcellular co-localization of Ap2^σ-mRFP with SlaB-GFP or SagA-GFP in sagAΔ or thiA_p-slaB backgrounds, respectively. Notice the relative depolarization of Ap2^σ-mRFP/SlaB-GFP in sagAΔ and of Ap2^σ-mRFP/SagA-GFP in thiA_p-slaB. Representative phenotypes selected from 20 hyphae for wt and mutant strains. Biological replicates: 2, Technical replicates: 10. DOI: http://dx.doi.org/10.7554/eLife.20083.015

Figure 6—figure supplement 1.. Epifluorescence microscopy following the in parallel localization of Ap2^σ-GFP and the TGN marker mRFP-PH^OSBP.

Notice the non-colocalization of the two markers, AP-2^σ-GFP mostly labeling the collar region and some small cortical foci, while mRFP-PH^OSBP labels typical trans-Golgi cytoplasmic foci. DOI: http://dx.doi.org/10.7554/eLife.20083.016

Figure 6—figure supplement 2.. Evidence for the functionality and proper subcellular localization of GFP- or mRFP-tagged ClaL.

(A) Colony growth phenotypes of ClaL-mRFP, ClaL-GFP, thiA_p-claL (repressed) strains compared to wild-type. (B) Epifluorescence microscopy of ClaL-GFP compared to ClaL-mRFP, showing very similar localization. (C) Epifluorescence microscopy depicting the collar region of a strain expressing ClaL-GFP. The image is overexposed for better visualization. In the magnified region of the endocytic collar, a rather weak association of ClaL with subcortical regions is visible. (D) Epifluorescence microscopy showing the subcellular localization of UapA-GFP under control or endocytic conditions (2 hr NH4+) in the wild-type background or in strains co-expressing ClaL-mRFP. Notice that the expression of ClaL-mRFP does not affect UapA-GFP endocytosis. DOI: http://dx.doi.org/10.7554/eLife.20083.017

Figure 7.. TIRF Microscopy confirms the non-colocalization of clathrin and AP-2.

(A) Epifluorescence images and respective TIRF (Total Internal Reflection Fluorescence) microscopy of ClaL-GFP, confirming that a fraction of ClaL is associated with the PM. The penetration depth for TIRF was set to 110 nm. Biological/Technical replicates: 2/10. (B) Additional subcellular localization experiments investigated the possible co-localization of Ap2^σ-GFP with ClaL-mRFP. Epifluorescence microscopy (EPI) confirms the very low cortical co-localization of AP-2 with ClaL, similar to that observed when the two proteins where inversely tagged (see Figure 6C). Respective TIRF microscopy shows no co-localization of the two proteins in the plasma membrane. The penetration depth for TIRF was set to 150 nm. Biological replicates: 3, Technical replicates: 8. BF: Brightfield. (C) Quantification of co-localization by calculating Pearson's Correlation Coefficient (PCC) for n = 3 hyphae, of a strain co-expressing GFP-tagged AP-2 and ClaL-mRFP. The corresponding P-values are p<0.0001 and p<0.05 for the PCCs calculated by epifluorescence microscopy and by TIRF, respectively. See Materials and methods for statistical analysis methods and statistical tests used. DOI: http://dx.doi.org/10.7554/eLife.20083.018

Figure 8.. AP-2 interacts genetically with endocytic factors and proteins involved in apical lipid maintenance.

(A) Growth phenotypes of single and double null mutants related to AP-2 and SagA, StoA, DnfA and DnfB. (B) Microscopic morphology of hyphal cells, stained with calcofluor, of strains shown in (A). Representative phenotypes selected from 100 hyphae for wt and 20–50 hyphae for mutant strains. Biological/Technical replicates: 4/25, 3/10, 3/15, 2/20, 2/15, 2/15, 3/10, 2/20, 2/15 and 2/15, respectively. (C) Apical deposition of calcofluor in strains shown in (A) and (B). Biological/Technical replicates as in (B). (D, E) Growth phenotypes and microscopic morphology of ap2^σΔ, basA1^tsand ap2^σΔ basA1^tsstrains. Inserts highlight the modification of calcofluor deposition from the collar region to the extreme apex in basA1^ts strains under the non-permissive temperature (42°C). Representative phenotypes selected from 45 hyphae for ap2^σΔ and mutant strains. Biological/Technical replicates: 3/15. (F) Localization at the extreme apex, rather than in the collar region, of AP-2 in a basA1^ts genetic background. Notice that a similarly modified localization of calcofluor (chitin) was obtained in the basA1^ts at 42°C (see relevant inserts in Figure 7E). Representative phenotypes selected from 30 hyphae. Biological replicates: 2, Technical replicates:15. (G–H) Quantitative analysis of growth types shown in (B) and (E), categorized as in Figure 2 in A, B or C. (G) Analysis of n = 100 hyphae of wild-type and n = 32, 58, 51, 40, 92, 43, 59, 58, 100 hyphae of mutant strains. (H) Analysis of n = 25, 77, 65 and n = 75, 42, 68 hyphae of ap2^σΔ, basA1, ap2^σΔ basA1 at 25^oC and 37°C respectively. DOI: http://dx.doi.org/10.7554/eLife.20083.019

Figure 9.. AP-2 is critical for ergosterol membrane localization.

(A) Apical filipin staining of ergosterol in hyphal cells of wild-type and mutants. Notice the significant alterations and loss or reduction in apical staining in ap2^σΔ, thiA_p-basA, thiA_p-slaB and stoAΔ genetic backgrounds. Representative phenotypes selected from n = 20 hyphae for wt and mutant strains. Biological replicates: 2, Technical replicates: 20. Scale bars represent 5 μm. (B) Quantitative analysis of fluorescence intensity of filipin staining along the surface of hyphae tips. The region measured is depicted in the cartoon on the left. DOI: http://dx.doi.org/10.7554/eLife.20083.020

Figure 10.. A speculative model highlighting the role of AP-2 in DnfA and DnfB endocytosis at the apical region of A. nidulans growing hyphal cells.

After reaching the PM, DnfA and DnfB diffuse laterally to the collar region where they are recognized by AP-2 and undergo actin polymerization-dependent endocytosis with the help of SagA and SlaB. Endocytic vesicles are sorted in the Early Endosomes (EEs) and from there DnfA and DnfB undergo retrograde traffic to the TGN, the Spitzenkörper (a vesicle sorting region in filamentous ascomycetes; Pantazopoulou et al., 2014) and eventually reach the PM (Schultzhaus et al., 2015; Peñalva, 2015). Extrapolating from the observation that AP-1 loss-of-function mutants are severely defective in polarity maintenance and growth, similar to AP-2 mutants, we predict that AP-1 is involved in the retrograde exocytosis (green arrows) of DnfA, DnfB and other cargoes essential for lipid or cell wall (for example, chitin synthases) maintenance. The model does not exclude that a fraction of DnfA and DnfB and other cargoes endocytosed by the AP-2 pathway would undergo degradation after being sorted into degradative EEs that are destined to the vacuole. The model also depicts that transporters and possibly other non-polar membrane proteins are not cargoes of the AP-2 pathway, but instead undergo clathrin- and α-arrestin-dependent endocytosis, followed by sorting into degradative EEs and eventual degradation in the vacuole (Vac). The model also suggests that SlaB and SagA endocytic factors might have roles in both the AP-2 and clathrin endocytic pathways. Finally, the model shows that SynA (V-Snare) is not a cargo of the AP-2 pathway. The roles of some other factors in exocytosis (Sec4, Rab11, Rab1, Rab6 and RabE) or sorting (RabB) of cargoes, is based on the work of the group of M.A. Peñalva (Peñalva, 2015). DOI: http://dx.doi.org/10.7554/eLife.20083.021