Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2016 Sep 29;7:12788.
doi: 10.1038/ncomms12788.

Enrichment of Hydroxylated C24- And C26-acyl-chain Sphingolipids Mediates PIN2 Apical Sorting at trans-Golgi Network Subdomains

Affiliations
Free PMC article

Enrichment of Hydroxylated C24- And C26-acyl-chain Sphingolipids Mediates PIN2 Apical Sorting at trans-Golgi Network Subdomains

Valérie Wattelet-Boyer et al. Nat Commun. .
Free PMC article

Abstract

The post-Golgi compartment trans-Golgi Network (TGN) is a central hub divided into multiple subdomains hosting distinct trafficking pathways, including polar delivery to apical membrane. Lipids such as sphingolipids and sterols have been implicated in polar trafficking from the TGN but the underlying mechanisms linking lipid composition to functional polar sorting at TGN subdomains remain unknown. Here we demonstrate that sphingolipids with α-hydroxylated acyl-chains of at least 24 carbon atoms are enriched in secretory vesicle subdomains of the TGN and are critical for de novo polar secretory sorting of the auxin carrier PIN2 to apical membrane of Arabidopsis root epithelial cells. We show that sphingolipid acyl-chain length influences the morphology and interconnections of TGN-associated secretory vesicles. Our results uncover that the sphingolipids acyl-chain length links lipid composition of TGN subdomains with polar secretory trafficking of PIN2 to apical membrane of polarized epithelial cells.

Figures

Figure 1
Figure 1. TGN subdomain labelled by SYP61 is enriched in hVLCFAs and sterols as compared with Golgi or TGN subdomain RAB-A2a enriched for clathrin.
(ag) Immunolocalization of CHC (b,e) in Arabidopsis root epithelial cells expressing either the TGN marker RAB-A2a–YFP (a) or the TGN marker SYP61–CFP (d). Strong co-localization between RAB-A2a and CHC is detected in merged images (c), whereas weaker co-localization is visualized between SYP61 and CHC (f,g). Statistical analysis show highly significant difference between RAB-A2a/CHC co-localization and SYP61/CHC co-localization (n=50 cells distributed over 10 roots for each experiment, 3 biological replicates). (h) Immunopurifications of SYP61–CFP-, MEMB12–YFP- and RAB-A2a–YFP-labelled compartments were performed by incubating a step-gradient-purified TM fraction with beads coated with anti-CFP/YFP antibodies. (i) Western blottings on IP SYP61–CFP-, RAB-A2a–YFP- and MEM12–YFP-labelled intact vesicles. IP, beads-IP fraction; TM, input, step-gradient-purified TM fraction. As compared with the input (TM), anti-CFP/YFP antibodies revealed that all protein markers (SYP61–CFP, MEMB12–YFP or RAB-A2a–YFP) are enriched in their targeted IP compartments. Sec21p and MEMB11 markers of the Golgi apparatus are enriched in IP MEMB12–YFP-labelled Golgi but not in IP SYP61–CFP-labelled TGN or RAB-A2a–YFP-labelled TGN. The ECHIDNA and SYP61 markers of TGN-associated secretory vesicles are enriched in IP SYP61–CFP-labelled TGN but not in MEMB12–YFP-labelled Golgi or RAB-A2a–YFP-labelled TGN. V-ATPase VHA-E, which traffic through the TGN, is not enriched in SYP61-immunopurified or MEMB12-immunopurified fraction but is slightly enriched in IP RAB-A2a–YFP-labelled TGN. The PMA2 and PM-ATPase markers for PM are not enriched in any IP compartments. (j,k) Acyl-chain composition of the total pool of FAs contained in IP fractions. (j) As compared with MEMB12–YFP-labelled Golgi, SYP61–CFP-labelled TGN shows a significant enrichment (about 2-fold, n=11 IPs for each compartment, 11 biological replicates for each compartment) in hVLCFAs h22:0, h24:1, h24:0, h26:0 and sterols. (k) As compared with MEMB12–YFP-labelled Golgi, RAB-A2a–YFP-labelled TGN does not display any enrichment in hVLCFAs or sterols (n=11 IPs for each compartment, 11 biological replicates for each compartment). Statistics were done by two-sided Wilcoxon's rank-sum test, **P-value<0.01, ***P-value<0.001. All scale bars, 5 μm.
Figure 2
Figure 2. hVLCFAs are almost specific of SLs and 50 nM metazachlor alters the composition of VLCFAs in the pool of SLs.
In untreated Arabidopsis roots, the pool of FAs of glycerophospholipids (a) contains very few hFAs (d). Metazachlor significantly reduces the 24:0-containing FAs but does not alter the global composition of FAs of glycerophospholipids, which are mainly composed of C16- and C18-containing FAs. Arrowed brackets display the sum of<24 carbon atom-containing FAs and>24 carbon atom-containing FAs. Each histogram is further divided in hFAs (α-h) and non-h FAs (non-α-h). The ratio between<24- and>24-FAs is not altered in the glycerophospholipids pool. (b,c) Contrastingly to glycerophospholipids, in untreated roots, FAs of GlcCer (b) and GIPCs (c) contain a substantial amount of hFAs (d). Interestingly, metazachlor strongly reduces α-hydroxylated h24:1, h24:0 and h26:0 and increases α-hydroxylated h16:0, h20:0 and h22:0 FAs of GlcCer and GIPCs pools. Arrowed brackets display the sum of<24 carbon atom-containing FAs and>24 carbon atom-containing FAs. Each histogram is further divided in hFA α-h and non-hFA non-α-h. The ratio between<24- and>24-FAs is drastically inverted in both GlcCer and GIPCs pool. (e) Metazachlor neither alters the global quantity of FAs of the glycerophospholipids pool nor the global quantity of FAs of the GlcCer and GIPCs pools. Moreover, global quantities of individual sterols are not affected by metazachlor. Statistics were done by two-sided Wilcoxon's rank- sum test, *P-value<0.05, n=4 for each experiment, 4 biological replicates. Errors bars are s.d.
Figure 3
Figure 3. Global FAs analysis reveals that metazachlor reduces all types of >24-FAs.
(a) Global FAs of Arabidopsis roots in untreated (black) and metazachlor-treated (red) conditions. Compounds of the root suberin, dicarboxylic acid DCA and fatty alcohols are not altered by a 50 nm metazachlor treatment. However, ω-hVLCFAs, which are components are the root suberin, are significantly reduced on metazachlor. Non-hydroxylated C22 and C24-FAs, which are more abundant in GIPCs than in glycerophospholipids, are significantly reduced on metazachlor. hVLCFAs h22:0, h24:0, h24:1, h25:0, h26:0 and h26:1, almost exclusively present in GIPCs, are strongly reduced on metazachlor. Accumulation of h16:0 and h20:0 is observed on metazachlor. (b) Contrastingly to VLCFAs, metazachlor does not alter the quantity of C16- and C18-containing FAs (n=4 for each experiment, 4 biological replicates). (c,d) Sums of <24 carbon atom-containing FAs (c) and>24 carbon atom-containing FAs (d). Each histogram is further divided in hFAs (α-h) and non-hFAs (non-α-h). On metazachlor, the ratio between <24- and >24-FAs is drastically inverted. Statistics were done by two-sided Wilcoxon's rank-sum test, *P-value<0.05, n=4 for each experiment, 4 biological replicates. Error bars are s.d.
Figure 4
Figure 4. Metazachlor alters root gravitropism and targets KCS9 KCS2 and KCS20.
(a) Root angle curvature towards the new gravity vector 24 h following a gravistimulation (turn the plate of 90°) is calculated, we then ranked the effective (n) into classes of 15° angles (0° was the exact direction of the new gravity vector) and represented each class of angles in a circular chart. (b) In untreated roots, reorientation of roots 24 h after a gravistimulation is very close to the gravity vector, whereas in 50 nM metazachlor-treated roots (c) this reorientation is much less efficient (n=110 roots per experiment). (di) As compared with untreated roots (d), the kcs2,20 double mutant (f) and kcs9 single mutant (h) do not display gravitropism phenotype. On 25 nM metazachlor treatment, kcs2,20 double mutant (g) and kcs9 single mutant (i) display obvious root gravitropism phenotype, whereas wild-type roots treated with 25 nM of metazachlor (e) do not react. Hence, kcs2,20 double mutant (f,g) and kcs9 single mutant (h,i) are hypersensitive to metazachlor in respect to root gravitropism (n=55 roots per experiment).
Figure 5
Figure 5. Metazachlor alters auxin distribution during root bending partly through the auxin efflux carrier PIN2.
(a,b) Dynamic auxin redistribution 60 min after a gravistimulation, visualized by DII-venus fusion, shows differential auxin distribution in untreated roots (a) with a higher concentration of auxin at the upper side of the root (a), whereas metazachlor-treated roots do not display this differential repartition of auxin after gravistimulation. (b,c) Quantification of signal intensities between the upper and lower side of gravistimulated roots clearly results in a highly significant difference (n=10 seedlings for each experiment over 3 biological replicates). (dg) The auxin efflux carrier mutant pin2-eir1 (d,e) and the auxin influx carrier mutant aux1-21 (f,g) show similar phenotype in untreated roots (d,f) (n=110 root per genotype over 3 biological replicates). In metazachlor-treated roots, the pin2-eir1 mutant (e) appears more resistant to metazachlor than the aux1-21 mutant (g) (n=110 roots per genotype over 3 biological replicates). Statistics were done by two-sided Welch's two sample t-test, ***P-value<0.001. All scale bars, 10 μm.
Figure 6
Figure 6. Reduction of VLCFAs alters apical polarity and secretory trafficking of PIN2 but not endocytosis and PM recycling of PIN2.
(ac) Compared with untreated cells (a), 50 nM metazachlor-treated cells (b) display intracellular accumulation of PIN2–GFP in endomembrane compartments. (c) Quantifications of fluorescence intensity ratios between the intracellular content and whole PM show a significant intracellular accumulation of PIN2–GFP in metazachlor-treated cells that is prevented by a pretreatment with 50 μM CHX from already 90 min pretreatment. (d) Quantifications of fluorescence intensity ratios between the apical-basal membranes and lateral membranes clearly indicate a significant loss of PIN2 polarity. (ei) Compared with wild-type roots treated with 25 nM metazachlor (e) kcs2,20 double mutant (f) and kcs9 single mutant (g) display intracellular accumulation of anti-PIN2 (Alexa647) in endomembrane compartments and loss of PM polarity of PIN2. (h) Quantifications of fluorescence intensity ratios between the intracellular content and whole PM. (i) Quantifications of fluorescence intensity ratios between the apical-basal membranes and lateral membranes. (jl) 50 μM BFA treatment, after a 50 μM CHX pretreatment, show no significant differences (l) in PIN2–GFP accumulation from the PM to the so-called intracellular ‘BFA bodies' between untreated cells (j,l) and metazachlor-treated cells (k,l), revealing that PIN2 endocytosis is not altered by metazachlor. (mo) Washout of BFA in presence of CHX after a 50 μM CHX pretreatment and 50 μM BFA treatment show no significant differences (o) in PIN2 redistribution at PM from ‘BFA bodies' between untreated cells (m,o) and metazachlor-treated cells (n,o). Statistics were done by two-sided Wilcoxon's rank-sum test, XP-value>0.05, *P-value<0.05, ***P-value<0.001, n=200 cells distributed over 20 roots for each experiment (3 biological replicates). All scale bars, 5 μm.
Figure 7
Figure 7. Metazachlor accumulates PIN2 at SVs sites of TGN.
(ao) Co-localization of endomembrane compartments labelled by PIN2–GFP (a,d,g,j,m) and either SYP61–CFP-labelled TGN-associated SVs (b), RAB-A5d-mCherry-labelled TGN-associated SVs (e), CLC-mOrange-labelled TGN-associated CCVs (h), MEMB12-mCherry-labelled Golgi apparatus (k) or SYP32-mCherry-labelled Golgi apparatus (n) on 50 nM metazachlor. (c,f,i,l,o) Merged pictures of corresponding pictures. (g) Quantification of co-localization events show a strong match of PIN2 with TGN-associated-SYP61/RAB-A5d-SVs, whereas weak co-localization is detected with MEMB12/SYP32-Golgi. (p) Co-localization values of SYP61/RAB-A5d with PIN2 and MEMB12/SYP32 with PIN2 are highly different. PIN2 co-localizes at medial level with TGN-associated-CLC-Clathrin vesicles but is significantly different from co-localization of PIN2 with SYP61/RAB-A5d. Statistics were done by two-sided Kruskal–Wallis rank sum test, XP-value>0.05, ***P-value<0.001, n=40 cells distributed over 10 roots for each experiment (3 biological replicates). All scale bars, 5 μm. Errors bars are s.e.m.
Figure 8
Figure 8. Metazachlor alters TGN-associated SVs morphology and tubular interconnections at TGN.
Transmission electron microscopy (TEM) of TGN membrane structures in Arabidopsis root treated (eh) or not (ad) with metazachlor. (a) Untreated cell showing Golgi apparatus (GA) and the SVs visible as a tubulo-vesiculated membrane network (SVs/TGN) at the trans-side of the Golgi. (e) Metazachlor-treated cell (50 nM) showing Golgi apparatus and swollen TGN-associated SVs. (b,f) Quantification show that the average diameter of SVs per TGN is around 80 nm in untreated cells (b), while being around 160 nm in metazachlor-treated cells (n=22 TGN for each for each experiment over 3 biological replicates, statistics were done by two-sided Welch's two-sample t-test, ***P-value<0.001). (f,c,d) Magnification from a displaying tubular interconnections (black arrows) between SVs at TGN in untreated cells. (g,h) Magnification from e displaying larger SVs without tubular interconnections detected between them. Scale bars, 100 nm (a,e) and 50 nm (c,d,g,h).

Similar articles

See all similar articles

Cited by 11 articles

See all "Cited by" articles

References

    1. van Meer G., Stelzer E. H., Wijnaendts-van-Resandt R. W. & Simons K. Sorting of sphingolipids in epithelial (Madin-Darby canine kidney) cells. J. Cell Biol. 105, 1623–1635 (1987). - PMC - PubMed
    1. Simons K. & van Meer G. Lipid sorting in epithelial cells. Biochemistry 27, 6197–6202 (1988). - PubMed
    1. Gleeson P. A., Lock J. G., Luke M. R. & Stow J. L. Domains of the TGN: coats, tethers and G proteins. Traffic 5, 315–326 (2004). - PubMed
    1. Rodriguez-Boulan E., Paskiet K. T., Salas P. J. & Bard E. Intracellular transport of influenza virus hemagglutinin to the apical surface of Madin-Darby canine kidney cells. J. Cell Biol. 98, 308–319 (1984). - PMC - PubMed
    1. Kleine-Vehn J. et al. . Recycling, clustering, and endocytosis jointly maintain PIN auxin carrier polarity at the plasma membrane. Mol. Syst. Biol. 7, 540 (2011). - PMC - PubMed

LinkOut - more resources

Feedback