close
Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Dec 15;130(24):4132-4143.
doi: 10.1242/jcs.212308. Epub 2017 Nov 1.

Giantin-knockout models reveal a feedback loop between Golgi function and glycosyltransferase expression

Nicola L Stevenson  1 , Dylan J M Bergen  1   2 , Roderick E H Skinner  2 , Erika Kague  2 , Elizabeth Martin-Silverstone  2 , Kate A Robson Brown  3 , Chrissy L Hammond  2 , David J Stephens  4
Affiliations

Giantin-knockout models reveal a feedback loop between Golgi function and glycosyltransferase expression

Nicola L Stevenson et al. J Cell Sci. .

Abstract

The Golgi is the cellular hub for complex glycosylation, controlling accurate processing of complex proteoglycans, receptors, ligands and glycolipids. Its structure and organisation are dependent on golgins, which tether cisternal membranes and incoming transport vesicles. Here, we show that knockout of the largest golgin, giantin, leads to substantial changes in gene expression but only limited effects on Golgi structure. Notably, 22 Golgi-resident glycosyltransferases, but not glycan-processing enzymes or the ER glycosylation machinery, are differentially expressed following giantin ablation. This includes near-complete loss of function of GALNT3 in both mammalian cell and zebrafish models. Giantin-knockout zebrafish exhibit hyperostosis and ectopic calcium deposits, recapitulating phenotypes of hyperphosphatemic familial tumoral calcinosis, a disease caused by mutations in GALNT3. These data reveal a new feature of Golgi homeostasis: the ability to regulate glycosyltransferase expression to generate a functional proteoglycome.

Keywords: GALNT3; Giantin; Glycosylation; Golgi; Hyperphosphatemic tumoral calcinosis; Zebrafish.

PubMed Disclaimer

Conflict of interest statement

Competing interestsThe authors declare no competing or financial interests.

Figures

Fig. 1.
Fig. 1.
Generation of a giantin-KO cell line. (A) Genomic sequence for CRISPR/Cas9 target site in WT and engineered KO RPE-1 cell line. Purple lines and scissors depict gRNA binding and cut sites. Blue nucleotides show the CRISPR PAM site. Green and red nucleotides are those deleted and inserted in the KO mutation, respectively. Amino acid translation shown underneath; asterisk indicates a premature stop codon. (B,C) Western blot analysis (B) and immunofluorescence staining (C) of giantin using three different antibodies raised against the C-terminus (C-term), N-terminus (N-term) and full-length (FL) protein. All immunoreactivity is lost in the KO cells. (D) WT and KO mixed population stained for giantin and other Golgi markers for direct comparison. Images are maximum projections. Scale bars: 10 µm.
Fig. 2.
Fig. 2.
Loss of giantin has no effect on Golgi structure or trafficking. (A) Representative images of WT and KO cells immunolabelled for two cis-Golgi markers. (B,C) The number of GM130-positive elements per cell (B) and their area (C) was found to be equivalent in WT and KO cells (n=3; 387 WT and 320 KO cells quantified; orange bars indicate median and interquartile range; statistics Mann–Whitney; fragments smaller than 0.5 μm2 excluded). (D) Co-labelling of cells with cis-Golgi (GM130) and trans-Golgi (TGN46) markers shows Golgi polarity is maintained in KO cells. (E) Representative images of WT and KO cells immunolabelled for early secretory pathway markers as indicated. In A,D,E, images are maximum projections. Scale bars: 10 μm. (F–H) Transmission electron micrographs of Golgi elements in WT and KO cells. The number of cisternae per stack (G) and length of Golgi cisternae (H) was quantified from experiments represented in F (n=3; total 30 cells per cell line; orange bars indicate median and interquartile range; statistics Mann–Whitney). (I–L) WT and KO cells expressing Str-Kdel/ManII-SBP-EGFP were treated with biotin and imaged live (I,J) or fixed at 0, 10, and 20 min post-biotin addition and immunolabelled for GM130 (K,L). (I) Single-plane images taken from representative movies at 5 min intervals. See Movies 1–6. Scale bar: 10 μm. Arrows show arrival of reporter at Golgi. (J) Quantification of the time at which fluorescence appears in the Golgi in movies represented in I (n=3; 15 WT cells and 23 KO cells quantified; bars show median and interquartile range; statistics Mann–Whitney). (K) Quantification of the number of GFP-positive Golgi at each time point in fixed cells (n=3; 378 WT and 310 KO cells quantified; mean and s.d. shown). (L) Representative single-plane images of fixed cells at each time point. Scale bars: 10 μm. (M) Western blot analyses of ER stress markers in lysates taken from WT and KO cells following treatment with Brefeldin A for the indicated time.
Fig. 3.
Fig. 3.
Loss of giantin leads to mild changes in Golgi mini-stack structure. (A) Representative maximum projection images of WT and giantin-KO cells incubated with 5 µm nocodazole as indicated and immunolabelled for cis-Golgi (GM130) and trans-Golgi (TGN46) markers or tubulin. In wash-out panels, cells were incubated with nocodazole for 3 h then washed and incubated in growth medium for time indicated. Scale bars: 10 μm. (B) Western blot analysis of golgin expression in WT and KO cells. (C) Quantification of blots represented in B (n=3, mean and s.d. shown). (D) Transmission electron micrographs of WT and KO cells incubated with 5 μM nocodazole for 90 min. Inserts show zoom of region denoted by black squares. (E–G) Quantification of experiments represented in D showing (E) cisternal length, (F) number of cisternae per stack and (G) the angle between lines drawn from each lateral rim of the stack to the centre (n=3; 27 WT and 21 KO cells quantified; E and G show median and interquartile range; F, mean and s.d.; statistics Mann–Whitney).
Fig. 4.
Fig. 4.
GALNT3 expression is lost in giantin-KO cells. (A) Western blot validating down-regulation of GALNT3 in KO cells (representative of n=5 biological replicates). (B) Maximum projection images of mixed populations of WT and KO cells immunolabelled for giantin, GM130 and GALNT3. Arrows highlight giantin-KO cells. (C) Representative projections of WT and KO cells expressing FLAG-tagged GALNT3 fixed and stained as indicated. Scale bars: 10 µm.
Fig. 5.
Fig. 5.
Giantin-KO zebrafish have reduced galnt3 expression and exhibit HFTC-like phenotypes. (A) Real-time qPCR pair-wise analysis of galnt3 expression at 60–63 dpf in two golgb1 mutant zebrafish lines normalised to gapdh mRNA levels as housekeeping gene. Bars show mean expression for each mutant line (n=3 per genotype group) relative to WT siblings (WT expression 1 AU depicted by dashed line). Each circle represents one individual (*P≤0.05, mean with s.d.; unpaired t-test was used as data were normally distributed). (B) Lateral views of micro-CT scans of 10-month-old WT and golgb1X3078/X3078 homozygous mutants, presented as isosurface renders. Boxed regions show enlarged regions in C. (C) Enlarged regions of the spine; white arrows demarcate intervertebral discs (IVDs), which in WT are not mineralised but in the mutant, ectopic mineralisation is seen manifesting as vertebral fusions. (D) Ventral (with high-resolution inset) and (E) lateral view micro-CT images showing craniofacial and spinal elements of a representative WT sibling (Q2948X line, n=3 females). (F–I) Three golgb1Q2948X/Q2948X female individuals showing ectopic calcium deposits in soft tissues (F, ventral view with high-resolution inset; G, lateral view of individual 1; H, ventral view of individual 2) and in spinal column (I, digital axial z-slices of individual 3). In D–I, red arrows indicate mandible joint and green arrows, ectopic deposits. Line Q2948X were imaged at 8 months post fertilisation. Scale bars: 100 µm.
See this image and copyright information in PMC

References

    1. Alvarez C., Garcia-Mata R., Hauri H.-P. and Sztul E. (2001). The p115-interactive proteins GM130 and giantin participate in endoplasmic reticulum-Golgi traffic. J. Biol. Chem. 276, 2693-2700. 10.1074/jbc.M007957200 - DOI - PubMed
    1. Asante D., Maccarthy-Morrogh L., Townley A. K., Weiss M. A., Katayama K., Palmer K. J., Suzuki H., Westlake C. J. and Stephens D. J. (2013). A role for the Golgi matrix protein giantin in ciliogenesis through control of the localization of dynein-2. J. Cell Sci. 126, 5189-5197. 10.1242/jcs.131664 - DOI - PMC - PubMed
    1. Bard F. and Chia J. (2016). Cracking the glycome encoder: signaling, trafficking, and glycosylation. Trends Cell Biol. 26, 379-388. 10.1016/j.tcb.2015.12.004 - DOI - PubMed
    1. Beecham G. W., Hamilton K., Naj A. C., Martin E. R., Huentelman M., Myers A. J., Corneveaux J. J., Hardy J., Vonsattel J.-P., Younkin S. G. et al. (2014). Genome-wide association meta-analysis of neuropathologic features of Alzheimer's disease and related dementias. PLoS Genet. 10, e1004606 10.1371/journal.pgen.1004606 - DOI - PMC - PubMed
    1. Bennett E. P., Mandel U., Clausen H., Gerken T. A., Fritz T. A. and Tabak L. A. (2012). Control of mucin-type O-glycosylation: a classification of the polypeptide GalNAc-transferase gene family. Glycobiology 22, 736-756. 10.1093/glycob/cwr182 - DOI - PMC - PubMed

Substances