Complex cascades of interactions between different molecules regulate every process in the body. Enzymes are critical for this, because they act as ‘catalysts’ to speed up chemical reactions in a cell. Each type of enzyme has a specific role. The enzyme ppGalNAc-T3, for example, attaches sugar molecules onto certain proteins via a process called glycosylation. This modification fine-tunes the activity of the proteins.

The enzyme ppGalNAc-T3 is implicated in at least two medically important pathways. It increases the amount of the hormone FGF23, which regulates phosphate levels in the bloodstream. Hormones are messenger molecules that regulate most processes that are crucial for life, and too much or too little of a hormone can lead to diseases. High levels of FGF23, for example, can cause serious and often fatal problems in patients with chronic kidney disease.

The enzyme ppGalNAc-T3 is also known to encourage and stimulate some cancer cells to spread from their original location to other parts of the body – a process known as metastasis – which ultimately leads to the vast majority of cancer deaths every year. Thus, a drug or molecule that blocks ppGalNAc-T3 could be used to lower FGF23 levels in patients with kidney disease and potentially prevent cancer cells from spreading. However, until now, it was unknown if such a molecule existed.

To identify a compound that specifically regulates ppGalNAc-T3, Song and Linstedt engineered human cells grown in the laboratory to become fluorescent when ppGalNAc-T3 was blocked. Out of the over 20,000 compounds screened, one compound named T3Inh-1 selectively blocked ppGalNAc-T3. Further experiments showed that the T3Inh-1 compound reduced FGF23 hormone levels in both tissue cells grown in the laboratory and mice, without causing any toxic side effects. It also prevented breast cancer cells grown in the laboratory from spreading. The results demonstrated that T3Inh-1 is the first drug-like inhibitor that can target this kind of enzyme.

An important next step will be to test the compound in animal models for chronic kidney disease and cancer metastasis.

Glycosylation, which fine-tunes the function of proteins, is the most abundant and diverse posttranslational modification (Schjoldager and Clausen, 2012; van der Post et al., 2013; Moremen et al., 2012). Despite many documented roles in health and disease, the enzymes that mediate mucin-type O-glycosylation in the Golgi apparatus have yet to be discovered as druggable targets. The initiating enzymes, a family of 20 ppGalNAc-transferase isozymes, determine which substrates are modified and at which sites. Significant questions remain regarding their specificity, regulation, targets and functions and the lack of a pharmacological approach has been a critical limitation. The only confirmed inhibitor of mucin-type O-glycosylation, benzyl-N-acetyl-α-galactosaminide, blocks elongation rather than initiation and requires millimolar concentrations that can be toxic (Kuan et al., 1989; Patsos et al., 2009). Not only are pan-specific modulators of the ppGalNAc-transferases lacking, there is nothing isoform-specific. The latter will be essential to restrict effects to particular pathways and could possibly lead to a new basis for therapeutics conceptually related to the widespread use of drugs targeting individual protein kinases.

ppGalNAc-T3 serves as an important test case as it is implicated in at least two medically important pathways: cancer metastasis and stabilization of FGF23 (Chefetz and Sprecher, 2009; Schjoldager et al., 2011; Kato et al., 2006; Peng et al., 2012; Kohsaki et al., 2000; Kitada et al., 2013; Gao et al., 2013; Brockhausen, 2006; Brooks et al., 2007). ppGalNAc-T3 is overexpressed in cancerous tissue often correlating with shorter survival (Kitada et al., 2013; Brockhausen, 2006; Harada et al., 2016; Mochizuki et al., 2013). Knockdown of ppGalNAc-T3 expression in cultured ovarian cancer cells inhibits their invasive capacities arguing that ppGalNAc-T3 has potential as a therapeutic target (Wang et al., 2014). ppGalNAc-T3 mediates glycan-masking of FGF23 in bone as part of a control mechanism determining the form of FGF23 that is secreted (Kato et al., 2006; Tagliabracci et al., 2014). When present, the added O-glycan blocks FGF23 cleavage by the furin protease resulting in secretion of intact FGF23 that activates FGF23 receptor complexes at the kidney and intestine (Rowe, 2015). In contrast, non-glycosylated FGF23 is cleaved and the cleaved C-terminal product competitively blocks these same receptors (Goetz et al., 2010). Significantly, elevated intact FGF23 occurs in chronic kidney disease and upon kidney transplant where it is directly linked to poor prognosis due to its effects on renal phosphate reabsorption and 1,25-dihydroxyvitamin D biosynthesis (Eckardt and Kasiske, 2009). Many studies have concluded that therapeutic control over FGF23 would be transformative in the clinic (Degirolamo et al., 2016; Fukumoto, 2016; Isakova et al., 2015; Smith, 2014).

As a step toward testing whether biologic discoveries and new therapeutics will result from developing small molecule modulators that control the activity of specific ppGalNAc-transferases we initiated high throughput cell-based screening of compound libraries. An inhibitor of ppGalNAc-T3 was identified and found to block breast cancer cell invasiveness as well as secretion of intact FGF23 promising new therapeutic approaches for cancer and chronic kidney disease, respectively.

HEK cell lines were engineered to express fluorescent sensors that are specific to ppGalNAc-T2 or ppGalNAc-T3 activity (Song et al., 2014). For each sensor, glycosylation of its isozyme-specific target site prevents furin protease from removing a blocking domain (Figure 1A). Thus, fluorescence increases upon ppGalNAc-transferase inhibition because removal of the blocking domain allows dimerization of a fluorogen activating protein domain so that it binds and activates the fluorescence of malachite green. They are ratiometric because the sensor backbone contains a green fluorescent protein as an internal control for expression. Each sensor showed clear activation after mutation of its glycan acceptor sites and these mutated constructs served as positive controls in the screen (Figure 1B, ∆glycan). The T3 sensor exhibited a background level of activation due to incomplete glycosylation (Song et al., 2014) but this was considered advantageous for the possible identification of enzyme activators along with the desired inhibitors. Consistent with our previous report (Song et al., 2014), sensor expression in HEK cell lines depleted of either ppGalNAc-T2 or T3 via zinc finger nuclease editing resulted in specific activation of the corresponding sensor confirming their isozyme selectivity (Figure 1B, HEK∆T2, HEK∆T3). Our screen included compounds based on structural diversity (21,710 compounds in total) with 6 hr treatments at 10 µM prior to flow cytometry to assay MG and GFP fluorescence on a cell-by-cell basis (Figure 1C). Each compound was tested in duplicate and against both sensors (Figure 1—source data 1). Because each sensor requires essentially identical cellular reactions- the only difference being which ppGalNAc-transferase isoform modifies the sensor- most off-target hits (such as sugar nucleotide transporters, extending enzymes, or furin) will alter both sensors, whereas directly acting, isoform-specific candidates will be sensor-specific. Using cut-off parameters for the MG/GFP ratios that excluded >99% of the compounds (Q ≥ 3 or Q ≤ −2.5), the screen yielded 72 sensor-specific hits with 18 increasing and 35 decreasing the T2 sensor fluorescence and 11 increasing and 8 decreasing the T3 sensor fluorescence (Figure 1D).

Figure 1 with 3 supplements see all

Download asset Open asset

Figure 1—source data 1

Primary screen data for HEK cells expressing T2 or T3 sensors.

The accompanying spreadsheet shows calculated Q values (see Methods) for each compound tested. Note that autofluorescent compounds are left blank.

https://doi.org/10.7554/eLife.24051.004

Download elife-24051-fig1-data1-v3.xlsx

Figure 1—source data 2

Secondary screen data (in vitro enzyme assays).

https://doi.org/10.7554/eLife.24051.005

Download elife-24051-fig1-data2-v3.xlsx

To determine which of these directly acted on the targeted enzymes, we carried out in vitro glycosylation assays in which the purified lumenal domains of ppGalNAc-T2 or ppGalNAc-T3 (containing catalytic and lectin domains) were incubated with peptide and UDP-GalNAc substrates in the presence of 50 µM of each compound. A second stage reaction (UDP-Glo) then converted the accumulated UDP product to ATP and then, via luciferase, to light. This resulted in 20 candidates that either reduced or increased the luminescence by a factor ≥50% relative to vehicle-only controls (Figure 1E). Of these, one compound (#1614) stood out as a strong and selective inhibitor of ppGalNAc-T3 and became the focus of this phase of the study. The compound is a quinoline of no known activity that we now refer to as ppGalNAc-T3 Inhibitor 1 or T3Inh-1 (Figure 1D, inset).

Importantly, T3Inh-1 exhibited no toxicity and did not affect HEK cell proliferation (Figure 1—figure supplement 1). Also, a 24 hr treatment did not affect staining intensity by the lectins Concanavalin A (ConA), Wheat germ agglutinin (WGA), Sambucus Nigra (SNA) or Vicia Villosa (VVA), which bind branched alpha-mannose, N-acetylglucosamine, sialic acid, or terminal GalNAc, respectively (Figure 1—figure supplement 2). This implies that the enzymes contributing to the abundant glycans of N- and O-glycosylation detected by these lectins were unaffected and is consistent with ppGalNAc-T3 modifying a relatively limited number of substrates (Schjoldager et al., 2015). Finally, there was no change in localization or expression level of ppGalNAc-T3, ppGalNAc-T2 or any other Golgi marker tested (Figure 1—figure supplement 3).

To determine the effective concentration of T3Inh-1 in cells and in vitro, it was retested at various doses against the sensors and against the purified enzymes. T3Inh-1 activated the T3 sensor with an apparent IC₅₀ of 12 µM and showed little or no activity towards the T2 sensor (Figure 2A). Similarly, T3Inh-1 was a potent and selective direct inhibitor of ppGalNAc-T3 (Figure 2B). Inhibition of ppGalNAc-T3 occurred with an IC₅₀ of 7 µM and was undetectable against ppGalNAc-T2. T3Inh-1 also lacked activity against ppGalNAc-T6 (Figure 2B), which is the isozyme considered most closely related to ppGalNAc-T3 (Bennett et al., 2012 ; Revoredo et al., 2016).

Figure 2

Download asset Open asset

Figure 2—source data 1

Panel A: sensor signals versus T3Inh-1 concentration.

https://doi.org/10.7554/eLife.24051.012

Download elife-24051-fig2-data1-v3.xlsx

Figure 2—source data 2

Panel B: enzyme activity versus T3Inh-1 concentration.

https://doi.org/10.7554/eLife.24051.013

Download elife-24051-fig2-data2-v3.xlsx

Figure 2—source data 3

Panel C: inhibitor effect versus peptide concentration.

https://doi.org/10.7554/eLife.24051.014

Download elife-24051-fig2-data3-v3.xlsx

Figure 2—source data 4

Panel D: inhibitor effect versus UDP-GalNAc concentration.

https://doi.org/10.7554/eLife.24051.015

Download elife-24051-fig2-data4-v3.xlsx

Figure 2—source data 5

Panel E and F: Fluorescence change versus T3Inh-1 concentration.

https://doi.org/10.7554/eLife.24051.016

Download elife-24051-fig2-data5-v3.xlsx

Towards characterizing the mechanism of inhibition we used the in vitro assay with purified ppGalNAc-T3 and individually varied both peptide and UDP-GalNAc substrate concentrations in the presence of 0, 7.5 or 15 µM T3Inh-1. The results were similar for both substrates (Figure 2C–D), where T3Inh-1 decreased the Vmax and increased the Km (Table 1) indicating a mixed-mode of inhibition in which the inhibitor most likely binds both free enzyme to reduce substrate binding and enzyme-substrate complexes to reduce turnover. Implied in this model of action is direct binding to the enzyme typically at an allosteric site. To test for direct binding, intrinsic tryptophan fluorescence of ppGalNAc-T3 was determined in the presence of increasing T3Inh-1 concentrations. At all concentrations, the compound itself yielded miniscule signals, whereas the compound had a profound and dose-dependent effect on the ppGalNAc-T3 emission spectrum (Figure 2E). These results confirmed direct binding with an apparent Kd of 17 µM (Figure 1F). The similarity in concentration dependence of sensor activation in cells, in vitro inhibition and direct binding argues that T3Inh-1 acts directly on cellular ppGalNAc-T3 and inhibits its activity.

Given its validation as a direct inhibitor of ppGalNAc-T3 without obvious off-target effects we turned to biologically relevant tests of T3Inh-1. As mentioned, overexpression of ppGalNAc-T3 is linked to cancer cell invasiveness as well as poor outcomes in patients (Kitada et al., 2013; Brockhausen, 2006; Harada et al., 2016; Mochizuki et al., 2013). Although no linkage to breast cancer has been reported, our analysis of publically available data for 1117 breast cancer patients (Szász et al., 2016) using Kaplan-Meier survival plots shows that high expression of ppGalNAc-T3 correlates with poor patient overall and metastasis-free survival (Figure 3—figure supplement 1A–B). Therefore, we carried out migration and invasion assays with the breast cancer cell line MDA-MB231, which expresses a relatively high level of ppGalNAc-T3 (Figure 3—figure supplement 1C), in the absence or presence of 5 µM T3Inh-1. Cells were cultured on uncoated (to assay migration) or Matrigel-coated (to assay invasiveness) Bioboat filters for 24 or 48 hr and those cells that moved to the underside of the filters were imaged and quantified. T3Inh-1 was strikingly effective, inhibiting migration by >80% (Figure 3A) and invasion by 98% (Figure 3B) while causing no discernable effect on cell proliferation (Figure 3C). To confirm that the effect was due to ppGalNAc-T3, the same experiment was carried out using MCF7 cells, which is a breast cancer cell line that expresses relatively low levels of ppGalNAc-T3 (Figure 3—figure supplement 1C). Critically, invasion by MCF7 cells was significantly increased by transfection with ppGalNAc-T3 and this increase was strongly blocked by T3Inh-1 (Figure 3D). Although O-glycosylation has been connected to cancer migration and invasion there is little evidence that it could be targetable for clinical purposes. Our results provide a ‘proof of concept’ and a strong starting point for developing the necessary tools.

Figure 3 with 1 supplement see all

Download asset Open asset

Figure 3—source data 1

Panel A: Cell counts in migration assay.

https://doi.org/10.7554/eLife.24051.019

Download elife-24051-fig3-data1-v3.xlsx

Figure 3—source data 2

Panel B: Cell counts in invasion assay.

https://doi.org/10.7554/eLife.24051.020

Download elife-24051-fig3-data2-v3.xlsx

Figure 3—source data 3

Panel C: Cell counts in proliferation assay.

https://doi.org/10.7554/eLife.24051.021

Download elife-24051-fig3-data3-v3.xlsx

Figure 3—source data 4

Panel D: Cell counts in MCF7 invasion assay.

https://doi.org/10.7554/eLife.24051.022

Download elife-24051-fig3-data4-v3.xlsx

As the relevant target(s) of ppGalNAc-T3 that drive metastatic-like cell behavior remain to be identified, we next tested whether T3Inh-1 could inhibit glycan masking of FGF23, a known ppGalNAc-T3 target. If so, we expected reduced secretion of intact FGF23. HEK cells co-expressing transfected FGF23 and ppGalNAc-T3 were treated with increasing concentrations of T3Inh-1 and secreted FGF23 was assayed by immunoblot. There was a clear dose-dependent loss of intact FGF23 (Figure 4A) and an increase in the ratio of cleaved/intact yielding a half-max of 14 µM for this effect (Figure 4B). For an unknown reason, perhaps related to its instability in media (Kato et al., 2006), the cleaved fragment did not show a corresponding increase. Rather, its recovery varied with the average over three experiments yielding a relatively small increase (Figure 4—figure supplement 1A–B). As expected, intact and cleaved FGF23 showed no change in cell lysates even for cells with lysosomal degradation inhibited by chloroquine (Figure 4—figure supplement 1C–E) arguing that T3Inh-1 affected cleavage just prior to secretion and did not affect FGF23 expression or cause intracellular routing to lysosomes. Importantly, we also tested the effect of T3Inh-1 on cleavage of ANGPTL3, which is controlled by ppGalNAc-T2-mediated glycan masking (Schjoldager et al., 2012, 2010). Secreted intact ANGPTL3 remained high at all concentrations, confirming the selectivity of T3Inh-1 towards ppGalNAc-T3 (Figure 4A–B). Reasoning that we might be able to see a similar effect on secreted FGF23 in an animal model, mice were injected intraperitoneally with T3Inh-1 and serum levels of cleaved FGF23 were determined. Three groups (0, 25 and 50 mg/kg T3Inh-1) of mice received either one or two injections separated by 24 hr, followed by blood collection after another 24 hr. There were no apparent ill effects on animal health. An ELISA assay with antibodies against the N- and C-terminal portions of FGF23 was used to determine the ratio of cleaved/intact FGF23 in the blood (Sun et al., 2015; Bai et al., 2016). Remarkably, T3Inh-1 caused a robust and statistically significant increase in this ratio at the tested 25 and 50 mg/kg concentrations (Figure 4C). These findings support the further development of T3Inh-1 toward mitigating the effects of elevated FGF23 signaling in chronic kidney disease patients.

Figure 4 with 1 supplement see all

Download asset Open asset

Figure 4—source data 1

Panel B: Cleaved/intact FGF23 and ANGPTL3 secreted at differing T3Inh-1 concentrations.

https://doi.org/10.7554/eLife.24051.025

Download elife-24051-fig4-data1-v3.xlsx

Figure 4—source data 2

Panel C: Cleaved/intact FGF23 in mouse serum after one two T3Inh-1 injections.

https://doi.org/10.7554/eLife.24051.026

Download elife-24051-fig4-data2-v3.xlsx

This study identifies an isozyme-selective inhibitor targeting ppGalNAc-T3. The compound binds directly conferring a mixed-mode of inhibition and is equally active in vitro and in cells. Its discovery paves the way for structural studies that will contribute to our understanding of the enzyme reaction mechanism and guide rational design of modified versions of T3Inh-1 to improve its binding affinity and efficacy. Forthcoming tests of disease models, possibly employing higher affinity versions, may strengthen the case for therapeutic uses of T3Inh-1. It is difficult to predict possible side effects because a full list of ppGalNAc-T3 substrates is not yet available. However, the known effects of its knockout in the mouse model are all attributed to FGF23 processing (Ichikawa et al., 2014). Thus, use of T3Inh-1 to reduce intact FGF23 (and increase the inhibitory cleaved fragment) to treat chronic kidney disease may have limited side effects. Clearly, the issue of multiple substrates has not been a major concern in successful therapies targeting protein kinases. To conclude, we anticipate rational design aided by T3Inh-1, as well as further screening using isozyme-specific sensors, to result in a panel of both isozyme- and pan-specific modulators targeting ppGalNAc-transferases. As individual ppGalNAc-transferase isozymes are associated with unique diseases (Schjoldager and Clausen, 2012; van der Post et al., 2013; Moremen et al., 2012), the result would be a new class of therapeutics capable of treating an array of differing diseases.

1. 1. Bai X
   2. Miao D
   3. Xiao S
   4. Qiu D
   5. St-Arnaud R
   6. Petkovich M
   7. Gupta A
   8. Goltzman D
   9. Karaplis AC

   (2016) CYP24 inhibition as a therapeutic target in FGF23-mediated renal phosphate wasting disorders

   Journal of Clinical Investigation 126:667–680.

   https://doi.org/10.1172/JCI81928

   • PubMed
   • Google Scholar
2. 1. Bennett EP
   2. Mandel U
   3. Clausen H
   4. Gerken TA
   5. Fritz TA
   6. Tabak LA

   (2012) Control of mucin-type O-glycosylation: a classification of the polypeptide GalNAc-transferase gene family

   Glycobiology 22:736–756.

   https://doi.org/10.1093/glycob/cwr182

   • PubMed
   • Google Scholar
3. 1. Brockhausen I

   (2006) Mucin-type O-glycans in human Colon and breast Cancer: glycodynamics and functions

   EMBO Reports 7:599–604.

   https://doi.org/10.1038/sj.embor.7400705

   • PubMed
   • Google Scholar
4. 1. Brooks SA
   2. Carter TM
   3. Bennett EP
   4. Clausen H
   5. Mandel U

   (2007) Immunolocalisation of members of the polypeptide N-acetylgalactosaminyl transferase (ppGalNAc-T) family is consistent with biologically relevant altered cell surface glycosylation in breast Cancer

   Acta Histochemica 109:273–284.

   https://doi.org/10.1016/j.acthis.2007.02.009

   • PubMed
   • Google Scholar
5. 1. Chefetz I
   2. Sprecher E

   (2009) Familial tumoral calcinosis and the role of O-glycosylation in the maintenance of phosphate homeostasis

   Biochimica Et Biophysica Acta (BBA) - Molecular Basis of Disease 1792:847–852.

   https://doi.org/10.1016/j.bbadis.2008.10.008

   • PubMed
   • Google Scholar
6. 1. Degirolamo C
   2. Sabbà C
   3. Moschetta A

   (2016) Therapeutic potential of the endocrine fibroblast growth factors FGF19, FGF21 and FGF23

   Nature Reviews Drug Discovery 15:51–69.

   https://doi.org/10.1038/nrd.2015.9

   • PubMed
   • Google Scholar
7. 1. Eckardt KU
   2. Kasiske BL

   (2009) Kidney disease: improving global outcomes

   Nature Reviews Nephrology 5:650–657.

   https://doi.org/10.1038/nrneph.2009.153

   • PubMed
   • Google Scholar
8. 1. Evan GI
   2. Lewis GK
   3. Ramsay G
   4. Bishop JM

   (1985) Isolation of monoclonal antibodies specific for human c-myc proto-oncogene product

   Molecular and Cellular Biology 5:3610–3616.

   https://doi.org/10.1128/MCB.5.12.3610

   • PubMed
   • Google Scholar
9. 1. Fukumoto S

   (2016) FGF23-FGF receptor/Klotho pathway as a New Drug target for disorders of bone and Mineral metabolism

   Calcified Tissue International 98:334–340.

   https://doi.org/10.1007/s00223-015-0029-y

   • PubMed
   • Google Scholar
10. 1. Gao Y
    2. Liu Z
    3. Feng J
    4. Sun Q
    5. Zhang B
    6. Zheng W
    7. Ma W

    (2013) Expression pattern of polypeptide N-acetylgalactosaminyltransferase-10 in gastric carcinoma

    Oncology Letters 5:113–116.

    https://doi.org/10.3892/ol.2012.980

    • PubMed
    • Google Scholar
11. 1. Goetz R
    2. Nakada Y
    3. Hu MC
    4. Kurosu H
    5. Wang L
    6. Nakatani T
    7. Shi M
    8. Eliseenkova AV
    9. Razzaque MS
    10. Moe OW
    11. Kuro-o M
    12. Mohammadi M

    (2010) Isolated C-terminal tail of FGF23 alleviates hypophosphatemia by inhibiting FGF23-FGFR-Klotho complex formation

    PNAS 107:407–412.

    https://doi.org/10.1073/pnas.0902006107

    • PubMed
    • Google Scholar
12. 1. Harada Y
    2. Izumi H
    3. Noguchi H
    4. Kuma A
    5. Kawatsu Y
    6. Kimura T
    7. Kitada S
    8. Uramoto H
    9. Wang KY
    10. Sasaguri Y
    11. Hijioka H
    12. Miyawaki A
    13. Oya R
    14. Nakayama T
    15. Kohno K
    16. Yamada S

    (2016) Strong expression of polypeptide N-acetylgalactosaminyltransferase 3 independently predicts shortened disease-free survival in patients with early stage oral squamous cell carcinoma

    Tumor Biology 37:1357–1368.

    https://doi.org/10.1007/s13277-015-3928-7

    • PubMed
    • Google Scholar
13. 1. Ichikawa S
    2. Gray AK
    3. Padgett LR
    4. Allen MR
    5. Clinkenbeard EL
    6. Sarpa NM
    7. White KE
    8. Econs MJ

    (2014) Genetic rescue of glycosylation-deficient Fgf23 in the Galnt3 knockout mouse

    Endocrinology 155:3891–3898.

    https://doi.org/10.1210/en.2014-1199

    • PubMed
    • Google Scholar
14. 1. Isakova T
    2. Ix JH
    3. Sprague SM
    4. Raphael KL
    5. Fried L
    6. Gassman JJ
    7. Raj D
    8. Cheung AK
    9. Kusek JW
    10. Flessner MF
    11. Wolf M
    12. Block GA

    (2015) Rationale and approaches to phosphate and fibroblast growth factor 23 reduction in CKD

    Journal of the American Society of Nephrology 26:2328–2339.

    https://doi.org/10.1681/ASN.2015020117

    • PubMed
    • Google Scholar
15. 1. Jesch SA
    2. Lewis TS
    3. Ahn NG
    4. Linstedt AD

    (2001) Mitotic phosphorylation of golgi reassembly stacking protein 55 by mitogen-activated protein kinase ERK2

    Molecular Biology of the Cell 12:1811–1817.

    https://doi.org/10.1091/mbc.12.6.1811

    • PubMed
    • Google Scholar
16. 1. Kato K
    2. Jeanneau C
    3. Tarp MA
    4. Benet-Pagès A
    5. Lorenz-Depiereux B
    6. Bennett EP
    7. Mandel U
    8. Strom TM
    9. Clausen H

    (2006) Polypeptide GalNAc-transferase T3 and familial tumoral calcinosis. secretion of fibroblast growth factor 23 requires O-glycosylation

    Journal of Biological Chemistry 281:18370–18377.

    https://doi.org/10.1074/jbc.M602469200

    • PubMed
    • Google Scholar
17. 1. Kitada S
    2. Yamada S
    3. Kuma A
    4. Ouchi S
    5. Tasaki T
    6. Nabeshima A
    7. Noguchi H
    8. Wang KY
    9. Shimajiri S
    10. Nakano R
    11. Izumi H
    12. Kohno K
    13. Matsumoto T
    14. Sasaguri Y

    (2013) Polypeptide N-acetylgalactosaminyl transferase 3 independently predicts high-grade tumours and poor prognosis in patients with renal cell carcinomas

    British Journal of Cancer 109:472–481.

    https://doi.org/10.1038/bjc.2013.331

    • PubMed
    • Google Scholar
18. 1. Kohsaki T
    2. Nishimori I
    3. Nakayama H
    4. Miyazaki E
    5. Enzan H
    6. Nomoto M
    7. Hollingsworth MA
    8. Onishi S

    (2000) Expression of UDP-GalNAc: polypeptide N-acetylgalactosaminyltransferase isozymes T1 and T2 in human colorectal Cancer

    Journal of Gastroenterology 35:840–848.

    https://doi.org/10.1007/s005350070021

    • PubMed
    • Google Scholar
19. 1. Kuan SF
    2. Byrd JC
    3. Basbaum C
    4. Kim YS

    (1989)

    Inhibition of mucin glycosylation by aryl-N-acetyl-alpha-galactosaminides in human Colon cancer cells

    The Journal of Biological Chemistry 264:19271–19277.

    • PubMed
    • Google Scholar
20. 1. Mochizuki Y
    2. Ito K
    3. Izumi H
    4. Kohno K
    5. Amano J

    (2013) Expression of polypeptide N-acetylgalactosaminyl transferase-3 and its association with clinicopathological factors in thyroid carcinomas

    Thyroid 23:1553–1560.

    https://doi.org/10.1089/thy.2012.0613

    • PubMed
    • Google Scholar
21. 1. Moremen KW
    2. Tiemeyer M
    3. Nairn AV

    (2012) Vertebrate protein glycosylation: diversity, synthesis and function

    Nature Reviews Molecular Cell Biology 13:448–462.

    https://doi.org/10.1038/nrm3383

    • PubMed
    • Google Scholar
22. 1. Mukhopadhyay S
    2. Bachert C
    3. Smith DR
    4. Linstedt AD

    (2010) Manganese-induced trafficking and turnover of the cis-Golgi glycoprotein GPP130

    Molecular Biology of the Cell 21:1282–1292.

    https://doi.org/10.1091/mbc.E09-11-0985

    • PubMed
    • Google Scholar
23. 1. Patsos G
    2. Hebbe-Viton V
    3. Robbe-Masselot C
    4. Masselot D
    5. San Martin R
    6. Greenwood R
    7. Paraskeva C
    8. Klein A
    9. Graessmann M
    10. Michalski JC
    11. Gallagher T
    12. Corfield A

    (2009) O-glycan inhibitors generate aryl-glycans, induce apoptosis and lead to growth inhibition in colorectal Cancer cell lines

    Glycobiology 19:382–398.

    https://doi.org/10.1093/glycob/cwn149

    • PubMed
    • Google Scholar
24. 1. Peng RQ
    2. Wan HY
    3. Li HF
    4. Liu M
    5. Li X
    6. Tang H

    (2012) MicroRNA-214 suppresses growth and invasiveness of cervical Cancer cells by targeting UDP-N-acetyl-α-D-galactosamine:polypeptide N-acetylgalactosaminyltransferase 7

    Journal of Biological Chemistry 287:14301–14309.

    https://doi.org/10.1074/jbc.M111.337642

    • PubMed
    • Google Scholar
25. 1. Puri S
    2. Bachert C
    3. Fimmel CJ
    4. Linstedt AD

    (2002) Cycling of early golgi proteins via the cell surface and endosomes upon lumenal pH disruption

    Traffic 3:641–653.

    https://doi.org/10.1034/j.1600-0854.2002.30906.x

    • PubMed
    • Google Scholar
26. 1. Revoredo L
    2. Wang S
    3. Bennett EP
    4. Clausen H
    5. Moremen KW
    6. Jarvis DL
    7. Ten Hagen KG
    8. Tabak LA
    9. Gerken TA

    (2016) Mucin-type O-glycosylation is controlled by short- and long-range glycopeptide substrate recognition that varies among members of the polypeptide GalNAc transferase family

    Glycobiology 26:360–376.

    https://doi.org/10.1093/glycob/cwv108

    • PubMed
    • Google Scholar
27. 1. Rowe PS

    (2015) A unified model for bone-renal mineral and energy metabolism

    Current Opinion in Pharmacology 22:64–71.

    https://doi.org/10.1016/j.coph.2015.03.006

    • PubMed
    • Google Scholar
28. 1. Röttger S
    2. White J
    3. Wandall HH
    4. Olivo JC
    5. Stark A
    6. Bennett EP
    7. Whitehouse C
    8. Berger EG
    9. Clausen H
    10. Nilsson T

    (1998)

    Localization of three human polypeptide GalNAc-transferases in HeLa cells suggests initiation of O-linked glycosylation throughout the golgi apparatus

    Journal of Cell Science 111:45–60.

    • PubMed
    • Google Scholar
29. 1. Schjoldager KT
    2. Clausen H

    (2012) Site-specific protein O-glycosylation modulates proprotein processing - deciphering specific functions of the large polypeptide GalNAc-transferase gene family

    Biochimica Et Biophysica Acta (BBA) - General Subjects 1820:2079–2094.

    https://doi.org/10.1016/j.bbagen.2012.09.014

    • PubMed
    • Google Scholar
30. 1. Schjoldager KT
    2. Joshi HJ
    3. Kong Y
    4. Goth CK
    5. King SL
    6. Wandall HH
    7. Bennett EP
    8. Vakhrushev SY
    9. Clausen H

    (2015) Deconstruction of O-glycosylation--GalNAc-T isoforms direct distinct subsets of the O-glycoproteome

    EMBO Reports 16:1713–1722.

    https://doi.org/10.15252/embr.201540796

    • PubMed
    • Google Scholar
31. 1. Schjoldager KT
    2. Vakhrushev SY
    3. Kong Y
    4. Steentoft C
    5. Nudelman AS
    6. Pedersen NB
    7. Wandall HH
    8. Mandel U
    9. Bennett EP
    10. Levery SB
    11. Clausen H

    (2012) Probing isoform-specific functions of polypeptide GalNAc-transferases using zinc finger nuclease glycoengineered SimpleCells

    PNAS 109:9893–9898.

    https://doi.org/10.1073/pnas.1203563109

    • PubMed
    • Google Scholar
32. 1. Schjoldager KT
    2. Vester-Christensen MB
    3. Bennett EP
    4. Levery SB
    5. Schwientek T
    6. Yin W
    7. Blixt O
    8. Clausen H

    (2010) O-glycosylation modulates proprotein convertase activation of angiopoietin-like protein 3: possible role of polypeptide GalNAc-transferase-2 in regulation of concentrations of plasma lipids

    Journal of Biological Chemistry 285:36293–36303.

    https://doi.org/10.1074/jbc.M110.156950

    • PubMed
    • Google Scholar
33. 1. Schjoldager KT
    2. Vester-Christensen MB
    3. Goth CK
    4. Petersen TN
    5. Brunak S
    6. Bennett EP
    7. Levery SB
    8. Clausen H

    (2011) A systematic study of site-specific GalNAc-type O-glycosylation modulating proprotein convertase processing

    Journal of Biological Chemistry 286:40122–40132.

    https://doi.org/10.1074/jbc.M111.287912

    • PubMed
    • Google Scholar
34. 1. Smith ER

    (2014) The use of fibroblast growth factor 23 testing in patients with kidney disease

    Clinical Journal of the American Society of Nephrology 9:1283–1303.

    https://doi.org/10.2215/CJN.10941013

    • PubMed
    • Google Scholar
35. 1. Song L
    2. Bachert C
    3. Schjoldager KT
    4. Clausen H
    5. Linstedt AD

    (2014) Development of isoform-specific sensors of polypeptide GalNAc-transferase activity

    Journal of Biological Chemistry 289:30556–30566.

    https://doi.org/10.1074/jbc.M114.599563

    • PubMed
    • Google Scholar
36. 1. Sun N
    2. Guo Y
    3. Liu W
    4. Densmore M
    5. Shalhoub V
    6. Erben RG
    7. Ye L
    8. Lanske B
    9. Yuan Q

    (2015) FGF23 neutralization improves bone quality and osseointegration of titanium implants in chronic kidney disease mice

    Scientific Reports 5:8304.

    https://doi.org/10.1038/srep08304

    • PubMed
    • Google Scholar
37. 1. Szász AM
    2. Lánczky A
    3. Nagy Á
    4. Förster S
    5. Hark K
    6. Green JE
    7. Boussioutas A
    8. Busuttil R
    9. Szabó A
    10. Győrffy B

    (2016) Cross-validation of survival associated biomarkers in gastric Cancer using transcriptomic data of 1,065 patients

    Oncotarget 7:49322–49333.

    https://doi.org/10.18632/oncotarget.10337

    • PubMed
    • Google Scholar
38. 1. Tagliabracci VS
    2. Engel JL
    3. Wiley SE
    4. Xiao J
    5. Gonzalez DJ
    6. Nidumanda Appaiah H
    7. Koller A
    8. Nizet V
    9. White KE
    10. Dixon JE

    (2014) Dynamic regulation of FGF23 by Fam20C phosphorylation, GalNAc-T3 glycosylation, and furin proteolysis

    PNAS 111:5520–5525.

    https://doi.org/10.1073/pnas.1402218111

    • PubMed
    • Google Scholar
39. 1. van der Post S
    2. Subramani DB
    3. Bäckström M
    4. Johansson ME
    5. Vester-Christensen MB
    6. Mandel U
    7. Bennett EP
    8. Clausen H
    9. Dahlén G
    10. Sroka A
    11. Potempa J
    12. Hansson GC

    (2013) Site-specific O-glycosylation on the MUC2 mucin protein inhibits cleavage by the Porphyromonas gingivalis secreted cysteine protease (RgpB)

    Journal of Biological Chemistry 288:14636–14646.

    https://doi.org/10.1074/jbc.M113.459479

    • PubMed
    • Google Scholar
40. 1. Wang ZQ
    2. Bachvarova M
    3. Morin C
    4. Plante M
    5. Gregoire J
    6. Renaud MC
    7. Sebastianelli A
    8. Bachvarov D

    (2014) Role of the polypeptide N-acetylgalactosaminyltransferase 3 in ovarian Cancer progression: possible implications in abnormal mucin O-glycosylation

    Oncotarget 5:544–560.

    https://doi.org/10.18632/oncotarget.1652

    • PubMed
    • Google Scholar

Author details

1. Lina Song

   Department of Biological Sciences, Carnegie Mellon University, Pittsburgh, United States

   Contribution

   LS, Conceptualization, Data curation, Formal analysis, Writing—original draft

   Competing interests

   The authors declare that no competing interests exist.

2. Adam D Linstedt

   Department of Biological Sciences, Carnegie Mellon University, Pittsburgh, United States

   Contribution

   ADL, Conceptualization, Supervision, Funding acquisition, Methodology, Project administration, Writing—review and editing

   For correspondence

   linstedt@andrew.cmu.edu

   Competing interests

   The authors declare that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0003-0754-1638

• 2,239

  views

• 474

  downloads

• 29

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.