University of Groningen Exploring and exploiting bacterial protein glycosylation systems Yakovlieva, Liubov

(1)

University of Groningen

Exploring and exploiting bacterial protein glycosylation systems

Yakovlieva, Liubov

DOI:

10.33612/diss.173544104

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.

Document Version

Publisher's PDF, also known as Version of record

Publication date: 2021

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):

Yakovlieva, L. (2021). Exploring and exploiting bacterial protein glycosylation systems. University of Groningen. https://doi.org/10.33612/diss.173544104

Copyright

Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.

(2)

561536-L-bw-Yakovlievla 561536-L-bw-Yakovlievla 561536-L-bw-Yakovlievla 561536-L-bw-Yakovlievla Processed on: 11-6-2021 Processed on: 11-6-2021 Processed on: 11-6-2021

Processed on: 11-6-2021 PDF page: 231PDF page: 231PDF page: 231PDF page: 231

C

HAPTER

8

Summary and Outlook

(3)

CHAPTER 8

224

The overall aim of the research presented in this Thesis was to investigate mechanistic aspects of two unusual bacterial N-glycosyltransferase systems – ApNGT from Actinobacillus pleuropneumoniae and HiNGT from

Haemophilus influenzae, and EarP from Pseudomonas aeruginosa. ApNGT and

HiNGT perform hyperglucosylation of bacterial adhesin protein, decorating asparagine residues with simple glucose moieties. This, in turn ensures adhesin stability and its tethering to the bacterial membrane during H. influenzae colonization. EarP modifies a single arginine residue in the cytosolic elongation factor P (EF-P) protein with the bacterial sugar L-rhamnose. This activates EF-P to rescue ribosomal stalling on polyproline stretches during protein synthesis. Both glycosylation systems contribute to virulence or fitness of bacterial pathogens, and represent highly unique protein glycosylation systems. Understanding the mechanistic basis of the enzymes involved will aid in developing these systems into promising antibacterial targets.

In Chapter 1 a general introduction was given to protein glycosylation and this Thesis. In Chapter 2 the phenomenon of processivity in bacterial glycosyltransferases was reviewed. Various examples of glycoconjugates synthesized by processive glycosyltransferases were discussed with specific focus on various methods to study processive mechanisms. A separate section was dedicated to processive protein glycosyltransferases, which are typically not investigated for processive behavior.

In Chapter 3 the mechanism of adhesin protein hyperglucosylation was investigated. Specifically, it was investigated whether the mechanism of hyperglycosylation was processive or distributive. Through examining reaction product profiles with intact protein mass spectrometry, and by performing kinetic and affinity studies it was determined that the mechanism of the NGT enzyme can be designated as semiprocessive. In the case of NGT-catalyzed hyperglucosylation of the adhesin HMW1ct, the initial fast phase of the reaction proceeds with significant processivity, but is then hampered by product inhibition, leading to a switch in the mechanism and slower progression to the final product. By performing proteomics experiments, the preferred glucosylation sites were identified to reside mainly in the exposed loops of the substrate protein, as well as in close proximity to one another. Lastly, molecular dynamics simulations of a glucopeptide (based on the sequence of HMW1ct) in the active site of the NGT enzyme revealed the possibility for bi-directional binding of the substrate and possibility to slide along the active site to achieve modification without dissociating. Together, these findings reveal the molecular basis for processive behavior of the NGT enzyme, and as such constitute the first

(4)

225 account of processive character in a protein N-glycosyltransferase. It potentially explains how sufficient levels of adhesin hyperglucosylation are achieved in vivo and opens the possibilities for various biotechnological applications.

The NGT-HMW1ct glycosylation system was further used in Chapter 4 to perform azidoglucose labelling of HMW1ct. The hyperglucosylated adhesin protein has been postulated to act as an exoantigen in inducing the autoimmune response in multiple sclerosis (MS).1_{In this Chapter azidoglucose-labelled}

HMW1ct was generated in vitro as a first step towards evaluating this labeled protein as a tool in the studies of MS antigen uptake and presentation. A panel of azidoglucoses was designed and synthesized to be used in combination with the ApNGT-HMW1ct system: 2AzGlc, 3AzGlc, 4AzGlc. UDP-6AzGlc was obtained commercially. Whereas UDP-2AzGlc and UDP-UDP-6AzGlc did not prove to be suitable donor substrates for ApNGT, UDP-3AzGlc and UDP-4AzGlc allowed incorporation of several azidoglucose moieties in the HMW1ct protein. Optimal conditions of 1:1 enzyme:protein (molar ratio) and incubation for three days resulted in a final product mixture of HMW1ct containing 2, 3, and 4 azidoglucose moieties. Proteomics analysis revealed incorporation of AzGlc into glycosylation sites that are located in the exposed loops of the HMW1ct predicted structure. To gain insight into the differences in reactivity of the UDP-AzGlc analogues in the glycosylation reaction, docking and molecular dynamics simulations were performed. These studies revealed conformational differences in how the different UDP-azidoglucoses bind in the ApNGT active site. Azidoglucose-labeled HMW1ct (a mixture of 2, 3, 4-times AzGlc incorporated) was isolated and is currently studied in MS antibody binding studies (ELISA assay).

If binding of MS antibodies to azidoglucose-labeled HMW1ct proves satisfactory, further studies into antigen uptake and presentation can be performed. Glycan moieties of glycoproteins have been shown to play an important role in antigen processing, either indirectly via stabilization of the peptide structure and protection from proteases, or directly as immunogenic epitopes in the T-cell recognition.2_{The steps of antigen processing can be}

visualized via a click reaction with fluorophore and proteolytic AzGlc-HMW1ct fragments can be pulled down by conjugation with affinity label. First, the in

vitro proteolysis of AzGlc-HMW1ct by lysosomal extracts can be investigated to

identify which (azidoglucose)peptide fragments are generated.3_{Further studies}

with human cells can be performed to identify antigenic peptides, as has been described previously for fluorescently labeled subtilisin.4

(5)

CHAPTER 8

226

Chapter 5 described preliminary studies to investigate multimeric

structures of the ApNGT/HiNGT-HMW1ct system. Based on the initial observation of a molecular weight for ApNGT in native protein MS that suggested a dimeric structure, further studies were performed to elucidate oligomeric states of ApNGT, HiNGT, (Glc-)HMW1ct and combinations thereof. Namely, blue native PAGE and SEC/SEC-MALLS experiments were performed. In blue native PAGE and SEC alleged multimers of ApNGT, HiNGT and (Glc-)HMW1ct were observed in comparison to protein standards included in the same experiments. Exact molecular weights of these proteins were measured by performing SEC-MALLS experiments. This analysis revealed a dimeric structure of ApNGT, a trimeric structure of HiNGT, and a mainly monomeric state of HMW1ct with a small fraction of oligomers of unidentified Mw. Further studies are needed to determine the architecture of the NGT and HMW1ct oligomers and their functional relevance to protein glycosylation.

As has been described in Chapter 1 and Chapter 3, ApNGT and HiNGT are unusual bacterial glycosyltransferases that perform unconventional protein

N-glycosylation in the cytoplasm. At the same time this protein glycosylation is

of clinical relevance as it renders nontypeable Haemophilus influenzae (NTHi) pathogens adherent.5_{This pathogen has become a major cause of upper}

respiratory tract diseases since the introduction of the vaccine against encapsulated Haemophilus influenzae (Hib).6_{Insight into the mechanism and}

structure of the NGT enzymes is valuable as it will allow development of novel inhibitors of protein glycosylation and potentially treatment against NTHi.

In Chapter 6 the method of Pd-catalyzed chemoselective oxidation of

glucopeptides was developed. The organometallic catalyst

[(neocuproine)PdOAc]2OTf2 has been used previously to achieve regioselective

C3-oxidation of glucosides7_{and the resulting keto-group could be further}

expanded into a variety of useful functionalities.8,9_{In this Chapter a panel of}

glucopeptides was treated with oxidation conditions and was shown to be successfully oxidized on the glucose moiety. Selectivity for oxidation of glucose was confirmed in the experiments with a similar galactopeptide where conversion to the oxidized product only took place under significantly longer reaction times and increased catalyst loading (10 eq for the galactopeptide vs 0.5 eq of Pd cat. for the glucopeptide). Limitations of this oxidation method primarily stemmed from the oxidation of threonine hydroxyl groups that resulted in fragmentation of the peptides with more complex sequences and Pd-mediated deglycosylation. The latter could be successfully suppressed by treatment with a Pd-catalyst scavenger. The resulting keto-group on oxidized

(6)

227 glucopeptides could be further transformed into an oxime functionality, which allows introduction of various groups of interest. The oxidation-oxime formation methodology was further expanded to the complex mixture of tryptic glucopeptides that was generated from Glc-HMW1ct. Preliminary screenings show that 1 eq of Pd-catalyst and 24 h incubation time was enough to already achieve significant levels of oxidation without inducing degradation. Furthermore, oxime formation has also proved possible and hydroxylamine oxime and biotin oxime could be formed in moderate amounts from the oxidized mixture of glucopeptides. Taken together, this Chapter outlined the methodology for late-stage modification of glucopeptides as well as selective oxidation and functionalization of tryptic glucopeptides for proteomics analysis.

Chapter 7 was dedicated to deciphering the substrate recognition

specificity of the bacterial arginine rhamnosyltransferase EarP. This enzyme is responsible for arginine glycosylation, which is a subtype of N-linked glycosylation that is exceedingly rare, with only a few examples identified in nature.10,11_{EF-P-rhamnosylation is important for bacterial fitness and when}

abolished in Pseudomonas aeruginosa and Neisseria meningitidis led to multiple detrimental effects, including greater susceptibility to antibiotics.12,13_{In this}

Chapter the molecular basis of substrate recognition by the EarP enzyme that governs arginine rhamnosylation was investigated. By testing short peptide mimics of the substrate protein EF-P in the rhamnosylation reaction it was revealed that especially the secondary structure plays a crucial role in recognition. Consequently, cyclic designs that mimic the native beta-hairpin motif were successfully rhamnosylated in vitro, whereas disordered and linear peptides were not suitable substrates for the EarP. Through screening several secondary structure stabilizing templates and various peptide lengths, the smallest EF-P fragment to be rhamnosylated to date was identified, which is a cyclic 11mer peptide. Several mutants of this peptide were tested to screen the promiscuity of EarP and revealed that EarP had a greater tolerance towards amino acid sequence variation than the absence of secondary structure elements. This is especially important for the rational design of inhibitors as well as for developing tools for the identification of other bacterial arginine rhamnosylation systems. Future work on the peptide designs described in this chapter may involve incorporation of an additional beta-strand (residues Val53, Phe54, Lys55) to improve binding affinity to EarP. Additionally, for the design of inhibitors an identified minimal beta-hairpin can be combined with the nucleotide part of the sugar donor substrate (TDP of TDP-Rha) to fully occupy the active site of EarP.

(7)

CHAPTER 8

228

1. Walvoort, M. T. C.; Testa, C.; Eilam, R.; Nuti, F.; Rossi, G.; Real-Fernandez, F.; Lanzilla, R.; Morra, V. B.; Lolli, F.; Rovero, P.; Imperiali, B.; Papini, A.M. Antibodies from multiple sclerosis patients preferentially recognize hyperglucosylated adhesin of non-typeable

Haemophilus influenzae. Sci. Rep. 2016, 6, 39430.

2. Wolfert, M. A.; Boons, G. -J. Adaptive immune activation: glycosylation does matter. Nat.

Chem. Biol. 2013, 9, 776-784.

3. Villadangos, J. A.; Bryant, R. A. R.; Deussing, J.; Driessen, C.; Lennon-Dumenil, A. M.; Riese, R. J.; Roth, W.; Saftig, P.; Shi, G. P.; Chapman, H.A.; Peters, C.; Ploegh, H. L. Proteases involved in MHC class II antigen presentation. Immunol. Rev. 1999, 172, 109-120. 4. Patil, N. S.; Wong, D. L.; Collier, K. D.; McDonald, H.C. Fluorescent derivatization of a protease antigen to track antigen uptake and processing in human cell lines. BMC

Immunol. 2004, 5, 12.

5. St Geme III, J. W.; Falkow, S.; Barenkamp, S. J. High-molecular-weight proteins of nontypable Haemophilus influenzae mediate attachment to human epithelial cells. Proc.

Natl. Acad. Sci. U. S. A. 1993, 90, 2875–2879.

6. Ladhani, S.; Slack, M. P. E.; Heath, P. T.; von Gottberg, A.; Chandra, M.; Ramsay, M. E.; European Union Invasive Bacterial Infection Surveillance participants. Invasive

Haemophilus influenzae Disease, Europe, 1996–2006. Emerg. Infect. Dis. 2010, 16, 455-463.

7. Jäger, M.; Hartmann, M.; de Vries, J. G.; Minnaard, A. J. Catalytic Regioselective Oxidation of Glycosides. Angew. Chemie Int. Ed. 2013, 52, 7809–7812.

8. Eisink, N. N. H. M.; Lohse, J.; Witte, M. D.; Minnaard, A. J. Regioselective oxidation of unprotected 1,4 linked glucans. Org. Biomol. Chem. 2016, 14, 4859–4864.

9. Marinus, N.; Tahiri, N.; Duca, M.; Mouthaan, L. M. C. M.; Bianca, S.; van den Noort, M.; Poolman, B.; Witte, M. D.; Minnaard, A.J. Stereoselective Protection-Free Modification of 3-Keto-saccharides. Org. Lett. 2020, 22, 5622–5626.

10. Singh, D. G.; Lomako, J.; Lomako, W. M.; Whelan, W. J.; Meyer, H. E.; Serwe M.; Metzger, J. W. beta-Glucosylarginine: a new glucose-protein bond in a self-glucosylating protein from sweet corn. FEBS Lett. 1995, 376, 61–64.

11. Li, S.; Zhang, L.; Yao, Q.; Li, L.; Dong, N.; Rong, J.; Gao, W.; Ding, X.; Sun, L.; Chen, X.; Chen, S.; Shao, F. Pathogen blocks host death receptor signalling by arginine GlcNAcylation of death domains. Nature 2013, 501, 242–246.

12. Rajkovic, A.; Erickson, S.; Witzky, A.; Branson, O. E.; Seo, J.; Gafken, P. R.; Frietas, M. A.; Whitelegge, J. P.; Faull, K. F.; Navarre, W.; Darwin, A. J.; Ibba, M. Cyclic Rhamnosylated Elongation Factor P Establishes Antibiotic Resistance in Pseudomonas aeruginosa. mBio 2015,

6, e00823.

13. Yanagisawa, T; Takahashi, H.; Suzuki, T.; Masuda, A.; Dohmae, N.; Yokoyama, S. Neisseria meningitidis Translation Elongation Factor P and Its Active-Site Arginine Residue Are

University of Groningen Exploring and exploiting bacterial protein glycosylation systems Yakovlieva, Liubov