Protein N-glycosylation is ubiquitous in the brain and is closely related to cognition and memory. Alzheimer’s disease (AD) is a multifactorial disorder that lacks a clear pathogenesis and treatment. Aberrant N-glycosylation has been suggested to be involved in AD pathology. However, the systematic variations in protein N-glycosylation and their roles in AD have not been thoroughly investigated due to technical challenges. Here, we applied multilayered N-glycoproteomics to quantify the global protein expression levels, N-glycosylation sites, N-glycans, and site-specific N-glycopeptides in AD (APP/PS1 transgenic) and wild-type mouse brains. The N-glycoproteomic landscape exhibited highly complex site-specific heterogeneity in AD mouse brains. The generally dysregulated N-glycosylation in AD, which involved proteins such as glutamate receptors as well as fucosylated and oligomannose glycans, were explored by quantitative analyses. Furthermore, functional studies revealed the crucial effects of N-glycosylation on proteins and neurons. Our work provides a systematic multilayered N-glycoproteomic strategy for AD and can be applied to diverse biological systems.

Group I metabotropic glutamate receptor (mGluR I) activation exerts a slow postsynaptic excitatory effect in the CNS. Here, the issues of whether and how this receptor is involved in regulating retinal ganglion cell (RGC) excitability were investigated in retinal slices using patch-clamp techniques. Under physiological conditions, RGCs displayed spontaneous firing. Extracellular application of LY367385 (10 µM)/MPEP (10 µM), selective mGluR1 and mGluR5 antagonists, respectively, significantly reduced the firing frequency, suggesting that glutamate endogenously released from bipolar cells constantly modulates RGC firing. DHPG (10 µM), an mGluR I agonist, significantly increased the firing and caused depolarization of the cells, which were reversed by LY367385, but not by MPEP, suggesting the involvement of the mGluR1 subtype. Intracellular Ca-dependent PI-PLC/PKC and calcium/calmodulin-dependent protein kinase II (CaMKII) signaling pathways mediated the DHPG-induced effects. In the presence of cocktail synaptic blockers (CNQX, D-AP5, bicuculline, and strychnine), which terminated the spontaneous firing in both ON and OFF RGCs, DHPG still induced depolarization and triggered the cells to fire. The DHPG-induced depolarization could not be blocked by TTX. In contrast, Ba, an inwardly rectifying potassium channel (Kir) blocker, and Cs and ZD7288, hyperpolarization-activated cation channel (I ) blockers, mimicked the effect of DHPG. Furthermore, in the presence of Ba/ZD7288, DHPG did not show further effects. Moreover, Kir and I currents could be recorded in RGCs, and extracellular application of DHPG indeed suppressed these currents. Our results suggest that activation of mGluR I regulates the excitability of rat RGCs by inhibiting Kir and I.

25Alzheimer's disease (AD) is one of the most common neurodegenerative diseases that 26 currently lacks clear pathogenesis and effective treatment. Protein glycosylation is 27 ubiquitous in brain tissue and site-specific analysis of N-glycoproteome, which is 28 technically challenging, can advance our understanding of the glycoproteins' role in 29 AD. In this study, we profiled the multilayered variations in proteins, N-glycosites, 30 N-glycans, and in particular site-specific N-glycopeptides in the APP/PS1 and wild 31 type mouse brain through combining pGlyco 2.0 strategy with other quantitative 32 N-glycoproteomic strategies. The comprehensive brain N-glycoproteome landscape 33 was constructed, and rich details of the heterogeneous site-specific protein 34 N-glycosylations were exhibited. Quantitative analyses explored generally 35 downregulated N-glycosylation involving proteins such as glutamate receptors, as 36 well as fucosylated and oligo-mannose type glycans in APP/PS1 mice versus wild 37 type mice. Moreover, our preliminary functional study revealed that N-glycosylation 38 was crucial for the membrane localization of NCAM1 and for maintaining the 39 excitability and viability of neuron cells. Our work offered a panoramic view of the 40 N-glycoproteomes in Alzheimer's disease and revealed that generally impaired 41 N-glycosylation promotes Alzheimer's disease progression. 42Keywords: Quantitative N-glycoproteomics, Alzheimer's disease, Glutamate receptor, 43 N-glycan, Neuron hyperexcitability 44 4 7 glycan patterns was constructed ( Fig. 1B). Of the identified proteins, 276 were 103 exclusively contributed by the N-glycoproteome data (Fig. 1C). Among the total 104 1,493 N-glycosites, 423 and 689 N-glycosites were only identified by "N-glycosites" 105 and "site-specific N-glycopeptides" strategies, respectively, while 381 glycosites were 106 co-identified by the two strategies (Fig. 1D). The multi-omic data provided a 107 panoramic view of the proteomes, N-glycoproteomes and glycomes in the APP/PS1 108 mouse and wild type mouse brain. 109 Complex and heterogeneous N-glycoproteome landscape of the110 APP/PS1 mouse and wild type mouse brain 111 The micro-heterogeneity of protein N-glycosylation was exhibited through analyzing 112 the distribution of glycosites and site-specific glycopeptides on the glycoproteins. 113About 58.4% of N-glycoproteins have one N-glycosite, while the 7.6% most severely 114 N-glycosylated proteins have more than 5 sites ( Fig. 2A). The average number of 115 N-glycosites per glycoprotein is 2. The distribution of N-glycoforms on each site was 116 shown in Fig. 2B. The average number of glycoforms per site is 3.3, while there are 117 35 N-glycosites had more than 14 N-glycans on each of them. Proteins having the 118 most glycosites and the most heterogeneous sites were listed in the Figure (Fig. 2A 119 and B). Combining both information, the most heterogeneous glycoproteins in mouse 120 brain were listed in Figure EV1A. AT1B2, THY1, EAA2, and NPTN exhibited the 121 severest micro-...