Impaired 26S proteasome causes learning and memory deficiency and induces neuroinflammation mediated by NF-κB in mice

Background 

The ubiquitin-proteasome system (UPS) is a pathway commonly associated with protein degradation in eukaryotic cells(1,2). Selected proteins are tagged with polyubiquitin chains and, once marked, these proteins will be submitted to the catalytic activities of the proteasome(1,2). The proteasome is a multimeric protein complex which consists of specialized portions. The 20S proteasome consists of two outer α-rings and two inner β-rings, which serve as a gateway to protein entry and allows proteolytic activity, respectively(1,2). The 19S proteasome is an outer regulatory portion of the proteasome which has several functions, such as protein unfolding and α-ring opening to allow the protein to properly enter the proteolytic portion(1,2). Together, 20S and 19S form the 26S proteasome, a protein complex responsible not only for protein degradation but also cell-cycle control, genetic regulation, DNA repair and immune response(1). 

The proteasome is also involved in the regulation of memory(2-6): it is required to allow long-term memory formation(3,4) and reconsolidation(5), as well as mediate spine outgrowth(6), influencing synaptic plasticity. The UPS is therefore a highly investigated pathway in neurodegenerative disorders, such as Alzheimer’s disease (AD)(7-9). Other than memory loss and cognitive deficit, one of the many features of AD is chronic neuroinflammation, which can be observed in post-mortem samples and mouse models of AD(10). Yet, the link between proteasome activity, memory deficits and inflammation has not been well established to this date, which is exactly what this preprint aims to do. 

Key findings

Genetically disrupting the 26S proteasome reduces proteasome activity and impairs animal learning and memory 

The preprint authors evaluated proteasome activity in all three proteolytic subunits. All activities were decreased in Psmc1 KO mice when compared to the control group. During the Radial Arm Water Maze (RAWM) test (which evaluates memory), KO mice made more mistakes while trying to find the platform and spent more time searching for it. This result suggests that impairing 19S proteasome function causes learning and memory deficits in mice. 

26S proteasome deficiency causes accumulation of inflammation-associated proteins

Using mass spectrometry, Psmc1 KO mice showed alterations in immune system-related functions. Given that result, the authors investigated the expression of proteins associated with inflammation, namely STAT1, TREM2 and NF-kB, as well as glial activation markers GFAP and Iba-1. Psmc1 KO mice showed increased protein levels of all targets, suggesting that this knockout mouse in fact has a higher inflammatory burden. 

Inhibition of NF-κB improves learning and memory in Psmc1 KO animals

To investigate if NK-kB mediates the observed memory impairment, the authors injected the mice with PDTC, a NF-kB inhibitor, daily for three weeks. NF-kB inhibition was able to reverse the memory deficit caused by proteasome activity impairment after the 8th block of experiments. 

Inhibition of NF-κB improves 26S proteasome deficiency-caused neuroinflammation 

After injections with PDTC for three weeks, forebrain samples of Psmc1 KO treated mice were prepared and the protein content was analyzed by western blotting. The inhibition of NK-kB not only prevented the increase of inflammatory markers observed in Psmc1 KO but also reversed it, displaying an apparent general anti-inflammatory effect. 

 

Graphical abstract:

Figure 1: Graphical summary of the Huber et al. preprint showing the evaluated effects of Psmc1 KO in mice. Illustration made with Biorender. 

Why I think this preprint is important

This preprint links the cognitive deficit caused by forebrain-specific Psmc1 KO with inflammatory mediators, suggesting that the UPS is responsible at least partially for regulating inflammation. This finding is not only relevant to inflammation and memory in the forebrain, but also to other cognitive related areas and cellular functions. In addition, this is to our knowledge the first research project that used Psmc1 knock out animals to study memory loss, a classic AD symptom, linking this specific proteasome gene to an AD-like phenotype. 

Questions and suggestions

Q1: In the “Materials and Methods” section, it would be important to include the following details about the animals: the Ethics Committee protocol number, the sex of the animals, and their lineage and sub lineage (Enriquez et al., 2019). Additionally, it would be good to provide information about the temperature and general conditions of the animal facility. These details are crucial as they can significantly impact the results, particularly concerning the genetic background of both control and KO animals, which should be kept as similar as possible. 

Q2: In the first set of experiments, the authors measured proteasome activity in the forebrain of KO mice. Given the fact that RAWM is a cognitive test that evaluates spatial learning and areas such as the hippocampus or entorhinal cortex are more associated with this type of learning, measuring the proteasome activity in those brain regions could be more interesting when making this association with the RAWM results. Considering that the authors knocked out Psmc1 only in the forebrain (which is related to working memory),

perhaps a Y-maze would be a more suitable test, since it evaluates working memory and others, becoming a more suitable cognitive test(11). 

Q3: Still regarding the RAWM experiments, were the RAWM results shown in Figure 1 performed in parallel with the ones shown in Figure 4? If yes, it would be interesting to include the control group results in Figure 4 B and C to evaluate if the group performance on this test was different when compared to KO + PDTC. By doing this, the authors can discuss whether PDTC completely or partially reverts the memory impairment caused by Psmc1 KO. 

Q4: In the “Synaptosome isolation” method description, there is a typographical error in the first sentence (“tperformed”). 

Q5: In the description of the Mass spectrometric analysis of synaptic proteins, it is worth mentioning that this experiment is essentially a proteomic analysis in the “Material and methods” section. 

Q6: Could the authors check that all graphs contain the unit of the measurement in the y-axis? Most of the figures do not display “% of control” or “relative to control” 

Q7: As part of Figure 3, ELISA protocols to detect proinflammatory cytokines such as IL-1β or IFN-γ could be interesting to confirm the inflammatory state caused by Psmc1 KO since the synthesis and liberation of cytokines are downstream events that confirm the inflammatory state. Although TREM2 and STAT1 do suggest inflammation, additional data would help to support these results. TREM2, for example, is highly expressed in disease-associated microglia (DAM), but not necessarily a marker of inflammation(12). The same for Iba-1 and GFAP, which by themselves imply cell activation. Also, it appears that the antibody used in Figure 3D stains total NF-Kb. It would perhaps be more suitable to use p65 NF-kB, which stains the active form of NF-kB, or at least evaluate if this nuclear factor collocates with DAPI in an immunohistochemistry assay (if it is indeed in the nucleus), which could be another indication that NF-kB is in fact inducing inflammation. 

Q8: Currently, the immunohistochemistry legends only inform the reader that the experiments were performed in the cortex. It should perhaps be made more explicit what brain region was sliced and stained. Considering the experiments were done in total cortices, it is probably more suitable to stain only forebrain samples. 

Q9: In the PDTC injection experiments, shown in Figure 4, it would be beneficial to include “control” and “control + PDTC” groups to ensure that the inhibitor is reversing the effect of Psmc1 KO. PDTC could by itself provide a general improvement of memory and surely would decrease inflammatory markers evaluated in Figure 5, so the suggested additional groups can be included to strengthen the authors’ conclusions. In addition, although PDTC has been described as a selective NFkB inhibitor, it may have off-target effects that should be taken in consideration when discussing the results.

Q10: One additional suggestion is to repeat the western blotting experiments loading the samples used in Figures 3 and 4. By doing this, the authors will be able to compare more experimental groups, and this could provide additional data to enrich the discussion of whether PDTC reverts or attenuates the effects of Psmc1 KO in the pathways and inflammatory/reactivity markers evaluated in these experiments. 

Q11: The authors could consider including full western blotting membranes as Supplementary Figures to ensure that unspecific binding of antibodies does not take place in their experiments. This transparency increases the reliability of the data and avoids further questions. 

Q12: In the current manuscript, the authors frequently mention AD but none of the results displayed are directly related to AD models. To bring AD into the discussion, the authors could perform at least one experiment that links Psmc1 KO with AD. One relatively simple suggestion is to harvest astrocytes and microglia of wild type and KO animals in culture dishes and treat them with amyloid-β oligomers. The comparison of protein expression between these two groups would allow the authors to observe the effects of the KO in a widely used AD model. If this or similar experiments are not possible, it would perhaps be a good idea to rephrase the parts of the discussion that mention AD since memory loss is only one symptom of a much more complex disease. 

Q13: Recently, Psmc1 KO was correlated to a rare neurological syndrome named Birk-Aharoni Syndrome (BKAS)(13). This autosomal recessive syndrome causes phenotypic alterations such as impaired general development, intellectual disability, spastic tetraplegia, hearing loss, male genital alterations and elevation of liver enzymes. Do the Psmc1 KO mice have any phenotype similarities to the ones described by Birk and Aharoni? If yes, the results displayed in this paper might have additional contributions that go beyond AD. 

 

References: 

1. Murata S, Yashiroda H, Tanaka K. Molecular mechanisms of proteasome assembly. Nat. Rev. Mol. Cell Biol. 2009; 10:104–115. doi: 10.1038/nrm2630 
2. Ribeiro FC, Cozachenco D, Heimfarth L, Fortuna JTS, de Freitas GB, de Sousa JM, Alves-Leon SV, Leite REP, Suemoto CK, Grinberg LT, De Felice FG, Lourenco MV, Ferreira ST. Synaptic proteasome is inhibited in Alzheimer’s disease models and associates with memory impairment in mice. Commun Biol. 2023 Nov 7;6(1):1127. doi: 10.1038/s42003-023-05511-9. PMID: 37935829; PMCID: PMC10630330. 
3. Dong C, Upadhya SC, Ding L, Smith TK, Hegde AN. Proteasome inhibition enhances the induction and impairs the maintenance of late-phase long-term potentiation. Learn Mem. 2008 Apr 25;15(5):335-47. doi: 10.1101/lm.984508. PMID: 18441292; PMCID: PMC2364605.
4. Lopez-Salon M, Alonso M, Vianna MR, Viola H, Mello e Souza T, Izquierdo I, Pasquini JM, Medina JH. The ubiquitin-proteasome cascade is required for mammalian long-term memory formation. Eur J Neurosci. 2001 Dec;14(11):1820-6. doi: 10.1046/j.0953-816x.2001.01806.x. PMID: 11860477. 
5. Artinian J, McGauran AM, De Jaeger X, Mouledous L, Frances B, Roullet P. Protein degradation, as with protein synthesis, is required during not only long-term spatial memory consolidation but also reconsolidation. Eur J Neurosci. 2008 Jun;27(11):3009-19. doi: 10.1111/j.1460-9568.2008.06262.x. PMID: 18588539. 
6. Hamilton AM, Oh WC, Vega-Ramirez H, Stein IS, Hell JW, Patrick GN, Zito K. Activity-dependent growth of new dendritic spines is regulated by the proteasome. Neuron. 2012 Jun 21;74(6):1023-30. doi: 10.1016/j.neuron.2012.04.031. PMID: 22726833; PMCID: PMC3500563. 
7. Keller JN, Hanni KB, Markesbery WR. Impaired proteasome function in Alzheimer’s disease. J Neurochem. 2000 Jul;75(1):436-9. doi: 10.1046/j.1471-4159.2000.0750436.x. PMID: 10854289. 
8. Harris LD, Jasem S, Licchesi JDF. The Ubiquitin System in Alzheimer’s Disease. Adv Exp Med Biol. 2020;1233:195-221. doi: 10.1007/978-3-030-38266-7_8. PMID: 32274758 
9. Li Z, Jansen M, Pierre SR, Figueiredo-Pereira ME. Neurodegeneration: linking ubiquitin/proteasome pathway impairment with inflammation. Int J Biochem Cell Biol. 2003 May;35(5):547-52. doi: 10.1016/s1357-2725(02)00384-9. PMID: 12672447. 
10. Fakhoury M. Inflammation in Alzheimer’s Disease. Curr Alzheimer Res. 2020;17(11):959-961. doi: 10.2174/156720501711210101110513. PMID: 33509069. 
11. Kraeuter AK, Guest PC, Sarnyai Z. The Y-Maze for Assessment of Spatial Working and Reference Memory in Mice. Methods Mol Biol. 2019;1916:105-111. doi: 10.1007/978-1-4939-8994-2_10. PMID: 30535688. 
12. Deczkowska A, Keren-Shaul H, Weiner A, Colonna M, Schwartz M, Amit I. Disease-Associated Microglia: A Universal Immune Sensor of Neurodegeneration. Cell. 2018 May 17;173(5):1073-1081. doi: 10.1016/j.cell.2018.05.003. PMID: 29775591. 
13. Aharoni S, Proskorovski-Ohayon R, Krishnan RK, Yogev Y, Wormser O, Hadar N, Bakhrat A, Alshafee I, Gombosh M, Agam N, Gradstein L, Shorer Z, Zarivach R, Eskin-Schwartz M, Abdu U, Birk OS. PSMC1 variant causes a novel neurological syndrome. Clin Genet. 2022 Oct; 102(4):324-332. doi: 10.1111/cge.14195. Epub 2022 Aug 3. PMID: 35861243; PMCID: PMC9541193.

Tags: biochemistry, inflammation, memory, proteasome

doi: https://doi.org/10.1242/prelights.38193

Read preprint (No Ratings Yet)