Astrocytes upregulate interleukin-15 in response to neuroinflammation during Alzheimer's disease

Abstract

Background

Interleukin (IL)-15 is a pleiotropic cytokine linked to cognition that is increased in the cerebrospinal fluid of Alzheimer's disease (AD) patients; however, the cellular source of IL-15 and IL-15-sensing cells in AD has not been described.

Objective

We sought to determine the cell types responsible for IL-15 secretion in AD and their spatial relationship with amyloid-β (Aβ) pathology.

Methods

We performed immunofluorescent labeling of human postmortem frontal cortex tissue from cognitively intact and dementia patients and used RNA-sequencing and in vitro assays to confirm the stimulus and source of IL-15. We also assessed the hippocampus of 6–7-month-old 5xFAD mice for cytokine levels using multiplex ELISA.

Results

In the prefrontal cortex of AD patients, we observed increased IL-15 colocalization with GFAP⁺ astrocytes (p = 0.0181) near Aβ plaques and demonstrate that astrocytes, but not neurons, upregulate IL-15 in response to interferon-γ and tumor necrosis factor-α in vitro (p = 0.0003 and p < 0.0001, respectively). We also show that neurons express increased IL2Rγ and IL2Rβ colocalization as Braak score increases (p = 0.0002) and that IL-15 is increased in the 5xFAD hippocampus compared to control animals (p = 0.0021).

Conclusions

Astrocytes are a source of IL-15 in AD, particularly near localized Aβ pathology, and neurons are poised to respond. The 5xFAD model recapitulates the hippocampal inflammatory milieu associated with AD and may be a useful tool to study the role of IL-15 in the context of Aβ pathology.

Keywords

Alzheimer's disease astrocytes cytokines neuroinflammation

Introduction

Alzheimer's disease (AD) has a complex etiology influenced by age, genetics, and environmental factors. Several studies have recently implicated immune system dysregulation as a contributor to AD incidence.¹ Interestingly, alterations in cytokine expression associated with immune dysfunction are correlated with a range of cognitive deficits seen across a variety of neurological disorders.² One such cytokine, interleukin (IL)-15, is a pleiotropic cytokine belonging to the common γ-chain (γc) family of cytokines.^3,4 IL-15 typically signals as a complex through IL-15 receptor (IL-15R)ɑ, either bound to the cell surface of transpresenting cells or as a cleaved soluble complex.^5,6 The most characterized intracellular pathways associated with IL-15 signaling are the JAK/STAT, PI3 K/AKT/mTOR, and MAPK pathways, which are responsible for a variety of critical cellular functions including cell survival, proliferation, differentiation, and metabolism.⁷

Genetic evidence suggests that IL-15 may be involved in regulating cognition as genetic mutations in IL-15 correlate with post-traumatic stress disorder severity,⁸ educational attainment,⁹ and AD biomarkers.^10,11 Studies in AD patients have shown that IL-15 levels in cerebrospinal fluid (CSF) are positively correlated with amyloid-β (Aβ) pathology¹² and CSF-associated biomarkers of AD including phospho-tau/Aβ₄₂ ratio, synaptosomal-associated protein 25, neurofilament light chain¹³ and total tau.¹⁴ Counterintuitively, Mini-Mental State Examination scores also have a positive correlation with IL-15, indicating better cognitive performance with increased IL-15, despite concomitant increases in biomarkers that indicate CNS pathology.¹³ Further, animal studies altering IL-15Rα signaling provide additional evidence for the potential importance of IL-15 in cognitive function as genetic deletion of IL-15Rα leads to deficits in learning and memory,¹⁵ increased anxiety-like phenotypes,¹⁶ and circadian rhythm dysfunction.¹⁷ These findings collectively underscore the significant role of IL-15 in modulating cognitive function.

Astrocytes are the most abundant glial cell in the CNS, executing a number of functions vital to the support and maintenance of CNS homeostasis.¹⁸ Such functions include modulation of the blood-brain barrier, regulation of ionic balance, neurotransmitter reuptake, and modulation of the synaptic environment.¹⁹ The heterogeneous nature of astrocytes allows them to reside on a spectrum of reactive states with the ability to engage in complex interactions with neighboring cells and the CNS microenvironment.²⁰ During periods of neuroinflammation, astrocytes are particularly sensitive to cytokines including tumor necrosis factor (TNF)α, interferon (IFN)γ, and IL-1β, all of which are readily expressed during AD.² In both postmortem human tissue and in mouse models, astrocytes are known to upregulate IL-15 during other neuroinflammatory conditions such as multiple sclerosis (MS),^21,22 ischemic stroke,²³ intracerebral hemorrhage,²⁴ and neuromyelitis optica.²⁵ Additionally, because IL-15 levels are elevated in the CSF of AD patients,^13,14 while serum IL-15 levels decline with age,²⁶ we hypothesized that IL-15 in AD originates from a CNS-specific source. We therefore sought to identify the CNS cell type(s) responsible for its production.

In this study, we investigated the source of IL-15 in the AD brain, specifically the prefrontal cortex. We first demonstrated that there is differential expression of IL-15 between cortical astrocytes and neurons depending on the spatial proximity to Aβ plaques. We went on to demonstrate that astrocytes produce IL-15 under neuroinflammatory conditions while cortical neurons express the receptor complex to initiate IL-15 signaling. We also show increased levels of cytokines, including IL-15, in the hippocampi of 5xFAD mice compared to controls, indicative of neuroinflammation and mirroring clinical findings in AD patients. Taken together, these data suggest astrocyte-neuronal crosstalk via IL-15 signaling during Aβ pathology.

Methods

Animals

The 5×FAD mice were obtained from The Jackson Laboratory (Stock No: 032883). Mice were housed four to five animals per cage and mice had access to food and water ad libitum. All procedures were approved by the Institutional Animal Care and Use Committee at the Cleveland Clinic in accordance with American Veterinary Medical Association guidelines.

Cell culture

Primary human hippocampal astrocytes were obtained commercially (ScienCell) and maintained at 5% CO₂ and 37°C for up to four passages before use. SH-SY5Y cells were obtained from the Cleveland Clinic Cell and Media Core and maintained in DMEM/F12 with 10% FBS and 50 U/ml penicillin, and 50 μg/ml streptomycin at 5% CO₂ and 37°C for up to five passages before use. Cells were stimulated with IL-1β, TNFα, IFNγ, IFNβ, IL-17, granulocyte-macrophage colony stimulating factor (GM-CSF) (PeproTech), or complete media as a control for 24 (RNA sequencing) or 48 h (ELISA).

ELISA assays

The multiplex ELISA assay is a custom UPLEX assay from MesoScale Discovery with assays for IL-1β, TNFα, IFNγ, IFNβ, IL-10, IL-15, GM-CSF, IL6, CCL3, and CXCL10. This assay was performed with 3.0–23.8 mg hippocampal tissue isolated from 5xFAD mice or littermate controls and homogenized in 4X volume of RIPA buffer (Thermo Fisher) with Halt protease and phosphatase inhibitor (Thermo Fisher). The assay was performed according to manufacturer's instructions by the Cleveland Clinic Laboratory Diagnostic Core Facility. Analytes were normalized to volume to account for differences in input. The IL-15 ELISA (R&D Systems) was performed with cell culture supernatants and normalized to number of cells per well at isolation. Cells were counted with trypan exclusion automatically using a Countess 3 (Thermo Fisher).

Immunofluorescence analysis of postmortem human tissue

Human postmortem frontal cortex tissue sections were FFPE-preserved and sectioned at 8 μm and were generously provided by the University of Kentucky Sanders-Brown Alzheimer's Disease Research Center. Clinical diagnoses of mild cognitive impairment (MCI) and dementia were determined by a consensus committee composed of neurologists and neuropsychologists following review of the available clinical data as described.²⁷ Subject demographics are provided in Table 1. Sections were deparaffinized and were then subjected to antigen retrieval in 10 μM citrate buffer at 95°C for 5 min (Decloaking Chamber NxGen, Biocare Medical). Sections were then permeabilized in PBS with 2% Triton-X100 for 30 min. Amyloid was stained using Amylo-Glo RTD (BioSensis) according to manufacturer's instructions then blocked with PBS with 5% normal goat serum and 0.3% Triton-X100 for 1 h. Sections were then incubated with appropriate primary antibodies (Table 2) and then with secondary antibodies (Invitrogen) conjugated to AlexaFluor 488, 594, or 647 for 1 h at room temperature. Sections were then washed and treated with TrueBlack lipofuscin quencher (Biotium) per manufacturer's instructions before mounting using Prolong Gold (Invitrogen). Sections were imaged using a Zeiss LSM800 confocal microscope or a Keyence BZ-X710 epifluorescence microscope. Anti-IL-15 antibody specificity was validated using a synthetic neutralizing peptide (Invitrogen) per manufacturer's instructions (Figure 1A and B).

Table 1.

Patient demographics.

Subject ID	Age at Death	Braak score	Sex	APOE	Last Clinical Diagnosis	Time Elapsed between Last Diagnosis and Death
R5515	100	5	Female	ε3/ε3	Dementia	4 years
R5513	68	6	Male	ε3/ε4	Dementia	10 months
R5511	84	5	Male	ε3/ε4	Dementia	13 months
R5508	91	5	Female	ε3/ε3	Dementia	6 years
R5505	75	6	Female	ε4/ε4	Dementia	3 years
R5504	92	5	Female	ε3/ε3	Dementia	9 months
R5516	97	2	Male	ε2/ε3	MCI	4 years
R5512	89	5	Male	ε3/ε3	MCI	21 months
R5510	88	3	Male	ε3/ε3	MCI	9 months
R5503	76	1	Male	ε2/ε3	MCI	11 months
R5491	91	3	Female	ε3/ε3	MCI	6 months
R5509	81	0	Male	ε2/ε3	Normal	4 months
R5506	83	1	Female	ε3/ε3	Normal	14 days
R5501	83	1	Male	ε3/ε4	Normal	9 months
R5500	89	1	Female	ε3/ε3	Normal	4 months
R5486	91	2	Male	ε3/ε3	Normal	10 months

Table 2.

Primary antibodies.

Target	Source	Catalog No.	Host	Dilution	Working Concentration
MAP2	Sigma	M2320	Mouse	1:1000	2 μg/ml
GFAP	Invitrogen	13–0300	Rat	1:250	2 μg/ml
Iba1	Aves Labs	IBA1-0100	Chicken	1:100	2 μg/ml
IL-15	Invitrogen	PA5-34511	Rabbit	1:100	10 μg/ml
IL2Rγ	Invitrogen	PA5-119907	Rabbit	1:100	10 μg/ml
IL2Rβ	Biolegend	123224	Rat	1:250	8.5 μ/ml

Image analysis

Images were analyzed using ImageJ v 1.54p. Colocalization was determined by Manders’ coefficient analysis using the BIOP Plugin.

RNA sequencing

Primary human hippocampal astrocytes were stimulated with individual cytokines for 24 h prior to RNA isolation (Qiagen). RNA samples were submitted to the Cleveland Clinic Genomics Core facility for library preparation (Illumina stranded mRNA library) and sequenced on a NovaSeq 6000 using an SP flow cell (Illumina) at a sequencing depth of 20,000,000 reads per sample. Reads were aligned to the human reference genome (GRCh38) using STAR aligner, and gene-level quantification was performed using RSEM. Differential gene expression analysis was conducted using the DESeq2 package in R. Genes with adjusted p-values (FDR) < 0.05 and absolute value of log₂ fold change > 1 were considered significantly differentially expressed. Data visualization, including principal component analysis, was performed using R. Raw data is available upon request.

Statistical analysis

Statistical analyses were performed using GraphPad Prism. Normally distributed data were analyzed by t test and the Jonckheere-Terpstra trend test was used for analysis of patient data in Figure 3F using the clinfun package in R.

Results

Astrocytes upregulate IL-15 near amyloid-β plaques

To determine which cell types express IL-15 during Aβ pathology, we obtained human postmortem frontal cortex tissue and performed immunofluorescent labeling for IL-15 colocalization analysis with glial fibrillary acidic protein (GFAP), microtubule associated protein (MAP)2, and ionized calcium binding adaptor molecule (Iba)1. We first validated the specificity of the anti-IL-15 antibody using a blocking peptide and observed no signal in the pre-adsorped sample, indicating that the antibody was indeed specific for the cytokine (Figure 1A and B). Since IL15 transcript is expressed by microglia during physiological conditions,²⁸ we first assessed IL-15 colocalization with Iba1 in patient tissue, but did not observe overlap with the CNS myeloid compartment (Figure 1C). This was confirmed and quantified via cell type-specific colocalization analysis (Supplemental Figure 1A-D). Notably, there was a significant upregulation in overall IL-15 expression in MCI and dementia patients relative to controls (Supplemental Figure 1E). We next evaluated IL-15 expression specifically in astrocytes (Figure 1D-G) and neurons (Figure 1H-K) within the dementia group. We further stratified our analysis by high powered fields either containing or without Aβ as previous reports have detailed differences in expression with regards to proximity to pathology during other neuropathological conditions, including MS and stroke.^21–23 Of note, there was no difference in IL-15 colocalization with either cell type when analyzing within or outside of 100 μm of Aβ (Supplemental Figure 2). We next evaluated IL-15 colocalization with either astrocytes or neurons in 10× fields that were either void of or contained Aβ (low and high Aβ fields were >320 μm apart). In fields lacking observable Aβ, we detected limited colocalization between IL-15 and GFAP (Figure 1D and G) but strong colocalization between IL-15 and MAP2 (Figure 1H and K). This indicates that in CNS regions with limited Aβ, neurons are the primary expressors of IL-15. In contrast, in fields with significant Aβ pathology, IL-15:GFAP colocalization increases (Figure 1E-G), while IL-15:MAP2 colocalization decreases (Figure 1I-K), as confirmed by Manders’ coefficient analyses (Figure 1G, K; p < 0.05 for both, t test). Together, this suggests that astrocytes in a relatively enhanced inflammatory microenvironment near Aβ plaques upregulate IL-15.

Astrocytes upregulate IL-15 in response to AD-relevant inflammatory factors

To confirm our finding that astrocytes upregulate IL-15 in response to inflammation, we turned to in vitro studies. We first stimulated primary human hippocampal astrocytes with cytokines known to be expressed across a variety of neurological disorders (IL-17, GM-CSF, IFNγ, IFNβ, IL-1β, or TNFα)² and performed bulk RNA sequencing. Whole transcriptome analysis revealed that astrocytes undergo substantial gene expression changes in response to IFNγ, IFNβ, IL-1β, and TNFα but not IL-17 or GM-CSF (Figure 2A). Limiting the analysis to interleukin genes, we observed a strikingly consistent upregulation of IL15 across the four cytokines which elicited a response and noted that it was one of the highest upregulated cytokines from each of the four stimuli (Figure 2B). Given that IL-15 production is known to be driven by cytokines including IFNγ²⁹ and TNFα,³⁰ but not by IL-1β,³¹ we next determined if IFNγ or TNFα exposure translated to secreted IL-15 protein. We stimulated either astrocytes or SH-SY5Y neurons with IFNγ or TNFα and performed an ELISA on cell culture supernatants. While primary human hippocampal astrocytes produced a robust IL-15 response to both stimuli, the SH-SY5Y neuronal cell line exhibited minimal IL-15 production in response to either IFNγ (Figure 2C; p < 0.001, t test) or TNFα (Figure 2D; p < 0.0001, t test). These data support our finding that astrocytes are the primary source of IL-15 during neuroinflammation.

Neurons express IL2Rγ and IL2Rβ and are poised to respond to IL-15 during AD

We next asked which cells were likely responding to increased IL-15 in the CNS during AD-associated neuroinflammation. IL-15Rα is rapidly cleaved into a soluble complex with IL-15 to enter circulation,³² where it can interact with cell surface IL2Rγ and IL2Rβ. Notably, the cytoplasmic domain of IL-15Rα is dispensable for IL-15 signal transduction, and the IL-2R heterodimer is sufficient for IL-15 signaling in the absence of IL-15Rα.³³ Although we found that similar to IL-15, IL-15Rα was induced by IFNγ and TNFα in vitro (Supplemental Figure 3), we focused on the IL-2R subunits that were responsible for IL-15 signaling on target cells during AD. We first determined if IL2Rγ expression was elevated in the MCI and dementia groups compared to controls. Since IL2Rγ upregulation appeared to be predominantly elevated in the dementia group (Supplemental Figure 4), we next determined the cellular source by labeling frontal cortex tissue from dementia patients for MAP2, GFAP, and IL2Rγ. This receptor was primarily observed on neurons (Figure 3A), decorating both the soma and proximal portions of the axon (Figure 3B). In comparison, minimal staining was observed on astrocytes (Figure 3C). We next demonstrated that neurons co-express both IL2Rγ and IL2Rβ, the IL-15 receptor heterodimer complex, in the frontal cortex (Figure 3D-E). We next wondered if IL-2R subunits colocalized as pathology accumulated, enabling neurons to more readily sense IL-15. Interestingly, we also observed that IL2Rγ and IL2Rβ colocalization increased as Braak score increased (Figure 3F; p = 0.0002, Jonckheere-Terpstra test). These data indicate that neurons are the predominant target for IL-15 in the CNS and coincides with previously published functional data indicating neurons undergo electrophysiological changes in response to IL-15³⁴ and exhibit decreased apoptosis in vitro upon IL-15 stimulation.³⁵

IL-15 levels are elevated in the 5×FAD hippocampus

Previous reports indicate that IL-15 is increased in patients with higher Aβ burden.^13,14 We further showed that astrocytes produce IL-15 under inflammatory conditions in vitro (Figure 2C-D). To substantiate our findings in vivo, we turned to the widely used 5xFAD murine model of AD. These mice overexpress the human amyloid precursor protein gene (APP) with three familial AD mutations as well as the human PSEN1 gene with two familial AD mutations, providing an Aβ model of AD in mice. Since the ability of neurons to respond to IL-15 increases with pathology (Figure 3F), we wondered if 5xFAD mice with prominent neurodegeneration and neuroinflammation had increased hippocampal IL-15 levels. After dissecting the hippocampus of 6–7-month-old mice, a region known to have high Aβ deposition, we performed a multiplex ELISA for cytokines relevant to AD. We found that TNFα (Figure 4A, p < 0.01, t test), IL-1β (Figure 4B, p < 0.05, t test), and IL-10 (Figure 4C, p < 0.05, t test) were upregulated in the 5xFAD hippocampus compared to controls, consistent with previous reports.^36,37 Importantly, we also observed robust upregulation of IL-15 in the 5xFAD hippocampus (Figure 4D, p < 0.001, t test), recapitulating clinical findings in AD patients.^13,14 Of note, we did not observe a sex difference in the levels of hippocampal IL-15 in these mice (Supplemental Figure 5) suggesting a generalizable role for IL-15 in vivo during Aβ-associated pathology.

Discussion

Studies using genetic knockout strategies in mice have demonstrated that abrogation of IL-15 signaling disrupts normal cognition, including deficits in learning and memory,¹⁷ depression,³⁸ and anxiety,³⁹ all of which are common morbidities associated with AD. Previous work in other neuroinflammatory diseases such as MS,^21,22 stroke,^23,24 and neuromyelitis optica²⁵ has demonstrated that IL-15 is upregulated during pathology, predominantly by astrocytes. Further, clinical reports have shown an increase in IL-15 in the CSF of AD patients;^13,14 however, the CNS intrinsic source of this elevated IL-15 during AD has not yet been confirmed. Here, we demonstrate that upregulation of IL-15 during AD-associated pathology is likely from astrocytes (Figures 1–2), with some contribution from neurons (Figure 1). We also show that the two receptor subunits necessary for sensing IL-15, IL2Rγ and IL2Rβ, are expressed predominantly by neurons in postmortem frontal cortex tissue of AD patients (Figure 3). Finally, these findings were recapitulated in the 5xFAD model of AD, revealing a robust increase in IL-15 levels in the hippocampi of affected mice compared to controls (Figure 4).

In this study, we show that astrocytes upregulated IL-15 near Aβ plaques in the prefrontal cortex of AD patients and confirmed this inflammation-induced increase in IL-15 expression in vitro both at the transcript and protein levels. Contrary to previous reports,^21–25 however, we also demonstrate that neurons are another source of IL-15 in human prefrontal cortex, particularly in areas with minimal observable Aβ pathology. Interestingly, neuronal expression of IL-15 decreases near Aβ plaques, opposing the astrocytic expression profile. The expression of IL-15 by neurons under homeostatic conditions may reflect a fundamental role in maintaining neuronal health and regulating neuroimmune interactions. Indeed, IL-15 is known to support the survival and maintenance of a protective, long-lived memory CD8⁺ T cell population for a rapid response following antigen exposure.⁴⁰ Since astrocytes are key CNS responders to inflammation and can orchestrate the immune response, the source of IL-15 may switch to astrocytes during pathology to preserve neurons similar to what has been observed during ischemic conditions.³⁵

We also show that during AD, adult human neurons express IL2Rγ and IL2Rβ, the two receptor subunits which form the functional receptor for the IL-15-IL-15Rα complex. Similarly, Nguyen et al. demonstrated that embryonic rat neurons expressed these receptors and that pretreatment with IL-15 protected cortical neurons from ischemic insult in vitro by preventing apoptosis.³⁵ This effect is likely mediated through the JAK2/STAT5 pathway, since IL-15 activates this signaling cascade—similar to erythropoietin⁴¹ and GM-CSF,⁴² both of which are also known to increase neuronal survival. The effect of IL-15 on neurons is likely multifactorial, however, as previous studies have also shown that IL-15 treatment of ex vivo brain slices led to acute electrophysiological changes.³⁴ Notably, we observed increased colocalization of IL2Rγ and IL2Rβ within MAP2⁺ neurons as Braak score increased. We hypothesize that as neuroinflammation and neurodegeneration accumulate, IL-2R subunits more readily colocalize enhancing the ability of neurons to sense IL-15 as a protective strategy.

Other reports using in vitro and murine models have shown that Aβ induces neuroinflammation, including the production of IL-1β and TNFα,^43–46 which we show here increased IL15 transcription. We used the 5xFAD mouse model to determine that, in addition to IL-1β and TNFα, hippocampal Aβ pathology also led to elevated IL-15. Given that IL-15 signaling loss impairs cognition in mice,^{15,34,38,39,47} yet elevated IL-15 was observed in pathological regions of the 5xFAD model, we hypothesize that IL-15 is protective. However, it is possible that a bell-shaped dose-response curve exists where IL-15 may be beneficial at moderate levels but detrimental when dysregulated.

In summary, we show that IL-15 is expressed by neurons in areas with a low Aβ burden but is downregulated in areas with overt Aβ pathology. In contrast, astrocytes in these pathological regions exhibited elevated levels of IL-15, likely driven by neuroinflammatory cytokines including IL-1β and TNFα. Notably, neurons in these areas express the receptors to sense this IL-15, which increases concomitantly with AD progression. We speculate that this IL-15 upregulation during neuroinflammation, particularly in the context of Aβ pathology, represents an adaptive response aimed at preservation of cognitive reserve.

A key strength of this study is the inclusion of human tissue, which allowed for the direct assessment of both the cellular sources of IL-15 and the responding cell types. However, several limitations should be considered. While there was a balance of ApoE genotypes and Braak score in the dementia patient group, additional patients would be beneficial in improving the statistical power of the study. Further, the MCI cohort was predominantly male, which may affect the generalizability of the findings. Additionally, although there was a range in Braak score within this group, the predominance of male samples from this subset of patients should be taken into account when interpreting the results. Further, we utilized undifferentiated SH-SY5Y cells for in vitro studies to overcome the lack of propagation potential of terminally differentiated primary neurons. The limitations of this cellular model in recapitulating neuronal function should be considered.

Supplemental Material

sj-docx-1-alz-10.1177_13872877251393275 - Supplemental material for Astrocytes upregulate interleukin-15 in response to neuroinflammation during Alzheimer's disease

Supplemental material, sj-docx-1-alz-10.1177_13872877251393275 for Astrocytes upregulate interleukin-15 in response to neuroinflammation during Alzheimer's disease by Benjamin C Shaw, Kaitlin E Kaiser, Brandon C Smith, J Payton Timken, Antoine Louveau and Jessica L Williams in Journal of Alzheimer's Disease

Footnotes

Acknowledgements

We thank Kimberly Peterson for assistance with the multiplex ELISA, Tom Jaramillo, PhD for assistance with behavioral testing, and Jihye Kim, PhD for assistance with RNA sequencing analysis. We also thank Pete Nelson, MD, PhD, Sonya Anderson, and the clinicians and research volunteers at the University of Kentucky Sanders-Brown Alzheimer's Disease Research Center for providing the postmortem Alzheimer's disease tissue and clinical data.

ORCID iD

Jessica L Williams

Ethical considerations

The ethics committee of the University of Kentucky waived the need for ethics approval and patient consent for the analysis and publication of the retrospectively obtained and anonymized data for this non-interventional study.

All procedures involving animal subjects were approved by the Institutional Animal Care and Use Committee at Cleveland Clinic Research (Approval No: 3430) on February 24, 2025, in strict accordance with American Veterinary Medical Association guidelines.

Author contribution(s)

Benjamin C Shaw: Conceptualization; Data curation; Formal analysis; Formal analysis; Investigation; Visualization; Writing – original draft; Writing – review & editing.

Kaitlin E Kaiser: Data curation; Formal analysis; Investigation; Writing – original draft; Writing – review & editing.

Brandon C Smith: Data curation; Formal analysis; Investigation; Writing – review & editing.

J Payton Timken: Data curation; Formal analysis; Investigation; Writing – review & editing.

Antoine Louveau: Investigation; Methodology; Resources; Writing – review & editing.

Jessica L Williams: Conceptualization; Formal analysis; Project administration; Resources; Supervision; Writing – review & editing.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by K00NS120365 (BCShaw), the Case Western Reserve University Dean's Scholars Program (BCShaw), Coins for Alzheimer's Research Trust (AL), R56AG079897 (AL), The Mayer Foundation (JLW), NMSS RFA-2203-39228 (JLW), and R01NS119178 (JLW). Human postmortem tissue was provided by the University of Kentucky Sanders-Brown Alzheimer's Disease Research Center funded by P30AG072946.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data availability statement

The data supporting the findings of this study are available on request from the corresponding author.

Supplemental material

Supplemental material for this article is available online.

References

Karch

Goate

. Alzheimer's disease risk genes and mechanisms of disease pathogenesis. Biol Psychiatry 2015; 77: 43–51.

Shaw

Anders

Tinkey

, et al. Immunity impacts cognitive deficits across neurological disorders. J Neurochem 2024; 168: 3512–3535.

Burton

Bamford

Peters

, et al. A lymphokine, provisionally designated interleukin T and produced by a human adult T-cell leukemia line, stimulates T-cell proliferation and the induction of lymphokine-activated killer cells. Proc Natl Acad Sci U S A 1994; 91: 4935–4939.

Grabstein

Eisenman

Shanebeck

, et al. Cloning of a T cell growth factor that interacts with the beta chain of the interleukin-2 receptor. Science 1994; 264: 965–968.

Chirifu

Hayashi

Nakamura

, et al. Crystal structure of the IL-15-IL-15Ralpha complex, a cytokine-receptor unit presented in trans. Nat Immunol 2007; 8: 1001–1007.

Xie

Jiang

Qin

, et al. Complement-activated interferon-gamma-primed human endothelium transpresents interleukin-15 to CD8+ T cells. J Clin Invest 2020; 130: 3437–3452.

Skariah

James

Swamy

. Signalling mechanisms driving homeostatic and inflammatory effects of interleukin-15 on tissue lymphocytes. Discov Immunol 2024; 3: kyae002.

Wang

Karstoft

Nievergelt

, et al. Post-traumatic stress following military deployment: genetic associations and cross-disorder genetic correlations. J Affect Disord 2019; 252: 350–357.

Okbay

Wang

, et al. Polygenic prediction of educational attainment within and between families from genome-wide association analyses in 3 million individuals. Nat Genet 2022; 54: 437–449.

10.

Sarnowski

Ghanbari

Bis

, et al. Meta-analysis of genome-wide association studies identifies ancestry-specific associations underlying circulating total tau levels. Commun Biol 2022; 5: 336.

11.

Wang

Yang

Schneider

, et al. Genome-wide interaction analysis of pathological hallmarks in Alzheimer's disease. Neurobiol Aging 2020; 93: 61–68.

12.

Janelidze

Mattsson

Stomrud

, et al. CSF Biomarkers of neuroinflammation and cerebrovascular dysfunction in early Alzheimer disease. Neurology 2018; 91: e867–e877.

13.

Daniilidou

Eroli

Alanko

, et al. Alzheimer's disease biomarker profiling in a memory clinic cohort without common comorbidities. Brain Commun 2023; 5: fcad228.

14.

Popp

Oikonomidi

Tautvydaite

, et al. Markers of neuroinflammation associated with Alzheimer's disease pathology in older adults. Brain Behav Immun 2017; 62: 203–211.

15.

Hsuchou

, et al. Interleukin-15 receptor is essential to facilitate GABA transmission and hippocampal-dependent memory. J Neurosci 2010; 30: 4725–4734.

16.

Nguyen

Bohlen

Stricker

, et al. Hippocampus-specific deficiency of IL-15Ralpha contributes to greater anxiety-like behaviors in mice. Metab Brain Dis 2017; 32: 297–302.

17.

Khan

, et al. IL-15 receptor deletion results in circadian changes of locomotor and metabolic activity. J Mol Neurosci 2010; 41: 315–321.

18.

Shaw

Williams

. A novel PSMB8 isoform associated with multiple sclerosis lesions induces P-body formation. Front Cell Neurosci 2024; 18: 1379261.

19.

Habean

Kaiser

Williams

. Orchestrating stress responses in multiple sclerosis: a role for astrocytic IFNgamma signaling. Int J Mol Sci 2024; 25: 7524.

20.

Escartin

Galea

Lakatos

, et al. Reactive astrocyte nomenclature, definitions, and future directions. Nat Neurosci 2021; 24: 312–325.

21.

Saikali

Antel

Pittet

, et al. Contribution of astrocyte-derived IL-15 to CD8T cell effector functions in multiple sclerosis. J Immunol 2010; 185: 5693–5703.

22.

Broux

Mizee

Vanheusden

, et al. IL-15 amplifies the pathogenic properties of CD4 + CD28- T cells in multiple sclerosis. J Immunol 2015; 194: 2099–2109.

23.

Yao

, et al. Astrocyte-derived interleukin-15 exacerbates ischemic brain injury via propagation of cellular immunity. Proc Natl Acad Sci U S A 2017; 114: E396–E405.

24.

Shi

, et al. IL (Interleukin)-15 bridges astrocyte-microglia crosstalk and exacerbates brain injury following intracerebral hemorrhage. Stroke 2020; 51: 967–974.

25.

Han

Ren

, et al. Astrocytic interleukin-15 reduces pathology of neuromyelitis optica in mice. Front Immunol 2018; 9: 523.

26.

Quinn

Anderson

Strait-Bodey

, et al. Serum and muscle interleukin-15 levels decrease in aging mice: correlation with declines in soluble interleukin-15 receptor alpha expression. Exp Gerontol 2010; 45: 106–112.

27.

Abner

Kryscio

Cooper

, et al. Mild cognitive impairment: statistical models of transition using longitudinal clinical data. Int J Alzheimers Dis 2012; 2012: 291920.

28.

Zhang

Sloan

Clarke

, et al. Purification and characterization of progenitor and mature human astrocytes reveals transcriptional and functional differences with mouse. Neuron 2016; 89: 37–53.

29.

Kim

Rha

Shin

. IFN-gamma induces IL-15 trans-presentation by epithelial cells via IRF1. J Immunol 2022; 208: 338–346.

30.

Ouyang

Hsuchou

Kastin

, et al. TNF Stimulates nuclear export and secretion of IL-15 by acting on CRM1 and ARF6. PLoS One 2013; 8: e69356.

31.

Cardozo

Proost

Gysemans

, et al. IL-1beta and IFN-gamma induce the expression of diverse chemokines and IL-15 in human and rat pancreatic islet cells, and in islets from pre-diabetic NOD mice. Diabetologia 2003; 46: 255–266.

32.

Bergamaschi

Bear

Rosati

, et al. Circulating IL-15 exists as heterodimeric complex with soluble IL-15Ralpha in human and mouse serum. Blood 2012; 120: e1–e8.

33.

Anderson

Kumaki

Ahdieh

, et al. Functional characterization of the human interleukin-15 receptor alpha chain and close linkage of IL15RA and IL2RA genes. J Biol Chem 1995; 270: 29862–29869.

34.

Di Castro

Garofalo

Mormino

, et al. Interleukin-15 alters hippocampal synaptic transmission and impairs episodic memory formation in mice. Brain Behav Immun 2024; 115: 652–666.

35.

Nguyen

Ameri

Huynh

, et al. Interleukin-15 modulates the response of cortical neurons to ischemia. Mol Cell Neurosci 2021; 115: 103658.

36.

Kantarci

Aytan

Palaska

, et al. Combined administration of resolvin E1 and lipoxin A4 resolves inflammation in a murine model of Alzheimer's disease. Exp Neurol 2018; 300: 111–120.

37.

Jian

Kwan

Bunting

, et al. Adiponectin suppresses amyloid-β oligomer (AβO)-induced inflammatory response of microglia via AdipoR1-AMPK-NF-κB signaling pathway. J Neuroinflammation 2019; 16: 110.

38.

Hsuchou

Kastin

, et al. Interleukin-15 affects serotonin system and exerts antidepressive effects through IL15Rα receptor. Psychoneuroendocrinology 2011; 36: 266–278.

39.

Hsuchou

, et al. Essential role of interleukin-15 receptor in normal anxiety behavior. Brain Behav Immun 2010; 24: 1340–1346.

40.

Castillo

Schluns

. Regulating the immune system via IL-15 transpresentation. Cytokine 2012; 59: 479–490.

41.

Chen

, et al. Erythropoietin rescues memory impairment in a rat model of chronic cerebral hypoperfusion via the EPO-R/JAK2/STAT5/PI3K/Akt/GSK-3β pathway. Mol Neurobiol 2018; 55: 3290–3299.

42.

Choi

Kim

Park

, et al. Granulocyte macrophage-colony stimulating factor shows anti-apoptotic activity in neural progenitor cells via JAK/STAT5-Bcl-2 pathway. Apoptosis 2011; 16: 127–134.

43.

Shi

Liang

Chen

, et al. Gx-50 reduces β-amyloid-induced TNF-α, IL-1β, NO, and PGE2 expression and inhibits NF-κB signaling in a mouse model of Alzheimer's disease. Eur J Immunol 2016; 46: 665–676.

44.

Akama

Van Eldik

. Beta-amyloid stimulation of inducible nitric-oxide synthase in astrocytes is interleukin-1beta- and tumor necrosis factor-alpha (TNFalpha)-dependent, and involves a TNFalpha receptor-associated factor- and NFkappaB-inducing kinase-dependent signaling mechanism. J Biol Chem 2000; 275: 7918–7924.

45.

Floden

Combs

. Beta-amyloid-stimulated microglia induce neuron death via synergistic stimulation of tumor necrosis factor alpha and NMDA receptors. J Neurosci 2005; 25: 2566–2575.

46.

Halle

Hornung

Petzold

, et al. The NALP3 inflammasome is involved in the innate immune response to amyloid-beta. Nat Immunol 2008; 9: 857–865.

47.

Petitto

Huang

Hartemink

, et al. IL-2/15 receptor-beta gene deletion alters neurobehavioral performance. Brain Res 2002; 929: 218–225.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

9.66 MB