The project aims to develop and validate advanced microphysiological models for studying the efficacy of combined pharmacological treatments in the context of active ageing. The focus will be on sarcopenia associated with chronic kidney disease (CKD). We have established in vitro models to simulate both physiological and pathological intestinal environments. Using these platforms, we investigated the impact of systemic inflammatory insults on intestinal permeability by measuring trans-epithelial electrical resistance (TEER). The results showed a significant decrease in barrier integrity under the tested conditions.

In the next phase, we will evaluate the therapeutic potential of SGLT2 inhibitors (glyphozines), assessing their ability to restore intestinal integrity and modulate inflammation and oxidative stress. These compounds, which are primarily used to treat diabetes and CKD, could represent a novel therapeutic strategy for preventing sarcopenia by acting on key metabolic and inflammatory pathways.

Our model mimics the 'leaky gut' conditions observed in CKD, providing a robust platform for the preclinical evaluation of personalised therapeutic strategies for ageing populations.

Context: Sarcopenia, a progressive decline in muscle strength, mass, and function, contributes significantly to disability in aging populations. Emerging evidence suggests a complex interplay between skeletal muscle and the nervous system—the so-called “muscle-brain axis.” Biomarkers traditionally associated with neurodegeneration, such as neurofilament light chain (NfL), GFAP, amyloid beta (Aβ40/42), and tau proteins (total tau and p-tau181), may also reflect processes linked to physical frailty.

Objectives: This study aims to assess the association between neurodegeneration biomarkers and sarcopenia components in older adults, and to explore whether these biomarkers can also predict functional decline, particularly in Activities of Daily Living (ADL) and Instrumental Activities of Daily Living (IADL).

Methods: The OPA study involves 200 older adults (65–90 years) recruited from geriatric, neurology, and internal medicine outpatient clinics at the University Hospital of Padua. Comprehensive assessments included physical (handgrip, chair test, ASMMI, calf circumference), cognitive (MMSE, MoCA), and functional (ADL, IADL) evaluations. Biomarkers were measured from blood samples. Follow-up interviews were conducted at six months to record functional changes, falls, hospitalizations, and death. Multivariate linear regression analyses were adjusted for age, sex, and diagnostic group (e.g., Parkinson’s, Alzheimer’s).

Results: Preliminary analyses from 143 participants (mean age: 74 years, 51.4% women) revealed that NfL was significantly associated with reduced muscle strength. Additionally, t-tau and p-tau181 levels were linked to worsening ADL scores at follow-up. GFAP, Aβ, and ASMMI were not significantly associated with sarcopenia parameters.

Implications: These findings suggest that neurodegeneration biomarkers, particularly NfL and tau, may serve as early indicators of musculoskeletal decline and functional deterioration. This reinforces the biological relevance of the muscle-brain axis and supports the integration of neurological and geriatric assessments to better predict and possibly intervene in age-related decline. Future directions include identifying inflammatory mediators and critical windows for rehabilitation.

With ageing, bone tissue undergoes profound changes, including bone mass reduction and skeletal fragility. Ageing also impacts bone marrow (BM) homeostasis, leading to decreased cellularity and alterations in hematopoietic cell compartments. Genetic mutations contribute to the development of bone disorders, including Paget’s Disease of Bone (PDB), a late-onset bone remodelling disease characterized by extensive bone erosion and dysregulated bone deposition by osteoblasts. Mutations in the ZNF687 and PFN1 genes have been identified in severe forms of early-onset PDB and are therefore implicated in bone remodelling. A missense mutation (P937R) in ZNF687 has been identified as responsible for a severe form of PDB associated with Giant Cell Tumor (GCT/PDB). A 4 bp deletion (c.318_321del) in the PFN1 gene has been identified in a severe form of PDB complicated by osteosarcoma. However, the specific effects of these genetic alterations on osteoclast biology and skeletal homeostasis are still poorly defined. Thus, the objective of this study was to investigate the roles of ZNF687 and PFN1 mutations in BM homeostasis, expanding the current knowledge of their contribution to PDB pathogenesis and skeletal ageing. To this aim, we generated and characterized three mouse models:

(i) a Zfp687 knockout (KO) model,

(ii) a knock-in Zfp687 model carrying the human P937R mutation,

(iii) and a knock-in Pfn1 harbouring the OS/PDB-associated deletion.

TRAP analysis performed on femur sections from 3-month-old KO mice revealed a reduced osteoclastogenic potential, as evidenced by decreased osteoclast surface/bone surface (Oc.S/BS) and osteoclast number/bone surface (N.Oc/BS) ratios (p = 0.0104; p = 0.0283). Microcomputed tomography (µCT) analysis of KO femurs at 8 months showed a 41% increase in trabecular bone volume (p = 0.0169) compared to wild-type (WT) controls. Interestingly, we also observed a marked reduction in BM cellularity in KO mice compared to WT controls (p = 0.0062), as well as a reduction in marrow area, as evaluated by µCT analysis (p = 0.0394). To assess whether the reduced osteoclastogenic potential was linked to changes in the composition of BM-derived myeloid cells, we performed flow cytometry analysis of BM osteoclast progenitors. KO mice showed significantly lower levels of these cells compared to WT mice (p = 0.0149) at 3 months of age. These data were paralleled by decreased macrophage levels (p = 0.0140) in 3-month-old KO mice, suggesting that ZNF687 modulates myeloid cells involved in osteoclastogenesis. Conversely, Zfp687^P937R/P937R mice showed elevated macrophage levels compared to WT (p = 0.0434), supporting previously published data linking this mutation to increased osteoclast activity and closely mirroring PDB features. To further investigate the mechanistic role of ZNF687 in hematopoietic lineage regulation, we performed single-cell RNA sequencing on hematopoietic progenitor stem cells. Preliminary analysis revealed a marked reduction in cluster numbers in KO samples, with 18 distinct cellular clusters identified in WT samples compared to only 13 in KO samples, suggesting that Zfp687 may modulate BM hematopoietic heterogeneity. Although the PFN1 mutation is implicated in PDB, Pfn1-KI mice exhibited only mild alterations in osteoclast progenitor and macrophage levels, suggesting a more nuanced and potentially indirect regulatory role in bone and BM homeostasis. Collectively, these findings uncover a pivotal role for ZNF687 in coordinating bone and bone marrow homeostasis, through the regulation of osteoclastogenesis and hematopoietic lineage specification. This highlights ZNF687 as a promising therapeutic target in PDB and skeletal disorders associated with ageing.

Aging and neuromuscular disorders significantly impact mobility, posture and overall quality of life. Sarcopenia, characterized by the progressive loss of muscle mass and strength, is among the most critical disease, affecting approximately 29% of the elderly population. State-of-the-art gold-standard diagnostic methods are limited by high costs, the need for specialized personnel and a lack of portability. To address these limitations, wearable sensor technologies have emerged as a promising alternative, enabling real-time assessment of muscle function and movement in both clinical and everyday environments. In particular, surface electromyography (sEMG) is well-suited for this purpose, offering direct insights into neuromuscular activity by detecting muscle activation patterns. Within Spoke 8 of Age-it Project, a portable system for sarcopenia diagnosis has been developed using commercial lightweight, wireless sEMG sensors combined with customized data acquisition software. Specific clinical protocols have been established in two healthcare facilities, located in Padua and Cremona, with the scientific and clinical support of the CNR-Institute of Neuroscience of Padua. These protocols focus separately on upper and lower limbs, providing complementary information on patients’ neuromuscular status. Designed to be quick and user-friendly, the data acquisition process can be carried out by non-specialized staff, promoting broader accessibility and potential cost reductions for the national healthcare system. Moreover, the portability and ease of use of the proposed system allow the application not only in hospitals and nursing homes but also in domestic settings, enabling continuous and frequent monitoring to support timely, personalized and adaptive therapeutic interventions.

The progressive ageing of the population intensifies the need for accessible tools to probe motor planning deficits via reach-to-grasp (RG) tasks. Traditional optical motion-capture systems, while accurate, are costly and spatially constrained. This study compares three established upper-limb musculoskeletal models—Holzbaur, Wu, and Seth—for reconstructing RG joint kinematics from wearable IMU data against a gold-standard marker-based system. Five healthy adults (27.8 ± 2.1 yr) performed isolated joint movements (shoulder adduction/flexion, forearm pronation-supination, elbow flexion) and two RG tasks (“place object into cup” and “pass object to partner”). Four IMUs sampled at 100 Hz and a 15-marker optical system at 200 Hz provided data input. IMU signals were combined with each established upper-limb model and compared with the marker data processed through the same models; resulting joint angles were evaluated via RMSE and R² against the optical standard. Performance varied by degree of freedom: Holzbaur achieved the lowest shoulder elevation RMSE (5.3°; R² = 0.65) and shoulder rotation RMSE (12.3°; R² = 0.59); Wu excelled in explained variance for elevation (R² = 0.81; RMSE 6.9°); Seth performed best in elbow flexion (RMSE 5.2°; R² = 0.77) but poorly in supination across all models. Models with explicit scapular DOFs (Wu, Seth) require a scapular IMU, but tracking scapular motion is challenging; in contrast, Holzbaur uses a regression equation linking scapular movement to shoulder elevation, simplifying the setup at the cost of longer kinematic reconstruction. Balancing accuracy, goodness-of-fit, and sensor simplicity, the Holzbaur model emerged optimal for IMU-based RG biomechanics. Its adoption may help research and clinical assessments of motor planning in aging populations.

Reading is a daily activity, yet it is a complex cognitive process that requires the eyes to move from left to right to extract meaning from text. With aging, several factors may impact reading efficiency, including declines in visual acuity and increased visual crowding. These age-related changes can alter the size of the visual span—the number of letters or words that can be processed in a single fixation—and may affect the quality of parafoveal processing, potentially slowing down reading and increasing reliance on contextual or semantic cues as compensatory strategies.
In this study, we explored how aging influences parafoveal and foveal word processing by
employing the Parallel Rapid Visual Presentation Paradigm (PRVPP; Primativo et al., 2022). This paradigm allows for the presentation of two words simultaneously—one in foveal and one in parafoveal region—thus providing a useful tool to disentangle age-related differences in reading.
We recruited a sample of younger adults (M age = 24.88 ± 2.52 years) and older adults (M age = 70 ± 2.1 years), ensuring normal or corrected-to-normal vision via preliminary optometric screening. Each participant read word pairs briefly presented (150 ms), in both semantically related and unrelated conditions. We measured response times and accuracy for both foveal and parafoveal words, as well as the individual visual span.

Our hypotheses were threefold: (1) older adults would show overall slower response times
compared to younger adults; (2) accuracy in foveal word recognition would be preserved, but parafoveal accuracy would decline with age; (3) semantic relatedness would facilitate word recognition in both age groups, possibly even more in older adults as a compensatory mechanism.
Preliminary findings show overall differences in speed and accuracy, with both groups exhibiting faster reading times and higher accuracy for semantically related pairs. Furthermore, visual span correlates with parafoveal accuracy in older adults. Our results suggest that aging affects some components involved in reading but not all, and preserved mechanisms can compensate for those that decline.
These results contribute to a growing body of evidence indicating that although aging impairs some perceptual and attentional aspects of reading, other mechanisms—such as semantic processing—remain intact or even become more prominent. Understanding the interplay between declining and compensatory processes in aging can inform the development of targeted interventions or reading aids for older adults, with implications for cognitive health and autonomy in late adulthood. Specifically, insights into the attentional mechanisms involved and the nature of semantic effects in older readers may guide the design of interventions aimed at promoting comfortable and efficient reading in healthy aging. These may include, for example, the development of video game-based tools to train oculomotor control, enhance visual attention, and strengthen compensatory strategies during reading.