Thesis defense Christian Martín Gil

All promotions can be followed live via this link: 
​https://www.uu.nl/en/organisation/utrecht-university-hall/schedule

Osteoarthritis (OA) is a highly prevalent joint disease and a major cause of disability in the aging population1. Although OA has traditionally been studied from a structural perspective - focusing on cartilage degeneration, osteophyte formation, and synovial changes - pain is the primary symptom leading patients to seek medical care2. Current treatments, including standard analgesics and non-steroideal anti-inflammatory drugs (NSAIDs) are inadequate to relieve OA pain effectively or induce adverse effects3-5. Moreover, pain often persists even if the damage knee has been replaced. This highlights the need to better understand the underlying biological mechanisms of pain in OA. Emerging evidence indicates that the immune is a major contributor to pain and is associated with maladaptive changes in both peripheral and central somatosensory pathways6-8. Immune cells, such as macrophages, infiltrate and accumulate in nervous tissue and regulate pain pathways in various models of OA, inflammatory and neuropathic conditions9-12. Moreover, activation of these nervous tissue immune cells, such as macrophages, neutrophils, and lymphocytes, but also glial cells, results in the release of proinflammatory mediators that play critical roles in the maintenance of pain6, 8, 10, 13-15. However, despite evidence that macrophages contribute to chronic pain in different conditions, including in osteoarthritis, the mechanisms by which they are recruited and adopt a pain-promoting phenotype remain poorly understood. Furthermore, it is unclear whether findings from animal models translate to human OA, and whether immune profiles—particularly in cerebrospinal fluid (CSF) or synovial tissue—reflect neuroimmune changes linked to pain. This thesis aimed to address these gaps by investigating how dorsal root ganglia (DRG) macrophages contribute to pain in experimental OA, and by exploring proteomic profiles in CSF and synovial tissue from OA patients to identify immune-related markers associated with pain.

In chapter 2, we investigated whether macrophages contribute to the regulation of persistent OA pain. We discovered that, over the progression of OA, macrophages accumulate in the DRG that contain the sensory neurons that innervate the affected knee. These macrophages had an inflammatory, M1-like, phenotype. Notably, these DRG macrophages are required for pain persistence in this mouse model of OA. In vitro experiments showed that DRG sensory neurons direct macrophages toward a pain-promoting M1-like phenotype. Nav1.8 sensory neurons - nociceptors that mediate inflammatory pain - however were unexpectedly not necessary for the persistence of OA pain nor the accumulation of M1-like macrophages in the DRG. This suggests that other types of sensory neurons are responsible for programming these macrophages. Lastly, suppressing pain-promoting DRG macrophages through intrathecal administration of either M2 macrophages or a novel fusion protein combining the cytokines IL4 and IL10 (IL4-10 FP) reduced persistent OA pain and the amount of M1-like DRG macrophages.

To gain a deeper understanding of what leads to the accumulation and programming of macrophages in the DRG, we investigated potential factors responsible for these events. In chapter 3, we identified myostatin and CXCL11 as key contributors to both the accumulation and programming of DRG macrophages in OA. Our findings revealed that although macrophages accumulate in the DRG early during the onset of OA, they are not essential for the initial development of pain, but only maintenance of pain. We discovered that myostatin is produced by sensory neurons, whereas CXCL11 is expressed by a subset of satellite glial 135 Chapter 6 cells. Whilst inhibition of both factors, prevented the development of persisting OA pain, CXCL11 neutralization reduced the number of DRG total macrophages, whilst myostatin inhibition diminished only the programming of F4/80+iNOS+ DRG macrophages. Whilst myostatin, alone was sufficient to drive macrophages into an M1 phenotype and induce pain-associated behaviors, CXCL11 did not. Based on these results, we concluded that a combination of factors mediates the interaction between neurons and macrophages for the maintenance of OA pain. CXCL11 facilitates the accumulation of DRG macrophages, and myostatin predominantly plays a critical role in programming them toward a pain-sustaining, pro-inflammatory phenotype. Finally, we identified that inhibiting myostatin alleviated chronic OA pain in mice, representing an unexpected therapeutic target for the treatment of chronic OA pain.

To further better understand driving factor of human OA , subsequent chapters focus on human samples to uncover immune-related mediators in OA pain. In chapter 4, we investigated potential biomarkers, which associate with pain severity and phenotype, in the CSF of patients with osteoarthritis (OA). Our analysis identified two distinct patient clusters based on the expression patterns of 274 unique proteins in CSF, with CA1 (Carbonic Anhydrase)1and 4E-BP1 (Eukaryotic Translation Initiation Factor 4E-Binding Protein 1) emerging as key distinguishing markers. Interestingly, among the cluster exhibiting elevated levels of these proteins, female patients demonstrated better daily functional outcomes and improved WOMAC scores, whereas no significant differences were observed between the two clusters in male patients. Through a supervised approach, we further uncovered molecular differences between patients experiencing moderate to severe pain at rest and those reporting minimal pain at rest, as well as between individuals with and without neuropathic pain-like symptoms. This analysis revealed associations between pain severity and several proteins, including CDH3 (Cadherin 3), KYNU (Kynureninase), FGF-19 (Fibroblast Growth Factor 19), and CCL23 (C-C Motif Chemokine Ligand 23). In contrast, the presence of neuropathic-like pain features was linked to increased levels of NAAA (N-Acylethanola mine Acid Amidase) and CXCL9 (C-X-C Motif Chemokine Ligand 9). Overall, these findings suggest that CSF protein profiles hold promise not only for classifying OA patient subtypes but may also help to provide insights into the molecular mechanisms underlying distinct pain phenotypes in OA.

Finally, we aimed to identify potential therapeutic targets within the synovial knee tissue that contribute to osteoarthritis (OA) pain. Our findings in chapter 5 revealed that both transcript levels and protein expression of several complement factors were significantly elevated in the synovial tissue of OA patients with moderate to severe pain, compared to those reporting mild or limited pain. To investigate the functional relevance of these observations, we employed OA models in mice deficient for C5aR1 and C2. These models demonstrated that these complement components contribute to the persistence of pain-related behaviors in OA. Importantly, treatment with anti-C2 IgG not only reduced free C2 levels but also alleviated established OA pain in two distinct mouse strains and OA models. Collectively, these results suggest that targeting C2 may offer a promising therapeutic approach for the treatment of chronic OA pain. Altogether, this thesis uncovers critical immune mechanisms underlying OA pain, highlighting the pivotal role of macrophages in the DRG and complement as key modulators of pain maintenance. We demonstrated that neuron- and glia-derived factors such as myostatin and CXCL11 orchestrate the accumulation macrophages in the DRG to 136 Summary & General discussion sustain chronic OA pain. Importantly, not only the accumulation of macrophages in DRG is sufficient, as macrophages needed to acquire an inflammatory M1 like phenotype that was driven by myostatin. Importantly, in addition this work bridges mechanistic insights from preclinical models to human patients by identifying CSF protein signatures associated with pain severity and distinct pain phenotypes, emphasizing the complexity and heterogeneity of OA pain. Furthermore, the discovery of elevated complement components in synovial tissue linked to persistent pain, along with evidence that targeting complement factor C2 alleviates established OA pain in vivo, points to a novel and promising new paths for treatment of pain in OA. Collectively, these findings provide a comprehensive framework for understanding the neuro-immune crosstalk in OA pain and highlight novel molecular targets that may enable more effective, personalized treatments for chronic OA pain.