17 mars 2025
Dolci et al. (BioRxiv)
Keywords
Tumor innervation
Cancer Neuroscience
Tumor-associated macrophages (TAMs)
Regeneration
Spinal cord injury
Osteopontin (OPN)/Secreted phosphoprotein 1 (SPP1)
mTORC2 signaling
Main Findings
Tumor-associated macrophages (TAMs) express a distinct “neural growth” gene signature and actively promote intratumoral nerve infiltration through secreted phosphoprotein 1 (SPP1), which activates neuronal mTORC2 signaling. Exploiting this neurotrophic capacity, adoptive transfer of in vitro-generated TAMs into a severe spinal cord injury model remodels the lesion’s ECM, improves tissue oxygenation, reduces chronic inflammation, and drives neural regrowth and partial motor recovery. Proteomic analysis and subsequent validation confirm that these reparative effects are mediated via mTORC2 signaling, revealing a previously unrecognized role for TAMs in tumor innervation and CNS repair.
1. TAMs express a unique “neural growth” gene signature.
By analyzing various publicly available single-cell and bulk RNA-sequencing datasets, the authors show that a pro-neurogenic gene signature is predominantly expressed by TAMs in multiple human cancers (including pancreatic, breast, endometrial, and colorectal) as well as in murine glioblastoma. This “neural growth” signature also appears in human and murine glioblastoma-associated microglia (GB-MG). To study TAMs in vitro, the researchers generate murine bone marrow-derived macrophages (mBMDMs) expressing tdTomato from 8–10-week-old reporter mice (B6.Cg-Gt(ROSA)26Sortm9(CAG-tdTomato)Hze/J). They then expose these mBMDMs to IL-4 to generate M2 macrophages, or to conditioned media from MN/MCA1 sarcoma cells to obtain TAMs. RNA-sequencing confirms that TAMs, but not M2 macrophages, upregulate several genes associated with neural growth.
2. TAMs enhance tumor innervation through Secreted Phosphoprotein 1 (SPP1).
In a murine MN/MCA1 sarcoma model, the intratumoral delivery of TAMs results in about a two-fold increase in nerve counts and NF200+ axons, indicating that TAMs promote tumor innervation. Additionally, in direct co-culture experiments, TAMs—but not M2 macrophages—stimulate neurite outgrowth in human iPSC-derived motor neurons (iPSC-MNs) and in mouse neurons generated from neural stem cells (mNSCs) or dorsal root ganglia (mDRGs). Further transcriptomic analysis identifies Spp1 as a gene consistently and highly upregulated in TAMs. By knocking down SPP1 (TAMsSPP1-KD) and using Parecoxib, which antagonizes the NR4A transcription factor essential for SPP1 expression, the authors show that SPP1 is necessary for the neurotrophic function of TAMs. These results are also validated in human in vitro-generated TAMs (hTAMs).
3. TAMs directly promote neurite outgrowth, axonal regrowth, and CNS repair in vivo.
Beyond their effects in vitro, TAMs support CNS regeneration in zebrafish. When administered to Tg(-3.1neurog1:GFP)sb2 zebrafish embryos after injury, TAMs reduce the damaged neural area, suggesting that TAMs promote axonal regeneration. SPP1 inhibition with Parecoxib, an antagonist of the NR4A transcription factor, further confirms the crucial role of SPP1 in TAM-mediated neuroprotection and regrowth.
4. TAMs facilitate regeneration, ECM remodeling, and immune modulation in spinal cord injury (SCI).
In an SCI mouse model, TAM treatment decreases the accumulation of extracellular matrix components (such as collagen and fibronectin) and yields smaller, fragmented cysts, implying that TAMs may help remodel the injured site. Adoptive transfer of TAMs also promotes angiogenesis, as shown by increased CD31+ vessel expression, reduced hypoxic regions, and formation of longer, more branched blood vessels. Additionally, TAM-treated mice display elevated numbers of Iba1+ and CD206+ microglia/macrophages, suggesting that TAMs contribute to shaping a tissue environment that supports regeneration.
5. TAM-induced neural regeneration depends on the RICTOR/mTORC2 pathway.
Proteomic profiling of TAM-treated SCI tissue reveals an upregulation of proteins related to neurons, synapses, and myelin. Rictor, a core component of the mTORC2 complex, is significantly upregulated following TAM administration. To determine the necessity of the Rictor pathway, the authors use a recombinant AAV9 vector to knock down Rictor in neurons (RictorKD), which in turn decreases downstream Rho family mediators. Rictor downregulation impairs locomotor recovery, negating the beneficial effect of TAMs on motor function, thus highlighting the importance of mTORC2 signaling in TAM-mediated repair.
6. Long-term safety analysis supports the therapeutic potential of TAMs in SCI.
To evaluate long-term safety, single and repeated intraparenchymal TAM injections were performed. After one year, no major issues—such as tumorigenicity, weight loss, or behavioral changes—were observed, and histopathological examinations indicated no significant abnormalities or malignancies. These findings suggest that TAM therapy may be both effective and safe over extended periods.
7. Validation of TAM pro-neurogenic activity in the human system.
The authors confirm key observations in human TAMs (hTAMs). Single-cell transcriptomics shows that hTAMs display the “neural growth” gene signature compared with M2 macrophages derived from human blood monocytes. When co-cultured with iPSC-MNs or SH-SY5Y-differentiated neuronal cells, hTAMs enhance neurite outgrowth more than human M2 macrophages. Pharmacological or RNA-based inhibition of Spp1 disrupts this effect, confirming that SPP1 is central to the neurotrophic role of hTAMs. Finally, inhibiting the mTORC2/Rictor pathway with JRAB2011 reduces hTAM-induced neurite growth, solidifying the requirement for Rictor signaling in TAM-driven neuronal outgrowth.
Limitations
Although the study offers strong evidence of TAM-mediated CNS repair, several questions remain.
The characterization of infiltrating macrophage and microglia in the SCI lesion after TAM transfer would benefit from additional markers such as CD49d/ITGA4 or lineage tracing. Moreover, using a combination of cell activation markers such as TREM2, CD163, CD80/86 would clarify whether these cells adopt anti-inflammatory and pro-regenerative phenotypes.
It would be interesting to compare the effects of direct SPP1 injections with TAM therapy, to understand whether SPP1 administration alone might be sufficient for neural regeneration, especially since endogenous macrophages already reside in injured tissues.
Data on TAM “tumor-associated” phenotype stability in vivo and/or whether continuous SPP1 supplementation might help maintain TAM function over time would be valuable.
The authors highlight increased expression of neurogenic genes in TAMs relative to monocytes and microglia; however, it would be important to also determine the effect of their distinct niches.
As tumor innervation can correlate with aggressiveness, it would be of interest to explore whether TAM-induced innervation affects tumor progression or metastatic spread. Also, the measure of “tumor volume” as an endpoint instead of “tumor area”, would reveal more subtle differences in tumor growth.
Significance/Novelty
The authors uncover a previously unrecognized role of TAMs in both tumor innervation and neural tissue repair, showing that TAMs orchestrate neural growth through SPP1 (Osteopontin), which activates the neuronal mTORC2/RICTOR pathway.
These findings indicate that SPP1 is produced by macrophages and microglia in certain malignancies, such as glioblastoma, potentially explaining the aggressive nature of highly innervated CNS tumors.
Importantly, TAMs’ pro-neurogenic effects translate into functional improvements in spinal cord injury models, with long-term safety data suggesting a viable strategy for tissue regeneration.
The ability of TAMs to reduce collagen and fibronectin deposition, thus remodeling the ECM, also hints at broader clinical applications, including the potential for treating fibrotic damage from radiation or other injuries.
This study’s year-long assessment of TAM therapy in SCI provides essential insight into both efficacy and safety, positioning TAMs—or possibly SPP1-based therapies—as promising candidates for regenerative interventions.
Credit
Reviewed by Austeja Baleviciute as part of the cross-institutional journal club of the Immunology Institute of the Icahn School of Medicine, Mount Sinai (U.S.A), The Center for Immuno-Oncology, University of Oxford (U.K.), Karolinska Institutet (Sweden), University of Toronto (Canada) and MD Anderson Cancer Center of University of Texas (U.S,A).