Functional synapses between small cell lung cancer and glutamatergic neurons

Introduction

Arising from pulmonary neuroendocrine cells of the lung (PNECs), small cell lung cancer (SCLC) is a malignancy featuring loss of tumour suppressor genes TP53 and RB1^1-3. Despite this near-universal genomic signature, single-cell transcriptional analyses of patient samples have revealed extensive intratumoral heterogeneity, which contributes to the transient and variable responses to current therapies such as chemotherapy, immune checkpoint blockade, and radiation⁴. Unfortunately, patients with SCLC have a very poor prognosis with a median overall survival of just one year⁴.

Cancer neuroscience is an emerging field that explores the interplay between nerves and tumour development. Ground-breaking studies in central nervous system (CNS) tumours like glioma have revealed extensive crosstalk mediated by the formation of synapses between brain tumour cells and neurons. Specifically, glioma cells expressing AMPA glutamate receptors receive excitatory inputs from pre-synaptic neurons, boosting tumour cell proliferation and malignant progression^5-7. Since current standard of care treatments are ineffective against SCLC in the long term, targeting the neuron-tumour interface represents a potential alternative strategy.

Using a genome-wide insertional mutagenesis screen in mouse models and analyses of human patient datasets, the authors discovered an enrichment of neuronal, synaptic, and glutamate signaling-related genes in SCLC. Through co-culture and patch-clamp electrophysiology experiments, the authors revealed the existence of glutamatergic synapses between neurons and SCLC cells. They further demonstrate that inhibiting glutamatergic signaling in vivo attenuates tumour growth, paving the way for future studies aiming to develop new treatment strategies for SCLC. Importantly, this preprint is the first to identify the formation of bona fide synapses between neurons and a non-CNS tumour type.

Key Findings

Genome-wide mutagenesis screen and cross-species genomic analysis reveals enrichment of synapse-related genes in SCLC

To find genes that are important for SCLC development, the authors performed a piggyBac transposon mutagenesis screen in an established mouse model of SCLC (Ad-CMV-Cre;Rb1^fl/fl;Tp53^fl/fl, also known as RP mice)^{8, 9}. After extracting and sequencing DNA from over 300 murine tumours, the authors found recurrent insertions in genes that are known to contribute to SCLC tumorigenesis, such as the tumour suppressor genes Crebbp and Pten. Unexpectedly, there were recurrent insertions in neuronal and synaptic genes such as Nrxn1 (neurexin-1), Nlgn1 (neuroligin-1), and Reln (reelin). Furthermore, gene ontology (GO) analysis of whole-genome (WGS) and whole-exome sequencing (WES) datasets composed of 456 human SCLC samples revealed a surprising enrichment of genes involved in synaptic membranes, glutamatergic synapses, and other neuronal functions. To check whether the identified gene sets are highly expressed in SCLC, the authors compared transcriptomes from previously existing SCLC datasets^{3, 10} and healthy tissues. Encouragingly, they found that many of the synaptic and neuronal genes identified by the piggyBac screen and human WGS dataset were highly expressed in SCLC, prompting the authors to explore the novel role of synaptic genes in lung cancer.

SCLC cells have functional glutamatergic synaptic inputs from neurons which bolster proliferation

After identifying aberrations in neuronal and synaptic genes in SCLC, the authors wanted to functionally validate their findings. Thus, they co-cultured human SCLC cells with mouse cortical neurons and observed a striking enrichment of neuronal axons expressing the presynaptic marker VGLUT1 in contact with clusters of tumour cells. To determine whether these synaptic-like structures were functional, the authors performed retrograde tracing experiments using a modified rabies virus expressing GFP (RABV-GFP) that only infects cells with the viral receptor TVA and can only be transmitted through synapses by cells expressing glycoprotein G¹¹. GFP+ neurons were often observed surrounding TVA+/G+ SCLC cells in co-culture, indicating retrograde RABV transmission from “post-synaptic” SCLC cells to “pre-synaptic” neurons.

During electrophysiological patch-clamp recordings, SCLC cells exhibited spontaneous post-synaptic currents (sPSCs) when co-cultured with neurons but did not exhibit any electrical activity in monoculture. Furthermore, sPSCs in tumour cells were silenced by different pharmacological agents such as CNQX and AP5 (which antagonize AMPA and NMDA receptors, respectively) or riluzole, a glutamate release inhibitor. Importantly, these experiments indicate that SCLC tumour cells can communicate with neurons through glutamatergic synapses.

What impact does tumour-neuron crosstalk have on tumour cell growth? To answer this question, the authors compared the proliferation of SCLC monocultures to SCLC-neuron co-cultures. In EdU labelling and cell counting experiments, the authors consistently observed that a greater proportion of SCLC cells were EdU+ in co-cultures and exhibited an increased proliferative rate compared to tumour cells in monocultures. Thus, neuron-SCLC crosstalk appears to grant a growth advantage to tumour cells.

Above: immunofluorescent staining reveals the presence of synapse-like structures between mouse cortical neurons (marked by MAP2 in green) and human SCLC cells expressing DsRed. Presynaptic boutons are marked by VGLUT1 in white.

SCLC cells form synaptic connections with neurons in vivo

To determine whether SCLC tumour cells can form synapses with neurons in vivo, the authors examined lung tumour tissue from RP mice. Using antibodies against neuronal markers, they observed that nerve fibers were present within the tumour body of small tumours but remained mainly at the border of larger lesions. Biopsies taken from human advanced stage SCLC patients resembled large tumours from the genetic mouse model, with neurofilaments and synapses mainly present at the tumour border.

To validate the functionality of neuron-tumour synapses in vivo, the authors used an allograft model where mouse SCLC cells were engrafted into the hippocampus of GFP-expressing recipient mice. After confirming that transplanted tumour cells developed contacts with surrounding neurons through immunostaining and confocal microscopy, the authors recorded the electrical activity of tumour cells in acute brain slice preparations via whole cell patch clamp. Consistent with their in vitro studies, the authors detected sPSCs in a subset of the tumour cells and found that treating the brain slices with glutamate receptor antagonists abolished this electrical activity. Together, this set of experiments demonstrate that SCLC cells can form synaptic connections with neurons in vivo.  

Pharmacological inhibition of glutamatergic signaling demonstrates efficacy in treating SCLC in mice

Given their previous results, the authors wondered whether disrupting glutamatergic crosstalk between neurons and tumour cells could curb tumour growth in the lung. Thus, the authors treated RP mouse models of SCLC with riluzole, or dicarboxyphenylglycine (DCPG), a selective agonist of GRM8 (an inhibitory metabotropic glutamate receptor identified by the piggyBac screen). Compared to vehicle-treated mice, DCPG- and riluzole-treated mice exhibited slower tumour progression and even tumour shrinkage in some cases. Furthermore, disruption of glutamatergic signaling with DCPG and riluzole prolonged the overall survival of treated mice.

To mimic the standard treatment given to SCLC patients, the authors treated tumour-bearing mice with cisplatin and etoposide. While standard treatment alone resulted in a mixed response as expected, combining riluzole with standard chemotherapy greatly extended survival. Importantly, these experiments demonstrate that glutamatergic inhibitors negatively impact tumour growth on their own and in combination with existing frontline chemotherapies for SCLC.

 Why I chose this preprint

I first became interested in cancer neuroscience after reading pioneering studies from Michelle Monje, Frank Winkler, and Thomas Kuner’s labs. Their labs showed for the first time that brain tumour cells could integrate into neural circuits and use their relationships with neurons to fuel their malignant growth. While it has long been observed that different types of tumours are innervated by peripheral nerves (such as in breast, prostate, and pancreatic cancers), this preprint is the first to demonstrate the formation of functional synapses between neurons and a non-CNS tumour. I believe that this preprint clearly demonstrates the need to investigate crosstalk that occurs between tumours and their healthy neighbours and may inspire other researchers who are working on highly heterogeneous tumours to look outward for new treatment strategies.

Questions for the authors

1. This study highlighted glutamatergic signaling from neurons to SCLC cells, but do the tumour cells provide anything for the neurons? Do they secrete factors or have other signals that recruit neurons to fuel their growth?
2. Do PNECs exhibit crosstalk with neurons during normal lung development, or is this type of behaviour only observed in transformed cells?
3. Is it possible to validate the functionality of the synapses in the autochthonous model of SCLC?

References

1. Hockman, D. et al. Evolution of the hypoxia-sensitive cells involved in amniote respiratory reflexes. Elife 6, doi:10.7554/eLife.21231 (2017).
2. Kuo, C. S. & Krasnow, M. A. Formation of a Neurosensory Organ by Epithelial Cell Slithering. Cell 163, 394-405, doi:10.1016/j.cell.2015.09.021 (2015).
3. George, J. et al. Comprehensive genomic profiles of small cell lung cancer. Nature 524, 47-53, doi:10.1038/nature14664 (2015).
4. Dingemans, A. C. et al. Small-cell lung cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up(☆). Ann Oncol 32, 839-853, doi:10.1016/j.annonc.2021.03.207 (2021).
5. Venkatesh, H. S. et al. Neuronal Activity Promotes Glioma Growth through Neuroligin-3 Secretion. Cell 161, 803-816, doi:10.1016/j.cell.2015.04.012 (2015).
6. Venkatesh, H. S. et al. Electrical and synaptic integration of glioma into neural circuits. Nature 573, 539-545, doi:10.1038/s41586-019-1563-y (2019).
7. Venkataramani, V. et al. Glutamatergic synaptic input to glioma cells drives brain tumour progression. Nature 573, 532-538, doi:10.1038/s41586-019-1564-x (2019).
8. Meuwissen, R. et al. Induction of small cell lung cancer by somatic inactivation of both Trp53 and Rb1 in a conditional mouse model. Cancer cell 4, 181-189 (2003).
9. Rad, R. et al. PiggyBac transposon mutagenesis: a tool for cancer gene discovery in mice. Science 330, 1104-1107, doi:10.1126/science.1193004 (2010).
10. Rudin, C. M. et al. Molecular subtypes of small cell lung cancer: a synthesis of human and mouse model data. Nat Rev Cancer 19, 289 297,doi:10.1038/s41568-019-0133-9 (2019).
11. Wickersham, I. R. et al. Monosynaptic restriction of transsynaptic tracing from single, genetically targeted neurons. Neuron 53, 639-647, doi:10.1016/j.neuron.2007.01.033 (2007).

 

Posted on: 13 February 2023

doi: https://doi.org/10.1242/prelights.33630

Read preprint (No Ratings Yet)