🧠 neuroscience

Teneurins Are SPARCL1 Receptors

This study identifies teneurins as the specific receptors for SPARCL1, demonstrating that its C-terminal follistatin-like domain mediates binding to teneurins while its Ca2+-binding domain is required to trigger synapse formation.

Original authors: Zhang, X., Chen, X., Miao, Y., Sudhof, T. C.

Published 2026-07-15

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Zhang, X., Chen, X., Miao, Y., Sudhof, T. C.

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Imagine the brain as a bustling city where neurons are the buildings and synapses are the bridges connecting them. For a long time, scientists knew that a construction worker named SPARCL1 (also called Hevin) was crucial for building these bridges. But there was a massive disagreement in the scientific neighborhood about how this worker actually did the job.

Some researchers thought SPARCL1 was like a master key that unlocked a specific door made of two other proteins, Neurexins and Neuroligins. Others thought it was a different kind of tool entirely. The clues were contradictory, and the blueprint was missing.

In this new study, a team at Stanford decided to stop guessing and start taking the machine apart to see how it worked. Here is what they found, using a playful but strict look at the evidence.

The Magic in the Pocket

SPARCL1 is a long, floppy protein. It has a huge, messy tail at the beginning (the N-terminus) and a compact, organized tool kit at the end (the C-terminus). This tool kit has two specific parts: a Follistatin-like (FS) domain and a Calcium-binding (EC) domain.

The researchers asked: "Do we need the whole long, messy protein, or just the tool kit?"
They built tiny, purified versions of just the tool kit (the FS and EC domains) and dropped them into cultures of mouse brain cells.
The Result: The little tool kit worked just as well as the full-length protein! It boosted the number of synapses (bridges) by about 40-50%. Even better, the electrical activity of the neurons (the city's traffic) got 30-40% more frequent and synchronized.

This suggests that the long, messy tail isn't needed for the actual construction work. The magic is entirely in that small C-terminal tool kit.

The Great Reveal: It's Not the Key, It's the Magnet

Here is where the plot twists. The old theory said SPARCL1 was a key that fit into Neurexins and Neuroligins. The new team tested this by putting SPARCL1 in a room with Neurexins and Neuroligins.
The Result: SPARCL1 basically ignored them. At the concentration where it builds synapses (50 nM), it didn't stick to Neurexins or Neuroligins at all. The "key" didn't fit the "lock."

Instead, SPARCL1 found a different target entirely: Teneurins.
Think of Teneurins as a different set of magnets on the surface of the neurons. When the researchers tested SPARCL1 against all four types of Teneurins (Tenm1 through Tenm4), it grabbed onto them tightly. It was like a magnet finding its perfect metal match.

The Two-Step Dance

Now, the scientists had a new puzzle. They knew the FS part of SPARCL1 was the "magnet" that grabbed the Teneurins. But does the magnet alone build the bridge?
They tested the FS part all by itself.
The Result: The FS part grabbed the Teneurins perfectly, but it did not build any synapses. It was like a magnet that stuck to the wall but couldn't hang a picture.

Then they tested the EC part (the Calcium-binding part) alone.
The Result: It didn't stick to anything, and it didn't build anything.

The Solution: The two parts need to work together. The FS domain acts as the anchor, grabbing onto the Teneurins to hold SPARCL1 in place. Once it's anchored, the EC domain swings into action to actually trigger the construction of the synapse.

What This Means (and What It Doesn't)

The paper suggests a clear model: SPARCL1 is recruited to the synapse by its FS domain sticking to Teneurins, and then it uses its EC domain to say, "Okay, build a bridge here!"

However, the authors are careful to point out what they haven't proven yet.

They haven't proven this happens in a living mouse brain (in vivo), only in a dish (in vitro).
They haven't deleted all Teneurins in mice to see if SPARCL1 stops working, because making mice without all four Teneurins is currently too hard.
They don't know exactly how the EC domain triggers the building process, though they suspect it might involve collagen or the extracellular matrix.

So, while the old idea that SPARCL1 is a key for Neurexins is effectively ruled out by this study, the new idea that it's a Teneurin-anchored builder is a strong, evidence-backed hypothesis. It's a new blueprint for how the brain's construction crew operates, but the final inspection of the whole city is still waiting to happen.

Technical Summary: Teneurins Are SPARCL1 Receptors

Problem Statement
SPARCL1 (also known as Hevin) is a secreted protein primarily produced by astrocytes and vascular endothelial cells that potently enhances synapse formation. However, the molecular mechanism by which SPARCL1 acts at synapses remains unresolved due to conflicting literature. Previous studies have proposed divergent models: some suggest SPARCL1 binds to neurexins and neuroligins to bridge pre- and postsynaptic membranes, while others indicate that the C-terminal domains of SPARCL1 inhibit synaptogenesis. Furthermore, genetic deletion of neurexins or neuroligins fails to impair SPARCL1's synaptogenic activity, suggesting these may not be its primary receptors. The specific domains of SPARCL1 responsible for receptor binding and synapse induction, as well as the identity of its physiological receptor, remain unclear.

Methodology
The authors utilized primary mixed hippocampal neuron-glia cultures derived from newborn male and female mouse embryos to investigate SPARCL1 function.

Protein Engineering: Recombinant proteins were generated, including full-length SPARCL1, a C-terminal fragment containing only the follistatin-like (FS) and Ca2+-binding (EC) domains (SPARCL1FS-EC), and isolated FS and EC domains.
Synaptogenesis Assays: Cultures were treated with 50 nM of these proteins. Synapse density and size were quantified via immunocytochemistry for presynaptic (vGluT1) and postsynaptic (Homer1) markers, alongside dendritic (MAP2) staining. Network function was assessed using Ca2+ imaging (GCaMP6m) to measure firing frequency, amplitude, and synchronicity.
Binding Assays: Cell-surface binding assays were performed on HEK293T cells expressing various isoforms of neurexins, neuroligins, and teneurins (Tenm1–4). Binding was visualized using immunocytochemistry at nanomolar concentrations (50 nM).
Affinity Measurements: Binding affinities ( $K_d$ ) were determined for full-length SPARCL1 and the SPARCL1FS-EC fragment against teneurin-1 (Tenm1) expressed on cell surfaces.
Bioinformatics: Single-cell RNA sequencing data from the Allen Brain Institute were analyzed to map the expression profiles of SPARCL1 and SPARC across brain cell types.

Key Results

C-Terminal Domains Drive Synaptogenesis: Contrary to previous findings suggesting the N-terminal region is essential, the study demonstrates that the C-terminal FS and EC domains (constituting ~35% of the sequence) are sufficient to increase synapse density by 40–50% and enhance network synchronicity, mirroring the effects of full-length SPARCL1.
Identification of Teneurins as Receptors: At synaptogenic concentrations (50 nM), SPARCL1 showed no robust binding to neurexins or neuroligins. Instead, it avidly bound to all four teneurin isoforms (Tenm1–4). The binding affinity for Tenm1 was approximately 73 nM for full-length SPARCL1 and 88 nM for the FS-EC fragment.
Domain-Specific Functions:
- FS Domain: The isolated follistatin-like (FS) domain alone binds strongly to all teneurins but fails to stimulate synapse formation.
- EC Domain: The isolated Ca2+-binding (EC) domain does not bind teneurins and cannot induce synapses.
- Synergy: Synaptogenic activity requires the combination of both domains. The FS domain mediates recruitment to teneurins, while the EC domain is necessary for the induction of synapses.
Expression Context: Single-cell RNA-seq analysis confirms SPARCL1 is highly expressed in astrocytes but also abundantly in endothelial/perivascular cells and specific neuronal subclasses, distinguishing its expression pattern from SPARC.

Key Contributions

Resolution of Mechanistic Discrepancies: The study resolves the conflict regarding SPARCL1's binding partners by demonstrating that at physiologically relevant nanomolar concentrations, SPARCL1 binds teneurins rather than neurexins or neuroligins.
Domain Mapping: The authors delineate the distinct roles of SPARCL1's domains: the FS domain acts as the teneurin-binding module, while the EC domain acts as the synaptogenic effector.
Receptor Identification: The paper identifies presynaptic teneurins as the high-affinity receptors for SPARCL1, providing a mechanistic explanation for SPARCL1's ability to induce synapses independent of neurexin/neuroligin complexes.

Significance Statement
The authors claim that their results establish a model wherein SPARCL1 is recruited to developing synapses via the interaction of its follistatin-like domain with presynaptic teneurins. Once localized, the Ca2+-binding domain of SPARCL1 induces synapse assembly. This finding redefines the understanding of SPARCL1's mode of action, suggesting it functions as a teneurin-dependent synaptogenic factor rather than a neurexin/neuroligin modulator. The study also highlights that while the N-terminal sequence of SPARCL1 is largely unstructured and non-conserved, the conserved C-terminal domains are sufficient for its potent synaptogenic activity, offering a refined target for understanding synaptic development and potential therapeutic interventions.

The Magic in the Pocket

The Great Reveal: It's Not the Key, It's the Magnet

The Two-Step Dance

What This Means (and What It Doesn't)

More like this