Terminal Schwann Cells Regulate Presynaptic Vesicle Homeostasis but Not Neuromuscular Junction Integrity in Mice

Reviewer #1 (Public review):

Summary:

This manuscript's major strength is the identification of Col20a1 as a novel marker for terminal Schwann cells (tSCs) and the generation of the Col20a1-CreERT2 knock-in mouse line, which represents a valuable new genetic tool for studying tSC biology at the neuromuscular junction. However, the central conclusion that terminal Schwann cells regulate presynaptic vesicle homeostasis is not sufficiently supported because the electrophysiological and ultrastructural analyses are based on limited sample sizes. Increasing the number of animals and NMJs analyzed would substantially strengthen the conclusions. In addition, further validation of Col20a1 expression using RNAscope or immunostaining, together with comparisons to established tSC markers such as Kir4.1 and NG2, would help establish its specificity. Finally, the developmental appearance of Col20a1-positive axonal Schwann cells is intriguing and warrants further investigation to determine whether these cells migrate and differentiate into terminal Schwann cells during postnatal development.

Strengths:

The authors identified Col20a1 as a specific marker of terminal Schwann cells (tSCs) at the neuromuscular junction (NMJ) in mice and generated a Col20a1-CreERT2 knock-in mouse line for in vivo labeling of tSCs. This represents a novel and significant technical advance for the NMJ field, providing a valuable genetic tool for studying the development, maintenance, and function of terminal Schwann cells in vivo.

Weaknesses:

The major weakness of this manuscript is that the sample sizes are too small to support the authors' conclusion that "Terminal Schwann Cells Regulate Presynaptic Vesicle Homeostasis but Not Neuromuscular Junction Integrity in Mice."

Although the authors state that "Quantification of the tdTomato-positive NMJ ratio showed an ablation efficiency of approximately 80%, with only ~20% of NMJs retaining escaper tSCs" (page 9), they do not provide the sample size or sufficient quantitative information for the Col20a1-tdTomato/DTA ablation experiment shown in Figure 4. This information is essential for evaluating the robustness and reproducibility of the ablation strategy.

The electrophysiological analyses are also based on very limited sample sizes. According to Figure 6, mEPP recordings were obtained from 10 NMJs from 3 control mice and 14 NMJs from 5 tSC-ablated mice. The EPP recordings were based on similarly small numbers (control, n = 10 NMJs from 3 mice; tSC-ablated, n = 14 NMJs from 5 mice). Thus, only approximately 2-3 NMJs were analyzed per tSC-ablated mouse on average. Given the inherent variability among individual NMJs and animals, these sample sizes are insufficient to support broad conclusions regarding the effects of terminal Schwann cell ablation on synaptic transmission.

In addition, the authors describe the phenotype as "leaky" presynaptic spontaneous release, but this terminology is not defined and lacks mechanistic explanation. It is therefore unclear what specific physiological alteration the authors intend to describe.

The sample sizes for the electron microscopy analyses (Figure 7) also appear to be limited, making it difficult to determine whether the reported changes in synaptic vesicle distribution are representative or statistically robust. Because the central conclusion relies heavily on these electrophysiological and ultrastructural data, the evidence presented is not sufficient to support the claim that terminal Schwann cells regulate presynaptic vesicle homeostasis. At present, this conclusion is overly broad and not adequately supported by the available data.

Minor comments:

(1) While several figures contain high-quality NMJ images (e.g., Figures 1B and 4), the image quality in other figures should be improved. For example, the S100B immunostaining appears overexposed in some panels, making it difficult to distinguish individual Schwann cells or visualize the boundaries between adjacent cells.

(2) Some neuromuscular junctions shown in Figure 2C appear to be partially denervated. The authors should clarify whether these represent normal variability, effects of the experimental manipulation, or imaging artifacts.

(3) The authors should specify the muscle preparation used in Figure 3, as this information is necessary for interpreting the results and comparing them with previous studies.

Reviewer #2 (Public review):

Summary:

Two types of Schwann cells (SCs) ensheath motor axons - myelinating SCs along the axonal length and terminal SCs (tSCs) that cover nerve terminals at the neuromuscular junction (NMJ). Therefore, the NMJ is, like other synapses, tripartite, with specialized presynaptic, postsynaptic, and glial cells. Many studies have shown that tSCs play roles in the development and function of the NMJ, but for some of these, interpretation is difficult because it is hard to manipulate tSCs without also manipulating myelinating SCs. To circumvent this problem, Kong et al. make use of a gene selectively expressed in tSCs, Col20a1 (Figure 1), to generate a knock-in mouse line, Col20a1-CreER, that gives them genetic access to tSCs. They cross this to Cre-dependent lines that mark tSCs with a red fluorescent protein (Figures 2 and 3) or ablate them by expression of diphtheria toxin along with the fluorescent protein (Figure 4). They show that ablation at postnatal day (P) 10 does not affect the overall structure or function of the NMJ (Figures 4 and 5). It does, however, affect some aspects of neuromuscular transmission over the following few weeks (Figures 6 and 7). Long-term effects cannot be studied by this method, however, because terminal SCs are replaced, presumably from the preterminal population (Figure 8).

Strengths:

The work is done to a high technical standard, including detailed characterization of the knock-in model. Results are presented clearly and illustrated beautifully. The finding that some early reports of synaptic alterations may result from concurrent loss of axonal SCs is important in rethinking the role of tSCs.

Weaknesses:

(1) The authors claim that tSCs are dispensable for some aspects of NMJ maturation, including synapse elimination (called pruning here), formation of "pretzel-like" postsynaptic topology, and generation of junctional folds in the postsynaptic membrane (lines 223 and 363). However, this conclusion is based on injection of tamoxifen to initiate tSC ablation at P10, which is necessary because Col20a1 is expressed in some preterminal SCs at earlier times. It presumably takes a few days for CreER to translocate to the nucleus and activate the toxin transgene, and some more time for the toxin to be generated and act. This is problematic because synapse elimination and other aspects of maturation mentioned occur during the first two postnatal weeks and are largely complete by P14. Therefore, one cannot conclude that these aspects "proceeded normally despite the loss of tSCs....".

(2) Effects on synaptic transmission are modest at best, being significant at a level of p<0.05 but not p<0.01 (Figure 6E, G, H and most of L). Effects on vesicle density are more robust (Figure 7).

(3) The authors use red fluorescent protein from the Col20a1 to label tSCs, and antibodies to S100b to label all SCs. This is appropriate in normal muscle and soon after tSC ablation. At later times, however, the NMJ is repopulated by S100+ Col20a1- SCs (Figure 8B). It is therefore important to show when this repopulation begins, because a modest recovery of SC coverage could have a big effect. For example, Figure 4C quantifies loss of NMJs with residual RFP+ cells but not S110+ cells; both should be quantified at this and slightly later stages.

Reviewer #3 (Public review):

Summary:

This manuscript reports a novel genetic model, Col20a1-CreERT2 knock-in mouse, to target terminal Schwann cells (tSCs) at mouse neuromuscular junctions in a cell-type-specific and temporal manner. The authors analyzed multiple publicly available single-cell transcriptome databases to identify Col20a1 as the tSC marker. The authors crossed Col20a1-CreERT2 and Rosa26-LSL-tdTomato to label tSCs successfully. In addition, the authors generated Col20a1-CreERT2; Rosa-tdT/ diphtheria toxin subunit A (DTA) to specifically ablate tSCs and analyze the role of tSCs in motor behavior, neuromuscular junction electrophysiological function, and the histology and ultrastructure of neuromuscular junctions.

Strengths:

The Col20a1-CreERT2 x Rosa26-LSL-tdTomato mice successfully labeled the tSCs at NMJs and reported the developmental distribution of the Col20a1-positive cell population. The Col20a1-CreERT2; Rosa-tdT/DTA mice successfully ablated tSCs, which did not cause changes in gross neuromuscular junction architecture, neuromuscular synapse physiology, or motor behavior.

Weaknesses:

The conclusion of this manuscript will be strengthened by additional analysis showing time-course data of tSC ablation and replacement by non-recombined Schwann Cells. Currently, it is not clear when and how long the tSCs are ablated, which makes it difficult to interpret the data and phenotype. Detailed review comments are provided to the authors in the "recommendations for the authors" section.