T-box transcription factor 21 is expressed in terminal Schwann cells at the neuromuscular junction

Abstract

Introduction/Aims:

Terminal Schwann cells (tSCs) are nonmyelinating Schwann cells present at the neuromuscular junction (NMJ) with multiple integral roles throughout their lifespan. There is no known gene differentiating tSCs from myelinating Schwann cells, making their isolation and investigation challenging. In this work we investigated genes expressed within tSCs.

Methods:

A novel dissection technique was utilized to isolate the tSC-containing NMJ band from the sternomastoid muscles of S100-GFP mice. RNA was isolated from samples containing: (a) NMJ bands (tSCs with nerve and muscle), (b) nerve, and (c) muscle, and microarray genetic expression analysis was conducted. Data were validated by quantitative real-time polymerase chain reaction (qRT-PCR) and immunofluorescent staining. To identify genes specific to tSCs compared with other NMJ components, analysis of variance and rank-order analysis were performed using the Partek Genomic Suite.

Results:

Microarray analysis of the tSC-enriched NMJ band revealed upregulation (by 4- to 12-fold) of several genes unique to the NMJ compared with muscle or nerve parts alone (P < .05). Among these genes, Tbx21 (or T-bet) was identified, which showed a 12-fold higher expression at the NMJ compared with sciatic nerve (P < .002). qRT-PCR analysis showed Tbx21 mRNA expression was over ninefold higher (P < .05) in the NMJ relative to muscle and nerve. Tbx21 protein colocalized with tSCs and was not noted in myelinating SCs from sciatic nerve.

Discussion:

We found TBX21 to be expressed in tSCs. Additional studies will be performed to determine the functional significance of TBX21 in tSCs. These studies may enhance the investigative tools available to modulate tSCs to improve motor recovery after nerve injury.

Graphical Abstract

1 ∣. INTRODUCTION

Glia are supportive cells for neurons throughout the body. In the peripheral nervous system (PNS), Schwann cells (SCs) serve as the major glial component, comprising myelinating and nonmyelinating SCs.¹ Nonmyelinating SCs have unique and specific roles.² At the neuromuscular junction (NMJ), nonmyelinating terminal Schwann cells (tSCs) regulate critical NMJ functions, namely synaptogenesis, maintenance, and, crucially, repair after nerve injury.^3-8 After nerve injury, tSCs elaborate cytoplasmic processes that guide regenerating axons to the NMJ, bridge NMJs, and induce axonal sprouting.^9,10 As a key regulator in the NMJ reinnervation process, tSCs represent an exciting, albeit poorly understood, element of motor recovery and a novel target for research and therapy.

In spite of their potential, significant limitations exist in the ability to research tSCs. Most notably, tSCs are difficult to isolate, and a unique genetic tSC marker has not yet been realized. S100, desert hedgehog, and proteolipid protein 1 are markers common to all glial cells, whereas myelin basic protein is expressed in myelinating SCs.^7,11-14 Elucidation of tSC-specific markers requires transcriptome analysis to identify unique genetic expression profiles relative to other SC types. Castro et al recently identified neuron-glia antigen 2 a unique marker of differentiated tSCs using RNA seq.¹⁵ In this study, we propose the T-box transcription factor 21 (TBX21 or T-BET) is expressed in tSCs at the NMJ in mice. This finding will augment the investigative tools available to modulate these cells to propel tSC investigation and enhance our understanding of these cells to better promote motor recovery after nerve injury.

2 ∣. METHODS

2.1 ∣. Tissue collection

To isolate NMJs with tSCs, young adult (both sexes) S100-GFP mice (Figure S1) were used in which all SCs express GFP under the control of the S100 promoter.¹⁴ Sternomastoid (SM) muscles with well-defined clusters of NMJs, which contain nonmyelinating tSCs, were harvested from both sides of the neck. The NMJ band consists of the muscle region with highest NMJ-and therefore tSC-density (Figure S1). For each muscle, a synapse-rich (NMJ, or endplate) band was cut out under a fluorescent dissecting stereomicroscope (Olympus). As the NMJ band consists of tSCs, muscle, and nerve, synapse-free segments of muscle and nerve (sciatic and cranial nerves) were also similarly harvested. All mice were housed in a central animal facility and were maintained in strict accordance with National Institutes of Health guidelines and according to protocols approved by the institutional animal research ethics committee at Washington University School of Medicine.

2.2 ∣. RNA qualification for microarray and data analyses

Total RNA was extracted from homogenized tissues using a Trizol and RNeasy MinElute Cleanup Kit (Qiagen, Hilden, Germany). RNA was first quantified with NanoDrop ND-1000 and only those samples with a 260 nm/280 nm ratio of between 1.6 and 2.0 were used for analysis. Total RNA quality was confirmed by RNA 6000 NanoChip bioanalysis (Agilent Model 2100 bioanalyzer; Agilent Technologies, Palo Alto, California), and only samples with a 28S/18S ratio of between 1.5 and 2.0 and an RNA integrity number higher than 8 were selected for microarray. RNA transcripts were amplified using an RNA amplification kit (MessageAmp; ABI-Ambion). Hybridization to mouse 4X44K Agilent microarray gene expression chips (WTA2; Agilent Technologies) was conducted at the Genome Technology Access Center at Washington University, according to the manufacturer's protocol. To obtain a list of genes uniquely expressed at the NMJ compared with surrounding muscle and nerve, microarray data were analyzed using the Partek Genomic Suite (Genome Technology Access Center) at the threshold value of 0.05, rank-order analysis, and analysis of variance (ANOVA). All common genes expressed in the muscle and nerves were subtracted from the list of NMJ genes (Figure S2). Tissues were analyzed from six independent experiments.

2.3 ∣. Quantitative real-time polymerase chain reaction and immunostaining

The microarray data were validated using quantitative real-time quantitative polymerase chain reaction (qRT-PCR) and immunostaining.¹¹ qRT-PCR was performed in a Step One Plus instrument using TaqMan Fast Universal PCR Master Mix and a specific TaqMan PCR primers/probes combination (Applied Biosystems, Foster City, California). Frozen sections of sciatic nerve or SM muscle with tSCs from S100-GFP, wild-type (WT), and Tbx21 knockout (Tbx21-KO) mice were immunofluorescently stained using anti-Tbx21 mouse monoclonal (4B10; sc21749, Lot No. B0315) and rabbit polyclonal (H-210; sc-21 003, Lot No. A1014) antibody (Santa Cruz Biotechnology, Santa Cruz, California).^16,17

3 ∣. RESULTS

3.1 ∣. Analysis of tissue-specific genetic profiles showing enhanced Tbx21 expression at NMJ

Microarray analysis of the tSC-enriched NMJ band revealed upregulation of several genes relative to either isolated muscle or nerve. Of the 39 429 unique probes (excluding controls and duplicates), 264 genes (pre-list) were upregulated by 1.12- to 8.36-fold in the NMJ band as compared with the non-NMJ band muscle tissue (P < .05) (Figure S2). When comparing transcription profiles (pre-list) of the NMJ band (with non-NMJ band muscle subtracted) with cranial and sciatic nerve individually, 39 genes each (lists 1 and 2) were differentially upregulated (P < .05), with fold changes between 1.15 and 7.08 (cranial nerve) and 1.18 to 85.49 (sciatic nerve). Cross-examining these lists together yielded a list of 28 genes significantly upregulated at the NMJ (Table 1). Among these genes was Tbx21, which showed a 12-fold higher expression at the NMJ compared with sciatic nerve (P < .002). Other genes expected to be found at the NMJ were also identified, such as acetylcholinesterase Q subunit (ColQ), which is located at the synaptic cleft. Subsequent qRT-PCR experiments confirmed higher Tbx21 expression levels in the NMJ band. Levels of the Tbx21 mRNA transcript were over ninefold higher (P < .05) in the NMJ compared with muscle and nerve (Figure 1A). These data suggest Tbx21 is specifically expressed at the NMJ band, as compared with either surrounding muscle or nerve tissue.

TABLE 1.

Transcripts enriched at NMJs vs muscle and nerves in sternomastoid muscle of young adult S100-GFP mice

Probe set ID	Gene name	Gene bank ID	P value (NMJ vs muscle and nerves)	Average fold change (NMJ vs muscle and nerves)
A55P2180470	Chrne	NM009603	.0000104	29.29
A52P533707	Chrna1	NM007389	.0000027	27.09
A55P2109326	Ache	NM009599	.0000621	24.06
A55P2110548	Chrnd	NM021600	.0000060	22.36
A51P264769	Dmp1	NM016779	.0000045	21.69
A55P2090077	Myh13	NM001081250	.0000000	21.69
A52P987201	Pdzrn4	NM001164593	.0000019	21.28
A55P2043337	Colq	NM009937	.0000007	16.45
A52P331981	Zp2	NM011775	.0000003	12.52
A51P501364	Tbx21	NM019507	.0000013	9.28
A51P166339	Avil	NM009635	.0000058	7.99
A51P278034	Gdnf	NM010275	.0004611	7.69
A51P278034	Ufsp1	NM027356	.0000009	7.41
A52P203560	Fzd10	NM175284	.0000079	6.97
A51P264695	Crym	NM016669	.0000064	5.63
A51P391495	Pola2	NM008893	.0000025	5.44
A51P160673	Kcne1l	NM021487	.0085126	5.37
A51P502437	Cacna2d3	NM009785	.0000488	4.56
A55P2076462	Lnx1	NM001159577	.0014890	4.30
A52P599317	Hs6st2	NM001077202	.0014650	3.39
A55P2107731	Col20a1	NM028518	.0000059	3.21
A51P306933	Avpr2	NM019404	.0462272	3.02
A51P371331	Mpp5	NM019579	.0041381	2.56
A52P294123	D2hgdh	NM178882	.0005245	2.23
A52P121342	Nrxn1	NM020252	.0003621	2.26
A52P473419	Epb4.1l4a	NM013512	.0065884	2.10
A55P2130600	Musk	NM001037127	.0326842	2.00
A51P434101	Herpud1	NM022331	3.18 × 10−50	2.00

Abbreviations: ID, identification; NMJ, neuromuscular junction.

FIGURE 1.

TBX21 is localized to the NMJ in the sternomastoid muscle from WT and S100-GFP young adult mice. A, Tbx21 mRNA expression is highest in the NMJ compared with NMJ-free muscle or nerve. B, Average pixel intensity for Tbx21 protein in muscle and the NMJ measured using ImageJ (National Institutes of Health, Bethesda, Maryland). C, Several NMJs showing Tbx21 (green, arrows) at acetylcholine receptors using Tbx21 rabbit polyclonal antibody. D, No Tbx21 is observed in muscle without NMJs. E, Tbx21 is not seen in myelinating SCs (yellow arrows) of sciatic nerve. F, Tbx21 (red) colocalizes with tSCs (green, arrows) at the NMJ. F′, Red channel (Tbx21) pseudo-colored gray from (F). G, No Tbx21 is observed without Tbx21 antibody (control, CTR). H, Representative image showing localization of Tbx21 (green) at the NMJ of WT mice. H′, Green channel (Tbx21) pseudo-colored in gray from (H). A few pixels (asterisk) localize to a muscle fiber or a portion of another NMJ. I, Tbx21 is not present in the muscle from a Tbx21 knockout (Tbx21-KO) mouse. Red: α-bungarotoxin (acetylcholine receptor, red); blue: DAPI-nuclear staining. Scale bar = 20 μm in (C)-(I). Data expressed as mean ± standard deviation. N = 6 mice. Abbreviations: CN, cranial nerve; NMJ, neuromuscular junction; SN, sciatic nerve; Tbx21, T-box transcription factor 21; WT, wild-type. *P < .05

3.2 ∣. Tbx21 colocalizes with tSCs at the NMJ

Given the increased genetic expression of Tbx21 at the NMJ, we evaluated localization of Tbx21 protein at the NMJ via immunostaining using rabbit polyclonal anti-Tbx21 antibody (Figure 1). We confirmed that Tbx21 is present at the NMJ of WT mice (Figure 1C,H,H') and colocalizes with S100-GFP-positive tSCs (Figure 1F,F'). Importantly, Tbx21 immunostaining was not seen in non-synapse-rich muscle (Figure 1D) or myelinating SCs of sciatic nerve of WT or S100-GFP mice (Figure 1E). Examination of muscle tissue in Tbx21-KO mice served as a negative control and confirmed a lack of Tbx21 staining, indicating appropriate antibody specificity (Figure 1I). Data were replicated with mouse monoclonal anti-Tbx21 antibody (Figure 2A-C) and in multiple different muscles (Figure 2D-G). Together these data demonstrate that enhanced Tbx21 mRNA expression at the NMJ corresponds to Tbx21 protein localized within tSCs.

FIGURE 2.

Localization of Tbx21 at the NMJ in different hindlimb muscles of the young adult WT (A,B) and S100-GFP (D-G) mice. A,B, Representative NMJ images from SM muscle stained with anti-Tbx21 mouse monoclonal antibody (mTbx21; red, arrows) and BTX. C, Bar graph shows the average pixel intensity for Tbx21 in NMJ-free muscle and NMJ area (based on BTX) measured as integrated density/area using ImageJ (National Institutes of Health, Bethesda, Maryland). Note that Tbx21 staining intensity is lower than that of anti-Tbx21 rabbit polyclonal antibody shown in Figure 1. D, Representative image shows colocalization of Tbx21 (red) with tSCs (green, arrows) at the NMJ in EDL muscle using anti-Tbx21 rabbit antibody (rbbTbx21). E, No Tbx21 staining is observed without antibody (no Ab; control, CTR). F,G, Representative images showing localization of Tbx21 at the NMJ in TA (F) and SOL (G) muscles. Small images below A, B, and D-G represent isolated color channels for NMJs marked with dotted lines. BTX (AChR, green) (A,B); DAPI-nuclear staining (blue). Scale bar = 20 μm. Data expressed as mean ± standard deviation. N = 6 mice in (A)-(C) and N = 3 mice in (D)-(G). Abbreviations: Ab, antibody; BTX, α-bungarotoxin; EDL, extensor digitorum longus; NMJ, neuromuscular junction; SM, sternomastoid; SOL, soleus; Tbx21, T-box transcription factor 21; TA, tibialis anterior; tSC, terminal Schwann cell; WT, wild-type. *P < .05

4 ∣. DISCUSSION

In this study we have reported that TBX21 is expressed in tSCs at the NMJ in a mouse model and may be helpful for investigation of this perisynaptic glial cell subtype. The TBX21 expression in tSCs was unexpected. Tbx21 was originally described as a transcription factor that controls the T-helper 1 (Th1) genetic program in naive CD4⁺ cells. Tbx21 directly activates interferon-gamma gene transcription and enhances development of Th1 cells.¹⁸ TBX21 also plays multiple roles in many subtypes of immune cells.^19,20 Although TBX21 expression is required for protection against pathogens, exuberant T-bet–regulated immune responses can be a driving force in inflammatory diseases.²¹ More recently, this transcription factor was shown to be expressed in diverse cell types, such as in the endometrium and brain, raising the possibility for a role in coordinating genetic expression and functions outside of the effector cells of the immune system.^17,22,23 With this study, we have proposed a novel role for Tbx21 in tSCs.

Compared with the myelinating subtype, tSCs are relatively understudied, likely due to investigative barriers. Of particular importance are the obstacles in isolating them, their relatively small numbers, and the historic lack of a unique genetic marker. The data from this project combined with those from Castro et al (2020) will enhance tSC investigation.¹⁵ Seemingly incongruous to this dearth of work is the importance of tSC functions, particularly their roles in NMJ reinnervation after motor nerve injury.^9,10,24 Although we have demonstrated the expression of TBX21 in tSCs, further work needs to be conducted to investigate its function within tSCs.

Given the known immune functions of TBX21, it has a potential role in tSCs to coordinate neuroinflammation after peripheral nerve injury. The importance of this process is highlighted in diseases in which the immune system is dysregulated or functioning aberrantly. Neurodegenerative diseases, such as Guillain-Barré syndrome, myasthenia gravis, and amyotrophic lateral sclerosis, are examples of the devastating effects of immune dysregulation.^25-27 Yet, the disruption of neuroinflammation is also problematic, as damaged nerves rely on inflammation, particularly macrophage function, for repair.^28,29 This interplay between nerves and the immune system is therefore critical for PNS health and recovery. Localization of TBX21 within tSCs differentiates this cell population from myelinating Schwann cells and will advance tSC investigation. The unique ability of TBX21 to interface with and coordinate components of the immune system, coupled with its expression at the NMJ, solidifies TBX21 as an important investigative target to improve recovery after motor nerve injury, potentially opening new therapeutic avenues for the many patients with nerve injury.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Abbreviations:

AChR

acetylcholine receptor

BTX

α-bungarotoxin

ColQ

acetylcholinesterase Qsubunit

GFP

green flourescent protein

NMJ

neuromuscular junction

PNS

peripheral nervous system

qRT-PCR

quantitative real-time polymerase chain reaction

SC

Schwann cell

SM

sternomastoid

TBX21

T-box transcription factor 21

Tbx21-KO

Tbx21 knockout

Th1

T-helper 1

tSC

terminal Schwann cell

WT

wild-type