Extraocular Muscle Characteristics Related to Myasthenia Gravis Susceptibility

Background The pathogenesis of extraocular muscle (EOM) weakness in myasthenia gravis might involve a mechanism specific to the EOM. The aim of this study was to investigate characteristics of the EOM related to its susceptibility to myasthenia gravis. Methods Female F344 rats and female Sprague-Dawley rats were assigned to experimental and control groups. The experimental group received injection with Ringer solution containing monoclonal antibody against the acetylcholine receptor (AChR), mAb35 (0.25 mg/kg), to induce experimental autoimmune myasthenia gravis, and the control group received injection with Ringer solution alone. Three muscles were analyzed: EOM, diaphragm, and tibialis anterior. Tissues were examined by light microscopy, fluorescence histochemistry, and transmission electron microscopy. Western blot analysis was used to assess marker expression and ELISA analysis was used to quantify creatine kinase levels. Microarray assay was conducted to detect differentially expressed genes. Results In the experimental group, the EOM showed a simpler neuromuscular junction (NMJ) structure compared to the other muscles; the NMJ had fewer synaptic folds, showed a lesser amount of AChR, and the endplate was wider compared to the other muscles. Results of microarray assay showed differential expression of 54 genes in the EOM between the experimental and control groups. Conclusion Various EOM characteristics appear to be related to the increased susceptibility of the EOM and the mechanism of EOM weakness in myasthenia gravis.


Introduction
Myasthenia gravis (MG) is characterized by muscle weakness and fatigability, which is attributed to the presence of autoantibodies against the acetylcholine receptor (AChR), leading to dysfunction of acetylcholine signal transduction [1]. However, in patients with ocular MG, the autoantibody titer is relatively low [2]. The extraocular muscles (EOM) are known to be susceptible in other autoimmune diseases such as Grave disease [3], suggesting the presence of a unique mechanism that dictates pathogenesis of EOM weakness in MG and other autoimmune diseases, which is not present in other skeletal muscles. There are relatively few studies investigating the mechanism of EOM weakness, a common manifestation in patients with MG.
Previous studies have revealed that EOM expresses the embryonic form of the AChR, which is different from the AChR found in skeletal muscle [4]. However, it is not clear if embryonic AChR is a target of immune system attack resulting in EOM weakness in patients with MG. In fact, patients with MG who experience EOM weakness do not have circulating antibodies against embryonic AChR or T-cell immunity specific to embryonic AChR [5]. Additionally, active or passive immunity induced by the embryonic AChR or its antibody does not induce symptoms of MG. These findings suggest the presence of specific characteristics that predispose the EOM to susceptibility to damage, making it a target in MG patients.
The EOM has special functions. Contraction is rapid, and there is a remarkable tolerability to fatigue. Under normal conditions, it is easy to maintain stable contraction of the EOM in the presence of high-frequency nerve impulses. Previous studies showed differential expression of a number of genes including immune response genes and complement genes in EOM, suggesting that EOM could be particularly susceptible to complement-mediated injury triggered in MG [6]. In the present study, we aimed to investigate factors associated with increased susceptibility of EOM in MG. In particular, based on factors that decrease the safety of the neuromuscular junction (NMJ) [7], we investigated the mechanism of EOM weakness in MG.

Animals and Preparation of AChR Monoclonal Antibody (mAb35)
All animal procedures were approved by the Ethics Committee of Tangdu Hospital, Fourth Military Medical University, and the protocols followed the ethical guidelines of this committee. Nude mice purchased from Animal research center, Fourth Military Medical (Shaanxi, China) were housed one per cage in our animal facility, with a 12:12 h light-dark cycle and ad libitum access to food and water.
The TIB175 hybridoma cell line (American Type Culture Collection, Manassas, VA) was maintained in Dulbecco modified Eagle medium (Life Sciences Corp., Grand Island, NY) containing 10% fetal bovine serum. Nude mice were intraperitoneally inoculated with cells (1610 6 cells/mouse) to generate the mAChR antibody (mAb35). One week after inoculation, ascites were collected, and IgG was extracted by the ammonium sulfate salting-out method. After dialysis, the Lowry method was used to determine protein concentration. IgG was subjected to 15% SDS-PAGE and the purity was 97%.
Female F344 rats aged 10 to 11 weeks (,160 g) were purchased from Shanghai Slaccas Laboratory Animal Co. (Shanghai, China) and used for microarrays according to previous reports [8]. Female Sprague-Dawley (SD) rats aged 10 to 11 weeks (,200 g) were purchased from the Animal Center of the Fourth Military Medical University. Rats were assigned to the experimental group or the control group. The experimental group received an intraperitoneal injection of Ringer solution (1 ml) containing mAb35 (0.25 mg/kg) to induce experimental autoimmune myasthenia gravis as previously described [9,10], while the control group received an intraperitoneal injection of Ringer solution alone. The animal facility had a 12:12 h light-dark cycle. Three rats were housed per cage and had ad libitum access to food and water for 48 h before experiments.

Assessment of EOM Symptoms and EOM Wet Weight
Eyeball movement and concomitant symptoms were observed. Rats were sacrificed two days after immunization and the EOM was collected and weighed (wet weight).

Determination of Creatine Kinase Activity in Serum
Blood was obtained from the orbital sinus of anesthetized mice and allowed to clot. After centrifugation, the serum was collected and stored at 270uC for creatine kinase determinations. Creatine kinase (CK) activity of serum was assessed using the appropriate ELISA test kit (Uscn Life Science, Wuhan,China No. E90109Ra) according to the manufacturer's instructions. Final results were read at a wavelength of 450 nm.

Tissue Staining for Light Microscopy
Sprague-Dawley rats were anesthetized with intraperitoneally administered 1% sodium pentobarbital (50 mg/kg) and were then perfused transcardially with 150 ml normal saline. Muscles (EOM, diaphragm, and tibialis anterior) were collected. For staining of succinate dehydrogenase (SDH), muscles were stored in liquid nitrogen with no antifreeze. For acetylcholinesterase (AChE) staining, rats were perfused transcardially with 150 ml normal saline and then with 500 ml 4% paraformaldehyde. Muscles were collected and fixed in 4% paraformaldehyde again for 2 h, and then placed in 25% sucrose for 24 h until they sank, in order to prevent intracellular ice crystal formation. The following were used for section generation (25-mm thickness): paraffin sections, RM 2135 (Leica; Nussloch, Germany); cryostat sections, CM1900 (Leica); vibrating blade microtome sections, VT1000S (Leica); ultramicrotome sections, LKB Nova (LKB, Stockholm, Sweden). Sections were placed onto uncoated slides.
For SDH staining, sections were incubated in SDH buffer (50 ml 0.2 M sodium succinate, 50 ml 0.2 M phosphate buffer, 100 mg nitroblue tetrazolium chloride) at room temperature for 1 h and fixed in 4% paraformaldehyde for 30 min. For AChE staining, sections were incubated in AChE buffer (0.5 ml of 0.1 mM sodium acetate, 1 ml of 30 mM copper sulphate; 5 mg/ ml acetylthiocholine iodide, 1 ml of 5 mM potassium ferricyanide, 6.5 ml of 0.1 M PBS) at room temperature for 1 h. For doublestaining, SDH-stained sections were also stained for AChE. Sections were observed under an upright, bright-field microscope (BX-60; Olympus, Tokyo, Japan). A total of 10 fields/animal/ muscle were randomly selected for analysis. F344 as well as SD rats showed similar pathological changes (data not shown).

AChR Fluorescence Histochemistry
Free-floating muscle sections were treated with biotin-conjugated a-bungarotoxin (a-BuTx-Biotin; 1:500; Molecular Probes, Eugene, OR) at room temperature overnight and then with 1:500 avidin-conjugated Texas Red (Sigma-Aldrich, St. Louis, MO) or avidin-conjugated fluorescein isothiocyanate (FITCavidin) (Sigma-Aldrich) at room temperature for 3 h. Sections were mounted onto slides with 50% glycerol in 0.1 M phosphatebuffered saline (PBS) and observed under a microscope (BX-51, Olympus; 100 W halogen lamp, 400-700 nm). Representative images were acquired with a digital camera (DP-70, Olympus). The fluorescence exposure time for each field was 20 seconds. We acquired all images first and measured fluorescence intensity later to avoid/decrease any bleaching effect.

Pre-embedding Immunocytochemistry
Samples were collected and fixed in buffer (0.25% glutaraldehyde, 4% polyoxymethylene, 15% saturated picric acid in pH 7.4, 0.1 mol/l PBS) for 2 h. Samples were then cut into 50-mm sections and treated with a-BuTx-Biotin and avidin-biotin complex (ABC, Vector Laboratories, Burlingame, CA) followed by visualization with diaminobenzidine (DAB) (1:200, final concentration 200 mg/ ml). Sections were immersed in 25% sucrose for 2 h and washed in PBS. After postfixation in 0.5% osmium tetroxide in water at room temperature for 1 hour, sections were dehydrated in an ethanol series and then incubated in propylene oxide and subjected to flatembedding in epoxide resin (Diglycidyl Ether of Bisphenol-A or DGEBA) with heat. Appropriate regions were selected and glued to a blank block for ultrathin sectioning (500 nm sections, diamond knife, coated copper grid) followed by lead-uranium treatment for contrast. Sections were examined by electron microscopy (JEM100SX, Jeol Ltd., Tokyo, Japan; 60-80 kV accelerating voltage, image recording on film).

Determination of the Ratio of Postsynaptic Membrane to Presynaptic Membrane
Sprague-Dawley rats were anesthetized and perfused transcardially with 150 ml normal saline, followed by fixation and processing for electron microscopy as described above, with the exception of a-BuTx-Biotin, ABC, and DAB staining. The area and length of the presynaptic membrane and postsynaptic membrane were measured as described previously [11].

Western Blot Analysis
Agrin and SV2 expressions were detected by Western blot according to the method described previously [12]. In brief, tissue homogenates were prepared by homogenization in a modified RIPA buffer (0.1 M phosphate-buffered saline (PBS)/1% NP-40/ 0.5%, Tween-20/0.1% sodium dodecyl sulfate (SDS), and complete cocktail inhibitor (Roche, Mannheim, Germany), and extracted on ice for 30 min. Protein concentrations were determined by BCA assay (Pierce, Rockford, IL). A 40 mg/well protein was fractionated on 7.5% SDS-PAGE gels, and transferred to Immobilon P (Millipore, Bedford, MA), then blocked with blocking buffer (Tris-buffered saline, pH 7.6 and 0.5% skimmed milk) and incubated with monoclonal anti-agrin antibody (1:500; Chemicon, Temecula, CA ) and monoclonal anti-synaptic vesicle protein 2 (anti-SV2) antibody (1:400; K. Buckley, Harvard Medical School, Boston, MA) at 4uC overnight. The membranes were then incubated with HRP-conjugated anti-mouse antibodies at room temperature for 1 h. The bound antibodies were visualized using LumiGLo reagent (Pierce, Rockford, IL) and the expression levels of SV2 and Agrin relative to b-actin were determined.

Detection of Differentially Expressed Genes by Microarray Assay
Total RNA was extracted from F344 muscles with TRIzol (Life Technologies Corp.), according to the manufacturer's instructions. mRNA was purified with an Oligotex mRNA Midi Kit (Qiagen, Chatsworth, CA) and cDNA was prepared by Reverse-transcription Kit according to the manufacturer's instructions (Shanghai Union Gene Co., Ltd., Shanghai, China). cDNA samples from the control and experimental groups were labeled with Cy3-dUTP and Cy5-dUTP and hybridized to a whole rat Genome Oligo Microarray (BioStarR-40S) from Shanghai Union Gene Co., Ltd. (Shanghai, China) according to the manufacturer's instructions. A ScanArray 3000 (General Scanning, Moorpark, CA) was used to scan the microarray, and Imagene 3.0 software was used to detect the intensity of Cy3 and Cy5 fluorescence signals. The ratio of Cy5 to Cy3 signals (Cy5/Cy3) was calculated and gene expression was normalized to an internal reference. A gene with Cy5/Cy3 of .2 or ,0.5 was defined as a differentially expressed gene. To minimize individual variation, common differentially expressed genes were selected from two microarray assays for analysis.

Statistical Analyses
Statistical analyses were performed with SPSS 15.0 statistics software (SPSS Inc, Chicago, IL). Comparability of the proportion of muscle fibers between the control and experimental groups was tested by chi-square test, and data are presented as number (n) and percent (%). Normally distributed continuous variables were compared by one-way analysis of variance (ANOVA) if there were more than two groups. When a significant difference between groups was apparent, multiple comparisons of means were performed with the Bonferroni procedure, along with type-I error adjustment. In addition, independent two-sample t tests were performed to determine the difference between the control and experimental groups. Data are presented as mean 6 standard deviation (SD). All statistical assessments were two-sided and evaluated at the 0.05 level of significant difference.

Assessment of Ocular Symptoms and EOM Wet Weight
The specificity of mAb35 was demonstrated in TE671 cells which express skeletal muscle nAchR and in rat brain tissue (data not shown). Circulating AchR antibodies were also detected in rats injected with mAb35, and undetectable in unimmunized animals (data not shown). Anti-cholinesterase inhibitor (neostigmine) relieved MG symptoms while a non-depolarizing muscle relaxant (vecuronium bromide) exacerbated MG symptoms (data not shown). Additionally, comparison of amplitude after stimulation with electrical pulses showed a significant reduction in amplitude in the experimental group compared to the control group (data not ahown).
All PTMG rats exhibited decreased activity and increased lassitude and lethargy compared to control rats. Eye movement was decreased, and the palpebral fissure was small. Rats with severe MG exhibited enophthalmos, congestion, and even bloody discharge. Figure 1A depicted the ocular symptoms of PTMG rats. The experimental groups of both SD and F344 rats had significantly lower EOM wet weights compared to the control groups (Fig. 1B). Serum creatine kinase (CK) levels is a well defined marker for muscular damage which is released into the bloodstream following muscle cell damage or degeneration [13]. http://www.sciencedirect.com/science/article/pii/ S0304394003011121 -BIB5The ELISA results showed that the CK levels of serum samples were roughly 6.5-fold higher than the control values (Fig. 1C). The results provide evidences for muscular damage in MG.

Assessment of the Proportion of Muscle Fibers in Different Muscles
SDH staining was used to compare the proportion of three muscle fiber types in muscles in the control and experimental groups (Table 1). There were significant (P,0.05) differences in the proportion of the three types of muscle fibers in EOM and diaphragm muscle between the control and experimental groups. There was also a significant (P,0.05) difference between the groups in the proportion of type I and type IIA fibers in tibialis anterior muscle but no significant difference in the proportion of type IIB muscle fibers (P = 0.079).

Assessment of Endplate Distribution in Different Muscles
Endplate distribution was assessed by AChE staining. At low magnification, the NMJ was found to be arranged in the middle segment of muscle fibers of different muscles, forming the endplate. The endplate was compared in different muscles. Results showed that the endplate of the EOM was wider than that of the diaphragm or tibialis anterior in both control and PTMG groups (Fig. 2). In addition, some small NMJs were also distributed in the EOM. In the experimental group, the NMJ of the endplate of different muscles was slightly smaller than those in the control group. We also found that the endplate of the EOM was significantly narrowed, but this change was not obvious in the diaphragm and tibialis anterior muscles (Fig. 2).

Comparison of NMJs from Different Muscles
AChE staining showed that the NMJ was brown, flat, or oblate and distributed along the border of the muscle fibers (Fig. 3). The size of the AChE-positive region reflected the size of the NMJ, and the density of this region reflected the AChE content. The NMJ area was smaller in the experimental groups when compared to the control groups, and this decrease was more obvious in the EOM than in diaphragm and tibialis anterior muscles. However, the AChE content remained largely unchanged (Fig. 3A). There was also a significant decrease in the NMJ area in different muscles in the experimental group compared to the control group. In the control group, the NMJ area of diaphragm muscle was slightly larger than that of EOM and tibialis anterior muscle (Fig. 3B). The AChE content in the NMJ was comparable between the experimental and control groups (Fig. 3C).

Comparison of AChRs in the NMJ of Different Muscles by Fluorescence Histochemistry
Fluorescently labeled, a-BuTx-tagged AChRs aggregated in the NMJ along the border of the muscle fibers. Fluorescence intensity reflected AChR content, and fluorescence area reflected AChR distribution. In the experimental group, the fluorescence area in the different muscles was decreased when compared to control group, and this was accompanied by a decrease in fluorescence intensity (Fig. 4A). There was also a significant (P,0.05) difference in AChR distribution in the NMJ between the two groups. In the control group, the AChR positive area of EOM was smaller than that of diaphragm or tibialis anterior muscle (Fig. 4B). In addition, there was a significant difference between the two groups in NMJ fluorescence density of the three types of muscles (Fig. 4C).

Comparison of the NMJ in Different Muscles by Double-Staining
Double-staining for SDH and AChE can be used to differentiate the NMJ of different muscle fibers. The background of PTMG sections in SDH-stained sections was lighter than that in control sections, which may be related to decreased SDH expression in MG. The NMJ area of the different muscles was decreased in the experimental group compared to the control group, and the NMJ area was smaller in sections containing relatively strong SDH staining. Type I muscle fibers were decreased in the experimental group compared to the control group, but the NMJ in type I muscle fibers was comparable among the different muscle types (Fig. 5A). The NMJ area in muscle fibers with relatively strong SDH staining was smaller than that in muscle fibers with weak SDH staining, which was characteristic of both EOM and diaphragm muscle (Fig. 5B).

Comparison of AChRs in the NMJ of Different Muscles by Transmission Electron Microscopy
a-BuTx-labeled AChRs in the NMJ were characterized by high electron-dense material in the wrinkled surface of the postsynaptic membrane. In the experimental group, the wrinkled surface of the postsynaptic membrane was eliminated, and the AChR amount was decreased compared to that in the control group (Fig. 6). In addition, the amount of AChR in the EOM was less than that in diaphragm and tibialis anterior muscle in both groups.

Determination of the Ratio of Postsynaptic Membrane to Presynaptic Membrane in Different Muscles
Normal EOM had a lower ratio of postsynaptic membrane to presynaptic membrane compared to diaphragm and tibialis anterior muscle, although the ratio of nerve terminal area to postsynaptic membrane area was greater. The experimental group exhibited a decrease in the wrinkled surface of the synapse in the NMJ of the different muscles, compared to the control group. This was accompanied by decreased number of NMJ. The ratio of postsynaptic membrane to presynaptic membrane was therefore smaller, and the ratio of nerve terminal area to postsynaptic membrane area was larger compared to the control group (Fig. 7A). Table 2 shows that the ratio of postsynaptic membrane to presynaptic membrane at the NMJ was markedly decreased in the experimental group compared to the control group, but the ratio of nerve terminal area to postsynaptic membrane area was larger than that in the control group (all P,0.001). In addition, the EOM of the control group had a smaller ratio of postsynaptic membrane to presynaptic membrane compared to the tibialis anterior muscle. The ratio of postsynaptic membrane to presynaptic membrane in the EOM of the experimental group was smaller than that seen in diaphragm and tibialis anterior muscle. In both groups, the ratio of nerve terminal area to postsynaptic membrane area in the EOM was larger than that in tibialis anterior muscle. Furthermore, 2 markers of pre-and postsynaptic complex of NMJ, SV2 and Agrin, were determined by Western blot. In the experimental group, the levels of SV2 and Agrin in EOM were decreased significantly when compared to the control   labeled with a-bungarotoxin in the neuromuscular junction (NMJ) of three muscles in the two groups. B. Determination of pixel areas in NMJ by an automatic analyzer. C. Determination of gray value in NMJ by an automatic analyzer. *Significant (P,0.05) difference between the control and experimental groups determined by independent two-sample t test. a Statistically significant difference between indicated group and EOM group. b Statistically significant difference between indicated group and diaphragm group. Scale bar = 10 mm. doi:10.1371/journal.pone.0055611.g004 group (Fig. 7B). The results demonstrated that presynaptic membrane and postsynaptic membrane were damaged in EOM.

Differentially Expressed Genes in the EOM
We showed that the experimental and control groups exhibited a total of 54 differentially expressed genes in the EOM of F344 rats. This included 25 genes that were significantly upregulated and 29 genes that were significantly downregulated. Some of these differentially expressed genes and their functions are listed in Table 3.

Differentially Expressed Genes Between EOM and Tibialis Anterior Muscle in Control F344 Rats
Microarray results showed a total of 725 differentially expressed genes between EOM and tibialis anterior muscle in control F344 rats. This included 339 genes that were significantly downregulated and 386 genes that were significantly upregulated. Some of these differentially expressed genes and their functions are listed in Table 4.

Discussion
In this study, we investigated EOM susceptibility in MG. We used a PTMG rat model based on previous reports that in PTMG, electrophysiological and ultrastructural changes at the endplate are similar to those seen in patients with acquired MG [14]. We showed that the NMJ in the EOM of PTMG rats was simpler in structure compared to NMJ structures in the diaphragm and tibialis anterior muscle. This was manifested by fewer synaptic folds in the postsynaptic membrane, and fewer AChRs. Although passive transfer studies have been previously used in mice, we used a rat model because the disease features are more obvious in rats and because mAb35 is a rat antibody. In addition of using mAb35, we plan to use an irrelevant mAb in future studies to serve as an additional control.
We used SDH staining to show a significant decrease in type I muscle fibers and a significant increase in type IIA muscle fibers in PTMG rats compared to control rats. The extent of change differed among the different muscles. The proportion of type IIB muscle fibers remained largely unchanged. Our data suggested that the skeletal muscle weakness, resulting from NMJ transduction disorder, could be due to a decrease in fatigue-resistant, oxidative type I fibers. Decreased SDH expression in the muscle cells could reflect a transformation of type I fibers with high SDH expression into type IIA fibers with low SDH expression. Interestingly, the diaphragm and tibialis anterior muscle had a smaller proportion of type I fibers compared to the EOM. This may be related to the higher proportion of type I fibers in normal EOM or to the different loads in distinct muscles.
We showed a decrease in the AChE-positive area in all three muscles in the experimental group compared to the control group, although the AChE density remained unchanged. Since AChE is mainly located in the basement membrane of the synaptic space of the NMJ, PTMG rats exhibited a decrease in the area of the synaptic space (area of the basement membrane), although AChE content in the synaptic space remained unchanged.
We used AChE staining to show that NMJ size was similar in the EOM and in tibialis anterior muscle. We also used a-BuTxlabeled AChRs to show a significant decrease in the amount of AChR in the NMJ of PTMG rats compared to control rats. EOM had a lower AChR density compared to the other two muscles in both groups, although this was more obvious in the PTMG group. The EOM of the control group also had significantly decreased AChR-positive area compared to the other two muscles, although no marked difference was found among the different muscles in the experimental group. The small amount of AChR in the NMJ makes the EOM susceptible to immune attack in the presence of even low titers of antibody.
It was previously suggested that the length of the synaptic fold in unit area could reflect the amount of voltage-dependent sodium channels and AChRs [15]. Consistent with earlier reports [16], we showed that the NMJ structure in the majority of the EOM was simpler than in the other two muscles. In the experimental group, only a few synaptic folds were found in the postsynaptic membrane, which was almost flat compared to normal EOMs  where a complex synaptic structure was observed. Although the EOM and other skeletal muscles were involved in MG, the susceptibility of the EOM differed from that of the other muscles. Following attack by the immune system, the tolerance of the EOM becomes relatively poor, although changes in the types of different fibers, AChR amount, and NMJ ultrastructure were similar among the three muscles.
We showed that the EOM endplate was wider than that of the diaphragm or tibialis anterior muscle. The length of the muscle fiber in the EOM was one-tenth that in the other skeletal muscles, and muscle fiber diameter in the EOM was half that in the other skeletal muscles. The larger ratio of endplate to muscle fiber reflects the fact that movement of the EOM is finer than that of the other muscles.
Although the ratio of postsynaptic membrane to presynaptic membrane in the EOM was significantly decreased compared to the other muscles in both groups, EOM had a significantly higher ratio of nerve terminal area to postsynaptic membrane area compared to the other muscles. Since the length and area of the postsynaptic membrane reflect the amount of voltage-dependent sodium channels and AChRs, our data suggested that the NMJ of the EOM had fewer sodium channels and less AChRs than the other muscles. Release of synaptic vesicles from the presynaptic membrane would result in decreased amount of sodium channels and decreased action potential. The decreased difference between action potential and membrane potential would lead to a low safety factor.   Expression profiling previously showed 338 genes that were differentially expressed in the EOM compared to quadriceps femoris limb muscle [17]. In the present study, we showed that immune response genes such as thymosin, CD9, CD74, C1, C2, and B factor were significantly upregulated in the EOM compared to that in tibialis anterior muscle. The immunoglobulin domain gene, AChR antibody kappa chain gene was also upregulated. We also showed downregulation of genes associated with neuromuscular synaptic transmission including myosin, tropomyosin, myosin-binding protein, and synaptophysin and upregulation of ACh metabolism-related acetyl synthase and ACh transporter genes. Genes for ACh metabolism-related acetyltransferase, acetyl reductase, and mitochondrial transporter were also downregulated. In addition, some cell metabolism and signal transductionrelated genes such as Rho, CDK5, and genes encoding proteins in the respiratory chain were differentially expressed. We would like to elucidate the role of these genes in the susceptibility of the EOM to MG. Of note, genes for complement regulatory factors such as CD55 (decay-accelerating factor; DAF) and CD59 remained unchanged between control rats and PTMG rats, although both have been previously found to be closely related to the susceptibility of the EOM to MG [18][19][20][21][22][23]. A single nucleotide polymorphism in the DAF regulatory region was previously shown to be associated with EOM paresis [23]. It will be interesting to evaluate gene polymorphisms in some of the differentially expressed genes in our study and explore their role in EOM paresis. Our study provides the basis for further investigation of the mechanisms underlying increased EOM susceptibility in MG.