**E**ff**ects of a Fat-Rich Diet on the Spontaneous Release of Acetylcholine in the Neuromuscular Junction of Mice**

#### **Carlos Gimenez-Donoso 1, Marc Bosque 2, Anna Vila 2, Gemma Vilalta <sup>2</sup> and Manel M Santafe 2,\***


**\*** Correspondence: manuel.santafe@urv.cat; Tel.: +34-97775-9300 (ext. 9343)

Received: 11 September 2020; Accepted: 16 October 2020; Published: 21 October 2020

**Abstract:** Western societies are facing a clear increase in the rate of obesity and overweight which are responsible for musculoskeletal pain. Some of the substances described in the environment of myofascial trigger points (MTrPs) are the same as those found in the skeletal muscle of obese people, such as cytokines. Furthermore, elevated neuromuscular neurotransmission has been associated with MTrPs. The main objective of this study is to assess whether obesity or overweight may be a facilitator of myofascial pain. The experiments were performed on male Swiss mice. One experimental group was given a typical "cafeteria" diet and another group a commercial high-fat diet for six weeks. Intramuscular adipocytes were assessed with Sudan III. The functional study was performed with electromyographic recording to determine the plaque noise and intracellular recording of miniature endplate potentials (MEPPs). The intake of a cafeteria diet showed the presence of more adipocytes in muscle tissue, but not with the fat-supplemented diet. Both experimental groups showed an increase in the plaque noise and an increase in the frequency of MEPPs that lasted several weeks after interrupting diets. In summary, the supply of a hypercaloric diet for six weeks in mice increases spontaneous neurotransmission, thus facilitating the development of MTrPs.

**Keywords:** electromyography; high-fat diet; myofascial pain syndrome; obesity; spontaneous neurotransmission

#### **1. Introduction**

At present, Western societies are having a clear increase in the rate of obesity and overweight [1]. From 1975 to 2016, the global obesity rate has tripled [2]. By 2030, over 38% of the world's adult population will be overweight and 20% will be obese [3]. Obesity and overweight are considered an epidemic related to the development of various pathologies such as diabetes, cardiovascular disease, metabolic syndrome and musculoskeletal pain [3].

There seems to be a strong relationship between obesity and pain [4]. Usually, the association between overweight and musculoskeletal pain has been attributed to an increase in the mechanical stress caused by overweight in load bearing joints. However, the literature also shows an association between pain and overweight in joints that do not support load [3,5–9]. For example, associations between overweight and symptomatic osteoarthritis of the hand [5], shoulder and neck pain [6], the number of episodes and intensity of migraine attacks [7,8], even with neuropathic pain [9] have also been described. Thus, the relationship between overweight and musculoskeletal pain appears to be at least, in part, independent of the mechanical overload and probably involves systemic phenomena.

Obesity is accompanied by a chronic inflammatory response with high production of adipokines (IL-6, TNF-α, adiponectin, leptin, and resistine) and macrophage infiltration into the adipose tissue [10]. This chronic inflammatory response has been linked to metabolic syndrome and musculoskeletal pain [11]. Moreover, musculoskeletal studies demonstrate the impact of these cytokines on muscle metabolism [12]. However, there are no studies that analyze the impact that obesity may have on other muscle functions or on muscle pain.

Myofascial pain syndrome (MPS) is the set of sensory (local and referred pain), motor (limited range of motion, weakness) and autonomic signs and symptoms caused by myofascial trigger points (MTrPs) [13]. The prevalence in the general population ranges from 20% to 90% and it is suggested that practically, all adults will suffer at least one episode of myofascial pain in their lifetime [14].

MTrPs have been proposed to be a neuromuscular dysfunction in which abnormal motor end plate function results in an excessive concentration of acetylcholine (ACh) in the synaptic cleft [15]. This excess ACh causes a localized contraction of the sarcomeres below the neuromuscular junction. Thereupon, a cascade of events that cause cellular suffering, local acidic pH, and the release of sensitizing substances from the nociceptive nerve terminals occur [16]. Altogether these changes are responsible for local pain. In 2005, Shah identified a high presence of several of these substances (pH, SP, IL-6, TNF-α, CGRP... ) in the environment of MTrPs [16]. On the other hand, in our laboratory, we experimentally cause the appearance of contracted subsynaptic sarcomere by increasing local ACh with anticholinesterase drugs [17]. This increase in ACh release can be recorded by needle electromyography as spontaneous low-voltage electrical activity (30–60 mV) like endplate noise [14,17,18].

Some of the substances that Shah [16] found in the environment of active MTrPs are the same as those found in the skeletal muscle of obese people: IL-6, IL-1B, TNF-α [12]. No one has ever established a relationship between obesity and MPS before. However, given that in both clinical situations, a similar inflammatory profile surrounding muscle tissue is found, we hypothesized that muscle neurotransmission in obese or overweight individuals is increased, thus constituting a predisposing factor for the development of MTrPs.

In the present study, intracellular recordings and electromyography were performed at the end of the diet exposure period and several weeks after. An increase in the body weight of the mice was paralleled by a significant increase in the spontaneous release of acetylcholine that lasts for several weeks after diet manipulations. All the results obtained suggest that obesity and overweight can cause myofascial muscle pain.

#### **2. Materials and Methods**

The mice were cared for in accordance with the guidelines of the European Community's Council Directive (2010/63/EU) and the Spanish Royal Decree 53/2013 for the humane treatment of laboratory animals. The Animal Research Committee of the Universitat Rovirai Virgili (Reference number: 0233) reviewed and approved all experiments on animals. The experiments were performed on young (45–50 days) adult Swiss male mice (Charles River, L'Arbresle, France). Mice were habituated to the facility for at least 1 week prior to studies and were housed in groups of four, with sawdust bedding and ad libitum access to water and food throughout the study. The animals' rooms were maintained at a temperature of 22 ± 2 ◦C, a relative humidity of 50 ± 10%, and a 12 h light/dark automatic light cycle.

#### *2.1. Animals and Dietary Protocol*

Mice were randomly divided into cages of 2 animals. The cages were randomly grouped into three groups (see Figure 1): a control group (CTR; N = 10 cages, 20 animals), a "cafeteria diet" group (CAD; N = 10 cages, 20 animals) and a high-fat diet group (HFD; N = 10 cages, 20 animals). During the experiment, the CTR group received a normal laboratory diet (SAFE Diets: 230 HF Rat & Mouse Diet, Augy, France). A "cafeteria diet" model (detailed below) was added to the CAD group and a fat-supplemented diet (230 HF Rat & Mouse Diet, SAFE, Augy, France) was added to the HFD group [19]. Exposure to this type of diet was carried out during a period of 6 weeks. After 6 weeks, both the cafeteria diet and the high-fat diet were withdrawn and all mice were fed exclusively with the regular rodent chow for 3 extra weeks. At the end of the diet exposure (6 weeks), 4 animals were sacrificed for histological studies and 4 animals for electrophysiological and electromyographical recordings. Then, during the next three weeks after diet exposure, 4 animals were sacrificed each week for electrophysiological and electromyographical recordings (Figure 1).

**Figure 1.** Timeline of the experimental procedure. CAD, cafeteria diet; HFD, high-fat diet; EMG, electromyography (endplate noise recordings); EFG, electrophysiology (intracellular recordings).

The diets supplied were:

All animals were provided with regular rodent chow (SAFE A04 diet, Panlab, Barcelona, Spain) ad libitum. The composition of this diet was shown in Table 1.


**Table 1.** Nutritional facts of the diets used.

The nutritional data are expressed per 100 g. Cafeteria diet (CAD). High-fat diet (HFD).

The cafeteria diet (CAD) used in this study consisted of industrial pastries rich in saturated fat (cakes and cookies filled with chocolate) and fried peanuts [19–23]. The average composition is shown in Table 1. Every 2 days, the chopped pastries and peanuts were introduced together in the cage for the animals to eat *ad libitum*. Each time, the old cafeteria diet leftovers were removed and a new ration supplied. At all times, the animals had free access to their usual feed, so the animals continued to ingest the necessary nutrients so as not to suffer any nutritional deficiency. To ensure the amount of cafeteria diet eaten by the mice, the food introduced into the cage was weighed and the food debris was reweighed when removed. Each cage of 2 animals consumed 56 g of the CAD diet per week.

The high-fat diet (HFD), the other type of diet used in this study to induce overweight in animals was a diet enriched in fat (SAFE Diets: 230 HF Rat & Mouse Diet, Augy, France). Unlike the cafeteria diet, this diet is free of additives, colorants, stabilizers and flavorings that could interfere with the results [24–26]. The composition of this diet is shown in Table 1. This type of diet was placed inside the cage to facilitate its availability. As with the cafeteria diet, the food was weighed before placing it in the cage and when removing the remains in the next supply. At all times, the animals also had free access to their usual feed to ensure that they continued to ingest the necessary nutrients and not incur any nutritional deficiencies. Each cage of 2 animals consumed 60 g of the HFD diet per week.

#### *2.2. Muscles*

Animals were deeply anesthetized with isoflurane before being euthanized by exsanguination. The *levator auris longus* (LAL) was excised and dissected on a Sylgard-coated Petri dish containing normal Ringer solution (containing (in mM): 135 NaCl, 5 KCl, 2.5 CaCl2, 1 MgSO4, 1 NaH2PO4, 15 NaHCO3 and 11 glucose) continuously bubbled with 95% O2/5% CO2. The LAL muscles were used for methylene blue staining and immunologically labeled. The gastrocnemius muscles were used for electromyographic recordings. LAL muscles were used for Sudan III fat staining. The LAL is a small, flat muscle located immediately under the skin of the murine skull and is extremely useful for intracellular recording techniques (to visualize the muscle fibers and localize the possible synapses requires flat, thin, and transparent muscles). It is also useful for histological techniques since it does not require microtomy, thus minimizing the appearance of the artifacts.

#### *2.3. Sudan III*

This histological classical staining was performed in the LAL muscle of all the experimental groups and the controls at the end of the period of exposure (6 weeks). Sudan III stains lipids orange-red [27]. The LAL muscles were extracted and fixed in formalin.

The LAL muscle is a flat, thin muscle that does not require microtomy. Whole LAL muscles were immersed in the Sudan III preparation (50 mL of 50% alcohol, 50 mL acetone, 1 g Sudan III—Sigma-Aldrich, Steinheim, Germany) for 5 min. After cleaning the excess dye with 50% alcohol, a methylene blue contrast stain was performed (1 min). After washing off the excess under tap water, it was mounted on glycerin for visualization.

#### *2.4. Endplate Noise Recordings*

Electromyography (EMG) recordings were obtained from an anesthetized animal at controlled room temperature (22 ◦C–25.8 ◦C). The muscle used for this study was the gastrocnemius because of its ease of access and suitability. Recordings were obtained with an electromyography system (MedelecMystro plus, GR20) using a monopolar EMG needle (Natus Manufacturing Limited, London, UK) [17]. The needle was slowly inserted into the muscle and once inside, it was moved in order to enable recording in all directions. The muscle was divided into twelve areas to cover both the entire muscle and avoid recording the same endplate noise twice [17]. The recording needle was introduced into the gastrocnemius until an audible change was heard. The electromyography screen was then checked and if correct (without an alternating current, artifacts, etc.), the endplate noise was recorded. The number of areas with endplate noise (maximum twelve) and the frequency (number of potentials per second that appeared, expressed in Hz) were recorded.

#### *2.5. Electrophysiology: Intracellular Recordings*

Spontaneous miniature endplate potentials (MEPPs) were recorded intracellularly with conventional glass microelectrodes filled with 3 M KCl (20–40 MΩ). Records were rejected if the membrane potential was <−50 mV or if it fell by more than 5 mV during the recording period.

The recording electrodes were connected to an amplifier (Tecktronics, AMS02, Tektronix, Inc., Beaverton, OR, US). A distant Ag–AgCl electrode connected to the bath solution via an Agar bridge (Agar 3.5% in 137 mMNaCl) was used as a reference. The MEPPs were digitized (DIGIDATA 1200 Interface, Axon Instruments Inc, San Jose, CA, USA), stored, and analyzed using a computer. The Axoscope 10.2 was used (Axon Instruments Inc.) for data acquisition and analysis. The MEPP frequency was recorded for 100 s from at least 15 different neuromuscular junctions and the mean values

were determined. The mean amplitude (mV) per fiber was calculated and corrected for non-linear summation [28], assuming a membrane potential of –80 mV.

#### *2.6. Statistical Procedure*

Values are expressed as the mean ± SEM. In some instances, the values are expressed as "percentage of change". This is defined as: (experimental value/control value) × 100. We used the two-tailed Welch's *t*-test for unpaired values because our variances were not equal. This test was chosen as it is more conservative than the ordinary *t*-test. Differences were considered significant at *p* < 0.05.

#### **3. Results**

#### *3.1. Body Weight Evolution*

All the mice in each of the three groups increased their body weight during the first 6 weeks. However, the two groups supplemented with the hypercaloric diets (CAD and HFD) increased their weight over the values of the control group (Table 2). At the end of the exposure, at 6 weeks, the group of mice subjected to a cafeteria diet increased their body weight by 52% more than the controls and the group of mice subjected to a high-fat diet increased their weight by 45% over the weight of the controls.



Values are expressed as the mean ± SEM. Cafeteria diet group, CAD. High-fat diet group, HFD. \*, *p* < 0.05 with respect to the weight of the control animals.

By suppressing the supplementation of the cafeteria and the high-fat diets and maintaining the usual rodent chow, both groups reduced their weight from the first week to match the weight of the controls of the same age (Table 2).

It is confirmed that exposing the mice to 6 weeks of either a cafeteria diet or a high-fat diet causes overweight. By eliminating the hypercaloric diet supplementation, the animals reduced the previously acquired overweight by the first week.

#### *3.2. Muscle Fat*

In the group subjected to a cafeteria diet, a greater amount of adipocytes appeared between the muscle fibers of the LAL (Figure 2B) than in the controls (Figure 2A) and the HFD group (Figure 2C). This technique was applied to four animals per group (control, CAD and HFD).

**Figure 2.** Intramuscular adipocytes. The adipocytes were stained with Sudan III. Fat looks orange. Methylene blue has been used as a contrast dye. (**A**) *Levator auris longus* (LAL) muscle from a control animal. (**B**) LAL muscle from an animal exposed to a cafeteria diet for 6 weeks. (**C**) LAL muscle from an animal exposed to a high-fat diet for 6 weeks. Initial magnification 400×.

#### *3.3. Electrophysiology: Intracellular Recording*

As shown in Figure 3, at the end of the 6 weeks of exposure to the cafeteria diet, a significant increase in the frequency of MEPPs was observed, which is maintained in the following three weeks after withdrawing from the cafeteria diet. On the other hand, at the end of the 6 weeks of exposure to the high-fat diet, a potent increase in the frequency of MEPPs was also observed, greater than in the CAD group (Figure 3). This increase in spontaneous neurotransmission decreased immediately from the first week after withdrawing from the high-fat diet, but remained high for the next 3 weeks.

**Figure 3.** Intracellular recordings. (**A**) Frequency of miniature endplate potentials (MEPPs) expressed as number of events per minute. (**B**) Mean amplitude of the MEPPs expressed in mV. Grey area, period in which the animals were exposed to the CAD or HFD diets. Values are expressed as the mean ± SEM. For each experimental series, *N* = 4 animals. \* *p* < 0.05 with respect to control values. Cafeteria diet group, CAD. High-fat diet group, HFD.

The size of the MEPPs did not change at any time in any of the experimental groups (Figure 3B). The data in Figure 3 are included in Supplementary Materials (Table S1).

#### *3.4. Electromyography*

Upon the hypercaloric diets, the number of areas with plaque noise increased similarly in both the CAD and HFD groups (Figure 4A). When the diet was withdrawn, the number of areas with plaque noise in the CAD group remain elevated for 2 weeks but in the HFD group, it was elevated for only one week.

**Figure 4.** Electromyography. (**A**) Number of average areas with plate noise. (**B**) Each area with endplate noise was analyzed by quantifying the number of events/s (Hz). Grey area, end of the period in which the animals were exposed to the diets. Values are expressed as the mean ± SEM. For each experimental series, *N* = 8 gastrocnemius muscles from 4 animals. \* *p* < 0.05 with respect to control values.

Regarding the number of events in each area with plaque noise, a similar increase was obtained in the MEPPs record (Figure 4B): it was more powerful in the HFD group than in the CAD group. When withdrawing the dietary supplementation, as with the MEPPs recording, the HFD group returned to the control values faster than the CAD group. However, at 3 weeks, the two groups achieved control values. The data in Figure 4 are included in Supplementary Materials (Table S2).

In summary, the increase in the release of ACh caused by the CAD and HFD diets tends to last longer in time once the supplementation is withdrawn.

#### **4. Discussion**

The main hypothesis of this study is that the accumulation of fat in the skeletal muscle of overweight individuals causes an increase in the spontaneous neurotransmission at the neuromuscular junction.

#### *4.1. Overweight*

Different types of diets useful in achieving overweight rodents have been described [19]. Initially, in the present study, a hypercaloric CAD was used. This type of diet has been shown to produce overweight mice by increasing the accumulation of fat in the different tissues [20]. However, this type of

diet contains several substances such as preservatives, colorings, salt, and processed sugars that could have an effect beyond the induction of overweight. In a bid to isolate the effects of being overweight, a second intervention group was used, which was administered as HFD, which has also been shown to induce overweight in mice [24], but which is free of the remaining substances that could interfere with the results.

Our results confirm that 6 weeks of exposure to both a CAD as used in our study, as well as a HFD, are sufficient to achieve a significant increase in the weight of the mice (50%) compared to individuals of the same age on a normal diet. Therefore, exposure to these diets is a suitable model for studying overweight.

Similarly, the weight gain achieved with both diets is quickly reversed to a normal diet from the first week. In this sense, other studies such as the one by Reynés et al. [21] have shown that it is possible to reverse the overweight caused by the administration of a cafeteria diet for 1.5 months following a normal diet for 1.5 months in rats.

Furthermore, the rapid weight reduction obtained in the first week of withdrawal from the cafeteria diet in the present study was consistent with the results obtained by other authors such as Lalanza et al. [22] These authors described that after one week of withdrawal from the cafeteria diet, the overweight caused in rats subjected to 2 months of the cafeteria diet was reversed. This rapid decrease in the weight of the animals was justified by Rogers in 1985 [23] by which the change from a more "tasty" diet to a diet based on feed causes severe hypophagia, especially in the first days after the change.

#### *4.2. Muscle Fat*

Once the weight gain of the mice in both groups was confirmed, it is interesting to see if this weight gain was accompanied by a greater presence of fat in the muscle tissue. The Sudan III technique allows identifying a greater number of adipocytes between the muscle fibers only in the group subjected to a cafeteria diet. However, in the group subjected to the fat diet, despite suffering the same weight gain, no differences were observed in the accumulation of adipocytes in the muscle tissue compared to the control group. Little variations in the diet composition are an important issue to take into consideration as the Cafeteria diet is higher in carbohydrates and levels of proteins are lower than those in the HFD.

There are previous studies that compare the cafeteria diet with a diet rich in fat for rats and describe, in addition to an increase in weight, an increase in visceral and subcutaneous fat. These studies showed that the changes caused by the cafeteria diet are greater than those caused by the high-fat diet [19,21]. It has been proposed that the cafeteria diet induces a greater hyperphagia in animals than the high-fat diets that cause a stabilization in the amount of daily intake after the first weeks [21]. In the present study, only muscle fat was studied and no assessment of other fat deposits was made. However, both groups significantly increased their weight, which suggests that HFD also caused accumulations of visceral fat, although, it did not infiltrate the muscle tissue.

Kahn et al. in 2015 [29] described an increased presence of intramuscular fat in mice which are subjected to a high-fat diet. However, these authors maintained the exposure to a high-fat diet for a much longer time—24 weeks—than in the current study. It is possible that to obtain an increase in muscle fat, the HFD group needs a longer exposure time to the diet. Note that in that study [29] an increase in macrophages in the muscle was demonstrated before the muscle fat was increased. In the present study, the study of the presence of fat 3 weeks after stopping the cafeteria diet was not carried out since weight is normalized.

#### *4.3. Spontaneous Neurotransmission*

There is an increase in spontaneous neurotransmission in mice subjected to both diets, this increase was more important in the HFD group than in CAD. When the diet was withdrawn, the HFD group maintained this increase for more weeks than the CAD group. The alterations registered intracellularly and those registered with EMG do not behave the same since the intracellular recording is a more sensitive technique than EMG and therefore detects more subtle changes.

In the intracellular registry, the amplitude of the MEPPs was not modified in any experiment. This suggests that the possible accumulation of intramuscular fat does not affect the functional integrity of muscle fiber. In this sense, Garcia et al. [30], exposing muscles to anti-motor neuron IgG, injured the axons, but the muscle remains preserved, in such a way as to obtain a pathological increase in MEPPs without any variation of amplitude.

The supplementation of both diets causes an increase in spontaneous neurotransmission at the neuromuscular junction that persists for a time longer than the exposure to the diet. This result confirms only part of the hypothesis in the present study, which states that high-calorie diets cause an alteration in spontaneous muscle neurotransmission. However, the results obtained do not corroborate a correlation of this increase in neurotransmission with being overweight and the accumulation of muscle fat, since an increase in neurotransmission was found in the HFD despite no accumulation of muscle fat. These results open new questions regarding the mechanism by which an increase in spontaneous muscle neurotransmission is brought about by this type of diet. However, the fact that coherent changes appear in muscles as disparate functional strengthens the hypothesis that the studied effects of these diets on neurotransmission must be systemic in nature.

A possible factor that may explain these results is the involvement of the sympathetic nervous system. In a 2012 review, Smith and Minson [31] stated that there is sufficient evidence to show that increased fat deposits correlate with increased activity of the sympathetic nervous system in certain tissues such as the kidneys or skeletal muscle. Previously, in 1994, Scherrer [32] demonstrated that individuals with a BMI > 27 have a rate of sympathetic discharge to skeletal muscle that is twice as high as in lean individuals. On the other hand, Chen et al. [33] demonstrated that the activity of the sympathetic nervous system blocked by phentolamine may decrease the spontaneous neurotransmission recorded in a PGM. In addition, McNultty and colleagues [34] demonstrated that psychological stress can increase the spontaneous neurotransmission recorded in a PGM. It should be remembered that the neuromuscular junction of skeletal muscle in mice is innervated by the sympathetic nervous system [35]. Based on these data, it can be suggested that the mechanism by which the CAD is capable of increasing spontaneous neurotransmission is due to an excitation of the sympathetic nervous system caused more by the accumulation of fat than by the inflammatory state of the muscle secondary to fat. On the other hand, the spontaneous neurotransmission elevated and maintained for several weeks after stopping the diet and normalizing the weight of the mice could also be due to the increased activity of the sympathetic nervous system, in this case, it is secondary to the stress suffered by the animals when being deprived of the hypercaloric diet. This last situation has been described by Lalanza and co-authors [22], who observed that by suppressing a cafeteria diet in rats, an increase in their anxiety levels is caused. In this sense, it is known that the following diets based on tasty foods rich in fat can produce an addictive effect and its suppression generates a withdrawal syndrome that partly explains this increased anxiety [36,37].

On the other hand, the increase in neurotransmission may be due to the development of a pro-inflammatory state in the muscle because of diet. In this sense, Khan et al. [29] demonstrated that after only 2 weeks of HFD, macrophages already appear in the muscle fat tissue of mice. Furthermore, Fink et al. [25] showed that for mice exposed to a diet rich in fat, abundant pro-inflammatory macrophages and neutrophils appear early and that it increases with time. Taken together, these data suggest that from the first weeks of exposure to high-fat diets, a pro-inflammatory state begins to develop in muscle tissue, even before significant weight gains are achieved. In other words, changes could occur in the release of ACh without accumulations of fat in the muscle, as it occurs for the mice exposed to HFD in the present study.

A common factor for the two types of diet used in this study is their high content of saturated fat. It is well known that saturated fats have a direct effect on the immune system by activating the release of pro-inflammatory cytokines [38]. In a recent study, Song et al. [26] found that subjecting rats to a diet

rich in fat produced an increase in the postoperative pain in them. This fact coincides with other similar studies evaluating other types of pain models such as inflammatory or neuropathic [39,40]. In addition, Song et al. [26] demonstrated that a single week of a high-fat diet was not capable of producing obesity, but an increase in the pain response, although of lesser magnitude. That is, the increase in pain seems to be more related to the diet itself than to obesity. In addition, we also observed that the changes caused by an HFD returned faster to normality than the changes produced by cafeteria diet, which are maintained for one more week. This can be related to fat accumulation in muscle. All the changes observed between the two diets deserve more investigation and provides a model to investigate systemic vs. local effects of fat accumulation.

Within the possible relationship between the nervous system and the increase in spontaneous neurotransmission, some authors have proposed that some of the persistent changes in obese individuals after losing weight may be due to phenomena related to synaptic neuroplasticity [40]. In the results obtained in the present study, this phenomenon could intervene in the maintenance of elevated spontaneous neurotransmission after stopping the diet.

Several studies correlate the administration of high-fat diets or cafeteria diets with being overweight and with significant increases in the pain response of different pain models: neuropathic, postoperative, and inflammatory [26,39,40]. Currently, there are no studies that assess whether this type of diet has any influence on myofascial pain. The central factor of MPS are the MTrPs and their essential characteristic is an increased spontaneous neurotransmission [18]. The present study suggests that this type of hypercaloric diets could facilitate the development of MTrPs by increasing the spontaneous neurotransmission they generate. However, it is not clear if the effects come exclusively from diet or if being overweight and accumulating fat may also play a relevant role.

#### **5. Conclusions**

Exposure for six weeks to a hypercaloric diet (cafeteria or highfat) causes overweight in mice, an increase in adipocytes in muscle tissues and an increase in spontaneous neuromuscular neurotransmission. Upon abandoning the diets, the mice recover their weight rapidly, but spontaneous neuromuscular neurotransmission remains elevated. In other words, the alteration of spontaneous neurotransmission is not exclusively related to being overweight or to an increase in muscle fat. Overall, it can be concluded that exposure to a hypercaloric diet for 6 weeks in mice may be a predisposing factor for the development of MPS and other muscle alterations aggravated by a maintained increase in spontaneous neuromuscular neurotransmission.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2072-6643/12/10/3216/s1, Table S1: Figure 3 data, Table S2: Figure 4 data.

**Author Contributions:**Methodology, investigation: C.G.-D.; conceptualization, formal analysis: M.B.; investigation, resources, roles/writing—original draft: A.V.; validation, visualization: G.V.; supervision, writing—review and editing: M.M.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** We greatly appreciate the suggestions of M.T. Colomina in the draft of this article.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

## l**-Cysteine and Vitamin D Co-Supplementation Alleviates Markers of Musculoskeletal Disorders in Vitamin D-Deficient High-Fat Diet-Fed Mice**

#### **Rajesh Parsanathan, Arunkumar E. Achari, Prasenjit Manna and Sushil K. Jain \***

Department of Pediatrics and Center for Cardiovascular Diseases and Sciences, Louisiana State University Health Sciences Center, 1501 Kings Highway, Shreveport, LA 71130, USA; rparsa@lsuhsc.edu or rajesh.uom@gmail.com (R.P.); bioarun1985@gmail.com (A.E.A.); pmanna2012@gmail.com (P.M.) **\*** Correspondence: sjain@lsuhsc.edu; Tel.: +1-318-675-6086

Received: 24 September 2020; Accepted: 3 November 2020; Published: 6 November 2020

**Abstract:** Vitamin D (VD) deficiency is associated with musculoskeletal disorders. This study examines whether co-supplementation of l-cysteine (LC) and VD is better than monotherapy with LC or VD at alleviating musculoskeletal dyshomeostasis in the skeletal muscle of VD-deficient high-fat diet (HFD-VD-) fed mice. Mice were fed a healthy diet or an HFD; for VD-deficient animals, the mice were maintained on a HFD-VD-diet (16 weeks); after the first 8 weeks, the HFD-VD-diet-fed mice were supplemented for another 8 weeks with LC, VD-alone, or the same doses of LC + VD by oral gavage. Saline and olive oil served as controls. Myotubes were exposed with high-glucose, palmitate, Monocyte Chemoattractant Protein 1 (MCP-1), and Tumor Necrosis Factor (TNF), to mimic the in vivo microenvironment. In vitro deficiencies of glutathione and hydrogen sulfide were induced by knockdown of GCLC and CSE genes. Relative gene expression of biomarkers (myogenic: MyoD, Mef2c, Csrp3; muscle dystrophy: Atrogin1, Murf1, and Myostatin; bone modeling and remodeling: RANK, RANKL, OPG) were analyzed using qRT-PCR. Co-supplementatoin with LC + VD showed beneficial effects on gene expression of myogenic markers and OPG but reduced markers of dystrophy, RANK/RANKL in comparison to LC or VD alone-supplementation. In vitro myotubes treated with glutathione (GSH) precursors also showed a positive effect on OPG and the myogenesis genes, and inhibited RANK/RANKL and muscle-dystrophy markers. This study reveals that the co-supplementation of LC with VD significantly alleviates the markers of musculoskeletal disorders in the skeletal muscle better than monotherapy with LC or VD in HFD-VD-fed mice.

**Keywords:** vitamin D deficiency; l-cysteine; glutathione; myogenic markers; dystrophy markers; skeletal muscle

#### **1. Introduction**

Vitamin D (VD) deficiency or insufficiency is associated with diseases affecting muscle and bone health [1–3]. Low blood levels of both 25-hydroxyvitamin D (25(OH)D) and glutathione (GSH) are positively associated with metabolic syndrome in human subjects [4–8]. Antioxidant GSH deficiency increases the oxidative stress that may favor endogenous protein oxidative modification, impairs cellular physiology, and leads to the disease's manifestation. Supplementation with GSH or its precursor, the sulfur-containing amino acid l-cysteine (LC), has been successfully used to improve the GSH status in blood and tissues, reducing immune-metabolic syndrome [4–6,8].

Skeletal muscle is the largest tissue in the body, and any loss of function or regenerative properties debilitates the musculoskeletal system [9]. Myogenic markers such as myoblast determination protein 1 (MyoD), myocyte enhancer factor 2C (Mef2c), and cysteine and glycine-rich protein 3 (Csrp3) are positive regulators and promote myogenesis, regeneration, and play an essential role in muscle function [10,11]. Conversely, skeletal muscle-specific F-box protein (Atrogin1), muscle RING-finger protein-1 (Murf1), and Myostatin (Mstn) are critical molecules involved in muscle atrophy [12,13]. Skeletal muscle dystrophy/atrophy is a debilitating consequence of many pathological conditions and diseases [14]. Receptor activator of nuclear factor-kB (RANK), its ligand RANKL, and the soluble decoy receptor osteoprotegerin (OPG) pathway control bone remodeling and homeostasis [15,16]. The effects of RANK/RANKL/OPG extend well beyond its classical functions; in skeletal muscle, interaction with RANKL/RANK causes atrophy and dysfunction, whereas OPG provides significant protection against muscle damage [15].

Studies have shown that high-fat diet (HFD)-induced obesity leads to skeletal muscle oxidative stress, inflammation, and muscle mass loss by decreasing myogenic markers such as MyoD, Mef2c, and Csrp3, increasing muscle dystrophy markers such as Atrogin1, Murf1, and Myostatin [17–19]. Supplementation with a cysteine/thiol-based antioxidant delays or attenuates muscular dysfunction [20–22]. Similarly, supplementation with N-acetyl cysteine decreases osteoclast differentiation and increases bone mass in obese diabetic mice [23]. VD also maintains a normal bone resorption rate and formation through the RANKL/OPG signal [24]. VD deficiency is detrimental to muscle function, independent of alterations in phosphate and calcium levels [25]. Observational studies of VD-deficiency also associate reduced muscle mass and weakness [1–3]. However, interventional trials and meta-analyses of VD deficiency have yielded contradictory findings [25]. Our previous preclinical studies demonstrate that GSH epigenetically regulates VD metabolism genes. Supplementation with the VD + LC combination was more successful at boosting 25(OH)D levels by improving the status of VD metabolism genes in the liver, kidney, and muscle [4–8,26]. However, as far as we know, no previous study has examined the effect of co-supplementation with LC+VD on musculoskeletal markers in the muscle of HFD-VD-mice.

This study reports that LC (a GSH precursor) co-supplementation with VD significantly alleviates dyshomeostasis of the skeletal muscle in VD-deficient high-fat diet-fed mice, suggesting that combined supplementation with the nutraceuticals LC + VD could be a better option for musculoskeletal system disorders rather than supraphysiological monotherapy with VD alone.

#### **2. Materials and Methods**

The reagents used in the study, and all other chemicals were of analytical and molecular grade unless otherwise mentioned, and were purchased from Sigma Chemical Co. (St. Louis, MO, USA).

#### *2.1. Animal Experimental Design and Treatment*

Male C57BL/6J mice (5 weeks old, 20–24 g) were procured from The Jackson Laboratory (Bar Harbor, ME, USA). Mice were given access to food and water ad libitum and housed for acclimation (1 week) in a temperature-controlled room (22 ± 2 ◦C) with light/dark cycles (12/12 h). They were maintained under standard housing conditions throughout the experiment. After receiving approval (P-15-006) from the Institutional Animal Committee, according to the guidelines of the institution's ethical standards, all the procedures were performed.

Mice were randomized, labeled in individual cages, and divided into various groups. The mice were fasted overnight and tested for hyperglycemia by measuring their blood glucose concentrations before starting the treatment plan. Fasting blood glucose was analyzed by the tail prick method using a glucometer (Accu-Chek, Boehringer Mannheim Corp., Indianapolis, IN, USA).

Control animals were fed a healthy diet (Ctrl; lower in fat), while animals in the high-fat diet group were fed a high-fat diet (HFD) for a total of 16 weeks. The mice were maintained on a VD-deficient HFD (HFD-VD-) for 16 weeks (to mimic the VD-deficient condition). After the first 8 weeks, the mice were supplemented by oral gavage for another 8 weeks with either 5 mg LC/kg BW daily (LC), 67 IU VD/kg BW (VD) alone, or the same doses of LC + VD co-supplemented to HFD-VD-mice. Additionally, two control groups of HFD-VD-mice were maintained on oral gavage with either saline (S-Ctrl) or the same dose of the vehicle olive oil used for dissolving cholecalciferol (OO-Ctrl). These diets—(healthy diet, HFD, and HFD-VD-) composition details and dose justification for LC and vitamin D—are given in our recent publication [4].

Sex steroids, estrous cycle, and hormonal impacts (sexual dimorphism) influence musculoskeletal markers [27–31]; therefore, we chose only male mice for this pilot study. Furthermore, choosing a 22 ± 2 ◦C housing temperature better represents humans living in colder regions that also lack environmental temperature controls (to mimic human thermal relation) [32,33].

As mentioned above, at the end of the supplementation, the animals were isoflurane euthanized and then perfused with cold saline. Skeletal muscle (gastrocnemius) was collected immediately, quickly diced, and frozen in liquid nitrogen at −80 ◦C. This model (HFD and HFD-VD-) of dietary-induced insulin resistance created fasting hyperglycemia, hyperinsulinemia, elevated proinflammatory cytokines, decreased glutathione, and VD deficiency, thus representing a reasonable model of the human condition [4–6,34].

#### *2.2. Cell Culture, Treatments, and RNA Interference of GCLC and CSE*

Mouse C2C12 myoblasts (ATCC® CRL-1772™, Manassas, VA, USA) were cultured and brought to myotubes differentiation with appropriate conditions following the methods used in our previous published studies [5,6,35,36]. High glucose (HG; 25 mM), palmitate (0.6 mM) for 24 h; MCP-1 (2.5 ng/mL) and TNF (250 pg/mL) for 6 h treated to the myotubes, the doses and time points justifications given in our previous published studies [5,6,35,36]. GCLC or CSE and Control siRNA-A (50 and 100 nM siRNA), transiently transfected as per the method described earlier [4–6,35,36]. GSH precursor (l-cysteine; 300 μM) or an H2S donor (NaHS; 20 μM) were treated following the methods used in our previous published studies [5,6,35,36] to boost the level of GSH cellular content and H2S production. Cell viability was determined using the Alamar Blue reduction bioassay in all the experimental conditions [5,6,35,36].

#### *2.3. Relative Gene Expression*

RNA isolated from cells or tissues, quality and concentration determined, and 1 μg of RNA samples were reverse transcribed, and the qPCR performed as per our previously published studies [5,6,35,36]. The following primer/probe sets from Applied Biosystems™ TaqMan™ Gene Expression Assays were used for qPCR; MyoD (Mm00440387\_m1), Mef2c (Mm01340842\_m1), Csrp3 (Mm00443379\_m1), Atrogin1/Fbxo32 (Mm00499523\_m1), MuRF1/Trim63 (Mm01185221\_m1), Mstn/Myostatin (Mm01254559\_m1), Rank/Tnfrsf11a (Mm00437132\_m1), RankL/Tnfsf11 (Mm00441906\_m1), Opg/Tnfrsf11b (Mm01205928\_m1) and Gapdh (Mm99999915\_g1). The results were expressed as the relative quantification (RQ, the fold of control).

#### *2.4. Statistical Analyses*

Data were generated from multiple repeats of different biological experiments to obtain the mean values and standard errors of the mean. Significance was set at *p* < 0.05, and statistical differences were evaluated using two-way ANOVA followed by Dunnett's multiple comparisons test between groups were conducted using GraphPad Prism 8.2.1 (GraphPad Software, La Jolla, CA, USA).

#### **3. Results**

#### *3.1. E*ff*ect of HFD, VD-Deficient HFD, and* l*-Cysteine and Vitamin D Co-Supplementation on Gene Expression of Musculoskeletal Markers in Mouse Skeletal Muscle*

The skeletal muscle of HFD-fed mice showed attenuated myogenic markers (MyoD, Mef2c, and Csrp3) (Figure 1a), but there were no significant alterations in muscle dystrophy markers such as Atrogin1, Murf1, and Myostatin (Figure 1b). Only osteoprotegerin was downregulated in the RANK/RANKL/OPG system (Figure 1c). VD-deficient HFD-fed mice's skeletal muscle showed downregulation of myogenic markers similar to those seen in the HFD-fed mice (Figure 1a). However, muscle dystrophy markers increased significantly in the skeletal muscle of the HFD-VD- group

compared to those in the HFD group (Figure 1b). Compared to skeletal muscle in HFD-fed mice, the mRNA level of RANK/RANKL increased significantly in the HFD-VD- group, but the level of OPG was significantly downregulated in HFD-VD- group (Figure 1c).

Groups supplemented with l-cysteine or vitamin D alone showed a partially significant beneficial effect on markers such as MyoD, Mef2c, and OPG (Figure 1a,c). However, supplementation with LC or VD alone, or co-supplementation, significantly suppressed muscle the dystrophy markers, RANK, and RANKL in mouse skeletal muscle compared to results in the HFD-VD- group (Figure 1b). LC and VD co-supplementation more significantly alleviated myogenic markers and OPG (Figure 1a,c) in mouse skeletal muscle compared to results in the HFD and HFD-VD- groups, including those supplemented with LC or VD alone.

These findings indicated that co-supplementing LC with VD enhanced the beneficial effects against musculoskeletal disorder marker gene expression in skeletal muscle compared to monotherapy supplementation with LC or VD.

#### *3.2. Impact of High Glucose, Palmitate, and Inflammatory Cytokines on Musculoskeletal Markers*

Myotubes were exposed to high glucose-mediated glucotoxicity, palmitate-mediated lipotoxicity, and inflammatory cytokines to mimic low-grade inflammation models, which is observed in both obesity and diabetes. Glucolipotoxicity significantly downregulated the mRNA levels of the myogenic markers (MyoD, Mef2c, and Csrp3), and OPG, but the levels of dystrophy markers (Atrogin1, Murf1, and Myostatin), RANK, and RANKL were elevated (Figure 2a) compared to the control group. Inflammatory cytokines did not alter the level of myogenic markers. Proinflammatory cytokines such as MCP-1 and TNF elevated the expression of dystrophy markers, RANK, and RANKL (Figure 2b) compared to that in the control group. Collectively, this result indicates that glucolipotoxicity negatively affects the myogenic markers, dystrophy markers, and RANK/RANKL/OPG system, while inflammatory cytokines in vitro induce dystrophy markers, RANK, and RANKL. They may also contribute to musculoskeletal disorders.

#### *3.3. The Deficiency of Transsulfuration Pathway Key Genes GCLC and CSE (Knockdown) in Myotubes A*ff*ects Musculoskeletal Markers*

The expression of myogenic markers and OPG was attenuated in the GCLC, and CSE siRNA treated myotubes (Figure 3a,b), but the levels of dystrophy markers, RANK, and RANKL increased significantly compared to those of the control group (Figure 3a,b). Altogether, these data demonstrate that inhibited flow in the rate-limiting sulfur-containing amino acid (l-cysteine) pathway leads to a deficiency in the physiological antioxidant glutathione (GSH) or hydrogen sulfide (H2S), which alone or synergistically, may alter the expression of the musculoskeletal marker genes.

#### *3.4. GSH and H2S Inhibit Muscle Dystrophy Markers and Positively Induce Myogenic Markers Genes*

The possible beneficial effect of H2S or GSH on the expression of genes involved in myogenesis, muscle dystrophy, and the RANK/RAKL/OPG system was explored with the antioxidant precursors l-cysteine (a GSH/H2S precursor) or NaHS (an H2S donor) following the methods used in previous publications [4,6]. Results showed that compared to levels in the control group, the mRNA levels of myogenic genes and OPG significantly increased following LC or NaHS treatment, which also decreased dystrophy markers, RANK, and RANKL (Figure 4a). These responses to treatment with LC and NaHS indicate that GSH and H2S may directly or indirectly affect these genes and suggest that H2S and GSH may have a beneficial effect on muscle physiology.

**Figure 1.** The effects of high-fat diet (HFD), vitamin D (VD)-deficient HFD (HFD-VD-), and l-cysteine and vitamin D co-supplementation on gene expression musculoskeletal markers in mouse skeletal muscle. Male C57BL/6J mice (5 weeks old) were fed with standard chow diet (Control; Ctrl), a high-fat diet (HFD), or a VD-deficient HFD for 16 weeks. Mice were gavaged with saline (S-Ctrl), Olive oil (OO-Ctrl), l-Cysteine (LC), Cholecalciferol (VD), or VD + LC during the last 8 weeks. The mRNA levels of myogenic marker genes: myoblast determination protein 1, myocyte enhancer factor 2C, and cysteine and glycine-rich protein 3 (MyoD, Mef2c, and Csrp3) (**a**); dystrophy marker genes: skeletal muscle-specific F-box protein, muscle RING-finger protein-1, and Myostatin (Atrogin1, Murf1, and Myostatin) (**b**); bone modeling and remodeling genes: receptor activator of nuclear factor-kB, receptor activator of nuclear factor-kB ligand, and osteoprotegerin (RANK, RANKL, and OPG) (**c**) were analyzed using qRT-PCR. Results are mean ± SEM (*n* = 4). Two-way ANOVA, followed by Dunnett's multiple comparisons test, was performed between groups. Significance at *p* < 0.05: Asterisk symbol (\*) represents a comparison between control (Ctrl) with all other groups, whereas the hash symbol (#) represents a comparison between HFD-VD- saline and olive oil control (S-Ctrl and OO-Ctrl) with LC, VD, VD + LC co-supplementation groups. MyoD: Myoblast determination protein 1, Mef2c: Myocyte enhancer factor 2C, Csrp3: Cysteine and glycine-rich protein 3, Atrogin1: skeletal muscle-specific F-box protein, Murf1: Muscle RING-finger protein-1, RANK: Receptor activator of nuclear factor-kB, RANKL: Receptor activator of nuclear factor-kB ligand, OPG: Osteoprotegerin.

**Figure 2.** Glucolipotoxicity and inflammatory cytokines (MCP-1 and TNF) affect musculoskeletal markers in myotubes. Myotubes were treated with high glucose (25 mM) or palmitate (0.6 mM) for 24 h. Mannitol was used as an osmolality control (**a**). In another set of experiments, myotubes were treated with MCP-1 (2.5 ng/mL) or TNF (250 pg/mL) for 6 h (**b**). The mRNA levels of target genes responsible for myogenesis, muscle dystrophy, bone modeling, and remodeling (MyoD, Mef2c, Csrp3, Atrogin1, Murf1, Myostatin, RANK, RANKL, and OPG) were analyzed using qRT-PCR (**a**,**b**). Results are mean ± SEM (*n* = 3). Two-way ANOVA, followed by Dunnett's multiple comparisons test, was performed between groups. A *p*-value of <0.05 for a statistical test was considered significant and represented as an asterisk symbol (\*) compared with the control group. MCP-1: Monocyte Chemoattractant Protein 1, TNF: Tumor Necrosis Factor.

**Figure 3.** The deficiency of transsulfuration pathway key genes (GCLC and CSE knockdown) in myotubes affects musculoskeletal markers. Myotubes were transfected with GCLC siRNA (GSH deficient) (**a**) or CSE siRNA (H2S deficient) (**b**). Scrambled siRNA, served as a control. The mRNA levels of target genes responsible for myogenesis, muscle dystrophy, bone modeling, and remodeling (MyoD, Mef2c, Csrp3, Atrogin1, Murf1, Myostatin, RANK, RANKL, and OPG) were analyzed using qRT-PCR (**a**,**b**). Results are mean ± SEM (*n* = 3). Two-way ANOVA, followed by Dunnett's multiple comparisons test, was performed between groups. A *p*-value of <0.05 for a statistical test was considered significant and represented as an asterisk symbol (\*) compared with the control group.

**Figure 4.** Glutathione (l-cysteine) or hydrogen sulfide (sodium hydrosulfide) supplementation alters musculoskeletal markers in myotubes. Myotubes were treated with either l-cysteine (LC; 300 μM) or sodium hydrosulfide (NaHS; 20 μM) for 6 h. The mRNA levels of target genes responsible for myogenesis, muscle dystrophy, bone modeling, and remodeling (MyoD, Mef2c, Csrp3, Atrogin1, Murf1, Myostatin, RANK, RANKL, and OPG) were analyzed using qRT-PCR (**a**). Results are mean ± SEM (*n* = 3). Two-way ANOVA, followed by Dunnett's multiple comparisons test, was performed between groups. A *p*-value of <0.05 for a statistical test was considered significant and represented as an asterisk symbol (\*) compared with the control group. Control animals were fed a healthy diet (Ctrl; lower in fat), while animals in the high-fat diet group were fed a high-fat diet (HFD) for a total of 16 weeks (not shown in scheme). The mice were maintained on a VD-deficient HFD (HFD-VD-) for 16 weeks (to mimic the VD-deficient condition). After the first 8 weeks, the mice were supplemented by oral gavage for another 8 weeks with either 5 mg LC/kg BW daily (LC), 67 IU VD/kg BW (VD) alone, or the same doses of LC + VD co-supplemented to HFD-VD-mice. The markers of Myogenic: MyoD, Mef2c, and Csrp3; Muscle dystrophy: Atrogin1, Murf1, and Myostatin; Bone modeling and remodeling: RANK, RANKL, and OPG were analyzed. Myogenic markers and OPG decreased in HFD-VD-mice muscle, whereas muscle dystrophy markers increased significantly. LC + VD co-supplementation to HFD-VD-mice ameliorate partially or entirely all the markers mentioned above at par with control groups (**b**).

#### **4. Discussion**

Vitamin D (VD) is a nutrient essential for maintaining good bone health and improving muscle strength [1–3,37,38]. VD deficiency or insufficiency is associated with various musculoskeletal disorders [39–41]. After multivitamins, vitamin D, by itself, is the second-highest vitamin supplement consumed by the public for better health and delay or prevent musculoskeletal disorders [41–44]. However, controlled clinical studies show that VD alone supplementations' have limited therapeutical benefits, despite the clinical association between VD deficiency and disease outcome [4,7,26,45]. This study examined the hypothesis that the co-supplementation of l-cysteine (LC) with VD is better compared to monotherapies with LC or VD at alleviating dyshomeostasis in the skeletal muscle of VD-deficient high-fat diet-fed (HFD-VD-) mice.

This study reports that LC+VD co-supplementation showed significant beneficial effects on vital myogenic markers such as MyoD, Mef2c, and Csrp3 in an animal model of HFD-VD-. Further, in vitro studies carried out in mouse myotubes demonstrated that, while H2S/GSH deficiency (oxidative stress or antioxidant deficient condition), high glucose, and palmitate (metabolic insults) decreased myogenic markers, inflammatory cytokines, such as TNF and MCP-1, did not affect the markers of myogenesis. Previously it has been shown that low physiological concentrations (a deficient state) of 1,25(OH)2D3 (active VD) induces transdifferentiation of muscle cells into adipose cells (adipogenesis), whereas higher (physiological and supraphysiological) concentrations attenuate this effect and promote myogenic cell differentiation [46,47]. Further, VD ameliorates fat accumulation with AMPK/SIRT1 pathway activation in myotubes [48,49]. Moreover, GSH depletion and chronic inflammation impair myogenic differentiation through redox-dependent and independent pathways, while these effects are reversible following NAC or GSH replenishment [50]. GSH and H2S levels affect the intracellular redox state. In vitro supplementation of LC and NaHS shows a positive effect on myogenic markers. Additionally, preclinical studies have shown that co-supplementation of LC + VD significantly reduced oxidative stress by boosting GSH and positively upregulating the VD-regulatory genes (VDBP/VD-25-hydroxylase/VDR) epigenetically in the liver of mice in a 25(OH)D deficiency mouse model [4–6]. GSH optimization, along with VD is a better approach to alleviate myogenic genes. Therefore, reduction or prevention of oxidative imbalance and VD in muscle is of vital importance for the maintenance of the myogenic pathway.

GSH is a critical redox factor that mitigates oxidative stress and oxidative damage to endogenous proteins, impairs cellular physiology and leads to the manifestation of the disease. Supplementation with the GSH rate-limiting amino acid precursor l-cysteine has been used successfully to improve the GSH status, VD metabolism genes, and lower the incidence of immune-metabolic syndrome [4–8]. In the present study, supplementation with LC, along with VD, suppressed skeletal muscle dystrophy markers such as Atrogin1, Murf1, and Myostatin and RANK and its ligand (RANKL). VD deficiency is associated with muscle atrophy. In vitro metabolic insults, exposure to inflammatory cytokines, and antioxidant deficient conditions (H2S/GSH deficiency) induced muscle dystrophy markers along with RANK/RANKL.

Conversely, OPG showed an inverse relationship, unlike RANK/RANKL. Previous studies have shown that the muscle content of Atrogin1 was the highest in patients deficient in VD and lowest in patients sufficient in it, whereas VD supplementation seemed to repel atrophic changes and systemic inflammatory markers [51]. Antioxidants maintain muscle homeostasis, whereas disturbed redox status, known as the major contributing factor towards atrophy. N-Acetyl cysteine (NAC) treatment reduces muscle atrophy through beneficial antioxidant, anti-inflammatory, and anti-fibrotic effects [52]. Further, GSH/NAC inhibits RANK-L induced osteoclastogenesis both in vitro and in vivo [53]. VD also maintains a regular rate of bone resorption and formation through the receptor activator of the nuclear factor κB (RANK)/RANK ligand (RANKL)/osteoprotegerin axis [54]. Our study shows that antioxidant supplementation with LC and NaHS suppresses muscle dystrophy markers, RANK/RANKL, and boosts the levels of OPG. Hence, LC + VD co-supplementation treatment was extremely effective against the myopathic changes contributing to pathophysiology.

#### **5. Conclusions**

In VD insufficiency/deficiency, muscle function and physical function may be impaired before clinical or biochemical signs of musculoskeletal disease are evident. Low circulating levels of both glutathione (GSH) and 25(OH)D are positively associated with metabolic syndrome and poor health in human subjects [4]. A possible explanation for the limited success of clinical trials with VD-alone could be the need to simultaneously optimize deficiencies in essential antioxidant nutrients such as LC along with VD. Therefore, an efficient novel therapeutic strategy would be using the combined nutraceuticals LC + VD, which could simultaneously antagonize cellular oxidative stress and inflammation and thus provide a better option for musculoskeletal system disorders than supraphysiological monotherapy with VD [7]. Current data obtained from preclinical studies in a 25(OH)D deficient mouse model suggest that LC + VD more effectively boosted the actions/efficacy of VD on musculoskeletal markers than monotherapies with LC or VD (Figure 4b). Our data support the need for a clinical trial of co-supplementation of LC with VD to achieve better health outcomes, including skeletal muscle functions.

**Author Contributions:** Conceptualization, R.P. and S.K.J.; methodology, R.P.; software, R.P.; validation, R.P. and S.K.J.; formal analysis, R.P. and S.K.J.; investigation, R.P., A.E.A. and P.M.; resources, S.K.J.; data curation, R.P., A.E.A. and P.M.; writing—original draft preparation, R.P.; writing—review and editing, R.P. and S.K.J.; visualization, R.P.; supervision, S.K.J.; project administration, R.P. and S.K.J.; funding acquisition, R.P. and S.K.J. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was supported by Malcolm W. Feist Cardiovascular Research Fellowship to R.P. and the Endowed Chair in Diabetes to S.K.J. from the Center for Cardiovascular Diseases and Sciences (CCDS), LSUHSC-Shreveport, as well as grants to S.K.J. from the National Institutes of Health/National Center for Complementary and Integrative Health (RO1 AT007442, 2013-16 and 1 R33 AT010637-01A1, 2020-2023).

**Acknowledgments:** We thank Paula Polk, Manager and Wiola Luszczek, Research Specialist at the Research Core Facility at LSUHSC-Shreveport for their expert technical assistance. We also thank William E. McLean and Christopher M. Stevens for lab assistance. The authors thank Georgia Morgan for excellent editing.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

#### **Abbreviations**


#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Review* **Metabolic and Nutritional Issues Associated with Spinal Muscular Atrophy**

#### **Yang-Jean Li 1, Tai-Heng Chen 2,3,\*, Yan-Zhang Wu <sup>2</sup> and Yung-Hao Tseng <sup>2</sup>**


Received: 31 October 2020; Accepted: 15 December 2020; Published: 16 December 2020

**Abstract:** Spinal muscular atrophy (SMA), the main genetic cause of infant death, is a neurodegenerative disease characterized by the selective loss of motor neurons in the anterior horn of the spinal cord, accompanied by muscle wasting. Pathomechanically, SMA is caused by low levels of the survival motor neuron protein (SMN) resulting from the loss of the *SMN1* gene. However, emerging research extends the pathogenic effect of SMN deficiency beyond motor neurons. A variety of metabolic abnormalities, especially altered fatty acid metabolism and impaired glucose tolerance, has been described in isolated cases of SMA; therefore, the impact of SMN deficiency in metabolic abnormalities has been speculated. Although the life expectancy of these patients has increased due to novel disease-modifying therapies and standardization of care, understanding of the involvement of metabolism and nutrition in SMA is still limited. Optimal nutrition support and metabolic monitoring are essential for patients with SMA, and a comprehensive nutritional assessment can guide personalized nutritional therapy for this vulnerable population. It has recently been suggested that metabolomics studies before and after the onset of SMA in patients can provide valuable information about the direct or indirect effects of SMN deficiency on metabolic abnormalities. Furthermore, identifying and quantifying the specific metabolites in SMA patients may serve as an authentic biomarker or therapeutic target for SMA. Here, we review the main epidemiological and mechanistic findings that link metabolic changes to SMA and further discuss the principles of metabolomics as a novel approach to seek biomarkers and therapeutic insights in SMA.

**Keywords:** spinal muscular atrophy; metabolomics; nutrition; therapeutics; biomarkers

#### **1. Introduction**

Spinal muscular atrophy (SMA) is a congenital neuromuscular disease characterized by progressive muscle weakness resulting from the degeneration of motor neurons (MN) in the spinal cord [1]. Although SMA is considered a rare disease and the global incidence of live births is estimated to be about 1/10,000, SMA is still the second most common autosomal recessive genetic disease and the most common monogenic disorder that causes early infant death [2]. The carrier frequency varies from 1 in 38 to 1 in 72 among different ethnic groups, with a pan-ethnic average of 1 in 54 [3,4].

In a pathological view, SMA is resulted from an insufficient level of a 38 kDa protein, called the survival motor neuron (SMN), as a result of homologous deletion or mutation of the *Survival of Motor Neuron 1* (*SMN1*) gene [5]. Subsequent studies showed that two genes encode SMN protein in humans: *SMN1* and a 99% identical copy in sequence, known as *SMN2*. Indeed, *SMN2* mainly differs from

*SMN1* by a single nucleotide (C-to-T) substitution in the exon 7 [6]. Such a critical variant results in exon 7 exclusion in most transcripts (90%) of *SMN2*, SMNΔ7. Unlike the *SMN1* gene, *SMN2* can only produce 10 % full-length (FL) SMN [7]. Given that the residual FL-*SMN2* transcripts can compensate for defect *SMN1* to a limited extent, the SMA severity is partially rescued by *SMN2* copy numbers [8]. However, the correlation between this phenotype and genotype is not absolute, and recent studies have pointed out that other potential cellular mechanisms may also be involved in modifying the clinical severity of SMA [9].

It is still unclear whether the pathogenesis of SMA is caused by a specific pattern or a combination of dysregulated effects. The cell-autonomous effects due to SMN deficiency are the main causes of MN degeneration; however, it cannot be explained for the full SMA phenotype, implicating not only dysregulated neural networks but other non-neuronal cell types involved in the SMA pathology [10,11]. Emerging research extends the pathogenic effect of SMN deficiency beyond the MN, including other cells inside and outside the central nervous system, so that many peripheral organs and non-neural tissues show pathological changes in preclinical SMA models and diseased patients (Figure 1) [12–14]. Furthermore, increasing evidence suggests metabolic abnormalities in patients with SMA, such as altered fatty acid metabolism, impaired glucose tolerance, and muscle mitochondria defects [15–17]. Recent studies also indicate that many SMA patients are either undernourished, underfed, or overfed [18]. Notably, in some SMA patients, metabolic dysregulations may even present before their first neuromuscular signs [19]. These findings suggest that SMN is essential for the survival of motor neurons and affects certain enzyme production in the metabolism.

**Figure 1.** Overview of non-neuromuscular systemic pathology in spinal muscular atrophy (SMA). A summary of multi-organ involvement has been reported in SMA animal models and/or patients [17,20–35].

Over the last few years, the increased life expectancy of SMA patients has been achieved through the invention of novel therapies and the standardization of clinical care. However, knowledge of the altered metabolism and nutrition in SMA remains limited. The impact of SMN deficiency on metabolic abnormalities has been recently proposed. Before and after the onset of the disease, metabolomics studies in SMA patients can provide valuable information about the direct or indirect effects of SMN deficiency on metabolic abnormalities [13]. The present review will discuss the current knowledge regarding the metabolic involvement in SMA and the role of metabolomics in pursuing potential biomarkers and therapeutic insights for SMA.

#### **2. Lipid Metabolic Abnormalities in SMA**

Abnormalities of lipid metabolism have been described in different motor neuron diseases, including SMA [36]. As shown in Table 1, dysregulated lipid metabolism is the first and most studied nutritional problem in SMA [37,38]. Compared with healthy controls and non-SMA motor neuron diseases with equally debilitating statuses, the abnormal lipid metabolism found in patients and animal models appears unique to SMA [39,40]. Abnormal levels of fatty acid oxidation metabolites, especially dicarboxylic aciduria and esterified carnitine, were first reported in several studies of patients with severe SMA type [38,41,42]. Subsequently, an increasing number of studies suggest that patients with SMA are likely to have metabolic defects involving fatty acid metabolism. Of note, increased fat mass, even though relatively low caloric consumption has been repeatedly reported in patients with SMA [40,43]. Several serum fatty acids and lipids have been found correlated to the motor function of patients with SMA, suggesting potential biomarker candidates for SMA [44]. It has recently been implicated that defects in fatty acid transport and mitochondrial β-oxidation may also contribute to muscle wasting in patients with a severe SMA phenotype [32]. Nevertheless, the exact mechanism of this lipid metabolism abnormality in SMA is still unclear, but it is suspected to be related to the absence of the SMN gene product, defects in neighboring genes, or the loss of a neural "trophic factor" [31,42,45].

Although abnormal levels of fatty acid metabolites have been reported, no direct evidence has substantiated a specific defect of mitochondrial β-oxidation in SMA patients. There are several differences in metabolomics between patients with SMA and patients with a genetic defect of fatty acid β-oxidation. SMA patients usually had a normal acylcarnitine profile [42], contrary to an increased acylcarnitine level always found in mitochondrial β-oxidation defects. Moreover, fasting patients with impaired fatty acid β-oxidation always have markedly decreased ketone bodies. However, patients with SMA usually present with a normal or even a high ketone body level (increased ketosis), especially under stress [45,46]. The ability to mount fasting ketosis means that the liver can utilize fatty acids normally, but it does not rule out that it may be caused by muscle-specific mitochondrial defects in β-oxidation [32]. Therefore, it is postulated that dysregulated fatty acid metabolism in SMA patients might be directly related to SMN deficiency but is not attributed to the consequence of major enzyme block of mitochondrial β-oxidation, disuse muscle atrophy, or denervation [13,42,47].









SMA: spinal muscular atrophy.

Fatty vacuolization with macro- or micro-vesicular steatosis of the liver has been found in early studies of SMA patients [38,41,42]. Of note, liver failure and Reye-like syndrome with diffuse vesicular steatosis have been recently reported in patients with type 1 or 2 SMA [48,49]. An updated study further reports an increased susceptibility to develop dyslipidemia in 37% of SMA patients, with evidence of liver steatosis in their pathological specimens [27]. Similarly, these human findings are reproduced in different SMA mouse models, of which a specific Smn 2B/<sup>−</sup> mice model developed the non-alcoholic fatty liver disease (NAFLD) before denervation. Hyperglucagonemia might also contribute to dyslipidemia and hepatic steatosis, possibly through the pancreas–liver axis, leading to peripheral lipolysis of white adipose tissue and an increase in circulating lipids. These findings imply that the liver-intrinsic SMN deficiency might also cause dysregulated metabolism of the hepatocytes [26,50], which could predispose the cells to fat accumulation. Noteworthily, subacute liver failure was recently reported in two patients with type 1 SMA following gene replacement therapy [49]. It is postulated that increased susceptibility to dyslipidemia and associated fatty liver disease could predispose the SMA patient to liver injury, which might be induced or exacerbated after the gene therapy. A thorough investigation of the lipid content in the liver of SMA patients and mouse models, before and after the onset of the disease [47], may provide further evidence for whether the direct or indirect effects of SMN deficiency affect metabolic abnormalities.

Since carnitine and its acyl esters (acylcarnitines) are cofactors for β-oxidation, abnormal lipid metabolism may also be reflected in their production, fractions, and transportation. Because carnitine is essential for intramitochondrial β-oxidation, reduced carnitine would limit β-oxidation. Acylcarnitines are known to play a crucial role in stabilizing neuronal membranes and neurotransmission [51]. Supplementation of acylcarnitine has shown beneficial effects in treating chronic degenerative diseases [52,53]. However, there are still controversies regarding the dysregulation of production, synthesis, and carnitine/acylcarnitine extraction in SMA patients. Early studies suggested that the integrity of nerve and motor neurons might influence carnitine transportation and lipid β-oxidation in muscles. Reduced muscle carnitine and decreased activity of β-oxidation have been observed in animal models after denervation [54,55]. Similarly, reduced carnitine and acylcarnitine levels in plasma and muscles and increased urine excretion of acylcarnitine have been reported in SMA patients [37,56]. However, normal or mild-to-moderate elevated serum acylcarnitines, particularly C5-OH acylcarnitine and C3 propionylcarnitine, were found in the following studies of SMA patients with a severe phenotype [41,42]. In contrast, an updated article reported an adolescent with type 2 SMA who showed a dramatically low serum carnitine/acylcarnitine level at a catabolic state [48]. This finding suggests impaired intramitochondrial β-oxidation associated with dysregulated carnitine metabolism in SMA would become more prominent, especially under stress.

In the fat metabolism of healthy individuals, longer-chain fatty acids are transported into the mitochondria for β-oxidation. Carnitine palmitoyltransferase 1 (CPT1) is an enzyme that converts long-chain acyl-CoA into long-chain acylcarnitine, thereby transporting long-chain fatty acids to the mitochondria. Decreased CPT activity has been reported in muscles of severe type 1 SMA patients, compared with aged-matched infants [56]. Recently, reduced CPT1 activity was also found in an SMA (Smn 2B/−) mice model [25]. Of note, an isoform of CPT1, called CPT1c, which mainly expresses in neurons, including motor neurons, shows biosynthetic activity in neuron-specific acyl-CoA. Reduced activity of CPT1c leads to motor function impairment and muscle weakness [57]. Interestingly, an updated study indicates that MN-specific CPT1C can interact with atlastin-1 encoded by the ATL1 gene, which is mutated in hereditary spastic paraplegia, a kind of motor neuron degenerative disorder [58].

Acylcarnitines can also interact with different proteins to influence signaling pathways of neuronal cells [52]. Growth-associated protein 43 (GAP43), a protein involved in neuronal development, neurotransmission, and neuroplasticity, is modified post-translationally by a long-chain acylcarnitine, possibly through the acylation pathway [59]. Interestingly, a recent study found that motor neurons from SMA mouse models showed reduced GAP43 protein levels in axons and growth cones [60,61]. SMN seems to be responsible for regulating the localization and translation of GAP43 mRNA in these

axons, and the restoration of GAP43 mRNA and protein levels rescued the defect of axon growth in SMA mice. Therefore, dysregulated acylcarnitine might also affect SMA phenotypes, possibly through the post-translational regulation of motor neuron-specific protein GAP43. Acylcarnitine plays a role in GAP43-related axon growth/repair pathways and may represent a promising SMA treatment target.

Nevertheless, the inconsistent findings of carnitine/acylcarnitine metabolites in SMA patients argue the pathomechanism of the impaired β-oxidation in SMA. Applying modern techniques for quantitative analysis of carnitine and acylcarnitine of various lengths in different samples (e.g., plasma, urine, and muscle) may help decipher this ambiguity [62,63]. However, similar studies in SMA patients are scarce, and the findings of changed carnitine/acylcarnitine levels in SMA patients with different *SMN2* copies have not been updated. The discovery of plasma and urinary metabolite patterns, specifically reflective of fatty acid catabolism, can help clarify biochemical pathways that link lipid metabolism and provide potential biomarkers monitoring disease progression.

#### **3. Glucose Metabolic Abnormalities in SMA**

The concern about glucose metabolism abnormalities was initially raised through clinical observations in mild-to-intermediate phenotypes of SMA patients (Table 2). Two studies of type 2 SMA patients suggested they might be more likely to experience hypoglycemia following fasting [64,65]. A recent study in type 1 SMA patients also showed a similar finding of hypoglycemia even after a short-term fasting (>4 h but <6 h) [66]. The presence of hypoglycemia after fasting has been postulated to have an association with reduced gluconeogenesis. Because skeletal muscle is an important source of gluconeogenic substrates during fasting, hypoglycemia must be considered for SMA patients with severe muscle wasting, especially during surgery and fever [65]. Therefore, it is recommended that patients with recurrent hypoglycemia episodes should be provided with regular meals based on carbohydrates and protein, including late-night meals.

In contrast, other studies have reported hyperglycemia during fasting in patients with type 2 and type 3 SMA, some of whom were diagnosed with diabetes and ketoacidosis (Table 2) [17,67]. The metabolic syndrome features of increased fat mass and decreased lean mass have been reported in patients with type 2 and type 3 SMA [40]. A recent study also indicated that, in a good state, obese children with SMA type 2 were at increased risk of insulin resistance and impaired glucose tolerance, with 50% of participants showing urinary ketones [16]. It has been postulated that as the skeletal muscle is a major target of insulin action, muscle wasting (sarcopenia) promotes insulin resistance with increased risk of hyperglycemia [68,69]. Additionally, hyperleptinemia has been observed in patients with SMA types 1 to 3, which implies an indirect link to insulin resistance [70]. Nevertheless, even if glucose and insulin metabolism show an increased risk of insulin resistance, HbA1c levels are usually normal in most SMA patients examined [16,69,70].





Similarly, perturbations of glucose metabolism affecting glucose sensitivity and pancreatic defects have been observed in the SMA mice model [17,39,71]. In particular, the metabolic defects in the SMA Smn 2B/<sup>−</sup> mice model were characterized by fasting hyperglycemia, glucose intolerance, hypersensitivity to insulin, and hyperglucagonemia [17]. In the same study, analysis of pancreatic tissue from infants with SMA type 1 has recapitulated similar pancreatic development defects. Reduced SMN protein levels may also affect the insulin-like growth factor 1 (IGF-1) pathway in the liver of SMA mouse model [72]. IGF-1 is an anabolic hormone with a molecular structure comparable to insulin, which shows myotrophic effects on muscle tissue. Dysregulation of the IGF-1 signaling pathway has also been reported in biopsies from patients with type 1 SMA [73]. A recent study further indicated that IGF1 status is associated with insulin resistance in young SMA patients with early-onset sarcopenia [69]. However, the authors concluded that the myotrophic effect of IGF-1 might be adversely affected by insulin resistance, so therapeutic interventions for dysregulated glucose metabolism in SMA should target insulin resistance.

Nevertheless, it has been suggested that SMA patients receiving partial SMN restoration therapy may increase the risk of having pancreatic and glucose metabolism defects [71]. Meticulous monitoring of glucose homeostasis in SMA patients is essential to clarify the role of SMN in glucose metabolism and pancreatic function.

#### **4. Altered Vitamin Level in SMA**

A previous study indicates that the activity of SMN depends on folic acid and vitamin B12, both of which are necessary for protein methylation [74]. SMN binds to certain proteins with arginine- and glycine-rich domains, which are modified to dimethylarginine. The binding of other proteins that interact with SMN can also be greatly enhanced by methylation. Inadequate intake of folic acid and vitamin B12 may lead to protein hypomethylation [75], and subsequently may affect the clinical severity of SMA.

The SMN protein may play an active role in bone remodeling or uptake of vitamin D and calcium [35]; therefore, patients with SMA are often accompanied by osteopenia and may contend with fractures due to minor injuries. Compared with other neuromuscular disorders, reduced bone mineral density seems more significant in patients with SMA, especially in those losing ambulatory function [76]. Suboptimal vitamin D intake is frequently observed in patients with all SMA types [18,77]. Low serum levels of vitamin D and 25-OH vitamin D have been reported in patients with type 2 or 3 SMA [34]. However, in a small group of patients with type 1 SMA, the corresponding serum vitamin D levels did not reflect insufficient consumption [78]. Low bone mineral density (BMD) and femur fractures are highly prevalent in all SMA subtypes from a young age; however, few patients met osteoporosis criteria [79]. Adequate bone health assessment and intervention may be an unmet medical need for patients with SMA. It is imperative to determine the natural trajectory of BMD changes at different skeletal sites, especially in adolescent and young adult patients with SMA, and determine if low BMD and propensity to fracture are related to immobility and muscle weakness or direct action of SMN on bone turnover. More work is required to identify effective interventions to delay the decline in BMD and prevent fractures in patients with SMA.

Besides vitamin D and calcium, vitamin E, vitamin K, and folate intakes have been reported below values of Recommended Dietary Allowance (RDA) in half the cohort of patients with SMA [77]. Further research is needed to determine the appropriate intake of vitamin D and other macro- and micro-nutrients in this population.

#### **5. Dietary Issues in SMA**

Patients with SMA are at higher risk of suboptimal nutrition intake, and nearly half of the cohort demonstrated either undernutrition (underweight) or overnutrition (overweight) over time [18,77]. Changes in body composition, especially the loss of lean body mass, can be particularly harmful to

SMA patients because it can impair the respiratory strength of already weak muscles [43]. Therefore, nutrition support is considered a core component of multidisciplinary care for SMA patients [15,80].

However, the specific nutritional challenges in this population are not well described, and a particular diet has not been scientifically evaluated. An early study showed that when the mother was fed a lipid-rich diet, the pups of SMA mice could have a longer survival period and improved motor function [39]. These findings suggested that higher fat content may confer protective benefits during motor neuron loss. However, an updated study reported that low-fat diets could nearly double survival in Smn 2B/<sup>−</sup> mice, independent of changes in SMN levels, liver steatosis, or enhanced hepatic functions [81]. Although both studies are in the preclinical phase, such controversies suggest a need to establish clinical nutrition guidance from evidence-based research to provide better care for SMA patients.

The advances in therapy for SMA have improved survival and quality of life, which poses new challenges. The survival of patients with severe SMA has generated new phenotypes, and long-term outcomes are unknown [82]. Noteworthily, nutritional management may have a significant impact on the clinical course and even prognosis. For example, previous studies indicated that nutritional support could affect the therapeutic effects of trial agents on different SMA mice models [83,84]. Although it is difficult to show a clear association between metabolic effects in SMA patients who received therapies at this time, it has been emphasized that nutritional care must also be revised and monitored according to individual needs, especially in the SMA therapeutic era [15]. Optimal nutritional management for patients with SMA includes longitudinal evaluation of weight and length and dietary analysis. Recent studies have demonstrated that a modified diet based on measured energy expenditure and optimized protein can improve ventilation and lean body mass in patients with SMA [18,85]. In the future, non-invasive approaches for body composition assessment, e.g., bioelectrical impedance analysis, can be used to evaluate the nutritional status of children with SMA. Further research is needed to assess the use of elemental and semi-elemental formulas in SMA management, including the optimal intake of macronutrients and micronutrients for nutritional support and the ideal fat content and composition.

#### **6. Conclusions**

Active nutrition support and metabolism surveillance are crucial for patients with SMA, and a comprehensive nutritional assessment could guide individualized nutrition therapy for this vulnerable population. With the emergence of new gene-targeted and disease-modifying therapies, which may affect the metabolism of SMA patients, personalized nutritional optimization may become particularly important. Metabolomics study in SMA patients, before and after the disease onset, may provide valuable information regarding the direct or indirect effect of SMN deficiency on metabolic abnormalities. Furthermore, identifying and quantifying the specific metabolites in biofluids of SMA patients may serve as an authentic biomarker or therapeutic target for SMA.

**Author Contributions:** Y.-J.L. contributed conceptualization, original draft preparation, and manuscript writing. T.-H.C. contributed conceptualization, data curation, original draft preparation, manuscript writing and final approval to this work. Y.-Z.W. contributed conceptualization, data curation, original draft preparation to this work. Y.-H.T. contributed conceptualization, data curation, original draft preparation to this work. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Kaohsiung Medical University Hospital in Kaohsiung, Taiwan, grant number KMUH-108-8R47.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland Tel. +41 61 683 77 34 Fax +41 61 302 89 18 www.mdpi.com

*Nutrients* Editorial Office E-mail: nutrients@mdpi.com www.mdpi.com/journal/nutrients

MDPI St. Alban-Anlage 66 4052 Basel Switzerland

Tel: +41 61 683 77 34 Fax: +41 61 302 89 18

www.mdpi.com

ISBN 978-3-0365-3352-0