Proteomic Characterization of Senescent Laryngeal Adductor and Plantaris Hindlimb Muscles

Abstract

Objectives:

The goals of this study were to 1) compare global protein expression in muscles of the larynx and hindlimb and 2) investigate differences in protein expression between aged and nonaged muscle using label-free global proteomic profiling methods.

Methods:

Liquid chromatography-mass spectrometry (LC-MS/MS) analysis was performed on thyroarytenoid intrinsic laryngeal muscle and plantaris hindlimb muscle from 10 F344xBN F1 male rats (5 old and 5 young). Protein expression was compared and pathway enrichment analysis performed for each muscle type (larynx and limb) and age group (old and young muscle).

Results:

Over 1,000 proteins were identified in common across both muscle types and age groups using LC-MS/MS analysis. Significant age-related differences were seen across 107 proteins in plantaris hindlimb and in 19 proteins in thyroarytenoid laryngeal muscle. Bioinformatic and enrichment analysis demonstrated protein differences between the hindlimb and larynx may relate to immune and stress redox responses and RNA repair.

Conclusion:

There are clear differences in protein expressions between the laryngeal and hindlimb skeletal muscles. Initial analysis suggests differences between the two muscle groups may relate to stress responses and repair mechanisms. Age-related changes in the thyroarytenoid appear to be less obvious than in the plantaris. Further in-depth study is needed to elucidate how aging affects protein expression in the laryngeal muscles.

INTRODUCTION

Our understanding of laryngeal muscle relies heavily on skeletal limb muscle studies.^1–6 Preliminary studies have identified muscle fiber composition and mitochondrial density differences between laryngeal and limb muscles—likely the result of their functional demand differences. However, global characterization of protein expression between these two muscle types as they relate to biological function remains to be elucidated.^1,7,8 This lack of knowledge makes it challenging to appraise applicability of limb literature to voice rehabilitation for the laryngeal muscles.

Furthermore, although age-related functional, structural, and molecular alterations are known to occur in skeletal limb muscle—resulting in loss of muscle mass (sarcopenia) and muscle strength (dynapenia)^9–15—mechanisms underlying sarcopenia in the laryngeal muscles are far less understood. This gap is nontrivial considering the 1) exponential growth of the aging population in the United States and across the world and 2) high prevalence of presbyphonia and presbylarynges (age-related vocal dysfunction and structural laryngeal changes, respectively) in the elderly.^16–18 Changes in protein expression, modification, synthesis, and turnover underlie sarcopenia and dynapenia in the limb muscles. Therefore, a logical next step in understanding sarcopenia and dynapenia in the larynx is to broadly characterize protein-related mechanisms underlying senescence in laryngeal muscle.^10,13

Proteomic investigation using mass spectrometry identifies and quantifies thousands of proteins from small amounts of tissue.¹³ Bioinformatic analysis and pathway enrichment of large-scale proteomics data sets can inform protein targets for subsequent hypothesis-driven molecular and biochemical validation studies. These state-of-the art techniques are significantly more sensitive than two-dimensional sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE) methods initially used for rat laryngeal muscle.^19,20 Considering the substantial gaps in our understanding of laryngeal muscle expression compared to limb muscles and the effects of biological aging on the larynx, global proteomics using mass spectrometry is especially advantageous.

The first objective of this study was to establish differences between laryngeal and limb muscle proteomes to provide further evidence to support or refute the need for distinct investigations between the two muscle groups.⁷ The second objective was to examine the effects of biological aging on the laryngeal muscle proteome and provide specific targets for subsequent analysis of functional roles of individual proteins underlying senescence. These objectives were accomplished using label-free quantitative mass spectrometry analysis in the thyroarytenoid (laryngeal) and plantaris (hindlimb) muscles of young and old rats. The experimental design was a cohort study with two independent variables: 1) age (old vs. young) and 2) muscle type (larynx vs. hindlimb), with protein expression as dependent variables. The thyroarytenoid intrinsic laryngeal muscle and plantaris limb muscle were compared because they both are composed of fast twitch type II myosin heavy chain fibers and thus are assumed to utilize similar bioenergetics.^21–23

Our first hypothesis was that there would be two distinct global proteomic profiles between the limb and laryngeal muscle, both in aged and nonaged muscle. This hypothesis was driven by functional differences between the two muscle groups (e.g., vocalization vs. locomotion). Second, we hypothesized significant protein expression differences between young and old laryngeal muscle would be associated with energy metabolism, injury repair, and muscle contraction. This hypothesis was informed by previous work showing similar patterns with proteomic analysis of denervated laryngeal muscle;²⁴ this hypothesis was also based on previous studies demonstrating similarities in morphological changes in denervated and senescent limb muscle.¹¹

MATERIALS AND METHODS

Sample Preparation

A total of 10 Fischer 344 × Brown Norway F1 (F344BN) rats—a well-established strain used to study biological aging from the National Institute of Aging (NIA)—were used for the study. Five of the rats were 9 months old (young adult age group) and 5 were 32 months old (senescent group). All animals were euthanized per approved methods by the Institutional Animal Care Use Committee (IACUC) at New York University Grossman School of Medicine. Whole thyroarytenoid intrinsic laryngeal muscle and plantaris hindlimb muscle tissue were extracted. In brief, the whole larynx was removed and hemidissected posteriorly. The vocal fold mucosa was then removed under a dissection microscope, starting at the anterior commissure and stripped all the way posteriorly to the vocal process until the muscle was completely exposed. The thyroarytenoid muscle was then clipped at its insertion and origin with microdissection scissors and dissected out of its cartilaginous framework. The plantaris was removed by exposing the gastrocnemius and cutting it away with a sterile razor blade. Individual muscle samples were placed into 1.5 mL Eppendorf tubes, flash frozen in liquid nitrogen, and stored in a −80°C freezer until further processing.

Extracted laryngeal and limb muscle samples were homogenized by probe sonication (100% amplitude, 5 × seconds cycles) in lysis buffer containing 5% (w/v) sodium dodecyl sulfate (SDS) and 50 mM triethylammonium bicarbonate (TEAB) (pH = 7.55) to lyse cells, extract, and denature the proteins. After the centrifugation step, supernatants were transferred into clean tubes. Protein concentrations were measured by bicinchoninic acid assay (BCA) assay and proteins disulfide bonds were reduced and alkylated in the presence of 10 mM tris-2(-carboxyethyl)-phosphine (TCEP) and 20 mM chloroacetamide (CAA) for 30 minutes at 95°C. Digestions were performed according to S-Trap procedure.²⁵ In brief, samples were acidified with 12% phosphoric acid, proteins were precipitated by six times dilution with S-Trap buffer (100 mM TEAB in 90% MeOH, pH = 7.1) and captured on porous micro S-Trap spin column filters. S-Trap columns were washed three times with 150 μL of S-Trap buffer followed by on-filter digestion in 20 μL of 50 mM ammonium bicarbonate solution containing 0.4 μg of trypsin (Promega). Digestions were completed in 2 hours at 47°C. Prior to liquid chromatography-mass spectrometry (LC-MS/MS) analyses, peptide samples were desalted on in-house StageTips (C18 material). Eluted peptides were dried on Speedvac vacuum concentrator, resuspended in 2% ACN 0.5% AcOH prior to LC-MS/MS analysis on a Q Exactive HF-X instrument.

LC-MS/MS Analysis

Resulting peptide mixtures were analyzed by nanoflow LC-MS/MS. LC separation was performed online on EASY-nLC 1000 (Thermo Scientific) utilizing Acclaim PepMap 100 (75 μm × 2 cm) precolumn and PepMap RSLC C18 (2 μm, 100A × 50 cm) analytical column. Peptides were gradient eluted from the column directly into an Orbitrap HF-X mass spectrometer using 136 minutes gradient from 5% to 26% B in 100 minutes followed by ramp up to 40% B in 20 minutes and final equilibration in 100% B for 15 minutes (A = 2% ACN 0.5% AcOH/B = 80% ACN 0.5% AcOH). Flowrate was set at 200 nL/min. Mass spectrometer was operated in data-independent acquisition mode (DIA)^26,27 with MS² fragmentation (resolution 30,000, automatic gain control (AGC) = 3e6, max fill time of and stepped collision energy of 22.5, 25, 27.5, profile mode) across 22 m/z windows after every MS¹ scan event (resolution 120,000, AGC = 3e6, max fill time of 60 msec, profile mode).

Data Analysis

Identification and quantification of proteins were performed in Spectronaut software (https://biognosys.com/shop/spectronaut) using directDIA mode (without spectral library) against the SwissProt subset of the Rattus Norvegicus Uniprot database (http://www.uniprot.org/) containing 8,060 entries. Database search was performed with search engine Pulsar integrated into Spectronaut. The enzyme specificity was set to trypsin with the maximum number of missed cleavages set to 2. Oxidation of methionine was searched as variable modification; carbamidomethylation of cysteines was searched as a fixed modification. The false discovery rate (FDR) for peptide, protein, and site identification was set to 5%. Protein quantification was performed on MS2 level using the three most intense fragment ions per precursor. Protein quantities were calculated as sum of the corresponding peptide precursors.

Subsequent data analysis on the obtained protein expression matrix was performed in Perseus²⁶ (http://www.perseus-framework.org/) and R language for statistical computing and graphics (http://www.r-project.org/). Protein intensity values within each muscle type were log2-transformed and normalized so that medians of their distributions were the same. Protein expression differences between muscle types within each age group were tested with two sample t-tests. Permutation-based corrections for multiple hypothesis testing were used in all comparisons; any proteins that were below 5% FDR were considered significant hits.²⁷ Results were represented on volcano plots.

Protein intensities were z-scored to further explore differences between the thyroarytenoid and plantaris muscles in the young group. z-Scores were calculated by comparing the average intensity of each protein across all five replicates to the grand mean of all protein intensities within the same muscle type. Scatterplots were then generated to visualize differences between z-score distributions between the two muscle types. This approach allowed comparisons of relative intensity of each detected protein between the thyroarytenoid and plantaris muscles. One-dimensional (1D) annotation enrichment analysis was also conducted on plantaris and thyroarytenoid muscle in young animals to determine differences in potential underlying processes between the two muscle groups.²⁸ Gene Ontology (GO) slim terms for cellular components and biological functions were assigned to all proteins within the same group across muscle type.

RESULTS

Overall Results

Across all 20 samples (young/old and thyroarytenoid/plantaris combined), 9,977 peptide sequences were identified from 1,122 protein groups spanning a dynamic range of ~6 orders of magnitude. Table I lists the number of protein groups and peptides found in each sample. The majority of protein groups (813) were detected in all 20 samples with an additional 217 protein groups detected across at least 16 samples; 92% of protein groups (1,030) were detected in at least one sample in each muscle type and age group. A very small number of proteins were only detected in one muscle type (eight detected in plantaris; six detected in thyroarytenoid). Principal component analysis (PCA) revealed one plantaris sample in the old group was an outlier and contained several uniquely low-abundant proteins (Fig. 1A). The sample was subsequently excluded from further analysis due to the likelihood these differences were due to tissue acquisition or processing errors. See Supporting Table 1 for a complete list of identified proteins, log2-transformed and normalized intensity per sample, and statistical differences between age groups and muscle type.

TABLE I.

Number of Protein Groups and Peptide Sequences Identified in Each Sample.

Muscle	Age Group	Sample	Protein Groups (#)	Peptide Sequences (#)
Thyroarytenoid	Young	01	1,091	9,274
		02	1,077	9,041
		03	1,076	8,929
		04	1,077	9,011
		05	1,085	9,078
	Old	06	1,086	9,164
		07	1,077	9,094
		08	1,089	9,221
		09	1,090	9,214
		10	1,098	9,246
Plantaris	Young	11	985	7,721
		12	1,095	9,003
		13	1,088	9,041
		14	1,100	9,335
		15	1,002	7,358
	Old	16	977	7,225
		17	1,061	8,577
		18	1,051	8,577
		19	1,058	8,877
		20	1,066	8,631

Fig. 1.

Principal component analysis (PCA) and volcano plots between muscle type. (A) PCA of all 20 samples illustrating a clear separation between laryngeal and plantaris muscle, regardless of age, with one outlier in the plantaris in the old group. A slight age separation is observed in the plantaris, but not in the thyroarytenoid. The young thyroarytenoid sample present in the upper right of figure (A) was removed from analysis due to tissue processing error. (B, C) Volcano plots of differences in protein expression between muscle type within each age group. (B) Thyroarytenoid versus plantaris muscle in young animals. Significant downregulated proteins to the left of the log2 fold change in young laryngeal muscle with upregulated proteins to the right of the log2 fold change in young laryngeal muscle relative to the young plantaris muscle. (C) Thyroarytenoid versus plantaris muscle in old animals. Significant proteins to the left of the log2 fold change are downregulated in old laryngeal muscle while proteins to the right of the log2 fold change are upregulated in old laryngeal muscle relative to the old plantaris muscle.

Age Differences Between Thyroarytenoid and Plantaris Muscles

Although the vast majority of protein groups were identified in both the thyroarytenoid and plantaris muscles, PCA revealed clear separation in expression levels between these two muscles (cf. Fig. 1A). In the young group, the intensity of approximately half (n = 538) of all detected protein groups was statistically different between muscle types, with 260 proteins upregulated and 278 downregulated in the thyroarytenoid muscle relative to the plantaris (Fig. 1B). In the old group, the intensity of two-thirds of detected protein groups (n = 738) was statistically different, with 432 upregulated and 306 downregulated in the thyroarytenoid relative to the plantaris (Fig. 1C).

Comparison of Relative Expression Levels in the Young Group

Due to the large number of statistical differences in protein expression levels between muscle type, protein intensities were z-scored and normalized values were used to compare relative differences between limb and larynx (Fig. 2).^29,30 Eighteen proteins had an absolute z-score difference greater than 1.5 (Table II). Notable high-abundant proteins in the plantaris relative to the thyroarytenoid included proteins involved in muscle structure and contraction (carbonic anhydrase 3 [CAH3], myosin regulatory light chain 2, ventricular/cardiac muscle isoform [MLRV], and myosin light chain 3 [MYL3]), and slow skeletal muscles troponin T [TNNT1] and troponin I [TNNI1]). The only high-abundant protein in the thyroarytenoid muscle (z-score > 1.5) relative to the plantaris (z-score of −1.3) was myosin-binding protein-H (MBPHL).

Fig. 2.

Normalized protein abundance between thyroarytenoid and plantaris muscles in nonaged younger rats. Protein MS intensities were converted to z-scores before plotting. Results show approximately 40 distinct proteins between the muscles of the limbs and larynx when intensities were normalized, with 18 of those proteins with a difference greater than +/− 1.5 between muscle type. Of note, the most abundant protein in the larynx compared to hindlimb was myosin-binding protein-H (MBPHL). This protein was previously detected in human skeletal muscle, but its functional role is unknown. However, this protein has been well characterized and found to be over-expressed in atrial tissue compared to noncardiac tissue.²⁹ Interestingly, α- and β-cardiac fibers have been observed to coexpress with other types of laryngeal muscle fibers.³⁰ The role of the MBPHL protein specific to the larynx requires further elucidation.

TABLE II.

Proteins With a z-Score Absolute Difference of Greater Than 1.5.

	TA z-score	PL z-score	Difference	Protein Group Name	UniprotID	Genes
Higher TA	0.25	−3.26	3.51	BPI fold-containing family B member 1	BPIB1	Bpifb1 Lplunc1
	0.45	−2.54	2.99	Mast cell protease 2	MCPT2	Mcpt2
	1.53	−1.32	2.85	Myosin-binding protein H-like	MBPHL	Mybphl
	0.51	−2.27	2.77	Vomeromodulin	VOME	Bpifb9 RYF3
	0.38	−1.92	2.30	Keratin, type II cytoskeletal 4	K2C4	Krt4 Kb4
Lower TA	−1.09	0.44	−1.54	ATP synthase-coupling factor 6	ATP5J	Atp5pf Atp5j
	−0.12	1.86	−1.98	Myosin light chain 3	MYL3	Myl3 Mlc1v
	0.05	2.10	2.04	Myosin regulatory light chain 2	MLRV	Myl2 Mlc2
	−4.49	−2.32	−2.17	Protein S100-A9	S10A9	S100a9 Mrp14
	−3.95	−1.76	−2.19	Sterol regulatory element-binding protein 2	SRBP2	Srebf2
	−1.83	0.38	−2.21	Apolipoprotein A-IV	APOA4	Apoa4
	−1.16	1.06	−2.22	Fibromodulin	FMOD	Fmod
	0.35	2.86	−2.51	Carbonic anhydrase 3	CAH3	Ca3
	−2.37	0.46	−2.83	Thrombospondin-4	TSP4	Thbs4 Tsp-4 Tsp4
	−1.80	1.53	−3.33	Troponin I, slow skeletal muscle	TNNI1	Tnni1
	−4.73	−1.24	−3.49	N-lysine methyltransferase SMYD2	SMYD2	Smyd2
	−4.89	−1.19	−3.70	Ubiquitin-conjugating enzyme E2 G1	UB2G1	Ube2g1 Ubc7 Ube2g
	−2.96	1.41	−4.37	Troponin T	TNNT1	Tnnt1 Fang2

z-score relative protein abundance.

PL = plantaris; TA = thyroarytenoid.

Results of 1D enrichment analysis revealed many common GO terms between the two skeletal muscles types, as expected, indicating potentially shared biological functions including metabolic carbohydrate, cellular ketone, cofactor, heterocycle, nucleotide, nucleobase-containing small molecules, nucleotide, organic acid, and small molecule processes. However, there were also differences observed. GO terms specific to the thyroarytenoid muscle were related to cytoplasmic processes, macromolecular metabolism and modification, protein modifications, response to oxidative stress, and RNA metabolic processes. Additional processes in the plantaris included alcohol metabolic processes.

Effects of Age Within Muscle Type

Our criteria of a 5% FDR cutoff resulted in 19 proteins significantly impacted by age within the thyroarytenoid, with 7 downregulated and 12 upregulated proteins in old muscle (Fig. 3A). Within the plantaris, a total of 107 proteins were significantly impacted by age, with 86 downregulated and 21 upregulated proteins in old muscle (Fig. 3B and Supporting Table 1). Biological functions of each significantly different protein between young and old animals in the thyroarytenoid muscle are listed in Table III. In brief, many of the proteins expressed differently between old and young laryngeal muscle relate to immune and inflammatory responses (T-kininogen 1 [KNT1], low affinity immunoglobulin gamma Fc region receptor III [IGG2B], Ig kappa chain C region, B allele [KACB], class I histocompatibility antigen, non-RT1.A alpha-1 chain [HA11], BPI fold-containing family B member 1 [BPIB1], and Nucleophosmin [NPM]), muscle function (voltage-dependent calcium channel gamma-1 subunit [CCG1], ferritin heavy chain [FRIH], nicotinamide phosphoribosyltransferase [NAMPT], calcineurin subunit B type 1 [CANB1], ADP/ATP translocase 2 [ADT2]), and muscle structure and repair (histone H1.5 [H15], nucleophosmin [NPM]). Proteins related to immune and repair responses in the larynx were not dramatically different between old and young plantaris. An opposite directional trend was also observed in four proteins between the hindlimb and larynx (FRIH, HA11, IGG2B, KNT1), which were upregulated in old thyroarytenoid but downregulated in old plantaris.

Fig. 3.

Volcano plots of impact of age within each muscle type. Comparison of old versus young within the (A) thyroarytenoid muscle and (B) plantaris muscle. Significant proteins to the left of the log2 fold change are downregulated in old animals while proteins to the right of the log2 fold change are upregulated in old animals. Approximately 60% of the 19 significant protein hits were upregulated in the laryngeal muscle of old animals, while only ~20% of the 107 proteins found in the hindlimb muscle of old animals were upregulated.

TABLE III.

List of Proteins Significantly Different Between Young and Old Laryngeal (Thyroarytenoid) Muscle.

Biological Function in Aged Versus Nonaged Laryngeal Muscle Tissue	UniProt Protein ID	UniProt Gene Name
(upregulated)
Trypsin inhibitor; inhibits blood coagulation factor Xa and tryptase	Q64240	AMBP
Respond to inflammatory stimuli by increasing vascular permeability; may be involved in immunosenescence	P01048	KNT1
Immunoglobulin receptor binder	P20761	IGG2B
Final component of the complement system to be added in the assembly of the membrane attack complex	Q62930	CO9
Immunoglobulin	P01836	KACA
Plays a role in excitation-contraction coupling	P97707	CCG1
Ig kappa chain C region, B allele	P01835	KACB
Collagen fiber assembly. Found in several connective tissues.	P47853	PGS1
Expressed predominantly in epithelium of uterus and oviduct	Q9QZT0	CUZD1
Involved in presentation of foreign antigens to the immune system	P15978	HA11
F-actin-capping proteins bind in a Ca(2 +)-independent manner to the fast growing ends of actin filaments (barbed end) thereby blocking the exchange of subunits at these ends. Unlike other capping proteins (such as gelsolin and severin), these proteins do not sever actin filament	Q3T1K5	CAZA2
Important for iron homeostasis; Stores iron in a soluble, nontoxic, readily available form. Also plays a role in delivery of iron to cells	P19132	FRIH
(downregulated)
Involved in chromatin remodeling, nucleosome spacing, and DNA methylation; the histone H1 protein binds to linker DNA between nucleosomes forming the macromolecular chromatin fiber	D3ZBN0	H15
Catalyzes the condensation of nicotinamide with 5-phosphoribosyl-1-pyrophosphate to yield nicotinamide mononucleotide, an intermediate in the biosynthesis of NAD. It is the rate limiting component in the mammalian NAD biosynthesis pathway	Q80Z29	NAMPT
Involved in calcium sensitivity	P63100	CANB1
Potent lipase inhibitor; Secreted by the serous (von Ebner’s) glands at the back of the rat tongue	P04634	LIPG
Catalyzes the exchange of ADP and ATP across the mitochondrial inner membrane	Q09073	ADT2
Involved in diverse cellular processes such a ribosome biogenesis, protein chaperoning, histone assembly, cell proliferation, and regulation of tumor suppression	P13084	NPM
May play a role in innate immunity in mouth, nose, and lungs	A0JPN3	BPIB1

Information derived from http://david.ncifcrf.gov.

DISCUSSION

The main objectives of this study were to 1) determine proteomic differences between skeletal muscles of the hindlimb and larynx and 2) determine the effects of biological aging on the thyroarytenoid muscle proteome using LC-MS/MS methods. Overall, the same 1,000+ proteins were found in both muscle types. These findings confirm tissue as muscle phenotypes and validate the use of our strategy for rat thyroarytenoid muscle proteome characterization. Within these data, however, it is clear the proteome between the thyroarytenoid and plantaris are unique in their intramuscular protein expression, both at young and old ages. The main differences between limb and laryngeal protein expression involved proteins related to intramuscular stress and repair—which were more abundant in the larynx—and certain muscle structure and contractile proteins—which were more abundant in limb muscle.

Validation studies on candidate proteins are required to define relationships between differentially expressed proteins in the limb and larynx and elucidate their biological functions. However, several interesting trends in this study can guide these future hypothesis-driven investigations. The first trend were the GO terms unique to the thyroarytenoid muscle related to tissue stress and repair. These patterns suggest the thyroarytenoid may either be more sensitive to oxidative stress compared to the plantaris or undergo repair more frequently than the limbs. Interestingly, tissue remodeling and repair has been shown to occur in the larynx even in the absence of mechanical or neurological insult. For example, every 24 hours, two new myonuclear additions are generated for every 1,000 myofibers in the laryngeal muscle.³¹ In contrast, remodeling and repair only occurs in the limbs with intramuscular injury.³² From a functional perspective, the proactive nature of intramuscular repair in the larynx aligns with the critical role of the larynx in life-sustaining vegetative tasks (e.g., breathing, deglutition, airway protection). Neuromuscular precautionary measures are required to prevent muscle decline or deficiencies that would otherwise lead to asphyxiation, aspiration, and ultimately death.

The second trend is the 19 differentially expressed proteins in old versus young thyroarytenoid muscle. Proteins related to muscle structure and repair (H15, NPM) were downregulated in oldn thyroarytenoid muscle. This pattern aligns with clinical presentations of loss of muscle bulk in vocal fold atrophy—commonly observed in the aging population. Proteins related to innate epithelial mucosal airway immune responses (BPIB1), protein chaperoning, histone assembly, cell proliferation, and tumor suppressor regulation (NPM), NAD biosynthesis (NAMPT), exchange of ADP to ATP (ADT2), and calcium sensitivity (CANB1) were also downregulated in old thyroarytenoid muscle. Possible connections between these downregulated proteins in senescent laryngeal muscle and age-specific disease need to be explored further. One example is the relationship between BPIB1 downregulation—shown to result in diminished airway mucociliary clearance—and respiratory diseases common in the geriatric population (e.g., pneumonia).³³ Another example is the role of NPM in head and neck cancers—which become more prevalent with age.^34–36 A third example is the relationship between NAMPT and vocal exercise in senescent laryngeal muscle.³⁷

The roles of upregulated proteins in senescent thyroarytenoid muscle related to vascular permeability (KNT1), antigen presentation for the immune system (HA11), muscle excitation-contraction coupling (CCG1), and iron delivery to cells (FRIH) are unclear and require elucidation. What is also unclear are the inverse relationships between several proteins in the limbs and larynx (FRIH, HA11, IGG2B, and KNT1) and their functional role in the larynx with biological aging.

The third trend were proteins related to muscle structure and contraction that were upregulated in the plantaris compared to the thyroarytenoid (CAH3, MLRV, MYL3, TNNT1, TNNI1). These findings may have to do with greater weight-bearing loads in limb muscle compared to the larynx. More structure- and contraction-specific proteins would be required to increase muscle fiber size (hypertrophy), and thus strength, to accommodate load. Conversely, weight-bearing load does not occur in the larynx. Although the larynx does not undergo hypertrophic changes with use, increased neuromuscular efficiency between the nerve and muscle occurs with higher vocal demands.³⁸

The obvious limitation of this work is that global proteomics provide a snapshot of all proteins present in the tissue but do not inform their function. As a result, theoretical cogitations of protein-specific biological function as they relate to muscle type and senescence should be treated with caution. Further study on the differentially expressed proteins between old and young laryngeal muscle discovered in this study is required to confirm these findings and elucidate potential roles these proteins and others may play in the senescent larynx. However, these findings established the unique proteomic profile of the thyroarytenoid muscle relative to skeletal limb muscle and serve as a foundational step in characterizing the impact of biological age on the laryngeal muscle on molecular levels.

CONCLUSION

This study established the feasibility of using label-free LC-MS/MS to quantify rat thyroarytenoid muscle proteome. Although laryngeal and limb muscles of F344BN rats are both skeletal muscles, they exhibit distinctly different protein expression profiles, both in young adult rats and in how they respond to senescence. Understanding key differences in intramuscular expression between the limbs and the larynx and the impact of senescence is critical to inform whether limb literature can be applied to vocal muscle training and rehabilitation. These results suggest this theoretical approach to voice rehabilitation should be used with caution.