Sage Journals: Discover world-class research

Abstract

Purpose of the study

The pathophysiological mechanism of medication overuse headache is uncertain; no distinctive markers have been described right now. The aim of this study was to conduct proteomic analyses on serum samples from patients with medication overuse headache and healthy individuals. Specifically, mono- (SDS-PAGE) and two-dimensional gel electrophoresis (2-DE) followed by liquid chromatography tandem mass spectrometry (LC-MS/MS) were used to evaluate changes in serum proteins.

Main findings

By SDS-PAGE, four over-expressed bands were revealed in patients, compared to controls. 2-DE combined with LC-MS/MS analysis allowed confirmation of some proteins preliminarily detected by SDS-PAGE: Hemopexin, alpha-1-acid glycoprotein 1, apolipoprotein A4 and haptoglobin. Moreover, other differential proteins were isolated, mostly increased in MOH patients: Alpha-1-antitrypsin, immunoglobulin heavy constant alpha 1, retinol binding protein and transthyretin. Only one protein, immunoglobulin kappa constant, was decreased in the patients’ group.

Conclusions

The investigation of the serum proteome can offer a better understanding about biological mechanisms underlying medication overuse headache. Specifically, medication overuse headache shares some serum biochemical markers with chronic pain conditions. Further studies might uncover the relevance of these proteins in medication overuse headache.

Keywords

Biomarker proteomics MOH serum biomarker mechanisms pathophysiology

Introduction

About 2% of the worldwide population is suffering from chronic migraine, a neurological disorder characterized by reiterated episodes of severe and disabling head pain. Chronic migraine often involves a large intake of analgesics and acute headache medications (AHMs), paradoxically increasing the frequency and intensity of headache to a secondary disorder termed medication overuse headache (MOH) (1). Ergotamine, triptans, nonsteroidal anti-inflammatory drugs (NSAIDs), paracetamol, opioids, and caffeine are widely prescribed to chronic migraineurs and carry a different risk profile for developing MOH (2,3). Overall, their discontinuation and the subsequent amelioration of headache suggest their causative role. However, the neuropathology appears more complex and the frequent intake of AHMs does not necessarily produce MOH (4). In animal models, the regular consumption of AHMs, regardless of its type, leads to abnormal trigeminal nociceptive processing, ultimately related to neuronal nitric oxide synthase and calcitonin gene-related peptide (CGRP) (5,6). Anyway, normal serum levels of CGRP were described in MOH patients (7).

Although the discovery of biomarkers is a cornerstone in understanding of most medical conditions, MOH is still on the lookout. Several studies have so far explored different markers associated with medication overuse, including genetic polymorphisms, neurophysiological and neuroimaging abnormalities (8). They might help in identifying those patients at risk for developing the disorder, as well as the selection and monitoring of treatments, but none of them demonstrated a comprehensive role in the disease. Now, proteomics is refocusing the medical and pain research. It is becoming more and more important because of its ability to identify candidate biomarkers for many diseases, including cancers and neurodegenerative diseases (9,10). Demonstrating the clinical utility of a biomarker requires a disciplined approach: Initial explorative studies are expected to prove that the biomarker is significantly associated with the clinical state of interest. Up to now, there are no details regarding the serum proteomic profile of patients with MOH. Accordingly, the aim of this study was to explore for the first time and evaluate, using mono- (SDS-PAGE) and two-dimensional gel electrophoresis (2-DE) coupled with mass spectrometry (MS), whether changes in serum proteins of patients with MOH occur with respect to healthy individuals.

Methods

Population of the study

Sixty-nine consecutive patients with MOH, referring to the Headache and Drug Abuse Centre at the University of Modena and Reggio Emilia, were enrolled prospectively in a large single-centre proteomic study between September 2016 and January 2018. All patients were diagnosed by trained physicians, according to the international criteria (11). After signing written informed consent, collection of clinical data and completion of questionnaires were completed on admittance, before withdrawal therapy was started. The blood sampling was performed on the second day of hospitalization, fasting from midnight. For each patient, the subsequent parameters were described: (a) Prevalent acute headache medication used, classified in triptans, NSAIDs and mixtures; (b) daily drug intake (DDI), number of acute headache medications consumed every day; (c) years of chronic symptoms (CHR), time elapsed since the beginning of MOH.

Parameters (a) and (b) have been estimated considering the 3 months preceding the collection of blood samples.

Forty-two age- and sex-matched healthy volunteers were enrolled as controls. On the experimental day, they completed questionnaires and performed blood sampling, fasting from midnight. For both patients and healthy volunteers, exclusion criteria were: Headache on admittance to the centre, altered results in routine laboratory tests, other non-cephalic chronic painful conditions, and autoimmune, hepatic, renal, oncological, cardiovascular, psychiatric or neurological relevant pathologies. Proteomic analyses illustrated in this paper were conducted on a random sample of 20 patients and 18 controls.

Ethical considerations

This study was performed following the Helsinki Declaration principles, approved by the local Ethical Committee of Modena (prot. 2073) and subsequently amended by the Area Vasta Emilia Nord Ethics Committee (prot. 0019446/18). All patients and healthy volunteers gave their written consent to participate in the study before any experimental procedure was performed.

Serum samples collection and storage

Venous blood was collected in vacutainer tubes and allowed to clot at room temperature for 1 h. Serum was obtained by centrifugation at 2000 × g for 10 min at 4°C; a mixture of protease inhibitors (Sigma-Aldrich) was added to prevent protein degradation and alteration. Samples were divided into aliquots and stored at −80°C until analysis.

Proteomic analysis

Firstly, the complexity of serum was reduced by depletion of albumin and IgG using the ProteoPrep® Immunoaffinity albumin and IgG depletion kit (Sigma-Aldrich, St. Louis, MO, USA), that selectively removes these two most abundant serum proteins (12). Then, proteomic analysis was performed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis according to Laemmli’s method, as previously reported (13). Briefly, serum samples (5 µg) were mixed with a solution composed of Laemmli buffer/2-mercaptoethanol, heated to 95°C for 5 min and then loaded in duplicate onto 4–12% precast gradient gel, Novex NuPAGE™ gels (Thermo Fisher Scientific, Waltham, MA, USA). Protein bands were stained with Coomassie Blue G-250 and gel images were acquired by a calibrated densitometer (Bio-Rad Laboratories, Hercules, CA, USA). To reveal differentially expressed protein bands, the QuantityOne 1-D analysis software, version 4.6.7 (Bio-Rad Laboratories) was employed, as previously described (14). This software allows the detection of increased or decreased protein bands based on stain intensity, returning the signal as optical density (OD). Furthermore, serum samples were subjected to 2-DE. In the first-dimension separation, serum proteins (130 µg) were diluted with rehydration buffer (6 M urea, 2 M thiourea, 4% CHAPS, 25 mM dithiothreitol (DTT), 0.2% ampholytes) and then loaded onto immobilized pH gradient strips, Ready IPG Strip™ (Bio-Rad Laboratories), 7 cm long, pH range 3–10. The isoelectric focusing (IEF) and the subsequent equilibration steps were carried out as previously described (16). The second-dimension separation was achieved on 8–16% polyacrylamide gradient gel using 1X Tris-Glycine-SDS, pH 8.3 (Bio-Rad Laboratories), as running buffer. Finally, proteins were stained with a highly sensitive silver nitrate staining protocol, based on the method reported by Shevchenko et al. (15). After the densitometric acquisition, each gel image was examined by the PDQuest 2-D image analysis software, version 7.3.1 (Bio-Rad Laboratories). This software detects the spot stain intensity as OD, and the spot area (mm²), providing a final value expressed as density (OD/mm²). The 2D gel maps from MOH patients and controls were compared to detect differentially expressed protein spots according to density values.

LC-MS/MS protein identification

The bands and spots of interest were subjected to an “in-gel” trypsin digestion protocol as reported earlier (12). Shortly, bands and spots were first de-stained with acetonitrile (ACN) and with a solution composed of 30 mM potassium hexacyano-ferrate/100 mM sodium thiosulphate, respectively. Then, they were reduced with DTT, alkylated with iodoacetamide and digested overnight at 37°C with Trypsin Gold Mass Spectrometry grade (Promega, Madison, Wisconsin, United States). After digestion, peptides were extracted by ACN/water/trifluoroacetic acid (50:49:1) and concentrated in a vacuum dryer, Concentrator plus (Eppendorf, Hamburg, Germany).

Before MS analysis, dried samples were resuspended in water/ACN/formic acid (95:3:2), sonicated at room temperature and centrifuged at 12,000 × g for 10 min. Protein identification was performed by an UHPLC-ESI-QExactive™ (Thermo Fisher Scientific), composed of UltiMate 3000 UHPLC System coupled to an ESI-QExactive™ Hybrid Quadrupole-Orbitrap™ mass spectrometer (LC-MS/MS-QO System), controlled by Xcalibur™ software, version 29 build 2926 (Thermo Fisher Scientific). The column, Zorbax SB-C18 RRHT, 2.1 × 50 mm, 1.8 µ particle size (Agilent Technologies, Santa Clara, CA, USA) was equilibrated with buffer A (0.1% formic acid) and 2% ACN at a flow rate of 0.3 mL/min. After sample injection (8 µL), ACN was raised at 3%, then linearly increased until 21% in 19 min; hence, the percentage of ACN was brought to 90% in 4 min and kept for 3 min before starting the reconditioning step. The total run-time per sample was 35 min.

The acquired raw data, converted into the Mascot generic format using MsConvert (version 3.0.10730, ProteoWizard tools) were searched against the UniProtKB database for peptide sequences and C-RAP for contaminants by Mascot MS/MS ion search, version 2.4 (Matrix Science, London, UK) setting the following parameters: Homo sapiens taxonomy (_HUMAN), trypsin as proteolytic enzyme, two max missed trypsin cleavages, mass tolerances ± 10 ppm for the precursor ions and ± 0.05 Da for the product ions, cysteine carbamidomethylation as fixed modification and methionine oxidation as variable modification. The highest score hits among Mascot were selected, where the score is [−10 × log(P)] and P is the absolute probability that the observed match between the experimental data and the database sequence is a random event. Proteins were considered identified when they showed at least two unique peptides. Moreover, an automatic decoy database search was used to estimate the false discovery rate (FDR), which was adjusted to < 1%. Protein identification was done in duplicate, using bands or spots cut from different gels.

Statistics

The comparison of quantitative variables between MOH patients and healthy individuals was performed using independent t-test, whereas the comparison of qualitative variables was performed using χ². All p-values were two-sided, with a level of significance of 1%. R software (version 3.4.3) was used to perform statistical analysis.

Results

Population characteristics

Blood samples of 20 patients with MOH and 18 healthy volunteers were analysed (Table 1). Patients were between 34 and 74 years old (mean = 52.8), most were females. On admittance, headache was daily or almost daily for all of them, for not less than 3 months. Each of the 20 patients had suffered from migraine before being diagnosed with MOH. They consumed about two AHMs every day (mean =2.0, range = 0.4–4.4), and they had been suffering from chronic headache for more than 15 years (mean = 15.6, range = 0.33–40). According to the prevalent use of AHMs, 12 patients consumed triptans (60%), fewer patients were taking NSAIDs (n = 4; 20%) or mixture drugs (n = 4; 20%). The latter are combination analgesics, containing caffeine, indomethacin and prochlorperazine. The control group of healthy volunteers was an equally female-dominated group, aged between 25 and 59 years (mean = 40.6). Compared to healthy individuals, MOH patients had higher Zung Self-Rating Depression Scale (ZUNG-D) scores and less prevalent alcohol consumption. No significant differences were found in the following parameters: Body mass index, systolic and diastolic blood pressure, the prevalence of coffee and cigarette consumption and blood tests on renal function. Data and statistical comparisons are summarized in Table 1.

Table 1.

Baseline characteristics of MOH patients and healthy individuals.

	MOH patients(n = 20)	Controls(n = 18)
Age, years	52.8 ± 11.3*	40.6 ± 13.5
Sex, female	90%	83%
BMI, kg/m²	23.4 ± 3.2	23.8 ± 3.8
Coffee	80%	89%
Alcohol	20%	83%**
Cigarettes	15%	11%
SBP, mmHg	125.4 ± 13.2	119.1 ± 10.7
DBP, mmHg	77.2 ± 7.1	75.6 ± 8.8
DDI	2.0 ± 1.5	–
CHR	15.6 ± 12.4	–
ZUNG-D	46.1 ± 9.8*	31.6 ± 5.7
LDQ	15.6 ± 7.0	–
Creatinine, mg/dl	0.78 ± 0.12	0.78 ± 0.13
eGFR	>60 ml/min	>60 ml/min
Urea nitrogen, mg/dl	29.7 ± 8.6	31.2 ± 10.1

Data are expressed as mean ± SD, or in percentage.

Statistical significance was evaluated using independent t test (*p < 0.01) or χ² test (**p < 0.01).

BMI: Body Mass Index; SBP: systolic blood pressure; DBP: diastolic blood pressure; DDI: daily drug doses; CHR: years of chronification; ZUNG-D: Zung Self-Rating Depression Scale; LDQ: Leeds Dependence Questionnaire; eGFR: estimated Glomerular Filtration Rate (CKD-EPI Formula).

Note: “Alcohol” refers to the consumption of alcoholic beverages, from occasional to daily, without ever meeting the criteria of addiction.

Mono-dimensional gel electrophoresis (SDS-PAGE)

By SDS-PAGE and QuantityOne 1-D analysis software, four bands differentially expressed were revealed between patients with MOH and healthy individuals (Figure 1). Specifically, the four bands were over-expressed in the patient group, as indicated by fold-change values reported in the last column of Table 2. Proteins included in each band were identified by LC-MS/MS-QO System; all MS data are provided in Table 2.

Figure 1.

SDS-PAGE separation. The differentially expressed bands detected in serum of MOH patients compared to control subjects are indicated by alphanumeric labels (C0–C3 in controls; P0–P3 in MOH patients). The bands labels correspond to those reported in Table 2. Proteins included in each band were identified by LC-MS/MS. SDS-PAGE: 4–12% gradient gel (Novex NuPAGE); Coomassie Blue G-250 staining.

Table 2.

Differentially expressed proteins in serum of MOH patients and controls detected by SDS-PAGE and identified by LC-MS/MS.

Band ID^(a)	Protein full name	Entry name^(b)	Score^(c)	Theor. mass^(d)	Sign. match.^(e)	Sign. seq.^(f)	emPAI^(g)	Fold-change^(h)
P0	Apolipoprotein B-100	APOB	742	516651	59	55	0.38	+1.54
C0	Apolipoprotein B-100	APOB	1137	516651	72	67	0.49
P1	Alpha-2-macroglobulin	A2MG	3257	164613	170	69	4.63	+1.51
	Complement factor H	CFAH	244	143680	18	16	0.40
C1	Alpha-2-macroglobulin	A2MG	2579	164613	132	55	3.26
C1	Complement factor H	CFAH	106	143680	8	8	0.17
P2	Complement C3 (fragm)	CO3	1746	188569	87	40	1.46	+2.82
	Hemopexin	HEMO	508	52385	43	17	2.29
	Serum albumin	ALBU	458	71317	32	27	2.06
	Alpha-1B-glycoprotein	A1BG	405	54790	20	10	0.96
	Complement component C9	CO9	103	64615	7	7	0.36
	Kininogen-1	KNG1	72	72996	6	6	0.26
	Vitronectin	VTNC	39	55069	3	2	0.05
C2	Complement C3 (fragm)	CO3	1497	188569	73	38	1.21
	Serum albumin	ALBU	753	71317	55	34	4.15
	Hemopexin	HEMO	538	52385	40	17	2.67
	Alpha-1B-glycoprotein	A1BG	317	54790	15	10	0.86
	Kininogen-1	KNG1	85	72996	7	6	0.26
	Vitronectin	VTNC	59	55069	4	3	0.17
P3	Haptoglobin	HPT	769	45861	40	16	3.97	+1.91
	Apolipoprotein A-IV	APOA4	332	45371	16	14	1.71
	Alpha-1-acid glycoprotein 1	A1AG1	279	23725	10	5	1.26
	Complement C3 (fragm)	CO3	259	188569	17	14	0.28
	Serum paraoxonase/arylesterase 1	PON1	120	39877	8	7	0.64
	Zinc-alpha-2-glycoprotein	ZA2G	98	34465	5	4	0.39
	Alpha-1-acid glycoprotein 2	A1AG2	89	23873	5	3	0.42
C3	Haptoglobin	HPT	524	45861	27	13	2.04
	Complement C3 (fragm)	CO3	193	188569	13	12	0.20
	Zinc-alpha-2-glycoprotein	ZA2G	118	34465	5	4	0.39
	Alpha-1-acid glycoprotein 1	A1AG1	32	23725	2	2	0.26

^(a)Alphanumeric code of the protein band, corresponding to those reported in Figure 1 (P0-P3 bands for MOH patients; C0-C3 for controls).

^(b)Protein entry name, according to the UniProtKB database.

^(c)The highest scores with MASCOT search engine.

^(d)Theoretical molecular weight (Da).

^(e)Number of significant peptides matching the identified protein.

^(f)Number of significant sequences.

^(g)Exponentially modified protein abundance index (emPAI).

^(h)Fold-change of band intensity, achieved as the ratio between the optical density (OD) values measured in the MOH patients’ group vs. the control group.

Two-dimensional gel electrophoresis (2-DE)

2-DE combined with LC-MS/MS analysis allowed confirmation of some proteins preliminarily detected and identified by SDS-PAGE (Table 3): Hemopexin (HEMO), alpha-1-acid glycoprotein 1 (A1AG1), apolipoprotein A4 (APOA4) and haptoglobin (HPT). Figure 2 shows two representative 2D gel maps obtained from serum samples of control subjects (Figure 2(a)) and MOH patients (Figure 2(b)), which evidence the differentially expressed protein spots revealed by the PDQuest software. Moreover, other proteins were also discovered, mostly increased in MOH patients: Alpha-1-antitrypsin (A1AT), immunoglobulin heavy constant alpha 1 (IGHA1), retinol-binding protein (RETBP) and transthyretin (TTHY). Only one protein, immunoglobulin kappa constant (IGKC), was decreased in the patients group compared to controls.

Table 3.

Differentially expressed serum proteins detected by 2-DE and identified by LC-MS/MS.

Spot ID^(a)	Protein full name	Acc. Nu.^(b)	Gene name	Score^(c)	Theor. MW^(d)	Pep./sig. pep^(e)	Seq/sig. seq^(f)	emPAI^(g)	Seq.cov.^(h)	Fold-change⁽ⁱ⁾	Protein main function^(l)
IGKC	Immunoglobulin kappa constant	P01834	IGKC	1353	11.9	81/60	10/9	35.16	93	(↓) 0.53	Antigen binding
A1AT	Alpha-1-antitrypsin	P01009	SERPINA1	639	46.8	56/36	28/21	4.43	62	(↑) 1.63	Protease inhibitor
HEMO	Hemopexin	P02790	HPX	1388	52.3	152/94	35/26	9.82	58	(↑) 2.85	Metal ion binding
IGHA1	Immunoglobulinheavy constant alpha 1	P01876	IGHA1	927	38.4	68/43	16/15	4.37	59	(↑) 2.47	Antigen binding
A1AG1	Alpha-1-acid glycoprotein 1	P02763	ORM1	392	23.7	19/14	11/8	2.21	47	(↑) 3.75	Inflammatory response
APOA4	Apolipoprotein A-IV	P06727	APOA4	1230	45.3	88/57	40/35	24.49	72	(↑) 1.56	Lipid binding
HPT⁽¹⁾	Haptoglobin	P00738	HP	1383	45.8	82/62	22/19	10.97	48	(↑) 1.65	Acute phase response
HPT⁽²⁾	Haptoglobin	P00738	HP	235	45.8	19/11	10/9	0.85	20	(↑) 2.18	Acute phase response
HPT⁽³⁾	Haptoglobin	P00738	HP	1532	45.8	97/70	12/9	3.97	22	(↑) 1.85	Acute phase response
RETBP	Retinol binding protein	P02753	RBP4	716	23.3	44/34	14/14	12.66	78	(↑) 1.58	Retinol binding
TTHY⁽¹⁾	Transthyretin	P02767	Ttr	1084	15.9	51/36	13/11	24.31	77	(↑) 2.08	Hormone binding
TTHY⁽²⁾	Transthyretin	P02767	Ttr	44	15.9	10/4	7/3	0.67	72	(↑) 2.77	Hormone binding

In bold, proteins previously identified by SDS-PAGE (Table 2) are expressed.

^(a)Protein entry name from UniProtKB database as spot ID.

^(b)Primary accession number from UniProtKB database.

^(c)The highest scores with MASCOT search engine.

^(d)Theoretical molecular weight (kDa).

^(e)Total number of matched peptides/significant matches.

^(f)Total number of sequences/significant sequences.

^(g)Exponentially modified protein abundance index (emPAI).

^(h)Sequence coverage: percentage of sequenced amino acids per protein.

⁽ⁱ⁾Fold-change of protein spot expression (MOH/controls): (↓) decrease in MOH; (↑) increase in MOH.

^(l)Principal molecular function or biological process derived from UniProtKB database.

Figure 2.

Representative 2-D gel maps of serum samples. In circles are evidenced the differentially expressed spots between control group (a) and MOH patient’s group (b). For each identified protein the corresponding entry name (which reflects that listed in Table 3) is reported.

Discussion

This study represents an exploratory evaluation performed on MOH patients, designed to identify potential serum protein markers related to the chronic headache condition. It enters into a rapidly evolving area of research, concerning the study of the human pain pathophysiology using proteomic techniques. Serum profiles of patients with different chronic pain conditions (such as fibromyalgia, chronic widespread pain and painful neuropathies) were associated with abnormal levels of varying protein molecules (16,17), some appear capable of diagnosing endometriosis in patients suffering from chronic pelvic pain (18). Also, in migraine, different proteins were detected in the attack period compared to the attack-free period using 2-DE and MS analysis (19).

Serum proteins over-expressed in MOH patients

We found different proteins at unusually high concentrations in the serum of MOH patients. A1AT, TTHY, APOA4, and HPT have already been identified in previous studies and in an animal model of neuropathic pain. Specifically, A1AT and TTHY were found at a high level in the urine of MOH patients (20), whereas A1AT, APOA4, HPT, and TTHY were increased in rat serum up to 5 weeks after sciatic nerve ligation (21). Moreover, in the present study, three additional serum proteins were detected at high levels: HEMO, A1AG1, and RETBP.

HEMO is a glycoprotein known to bind heme, and transports it to the liver for breakdown and iron recovery. It is produced by several sources, including hepatocytes, mononuclear blood cells and the human brain (22), and it was previously linked to nerve injury (23,24). In patients with painful intervertebral disc degeneration, the cerebrospinal fluid level of HEMO correlated with pain intensity and physical disability (25). A1AG1 is a serum protein synthesized in the liver, which binds synthetic drugs and influences their distribution and availability in the body. Unlike other drugs-binding proteins present in serum, such as albumin, A1AG1 levels were found to be increased in chronic pain and several stress-related clinical situations (26,27). Thus, it may reduce the efficacy of drugs administered in patients who are chronically stressed, accounting for some aspects of apparent tolerance and reduced efficacy for tricyclic antidepressants and analgesics (28). Finally, RETBP is a 21 kDa protein that originates from the liver and circulates in the blood bound to TTHY. It mediates retinol transport in blood plasma and delivers retinol to the peripheral tissues. RETBP concentrations in the peripheral blood are about 50 mg/L and are increased in renal impairment. Its reabsorption by the kidney is nearly complete, so its increase in concentration is an indicator of impaired tubular function (29). A previous proteomic assessment of serum from fibromyalgia patients showed up-regulated levels of RETBP compared to healthy individuals (18). Moreover, its plasma concentration was increased in trigeminal neuralgia patients and decreased after surgery (30).

Does MOH have some neuropathic traits?

Several serum proteins were observed at unusually high concentrations in MOH patients, but the reason is unclear. Based on the results of previous studies, it could be assumed that MOH and neuropathic pain might have some similarities. The definition of neuropathic pain is still evolving and remains an issue of controversy. According to one of its most recent definitions (31), it consists of the following features: a) pain and sensory symptoms persisting beyond the healing phase; b) neurological sensory signs in variable degrees, manifesting as negative and/or positive sensory phenomena; and c) other neurological signs, including negative and/or positive motor dysfunction or autonomic symptoms. MOH shares nearly all the aforementioned clinical aspects of neuropathic pain, including hyperalgesia, allodynia, positive and negative sensory phenomena (32). Most closely related to MOH, the impact of prolonged exposure to AHMs might result in enduring neural damage and sensitization of the trigeminal nociceptive pathways, including increased facilitation and/or a diminished inhibition of pain (5,6,33). Medications typically used for neuropathic pain, such as tricyclic antidepressants and antiepileptics, show some effectiveness in MOH. In addition, the variable analgesic response of MOH with opioids appears to parallel the variable outcomes observed when opioids are administered for the treatment of neuropathic pain (34,35). Anyway, MOH is still considered more related to medication overuse and does not possess enough evidence to be regarded as a neuropathic pain manifestation; further efforts will deepen such relationship.

Limitations

The identification of candidate serum markers of MOH is only explorative. At the current status, the detected proteins provide no information about future risk of the disorder. Moreover, healthy subjects appeared significantly younger than patients after the random selection of samples. However, data from our study and those reported in the scientific literature do not suggest significant proteomic changes at these ages (52.8 vs. 40.6). A small increase in serum concentration has been described for A1GA1 (36), but is irrelevant with respect to the fold-change we observed.

Conclusions

The present study provides evidence that serum proteome can renew information about possible mechanisms underlying MOH. Specifically, the disease is associated with increased levels of serum proteins, some of which are also highly concentrated in the urine and serum of animals with neuropathic pain. Most proteins have been associated with chronic pain syndromes, suggesting that MOH may share some neuropathological mechanisms with neuropathic pain. The current findings might contribute to a better understanding of the pain sensitization processes, as well as widening the knowledge of the biological mechanisms involved in MOH. However, large-scale studies are needed to validate and extend the current results, using additional models and high-throughput approaches to demonstrate proper sensitivity and specificity for potential clinical use of some protein markers.

Clinical implications

Proteomics revealed high-level proteins in the serum of patients with MOH compared to healthy individuals.

Serum and urinary proteomic patterns of MOH share specific traits, also found in other chronic pain syndromes and in animal models of neuropathic pain.

MOH and neuropathic pain share some markers and clinical manifestations, but their relationship deserves further investigation.

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Abrams

BM.

Medication overuse headaches. Med Clin North Am 2013; 97: 337–352.

Mathew

NT.

Transformed migraine. Cephalalgia 1993; 13: 78–83.

Bigal

Serrano

Buse

, et al. Acute migraine medications and evolution from episodic to chronic migraine: A longitudinal population-based study. Headache 2008; 48: 1157–1168.

Scher

Rizzoli

Loder

EW.

Medication overuse headache: An entrenched idea in need of scrutiny. Neurology 2017; 89: 1296–1304.

De Felice

Ossipov

Porreca

Persistent medication-induced neural adaptations, descending facilitation, and medication overuse headache. Curr Opin Neurol 2011; 24: 193–196.

De Felice

Ossipov

Wang

, et al. Triptan-induced enhancement of neuronal nitric oxide synthase in trigeminal ganglion dural afferents underlies increased responsiveness to potential migraine triggers. Brain 2010; 133: 2475–2488.

Munksgaard

Ertsey

Frandsen

, et al. Circulating nociceptin and CGRP in medication-overuse headache Acta Neurol Scand 2019; 139: 269–275.

Diener

Holle

Solbach

, et al. Medication-overuse headache: Risk factors, pathophysiology and management. Nat Rev Neurol 2016; 12: 575–583.

Fredriksson

Dixon

, et al. Multiplexed protein detection by proximity ligation for cancer biomarker validation. Nat Methods 2007; 4: 327–329.

10.

Fredriksson

Horecka

Brustugun

, et al. Multiplexed proximity ligation assays to profile putative plasma biomarkers relevant to pancreatic and ovarian cancer. Clin Chem 2008; 54: 582–589.

11.

Headache Classification Committee of the International Headache Society: The International Classification of Headache Disorders, 3rd edition (beta version). Cephalalgia 2013; 33: 629–808.

12.

Bellei

Bergamini

Monari

, et al. High-abundance proteins depletion for serum proteomic analysis: Concomitant removal of non-targeted proteins. Amino Acids 2011; 40: 145–156.

13.

Bellei

Monari

Cuoghi

, et al. Discovery by a proteomic approach of possible early biomarkers of drug-induced nephrotoxicity in medication-overuse headache. J Headache Pain 2013; 14: 6.

14.

Bellei

Cuoghi

Monari

, et al. Proteomic analysis of urine in medication-overuse headache patients: Possible relation with renal damages. J Headache Pain 2012; 13: 45–52.

15.

Shevchenko

Wilm

Vorm

, et al. Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal Chem 1996; 68: 850–858.

16.

Ruggiero

Era

Cacace

, et al. A preliminary study on serum proteomics in fibromyalgia syndrome. Clin Chem Lab Med 2014; 52: e207–e210.

17.

Uçeyler

Rogausch

Toyka

, et al. Differential expression of cytokines in painful and painless neuropathies. Neurology 2007; 69: 42–49.

18.

Irungu

Mavrelos

Worthington

, et al. Discovery of non-invasive biomarkers for the diagnosis of endometriosis. Clin Proteomics 2019; 16: 14.

19.

Yuasa

Nagata

Fujii

, et al. Serum apolipoprotein E may be a novel biomarker of migraine. PLoS One 2018; 13: e0190620.

20.

Bellei

Monari

Bergamini

, et al. Validation of potential candidate biomarkers of drug-induced nephrotoxicity and allodynia in medication-overuse headache. J Headache Pain 2015; 16: 77.

21.

Bellei

Vilella

Monari

, et al. Serum protein changes in a rat model of chronic pain show a correlation between animal and humans. Sci Rep 2017; 7: 41723.

22.

Morris

Candy

Edwardson

, et al. Evidence for the localization of haemopexin immunoreactivity in neurones in the human brain. Neurosci Lett 1993; 149: 141–144.

23.

Jimenez

Stam

, et al. Proteomics of the injured rat sciatic nerve reveals protein expression dynamics during regeneration. Mol Cell Proteomics 2005; 4: 120–132.

24.

Madore

Camborieux

Bertrand

Regulation of hemopexin synthesis in degenerating and regenerating rat sciatic nerve. J Neurochem 1999; 72: 708–715.

25.

Lim

TKY

Anderson

Hari

, et al. Evidence for a role of nerve injury in painful intervertebral disc degeneration: A cross-sectional proteomic analysis of human cerebrospinal fluid. J Pain 2017; 18: 1253–1269.

26.

Colley

Fleck

Goode

, et al. Early time course of the acute phase proteins response in man. J Clin Pathol 1983; 36: 203–207.

27.

Fukui

Hameroff

Gandolfi

Alpha-1-acid glycoprotein and beta-endorphin alterations in chronic pain patients.

Anesthesiology 1984; 60: 494–496.

28.

Piafsky

DM.

Disease-induced changes in the plasma binding of drugs. Clin Pharmacokinet 1980; 5: 246–262.

29.

Warrington

Ravic

Dawnay

Renal and general tolerability of repeated doses of nimesulide in normal subjects. Drugs 1993; 46: S263–S269.

30.

Farajzadeh

Bathaie

Arabkheradmand

, et al. Different pain states of trigeminal neuralgia make significant changes in the plasma proteome and some biochemical parameters: A preliminary cohort study. J Mol Neurosci 2018; 66: 524–534.

31.

Backonja

MM.

Defining neuropathic pain. Anesth Analg 2003; 97: 785–790.

32.

Biondi

DM.

Is migraine a neuropathic pain syndrome?

Curr Pain Headache Rep 2006; 10: 167–178.

33.

Meng

Dodick

Ossipov

, et al. Pathophysiology of medication overuse headache: Insights and hypotheses from preclinical studies. Cephalalgia 2011; 31: 851–860.

34.

Arner

Meyerson

BA.

Lack of analgesic effect of opioids on neuropathic and idiopathic forms of pain.

Pain 1988; 33: 11–23.

35.

Johnson

Hutchinson

Williams

, et al. Medication-overuse headache and opioid-induced hyperalgesia: A review of mechanisms, a neuroimmune hypothesis and a novel approach to treatment. Cephalalgia 2013; 33: 52–64.

36.

Abernethy

Kerzner

Age effects on alpha-1-acid glycoprotein concentration and imipramine plasma protein binding. J Am Geriatr Soc 1984; 32: 705–708.

Identification of candidate proteomic markers in the serum of medication overuse headache patients: An exploratory study

Abstract

Purpose of the study

Main findings

Conclusions

Keywords

Introduction

Methods

Population of the study

Ethical considerations

Serum samples collection and storage

Proteomic analysis

LC-MS/MS protein identification

Statistics

Results

Population characteristics

Mono-dimensional gel electrophoresis (SDS-PAGE)

Two-dimensional gel electrophoresis (2-DE)

Discussion

Serum proteins over-expressed in MOH patients

Does MOH have some neuropathic traits?

Limitations

Conclusions

Clinical implications

Footnotes

Declaration of conflicting interests

Funding

References