Association of angiotensin converting enzyme type 2 serum level and gene polymorphisms with multiple sclerosis

Abstract

Multiple sclerosis (MS) is the most prevalent autoimmune neurodegenerative heterogeneous disease affecting young adults, and angiotensin-converting enzyme 2 (ACE2) is a potential biomarker in MS. To compare serum levels of ACE2 between MS patients and healthy controls, and to investigate the relationships between ACE2 gene polymorphisms (rs2074192 and rs2285666) and MS susceptibility and clinical characteristics in a Jordanian population. A case-control study was conducted. Enzyme-linked immunosorbent assay (ELISA) was used to measure ACE2 levels in the serum of 88 MS patients and 87 controls. Genotyping of rs2074192 and rs2285666 polymorphisms was performed for 498 MS patients and 504 healthy controls by Illumina HiSeq xTen system (Illumina Platform) technique. Genetic analyses were sex-stratified and used X-chromosome–appropriate coding, with Hardy–Weinberg equilibrium assessed in females only. After adjustment for covariates, log(ACE2) serum levels were significantly higher in MS patients compared to healthy controls (p < 0.001). The ACE2 rs2074192 TT genotype (p = 0.003), and T allele (p < 0.001) are significantly associated with cases. Significant associations were found between cases and the ACE2 rs2285666 CC genotype (p = 0.037), and C allele (p = 0.040). Significant associations were revealed between rs2074192 genotype and EDSS level (p = 0.036), and between rs2285666 genotype and current treatment with DMT (p = 0.022). MS patients had higher ACE2 serum levels than healthy controls and ACE2 is a susceptibility gene for MS in the Jordanian population.

Introduction

Multiple sclerosis (MS) is the most common autoimmune neurodegenerative disease affecting young adults with a mean age of 20 to 30 years¹. In both developed and developing countries, the incidence of MS increased from 2.1 million in 2008 to 2.9 million in 2023, with being two-three times more prevalent among females compared to males².

A great heterogeneity exists among MS patients in terms of morphological changes in the brain, clinical presentation, and response to treatment³. This heterogeneity is believed to be caused by the complex nature of MS, which results from the interaction between multiple predisposing factors; most of them are related to immune function, in addition to several environmental factors^4,5. It is postulated that an adaptive immune response against myelin causes inflammation of the central nervous system (CNS) and the development of lesions (plaques) in the brain⁶. This immune responses involve the activation of CNS resident microglia as well as the infiltration of peripheral immune cells like T and B cells, neutrophils, and monocytes⁷. Some studies showed that the autoimmune status in MS patients results from the involvement of autoreactive T helper 1 cells (Th1) and T helper 17 cells (Th17) causing injury to axons and destruction of myelin sheathes^8,9.

The identification of new biomarkers in MS is unequivocal to allowing the earlier diagnosis, better prediction of disease severity and progression, as well as the personalization of therapy in complement with magnetic resonance image (MRI) and clinical markers of MS. Moreover, such biomarkers may represent new therapeutic targets in MS. The renin-angiotensin system (RAS), classically known for its role in blood pressure regulation, was also found to affect the immune system function. It has two opposing axes that are active throughout the body, including the CNS^10,11. The classic RAS, recently known as the pathologic arm, consists of angiotensin II (AngII), angiotensin type 1 receptor (AT1R), and angiotensin-converting enzyme (ACE) which converts the angiotensin I (Ang I) to Ang II^10,11. Another axis called the protective arm is composed of angiotensin 1–7 (Ang (1– 7)), its Mas receptor (MasR), angiotensin type 2 receptor (AT2R), and angiotensin converting enzyme 2, (ACE2) which is responsible for the formation of Ang (1–7) from AngII with a minor efficiency to metabolize Ang I into angiotensin (1–9) [Ang-(1–9)]^10,11. A growing body of literature reported the role of the RAS in MS^12,13. ACE, the key enzyme in RAS, was suggested to have proinflammatory effects on T cells via stimulating the production of antigenic peptides^14,15. In experimental allergic encephalomyelitis (EAE), an in vivo model of MS, inhibition of ACE exhibited a beneficial effect by ameliorating the clinical course of the EAE and suppressing the disease^16,17. Additionally, proteomic analysis revealed that ACE and AT1R inhibition can suppress the Th17 cells¹⁸. In patients with MS, higher ACE activity and cerebrospinal fluid (CSF) levels were observed as compared to controls^19,20.

Little is known about ACE2 in MS, whereas it is recognized as another crucial RAS regulator. Ang (1–7), the downstream peptides of the ACE2 pathway, and their MasR were associated with neuroprotective activity by reducing oxidative stress and attenuating neural apoptosis^21,22. Ang (1–7) play a crucial role in maintaining the tight junctions of the blood-brain barrier (BBB), which is a key factor in MS, as inflammation leads to the infiltration of inflammatory mediators through the BBB into the CNS²³. As compared to controls, CSF levels of ACE2 were significantly lower in MS patients²⁰. Serum ACE2 was measured in different patient groups such as patients with hypertension²⁴, COVID-19²⁵, asthma²⁶, and systemic sclerosis²⁷. The results of previous studies support the importance of ACE2 as a regulatory enzyme of RAS in MS.

The human ACE2 gene is located at position p22.2 on chromosome X, where a large number of single nucleotide polymorphisms (SNPs) and other genetic variants have been found²⁸. Previous studies showed that rs2285666 and rs2074192 are potential risk factors for many diseases such as hypertension, severe COVID-19, metabolic syndrome, and type 2 diabetes mellitus^29–33. The ACE2 rs2285666 is located in intron 3 and is thought to create a new donor splice site, alter gene expression and serum protein levels^34,35. The ACE2 rs2074192 is located in intron 16 and was shown to be associated with increased donor site and protein features³⁶.

While previous studies examined CSF levels of ACE2, as far as we know, there are no studies investigating the serum levels of ACE2 in MS patients. Therefore, we aimed to evaluate the levels of circulating ACE2 in MS patients as compared to healthy controls and explore the association between these levels and patient clinical characteristics. We additionally aimed to examine the relationship between ACE2 gene SNPs, rs2285666 and rs2074192, and patient clinical features and MS susceptibility.

In this study we provide new evidence that ACE2 could serve as a potential diagnostic biomarker and susceptibility factor for MS.

Methods

Study design

This was a case-control study. Patients with MS disease were recruited from the major hospitals in Jordan: Al-Bashir hospital, Princess Basma hospital, and King Abdullah University Hospital, from January 31, 2021, to January 15, 2022. All eligible patients with MS who were attending neurology clinics at the participating sites and who fulfilled the McDonald’s criteria³⁷, were invited to participate in this study. Patients who had hypertension or were taking ACE inhibitors were excluded from the study. The study procedure and purpose were explained to the approached patients, and an informed consent was received. Clinical and demographic data were collected from patients including age, weight, sex, family history of MS or any other autoimmune disorder. The hospital identification number was obtained for each patient to have access to their medical profiles and obtain information related to the age of onset, number of relapses within the last 24 months, type of MS, and the type of disease modifying therapy (DMT) used. The Expanded Disability Status Scale (EDSS) was used as the rating scale of disability and impairment of MS patients³⁸. According to the EDSS level, patients were subdivided into three groups; EDSS 0-1.5: normal-no disabilities with some abnormal signs, EDSS 2-2.5: minimal disability, and EDSS ≥ 3: moderate-severe disability.

Patients were divided into two groups based on the MS subtype; relapsing-remitting MS (RRMS) and progressive MS (PRMS; consisting of both primary and secondary subtypes). Patients with RRMS who had a history of relapse in the previous six months were excluded from the study. Relapses were defined as focal neurological disturbances persisting for more than 24 h without any other plausible cause³⁹.

Additionally, healthy controls were recruited, including blood bank donors as well as non-affected relatives and friends of patients who visited the clinics. Those with autoimmune disorders, neurological diagnoses, psychotic diagnoses, chronic medical illnesses, or a family history of MS were excluded from the study.

ACE2 ELISA

The serum levels of ACE2 (CLOUD-CLONE CROP, USA, cat. #SEB886Hu) were measured using human enzyme-linked immunosorbent assay (ELISA) technique. All kit components were brought to room temperature (18–25 °C) before starting the procedure. The procedure was performed according to the manufacturer’s instructions. An EZ Read 400 ELISA microplate reader was utilized, and serum protein levels were measured in pg/ml. The samples from cases and controls were randomly mixed, and the laboratory analysis was conducted blindly.

Genotyping

DNA was extracted from peripheral blood samples collected in EDTA tubes of each participant according to the manufacturer’s protocol of DNA extraction kit (QIAamp^® Blood Mini Kit – QIAGEN). The DNA concentration for all extracted samples was measured using NanoDrop 1000 spectrophotometer (Thermo Scientific Inc, Wilmington, USA). The OD260/OD230 ratio was also measured. The quality of DNA libraries obtained was assessed by capilar electrophoresis by the QIAxcel Advanced System (Qiagen). All the DNA samples were normalized to 20 ng/µl. The SNP variants investigated in ACE2 gene were tested with amplicon sequencing in the Illumina HiSeq xTen system (Illumina Platform). This method uses oligonucleotide probes designed to target and capture regions of interest, followed by next-generation sequencing (NGS).

Primer sequences used in the study were as follows:

rs2074192 (chrX:15564667 (GRCh38.p14)).

Forward primer: 5’-CTTTCATGCCTTGCAACCTAGA-3’.

Reverse primer: 5’-AACTTTGCCCTTAAACACAGCA-3’.

rs2285666 (chrX:15592225 (GRCh38.p14)).

Forward primer: 5’-TCTGCAATCATTTTTAAAATCTGAGAGA-3’.

Reverse primer: 5’-ACCCAGATAATCCACAAGAATGCT-3’.

Statistical analysis

Statistical analysis was performed using the Statistical Package for Social Sciences SPSS software system (IBM SPSS Statistics 25, USA), GraphPad Prism 5^®, and R version 4.4.1. Categorical variables were summarized as frequencies and percentages, and continuous variables as medians with interquartile range (IQR). The Kruskal-Wallis’s test was used to compare findings from several groups, followed by Dunn’s test. Serum ACE2 levels demonstrated right-skewed distribution and were therefore log-transformed prior to analysis. Comparisons of ACE2 concentrations between MS patients and controls were performed using multivariable linear regression, with log-ACE2 as the dependent variable and MS status, age, sex, and BMI included as covariates. Additional MS-specific clinical associations (e.g., EDSS, disease duration) were evaluated using regression models restricted to the MS group.

Diagnostic performance of serum ACE2 was evaluated using receiver operating characteristic (ROC) analysis. Optimal cut-off values were derived using Unal’s index and subsequently entered into binary logistic regression models to assess predictive ability.

Genetic analysis (X-chromosome-appropriate)

Because ACE2 is located on the X chromosome, all genetic analyses accounted for sex-specific genotype representation. Genotypes were recoded into diploid categories (“major homozygous”, “heterozygous”, “minor homozygous”). Hardy–Weinberg equilibrium (HWE) was assessed only in female participants, as males are hemizygous. Standard chi-square–based HWE tests can be unreliable for numerically encoded genotypes or large samples; therefore, we used the exact HWE test (HWExact, HardyWeinberg R package), which provides stable estimates without asymptotic assumptions.

To evaluate associations between ACE2 SNPs and MS, sex-stratified logistic regression models were applied. In males, genotypes were coded as hemizygous (0 = major allele, 1 = minor allele). In females, additive models (0/1/2 copies of the minor allele) were used. All models were adjusted for age and BMI.

Haplotype analysis

Haplotype estimation for rs2074192 and rs2285666 was performed using the haplo.em and haplo.glm functions (haplo.stats package). Haplotypes with frequencies < 5% were excluded to avoid instability. Multiple-testing correction was applied using the false discovery rate (FDR).

A two-tailed p-value < 0.05 was considered statistically significant.

Results

Baseline characteristics of MS patients and healthy controls

A total of 173 participants were included in the ACE2 serum concentration analysis, included 88 MS patients, of whom 80 (90.9%) had RRMS, 8 (9.1%) had PRMS, and 87 healthy controls who were age-and gender-matched. The median age of the MS group was 30 years (IQR: 23.25–39.75 years) and of the control group was 30 years (IQR: 24–39). About 74% of MS patients were females and 13.6% had comorbid conditions. The comorbidities included diabetes mellitus, thyroid gland problems, hypertension, G6PD deficiency, and urinary incontinence.

Detailed demographics, clinical characteristics, and the serum levels of ACE2 of MS patients and healthy controls are summarized in Table 1.

Table 1.

Baseline characteristics of MS patients and healthy controls.

Protein study				Genotyping study
Characteristic	Patient group (N = 88)	Control group (N = 87)	P-value	Patient group (N = 498)	Control group (N = 504)	p-value
Age, yearsa	30 [23.25–39.75]	30 [24.00–39.00]	0.912 b	35.00 [27.00–42.00]	24.50 [22.00–31.00]	< 0.001 b
Gender			0.726 c			< 0.001 c
Male	23 (26.1)	20 (23.0)		135 (27.1)	221 (43.8)
Female	65 (73.9)	67 (77.0)		363 (72.9)	283 (56.2)
BMIa	23.34 [21.29–28.37]	26.22 [22.84–29.95]	0.012 b	24.22 [21.81–28.00]	25.15 [22.10–29.30]	0.028 b
Comorbidity			–			–
No	76 (86.4)	87 (100.0)		395 (79.3)	504 (100.0)
Yes	12 (13.6)	0 (0.0)		100 (20.1)	0 (0.0)
Missing	–			3 (0.60)	0 (0.0)
Age at MS onset, yearsa	25.00 [20.00–33.75]	–		27.00 [21.00–34.00]	–	–
Age at diagnosisa	26.00 [20.00–34.75]	–		28.00 [22.00–35.00]	–	–
MS disease duration, monthsa	36.00 [12.00–81.00]	–		48.00 [12.00–108.00]	–	–
MS type		–	–		–	–
CIS	9 (10.2)			57 (11.45)
RR-MS	71 (80.7)			386 (77.51)
PR-MS	8 (9.1)			47 (9.44)
Missing	0 (0.0)			8 (1.61)
EDSS scorea	1.00 [1.00–2.00]	–		2.00 [1.00–3.00]	–	–
EDSS level		–			–	–
Normal-mild disability	67 (76.1)			310 (62.2)
Moderate-severe	21 (23.9)			180 (36.1)
Missing	0 (0.0)			8 (1.6)
Previous relapse in the last 24 months		–			–	–
No	50 (56.8)			216 (43.4)
Yes	38 (43.2)			272 (54.6)
Missing	0 (0.0)			10 (2.0)
Number of relapses in the last 24 monthsa	1.00 [1.00–2.00]	–		2.00 [1.00–3.00]	–	–
Currently on DMT
No	31 (35.23)			173 (34.74)
Yes	57 (64.77)			317 (63.65)
Missing	0 (0.0)			8 (1.61)
Type of DMT		–			–	–
Interferon beta 1-a (Avonex®, IM once weekly)	2 (3.5)			17	–	–
Interferon beta 1-a (Rebif®, SC three times weekly)	13 (22.8)			72	–	–
Interferon beta 1-b (Betaferon®)	16 (28.1)			73	–	–
Natalizumab (Tysabri®)	1 (1.8)			12	–	–
Fingolimod (Gilenya®)	24 (42.1)			135	–	–
Dimethyl fumarate	1 (1.8)			6	–	–
Missing	0 (0.0)			8	–	–
ACE2 levela	1884.85 [1541.15–2349.25]	1059.72 [825.90–1451.42]	< 0.001b	–	–	–

All data are presented as n (%) unless otherwise indicated.

^aMedian with the interquartile range in brackets [25–75 percentile].

^bMann–Whitney test.

^cChi-Square test.

n= number, MS: Multiple Sclerosis, RRMS: relapsing-remitting multiple sclerosis, PRMS: progressive multiple sclerosis, CIS: Clinically Isolated Syndrome, EDSS: Expanded Disability Status Scale, IM: intramuscular, SC: subcutaneous, DMT: disease modifying therapy, ACE2 Angiotensin-Converting Enzyme 2.

For the genotype study, 498 MS participants, and 504 controls were successfully genotyped for ACE2 variants rs2074192 and rs2285666. The median age of the patient group was 35 years, and of the control group was 24.5 years. 72.9% of the patients were females. About 20% of the patients had other comorbidities. Most of the patients had RRMS (77.5%). About 64% of the patients were taking DMT, 42.1% of them were on Fingolimod.

Serum levels of ACE2 can discriminate between MS patients and controls

Serum ACE2 concentrations were right-skewed and were therefore log-transformed for analysis. In the multivariable linear regression model adjusting for age, sex, and BMI, log-transformed ACE2 levels [log(ACE2)] were significantly between MS patients and controls (p < 0.001).

Raw serum ACE2 concentrations showed marked right-skewness (Fig. 1A), with skewness = 1.60 and kurtosis = 3.40. QQ plots further confirmed deviation from normality (Fig. 1B). Shapiro–Wilk test confirmed non-normal distribution of raw ACE2 levels (p = 5.8 × 10⁻¹¹). Log-transformation improved distributional properties substantially (Fig. 1C and D), confirmed by Shapiro–Wilk test (p = 0.056), and was therefore used for all subsequent regression analyses.

Fig. 1.

Distribution and Normalization of Serum ACE2 serum Levels. Raw ACE2 concentrations demonstrated right-skewed distribution, as shown by histogram (A) and Q–Q plot inspection (B). Log-transformation improved normality substantially (C, D).

The median ACE2 level in MS patients was 1884.850 pg/ml compared to 1059.720 pg/ml in healthy controls (p < 0.001). Log(ACE2) levels were significantly higher in MS patients compared to controls (Fig. 2A), and this difference remained significant after adjustment for age, sex, and BMI.

Fig. 2.

Comparison of log-transformed ACE2 serum levels between MS patients and controls. (A) Box blot displaying the median values and interquartile ranges of serum levels of log(ACE2) in the cases and healthy controls groups. (B) Receiver operating characteristic (ROC) curve displaying the validity of log(ACE2) as a diagnostic marker for multiple sclerosis (MS).

ROC analysis of ACE2 serum levels demonstrated fair discriminatory performance (AUC = 0.783, 95% CI 0.712–0.860), confirming that ACE2 serum level can distinguish MS patients from controls (Fig. 2B). According to the Unal’s method the optimal cut-point value was determined and it corresponded to a value that is close to the AUC and where the absolute difference between specificity and sensitivity is the least⁴⁰. The cut-point was determined as 1459.03 pg/ml with positive and negative predictive values of 76.7% and 77.6%, respectively. The discrimination of MS patients from healthy controls based on ACE2 levels less than 1459.03 pg/ml (p < 0.001) resulted in odds ratio of 0.088 (95% CI: 0.043–0.178).

Relationship between log(ACE2) serum levels and demographics and clinical characteristics of MS patients and healthy controls

As shown in Tables 2 and 3, there is a significant association between gender and log(ACE2) serum levels in both cases and control groups. Significant correlations were also revealed between log(ACE2) serum levels and age (p = 0.018) and BMI (p = 0.004) of control participants.

Table 2.

Correlation between log(ACE2) serum level and demographics and clinical characteristics of MS patients and healthy controls.

Characteristic	MS patients (N = 88)		Healthy controls (N = 87)
	Spearman correlation	P-value	Spearman correlation	P-value
Age, years	0.03	0.783	0.25	0.018
BMI	0.15	0.181	0.31	0.004
Age at MS onset, years	0.058	0.599	–	–
Age at diagnosis	0.059	0.591	–	–
MS disease duration, months	-0.058	0.597	–	–
EDSS score	-0.078	0.476	–	–
Number of relapses in the last 24 months	-0.046	0.778	–	–

Significant values are in bold.

Table 3.

Association between log(ACE2) serum level and demographics and clinical characteristics of MS patients and healthy controls.

Characteristic	MS patients		Healthy controls
	Median [IQR]	p-value	Median [IQR]	p-value
Gender		0.04		< 0.001
Male	1583.28 [1359.64–2085.16]		1758.25 [1088.67–2816.53]
Female	2035.08 [1609.00–2394.38]		954.64 [799.62–1249.26]
Comorbidity		0.419	–	–
No	1864.26 [1472.24–2341.72]
Yes	2038.74 [1609.18–2507.99]
MS type		0.403	–	–
CIS	2157.70 [1504.51–2423.44]
RR-MS	1850.60 [1516.92-2276.30]
PR-MS	2174.96 [1599.59-3714.02]
EDSS level		0.614	–	–
Normal-mild disability	1902.02 [1557.78-2386.56]
Moderate-severe	1850.60 [1452.69-2231.53]
Previous relapse in the last 24 months		0.963	–	–
No	1859.14 [1569.71-2369.28]
Yes	1947.32 [1427.37-2354.73]
Currently on DMT		0.057	–	–
No	2092.36 [1657.46-2575.18]
Yes	1799.80 [1459.03-2239.02]

Significant values are in bold.

A near significant association was observed between log(ACE2) serum level and the current use of DMT (p = 0.057).

In multivariable regression (Table 3), MS status was not an independent predictor of log( ACE2) serum levels (p = 0.278). Age was positively associated with log(ACE2) concentrations, while BMI showed a marginal inverse association.

Allele and genotype frequencies of ACE2 polymorphisms in MS and control groups

Table 4 shows the results of comparing ACE2 alleles, genotypes, and haplotypes between MS patients and healthy controls.

Table 4.

ACE2 allele and genotype frequencies in MS patients and healthy controls.

SNP	Allele/genotype	MS patients (N = 498) N (%)	Healthy controls (N = 504) N (%)	Univariate analysis		Multivariate analysis
				P	OR (95% CI)	P	OR (95% CI)
rs2074192	T C	402 (40.4) 594 (59.6)	323 (32.0) 685 (68.0)	< 0.001	0.697 (0.580–0.837)	–	–
	TT CT CC	114 (22.9) 174 (34.9) 210 (42.2)	82 (16.3) 159 (31.5) 263 (52.2)	0.003 0.188 0.001	Ref 1.270 (0.890–1.814) 1.741 (1.243–2.438)	< 0.001 < 0.001 < 0.001	Ref 2.297 (1.488–3.544) 1.991 (1.367–2.898)
	TT CT + CC	114 (22.9) 384 (77.1)	82 (16.3) 422 (83.7)	0.008	1.528 (1.114–2.095)	< 0.001	Ref 2.078 (1.448–2.981)
rs2285666	T C	179 (18.0) 817 (82)	218 (21.6) 790 (78.4)	0.040	1.26 (1.01–1.57)	–	–
	TT CT CC	34 (6.8) 111 (22.3) 353 (70.9)	53 (10.5) 112 (22.2) 339 (67.3)	0.114 0.091 0.037	Ref 0.647 (0.391–1.072) 0.616 (0.391–0.972)	0.441 0.818 0.580	Ref 1.072 (0.593–1.939) 0.865 (0.518–1.444)
	TT CT + CC	34 (6.8) 464 (93.2)	53 (10.5) 451 (89.5)	0.040	0.624 (0.398–0.978)	0.659	Ref 0.891 (0.535–1.485)
Haplotypes rs2074192/ rs2285666	C/C T/C C/T T/T	210 (42.2) 200 (40.0) 87 (17.4) 2 (0.4)	236 (46.9) 159 (31.4) 106 (21.0) 3 (0.6)	0.026 < 0.001 0.049 –	0.818 (0.685–0.976) 1.448 (1.205–1.741) 0.798 (0.638–0.997) –	–	–

Significant values are in bold.

Allele and genotype frequencies are presented as absolute numbers with percentage in parentheses.

OR – odds ratio; CI – confidence interval. Univariate analysis is based on χ2 test. Multivariate analysis is adjusted for gender and age.

In the control group, the genotypic prevalence of rs2074192 polymorphism was 52.2% (n = 263) for CC, 31.5% (n = 159) for CT, and 16.3% (n = 82) for TT genotype. Among MS patients, CC was 42.2% (n = 210), CT was 34.9% (n = 174), TT was 22.9% (n = 114). As shown in the table, TT genotype, and T allele significantly associated with cases, showing a significant association between the risk of MS and the inheritance of TT genotype (p = 0.003).

Regarding rs2285666, among controls, the genotypic prevalence was 67.3% (n = 339) for CC, 22.2% (n = 112) for CT, and 10.5% (n = 53) for TT genotype. In the case group, CC was 70.9% (n = 353), CT was 22.3% (n = 111), and TT was 6.8% (n = 34). As shown in the table, CC genotype, and C allele significantly associated with cases, showing a significant association between the risk of MS and the inheritance of CC genotype (p = 0.037).

Haplotype analysis of ACE2 polymorphisms in MS patients and controls

Four haplotypes between ACE2 SNPs rs2074192 and rs2285666 were constructed. A significant difference in the overall frequencies of the haplotypes was observed between MS patients and controls (Table 4). In general, the C/C haplotype was more common in the controls (46.9%) than in the MS patients (42.2%) (OR 0.818, 95% CI 0.685–0.976, p = 0.026), the T/C haplotype was more common in MS patients (40.0%) than in controls (31.4%) (OR 1.448, 95% CI 1.205–1.741, p < 0.001), and the C/T haplotype was more common in controls (21.0%) than in MS patients (17.4%) (OR 0.798, 95% CI 0.638–0.997, p = 0.049).

HWE was evaluated only in females using the exact test (HWExact), as ACE2 is located on the X chromosome. Both ACE2 SNPs in MS patients and controls were in Hardy–Weinberg equilibrium among females, supporting high-quality genotyping and no deviation from expected equilibrium patterns. Significant LD was found between rs2074192 and rs2285666 SNPs (p < 0.01, Fig. 3), separated by 3 kb, and a slightly higher correlation was observed in MS patients (r² = 0.131) compared with controls (r² = 0.108). Gender-based differences.

Fig. 3.

Linkage disequilibrium (LD) analysis showing a strong LD between rs2874192 and rs2205666; (D statistic = − 0.0665, D′ statistic = 0.9278) (r statistic = − 0.3472, r2 = 0.0121).

Patient and control groups were analyzed according to gender. As shown in Table 5, In the control group, female patients had significantly lower BMI levels than male patients (p < 0.001). Females had also significantly lower EDSS scores than males (p = 0.010) as 66.8% of them were in the normal-mild disability group (p = 0.011). In terms of serum ACE2 levels, in the case group, females had significantly higher levels than males (2035.08 pg/ml vs. 1583.28 pg/ml; respectively, p = 0.040). The opposite was evident in the control group (954.64 pg/ml vs. 1583.28 pg/ml; respectively, p < 0.001).

Table 5.

Comparison of demographics, clinical characteristics, and serum log(ACE2) level and ACE2 gene polymorphisms according to gender.

Characteristic	MS patients			Healthy controls
	Males	Females	p-value	Males	Females	p-value
Age	33.00 [28.00–41.00]	35.00 [26.00–43.00]	0.543a	27.00 [23.00–35.00]	24.00 [21.00–28.00]	< 0.001 a
BMI	24.36 [21.90-27.43]	24.22 [21.71–28.19]	0.778a	26.37 [23.84–30.60]	23.92 [21.55–27.94]	< 0.001 a
Age at first symptoms	26.00 [22.00-32.75]	27.00 [21.00–34.00]	0.592a	–	–	–
Age at diagnosis	28.00 [22.50–33.50]	29.00 [22.00–36.00]	0.541a	–	–	–
Disease duration (months)	36.00 [12.00–108.00]	60.00 [12.00–108.00]	0.361a	–	–	–
Number of relapses in the previous 24 months	2.00 [1.00–3.00]	2.00 [1.00–3.00]	0.693a	–	–	–
EDSS score	2.00 [1.00–4.00]	2.00 [1.00–3.00]	0.010 a	–	–	–
EDSS level			0.011 b	–	–	–
Normal-mild disability	71 (53.8)	239 (66.8)
Moderate-severe disability	61 (46.2)	119 (33.2)
Log(ACE2) level	1583.28 [1359.64–2085.16]	2035.08 [1609.00–2394.38]	0.040 a	1758.25 [1088.67–2816.53]	954.64 [799.62–1249.26]	< 0.001 a
rs2074192
TT	53 (39.3)	61 (16.8)	< 0.001 b	58 (26.2)	24 (8.5)	< 0.001 b
CT	3 (2.2)	171 (47.1)		15 (6.8)	144 (50.9)
CC	79 (58.5)	131 (36.1)		148 (67.0)	115 (40.6)
TT	53 (39.3)	61 (16.8)	< 0.001 b	58 (26.2)	24 (8.5)	< 0.001 b
CT + CC	82 (60.7)	302 (83.2)		163 (73.8)	259 (91.5)
rs2285666
TT	21 (15.6)	13 (3.6)	< 0.001 b	41 (18.6)	12 (4.2)	< 0.001 b
CT	5 (3.7)	106 (29.2)		7 (3.2)	105 (37.1)
CC	109 (80.7)	244 (67.2)		173 (78.3)	166 (58.7)
TT	21 (15.6)	13 (3.6)	< 0.001 b	41 (18.6)	12 (4.2)	< 0.001 b
CT + CC	114 (84.4)	350 (96.4)		180 (81.4)	271 (95.8)

^aMann-Whitney test.

^bChi-square test.

Significant values are in bold.

A statistically significant difference in the ACE2 genotypes between males and females was found in the case and control groups. In the case group, the frequencies of rs2074192 TT, CT, and CC genotypes were 39.3%, 2.2%, and 58.5% in males and 16.8%, 47.1%, and 36.1% in females; respectively (p < 0.001). The frequencies of rs2285666 TT, CT, and CC genotypes were 15.6%, 3.7%, and 80.7% in males and 3.6%, 29.2%, and 67.2% in females; respectively (p < 0.001). In the control group, the frequencies of rs2074192 TT, CT, and CC genotypes were 26.2%, 6.8%, and 67.0% in males and 8.5%, 50.9%, and 40.6% in females; respectively (p < 0.001). The frequencies of rs2285666 TT, CT, and CC genotypes were 18.6%, 3.2%, and 78.3% in males and 4.2%, 37.1%, and 58.7% in females; respectively (p < 0.001).

Sex-stratified logistic regression revealed no significant association between rs2285666 and MS in either gender. For rs2074192, a modest protective effect appeared in females in the additive model (OR = 0.75, p = 0.034), although this did not remain significant after multiple-testing correction.

Relationship between ACE2 polymorphisms and demographics and clinical characteristics of participants

As shown in Table 6, a significant association was found between gender and the two gene polymorphisms: rs2074192 and rs2285666 (p < 0.001, Fig. 4A and B). Significant associations were revealed between rs2074192 genotype and EDSS level (p = 0.036, Fig. 4C), and between rs2285666 genotype and current treatment with DMT (p = 0.022, Fig. 4D). There was a significant association between the two gene polymorphisms (p < 0.001, Fig. 4E).

Table 6.

Relationship between ACE2 polymorphisms and demographics and clinical characteristics of MS patients.

Characteristic	rs2074192		rs2285666
	Test	p-value	Test	p-value
Age, yearsa	2.254	0.324	0.149	0.928
Gender	114.107	< 0.001 b	51.907	< 0.001 c
BMIa	1.284	0.526	4.156	0.125
Comorbidity	5.251	0.074c	0.337	0.843c
Age at MS onset, yearsa	2.033	0.362	0.278	0.870
Age at diagnosisa	2.246	0.325	0.531	0.767
MS disease duration, monthsa	0.561	0.755	0.196	0.907
MS type	5.283	0.260c	2.355	0.666b
EDSS scorea	1.106	0.575	5.334	0.069
EDSS level	6.645	0.036 c	4.883	0.084c
Previous treatment	0.503	0.781c	1.359	0.516c
Previous relapse in the last 24 months	2.529	0.289c	0.729	0.696c
Number of relapses in the last 24 monthsa	1.214	0.545c	0.970	0.616c
Currently on DMT	4.063	0.132c	7.566	0.022 c
rs2285666	102.65	< 0.001 b	–	–

^aKruskal-Wallis Test.

^bFisher’s Exact Test.

^cChi-square test.

Significant values are in bold.

Fig. 4.

Relationship between ACE2 polymorphisms and demographics and clinical characteristics of MS patients.

In MS patients, no associations were found between log(ACE2) serum level and rs2074192 and rs2285666 genotypes (p = 0.186 and 0.687, respectively). Similarly, no such associations were observed in the control group (p = 0.347 and p = 0.081, respectively; Fig. 5A and B).

Fig. 5.

Relationship between log(ACE2) serum level and ACE2 gene polymorphisms: (A) rs2074192 and (B) rs2285666.

Discussion

RAS is classically known for its role in controlling blood pressure and maintaining homeostatic balance. However, in recent years there has been a notable increase in research on the role of the RAS system in neurodegenerative diseases. ACE2 is considered one of the key enzymes that activate the RAS system, plays a role as anti-inflammatory and neuroprotective molecule. Our study was designed to investigate the serum levels of ACE2 and its genetic variations in Jordanian patients with MS and compare them with healthy controls and relate them to disease and patient’s characteristics.

ACE2, the first recognized human homolog of ACE sharing 42% of amino acids in the catalytic domain⁴¹, was investigated in our study. Serum ACE2 levels were significantly higher in MS patients than in healthy controls. In contrary to our result, Kawajiri et al. revealed significantly lower CSF ACE2 levels in MS patients than healthy controls²⁰, a disparity that can be related to differing sample sizes (20 patients versus 17 controls) and the variation between serum and CSF properties. CSF/brain ACE2 reflects central neuroinflammatory processes⁴², while serum ACE2 reflects peripheral renin-angiotensin system activity. This distinction contextualizes why serum ACE2 may behave differently from central ACE2 observed in earlier studies. Another explanation is the differential expression of various ACE2 species in CSF versus blood in MS patients, full-length versus truncated forms, a finding that was reported previously in patients infected with COVID-19^43,44. COVID-19 patients with encephalopathy displayed high CSF levels of full-length and one type of the ACE2 cleaved fragments over the another, suggesting a role of ACE2 beyond the renin-angiotensin system⁴³.

As ACE2 is considered one of the key components of the protective axis of RAS, its upregulation in MS patients could reflect the effort of the body to slow the pathological process in the early stages of MS. High ACE2 levels in MS patients, 72.7% of whom are RRMS, suggests their counter-regulatory mechanism of RAS dysregulation in MS patients compared to healthy controls. As evidenced by Stone et al., increased expression of RAS protective elements was observed in the EAE mouse model and brain tissues of MS patients at a stage of the disease when the regression or repair mechanisms were active¹¹. Another explanation could be an indication of the effectiveness of the therapy taken by about 65% of the included MS patients. Nevertheless, there was a trend of lower serum ACE2 levels in patients taking DMT (p = 0.057).

Future studies to unravel the molecular mechanisms underlying the higher serum ACE2 levels in patients with MS are required.

To evaluate the diagnostic power of ACE2 we performed the ROC analysis. It is commonly known that an AUC value equal to one is a perfect value to classify 100% of subjects correctly whereas an AUC of 0.5 indicates that subjects were randomly classified. As suggested by Xia et al., biomarkers could be classified based on AUC value: AUC = 0.9-1.0: excellent biomarker; 0.8–0.9: good biomarker; 0.7–0.8: fair biomarker; 0.6–0.7: poor biomarker; and 0.5–0.6: fail biomarker⁴⁵. Accordingly, in our study ACE2 serum concentration is considered a fair biomarker to discriminate MS patients from healthy controls.

Gender-based differences in ACE2 levels were also observed where MS females exhibited higher serum levels than MS males, while the opposite result was observed among healthy control group. One previous study that measured the ACE2 expression in the human carotid plaques reported significantly higher levels of this protein in females than in males⁴⁶. On the other hand, previous studies on other diseases revealed increased ACE2 activity and expression in males as compared to females^47–50. The higher ACE2 levels in MS females as compared to MS males may add an explanation to the differential MS disease prognosis between the two sexes^51,52. Because the human ACE2 protein is encoded by an X-linked gene that escapes X-chromosome inactivation⁵³, males will be more affected than females by rare variants potentially making them more prone to certain phenotypes caused by these variants.

Understanding host genetics can help explain clinical reactions in MS patients and inform successful management strategies. In the current study rs2074192 and rs2285666 ACE2 gene polymorphisms were assessed. To our knowledge no studies were conducted on ACE2 polymorphic genotype variants among Jordanian patients and their association with MS disease. Our genotype analysis revealed that rs2074192 CC genotype was the most frequent among MS patients and healthy controls. Similar results were obtained with rs2285666 genetic polymorphism.

There were statistically significant differences between patients and control groups regarding the rs2074192 CC and TT genotypes. The T allele was more common in the patients’ group, while the C allele was more common in the controls group. There were also statistically significant differences between patients’ and controls’ groups regarding the rs2285666 CC and TT genotypes. The T allele was more common in the controls’ group, while the C allele was more common in the patients’ group. When analyzed by gender, rs2074192 CC and CT genotypes were the most frequent among male and female patients, respectively. rs2074192 CT and TT genotypes were the least frequent among male and female patients, respectively.

Regarding rs2285666 genetic polymorphism, CC genotype was the most common among male and female patients, while CT and TT genotypes were the least common among male and female patients, respectively.

The rs2074192 CC genotype was associated with moderate-severe EDSS level. A trend toward the same finding but did not reach statistical significance (p = 0.069) was found with the rs2285666 CC genotype. Most MS patients who were on treatment had the rs2285666 CC genotype. It was also revealed that about 80% of the patients (27/34) who had the rs2285666 TT genotype were taking DMT, which may indicate genotype-treatment associations that require further investigation through prospective longitudinal studies.

A significant association was revealed between rs2074192 and rs2285666, indicating linkage disequilibrium between the two SNPs.

The present study is not without limitations. Firstly, the small sample size in the protein-study and we could not measure serum ACE2 levels in all participants. Secondly, no longitudinal samples were collected, thus we could not evaluate the clinical value of ACE2 in predicting relapses in RRMS patients, disease progression, and/or response to DMT. Thirdly, only two of the ACE2 SNPs were analyzed, and this incomplete genotyping may not represent all the haplotypes in ACE2.

Conclusion

As a conclusion, serum ACE2 levels were significantly higher in MS patients compared to healthy controls, and it may represent a diagnostic biomarker to discriminate MS patients from healthy individuals. Our findings encourage future studies to evaluate if higher serum ACE2 levels can be associated with the development and progression of MS disease through performing prospective longitudinal studies while focusing on including a higher proportion of patients with progressive diseases. We also in the current study investigated for the first time the frequency of ACE2 genotypes and alleles in patients with MS. Our results confirm the rs2074192 and rs2285666 genotypes as strong risk factors for MS susceptibility in the Jordanian population.

Author contributions

All authors contributed to the study conception and design. **Maha S. Al-Keilani** : Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing-Original draft. **Heba M. Abdelrazeq: ** Data curation, Methodology, Formal analysis, Writing original draft., **Nagham N. Hendi: ** Data curation, Methodology, Formal analysis, Writing-review & editing, **Saied A. Jaradat** : Data curation, Writing-review & editing.

Funding

This study was funded by the deanship of research at Jordan University of Science and Technology (grant number: 20210030).

Data availability

The datasets analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethical approval

This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Institutional Review Board (IRB) at Jordan University of Science and Technology (JUST) (approval number 98/136/2020, date of approval December 6, 2020).

Consent to participate

A written informed consent was obtained from all individual participants included in the study.