Screening of specific metabolic biomarkers for brucellosis and construction of a differential diagnostic model based on serum metabolomics

doi:10.3969/j.issn.1674-8115.2026.07.012

Abstract

Abstract:

Objective ·To analyze the correlations between differential metabolic features and clinical indicators in patients with brucellosis and rheumatoid arthritis, and to identify biomarkers with potential for differential diagnosis. Methods ·Clinical data were collected from 70 patients with brucellosis and 75 patients with rheumatoid arthritis, and baseline characteristics and clinical laboratory indicators were compared between the two groups. Serum metabolomic profiles were obtained using nanoparticle-enhanced laser desorption/ionization mass spectrometry (NPELDI-MS) to identity differentially expressed metabolites between the groups. Furthermore, Spearman′s rank correlation analysis was performed to identify brucellosis-specific metabolites significantly correlated with clinical indicators in the brucellosis group, followed by pathway enrichment analysis. A neural network-based diagnostic model was constructed based on these metabolites. Results ·Compared with the rheumatoid arthritis group, the brucellosis group demonstrated significantly lower erythrocyte sedimentation rate, neutrophil count, and serum creatinine levels (all P<0.05). In contrast, red blood cell count, hemoglobin level, and liver function-related indicators (aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, and total protein) were all significantly higher in the brucellosis group (all P<0.05). Among the 273 metabolic features obtained by NPELDI-MS, 133 showed significant differences between the two groups. Correlation analysis of these 133 differential metabolic features with the significantly different clinical indicators identified nine brucellosis-specific metabolites: taurine, pyroglutamic acid, uric acid, L-aspartic acid, glutamine, phenylalanine, urocanic acid, dihydroxyacetone phosphate, and pyrophosphate. These metabolites showed significant negative correlations with liver function-related indicators (aspartate aminotransferase, alanine aminotransferase, and alkaline phosphatase) and the inflammatory indicator (neutrophil count) (all P<0.05). Pathway enrichment analysis demonstrated that these metabolites were significantly enriched in several key metabolic pathways, including histidine metabolism; alanine, aspartate, and glutamate metabolism; phenylalanine, tyrosine, and tryptophan biosynthesis; phenylalanine metabolism; and taurine and hypotaurine metabolism. The diagnostic model constructed based on the combination of the nine metabolites achieved an area under the receiver operating characteristic curve of 0.852 in the training set and 0.860 in the validation set. Conclusion ·The nine‑metabolite combined diagnostic model shows favorable diagnostic efficacy for differentiating brucellosis from rheumatoid arthritis, providing a potential auxiliary biomarker panel for clinical application.

Key words: brucellosis, rheumatoid arthritis, differential diagnosis, metabolomics

CLC Number:

R378.5

Zhou Jinping, He Xiaoyan, Song Wen, Liu Yumei, Huang Lin, Ma Xiumin. Screening of specific metabolic biomarkers for brucellosis and construction of a differential diagnostic model based on serum metabolomics[J]. Journal of Shanghai Jiao Tong University (Medical Science), 2026, 46(7): 938-945.

Figures/Tables 6

TrendMD

References 25

[1]	Jin M, Fan Z X, Gao R F, et al. Research progress on complications of brucellosis [J]. Front Cell Infect Microbiol, 2023, 13: 1136674.
[2]	Laine C G, Johnson V E, Scott H M, et al. Global estimate of human brucellosis incidence [J]. Emerg Infect Dis, 2023, 29(9): 1789-1797.
[3]	Zheng R J, Xie S S, Lu X B, et al. A systematic review and meta-analysis of epidemiology and clinical manifestations of human brucellosis in China [J]. Biomed Res Int, 2018, 2018: 5712920.
[4]	Shi Q N, Qin H J, Lu Q S, et al. Incidence and warning signs for complications of human brucellosis: a multi-center observational study from China [J]. Infect Dis Poverty, 2024, 13(1): 18.
[5]	Zhang S W, Hu J, An S Q, et al. Prevalence and relevant factors of positive RF in brucellosis patients with arthralgia [J]. PLoS Negl Trop Dis, 2021, 15(9): e0009749.
[6]	Wu Q, Sun C, Guo L P, et al. Preparation and evaluation of Brucella T4SS recombinant proteins in serodiagnosis of human brucellosis based on TMT-based proteomics technology [J]. Front Cell Infect Microbiol, 2024, 14: 1514046.
[7]	冯强生, 宋月娟, 哈小琴, 等. 187例布鲁菌病患者的临床特征和感染指标分析 [J]. 中华实验和临床感染病杂志(电子版), 2020, 14(1): 46-50.
	Feng Q S, Song Y J, Ha X Q, et al. Clinical characteristics and infection biomarkers of 187 patients with brucellosis [J]. Chinese Journal of Experimental and Clinical Infectious Diseases (Electronic Edition), 2020, 14(1): 46-50.
[8]	Qiu S, Cai Y, Yao H, et al. Small molecule metabolites: discovery of biomarkers and therapeutic targets [J]. Sig Transduct Target Ther, 2023, 8: 132.
[9]	Cai X L, Jin J Y, Ye H, et al. Altered serum metabolome is associated with disease activity and immune responses in rheumatoid arthritis [J]. Clin Rheumatol, 2024, 43(12): 3669-3678.
[10]	Thomas T, Stefanoni D, Reisz J A, et al. COVID-19 infection alters kynurenine and fatty acid metabolism, correlating with IL-6 levels and renal status [J]. JCI Insight, 2020, 5(14): e140327.
[11]	中华人民共和国国家卫生健康委员会, 国家中医药管理局. 布鲁氏菌病诊疗方案(2023年版) [J]. 中国感染控制杂志, 2024, 23(5): 661-664.
	National Health Commission of the People′s Republic of China, National Administration of Traditional Chinese Medicine. Diagnosis and treatment scheme for Brucellosis (2023 Edition) [J]. Chinese Journal of Infection Control, 2024, 23(5): 661-664.
[12]	Aletaha D, Neogi T, Silman A J, et al. 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative[J]. Arthritis Rheum, 2010, 62(9): 2569-2581.
[13]	Luo B, Wang Q, Yang S T, et al. Epidemiological, clinical, and laboratory characteristics of 581 human brucellosis cases in Xinjiang, China [J]. Front Microbiol, 2025, 16: 1541277.
[14]	Sellami M, Saidane O, Mahmoud I, et al. Etiological features of liver involvement in rheumatoid arthritis [J]. Curr Rheumatol Rev, 2020, 16(4): 332-336.
[15]	Fu L, Zhang H, Dai Y Y, et al. Revealing metabolic alterations in brucellosis patients by targeted metabolomics [J]. J Pharm Biomed Anal, 2024, 249: 116370.
[16]	Guzelcicek A, Bayraktar N, Bayraktar M. Amino acids profile in children with acute brucellosis [J]. Curr Pediatr Rev, 2024, 20(2): 188-193.
[17]	Arocena G M, Zorreguieta A, Sieira R. Expression of VjbR under nutrient limitation conditions is regulated at the post-transcriptional level by specific acidic pH values and urocanic acid [J]. PLoS One, 2012, 7(4): e35394.
[18]	Luzader D H, Kendall M M. Commensal 'trail of bread crumbs' provide pathogens with a map to the intestinal landscape [J]. Curr Opin Microbiol, 2016, 29: 68-73.
[19]	Aydın H, Tekin Y K, Korkmaz İ, et al. Glutamine-driven metabolic adaptation to COVID-19 infection [J]. Indian J Clin Biochem, 2023, 38(1): 83-93.
[20]	Gamarra Y, Santiago F C, Molina-López J, et al. Pyroglutamic acidosis by glutathione regeneration blockage in critical patients with septic shock [J]. Crit Care, 2019, 23(1): 162.
[21]	van Eijk L E, Offringa A K, Bernal M E, et al. The disease-modifying role of taurine and its therapeutic potential in coronavirus disease 2019 (COVID-19) [J]. Adv Exp Med Biol, 2022, 1370: 3-21.
[22]	Zhang Y T, Guo S Y, Fu X D, et al. Emerging insights into the role of NLRP3 inflammasome and endoplasmic reticulum stress in renal diseases [J]. Int Immunopharmacol, 2024, 136: 112342.
[23]	McCormick N, Rai S, Yokose C, et al. op0306 anaerobicidal antibiotic use and risk of gout flares: target trial emulation for potential role of the microbiome in gout with chronic kidney disease (ckd) [J]. Ann. Rheum Dis, 2025, 84: 244-245.
[24]	Schenker R B, Ramirez C B, Jang C, et al. Dihydroxyacetone phosphate is a novel predictor of hepatic fibrosis in Latino adolescents with obesity [J]. J Pediatr Gastroenterol Nutr, 2025, 80(1): 174-181.
[25]	Rattazzi M, Bertacco E, Iop L, et al. Extracellular pyrophosphate is reduced in aortic interstitial valve cells acquiring a calcifying profile: implications for aortic valve calcification [J]. Atherosclerosis, 2014, 237(2): 568-576.

Variable	Brucellosis (n=70)	Rheumatoid arthritis (n=75)	P value
ESR/(mm·h^-1)	20.500 (11.750, 33.000)	34.000 (17.000, 54.000)	0.002
CRP/(mg·L^-1)	11.500 (3.400, 28.000)	10.270 (2.690, 25.640)	0.789
WBC/(×10⁹·L^-1)	5.640 (4.325, 6.913)	6.200 (5.240, 7.920)	0.060
NEUT/(×10⁹·L^-1)	2.985 (2.168, 4.080)	3.740 (2.700, 5.240)	0.006
LYM/(×10⁹·L^-1)	1.925 (1.338, 2.508)	1.740 (1.180, 2.290)	0.227
RBC/(×10¹²·L^-1)	4.682±0.059	4.294±0.056	<0.001
Hb/(g·L^-1)	130.600±1.760	125.133±2.000	0.043
PLT/(×10⁹·L^-1)	256.000 (213.000, 303.500)	261.000 (223.000, 318.000)	0.420
AST/(U·L^-1)	21.950 (17.700, 32.600)	19.100 (15.300, 23.200)	0.004
ALT/(U·L^-1)	26.500 (17.630, 47.450)	16.800 (12.300, 24.400)	<0.001
ALP/(U·L^-1)	109.350 (89.550, 140.950)	78.200 (61.400, 101.100)	<0.001
TP/(g·L^-1)	73.201±0.930	67.943±0.632	<0.001
SCr/(μmol·L^-1)	54.300 (46.975, 62.100)	62.270 (55.540, 71.780)	<0.001
BUN/(mmol·L^-1)	5.090±0.199	5.306±0.153	0.389

Variable	Brucellosis (n=70)	Rheumatoid arthritis (n=75)	P value
ESR/(mm·h^-1)	20.500 (11.750, 33.000)	34.000 (17.000, 54.000)	0.002
CRP/(mg·L^-1)	11.500 (3.400, 28.000)	10.270 (2.690, 25.640)	0.789
WBC/(×10⁹·L^-1)	5.640 (4.325, 6.913)	6.200 (5.240, 7.920)	0.060
NEUT/(×10⁹·L^-1)	2.985 (2.168, 4.080)	3.740 (2.700, 5.240)	0.006
LYM/(×10⁹·L^-1)	1.925 (1.338, 2.508)	1.740 (1.180, 2.290)	0.227
RBC/(×10¹²·L^-1)	4.682±0.059	4.294±0.056	<0.001
Hb/(g·L^-1)	130.600±1.760	125.133±2.000	0.043
PLT/(×10⁹·L^-1)	256.000 (213.000, 303.500)	261.000 (223.000, 318.000)	0.420
AST/(U·L^-1)	21.950 (17.700, 32.600)	19.100 (15.300, 23.200)	0.004
ALT/(U·L^-1)	26.500 (17.630, 47.450)	16.800 (12.300, 24.400)	<0.001
ALP/(U·L^-1)	109.350 (89.550, 140.950)	78.200 (61.400, 101.100)	<0.001
TP/(g·L^-1)	73.201±0.930	67.943±0.632	<0.001
SCr/(μmol·L^-1)	54.300 (46.975, 62.100)	62.270 (55.540, 71.780)	<0.001
BUN/(mmol·L^-1)	5.090±0.199	5.306±0.153	0.389

No.	Variable	Metabolite	Brucellosis		Rheumatoid arthritis
No.	Variable	Metabolite	r	P value (FDR-adjusted)	r	P value (FDR-adjusted)
1	NEUT	PPi	-0.346	0.021	-0.213	0.163
2	NEUT	Gln	-0.380	0.010	-0.078	0.664
3	NEUT	UA	-0.467	0.001	-0.191	0.221
4	AST	L-Asp	-0.374	0.012	-0.065	0.716
5	AST	Tau	-0.394	0.007	0.193	0.218
6	AST	pGlu	-0.403	0.006	-0.106	0.526
7	ALT	pGlu	-0.321	0.033	-0.014	0.943
8	ALT	DHAP	-0.322	0.033	0.043	0.822
9	ALT	UA	-0.332	0.027	-0.086	0.624
10	ALT	Tau	-0.425	0.003	0.130	0.431
11	ALP	UCA	-0.343	0.023	-0.134	0.414
12	ALP	Phe	-0.360	0.016	-0.044	0.813
13	ALP	pGlu	-0.362	0.015	-0.112	0.504
14	ALP	Tau	-0.384	0.009	-0.004	0.982

No.	Variable	Metabolite	Brucellosis		Rheumatoid arthritis
No.	Variable	Metabolite	r	P value (FDR-adjusted)	r	P value (FDR-adjusted)
1	NEUT	PPi	-0.346	0.021	-0.213	0.163
2	NEUT	Gln	-0.380	0.010	-0.078	0.664
3	NEUT	UA	-0.467	0.001	-0.191	0.221
4	AST	L-Asp	-0.374	0.012	-0.065	0.716
5	AST	Tau	-0.394	0.007	0.193	0.218
6	AST	pGlu	-0.403	0.006	-0.106	0.526
7	ALT	pGlu	-0.321	0.033	-0.014	0.943
8	ALT	DHAP	-0.322	0.033	0.043	0.822
9	ALT	UA	-0.332	0.027	-0.086	0.624
10	ALT	Tau	-0.425	0.003	0.130	0.431
11	ALP	UCA	-0.343	0.023	-0.134	0.414
12	ALP	Phe	-0.360	0.016	-0.044	0.813
13	ALP	pGlu	-0.362	0.015	-0.112	0.504
14	ALP	Tau	-0.384	0.009	-0.004	0.982