Proteomics profile of serum and liver samples in women with morbid obesity and metabolic dysfunction-associated steatohepatitis

doi:10.21203/rs.3.rs-4996622/v1

Download PDF

Research Article

Proteomics profile of serum and liver samples in women with morbid obesity and metabolic dysfunction-associated steatohepatitis

https://doi.org/10.21203/rs.3.rs-4996622/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Non-invasive methods are necessary for the diagnostic and follow-up of metabolic dysfunction-associated steatohepatitis (MASH). This study aims to perform a proteomic analysis on serum and liver samples from morbid obese (MO) women to identify key mediators of MASH. HPLC-MS/MS proteomics was conducted on serum and liver samples from a cohort of 174 MO women classified by liver histology: 44 normal liver (NL), 66 simple steatosis (SS) and 64 MASH. Serum proteomics identified 257 proteins. The MASH individuals had 13 altered proteins, 11 upregulated and 2 downregulated. Altered proteins are primarily involved in molecular pathways of the initial triggering and complement cascade (50%). Liver proteomics identified 2081 proteins, with 72 upregulated and 84 downregulated in MASH. These proteins are mostly involved in molecular pathways of amino acid metabolism (31,25%), antimicrobial peptides (20%), fatty acid metabolism (17,5%). We identified 13 altered proteins in serum of MASH: increased levels of fructose-bisphosphate aldolase, clusterin, collectin-10 and − 11, scavenger receptor cysteine-rich-M130, attractin, pigment epithelium-derived factor, vitronectin, complement factor-H, thrombospondin-4 and apolipoprotein-AIV and decreased levels of sex hormone-binding globulin and adiponectin. These proteins can be part of a panel of biomolecules for the diagnosis or follow-up of MASH.

Applied Biochemistry

proteomics

metabolic dysfunction-associated steatohepatitis

obesity

biomarker

Metabolic dysfunction-associated steatotic liver disease (MASLD) was recently termed to define the presence of hepatic steatosis in conjunction with one metabolic risk factor without alcohol consumption [1]. MASLD encompasses a range of liver conditions spanning from simple steatosis (SS) to the more severe form known as metabolic dysfunction-associated steatohepatitis (MASH), characterized by hepatocyte ballooning and lobular inflammation with or without liver fibrosis [2]. Approximately 20% of MASLD patients have MASH, which can progress to cirrhosis or hepatocellular carcinoma (HCC) [3]. Furthermore, MASH is associated with an increased risk of cardiovascular disease, the primary cause of morbidity and mortality in patients with MASLD [4]. Although liver biopsy remains the gold standard for diagnosing MASH, its invasive nature and qualitative interpretation pose significant limitations [5]. Therefore, there is a clear need for developing non-invasive screening approaches to accurately diagnose and follow-up MASH patients [6].

Proteomics is an omics technology that identifies and quantifies peptides and proteins in a biological sample, defining a disease phenotype. The study of proteomes plays an essential role in various research areas, including biomarker discovery, drug development and understanding disease mechanisms [7, 8]. To date, several proteomic analyses in human subjects have been previously published regarding MASLD and MASH [9]. Some studied identified proteins significantly altered in serum and liver samples of MASLD subjects compared to metabolically healthy controls [10, 11]. Other authors suggested a model to identify key histological phenotypes of MASH such as steatosis, fibrosis, inflammation and ballooning using a set of proteins in serum samples [12]. Moreover, another study defined groups of proteins related to MASH and HCC in serum samples [13]. However, no panel has yet been established for the diagnosis of MASH, so further proteomic studies on serum and liver samples in a considerable and homogeneous cohort are necessary.

In this context, we previously performed metabolomic and lipidomic analyses in serum samples of a cohort of women with morbid obesity (MO) and different hepatic histological diagnoses and found some lipid metabolites associated to MASH [14, 15]. In this study, to take a step forward in this research, we now aim to perform a proteomic analysis using serum and liver samples of a homogeneous cohort of women with MO to identify key protein mediators of MASH with the intention of developing a multi-omics panel for non-invasive MASH diagnosis and follow-up.

Subjects

In this work, we recruited a homogeneous cohort of 174 women with MO (body mass index, BMI > 40kg/m2) from the bariatric surgery unit. Given that men and women differ in terms of metabolic parameters, we recruited only women for this study [16, 17].

The sample size was calculated accepting an alpha risk of 0.05 and a beta risk of less than 0.2 in a bilateral contrast. We needed a minimum of 68 cases (MASLD) and 34 controls (normal liver, NL) to detect at least an odds ratio of 0.15. It is assumed that the exposure rate in the control group would be 0.333 (POISSON approximation, GRANMO sample size calculator v.7.04).

As exclusion criteria, we avoided patients who had an acute illness, acute or chronic inflammatory or infective diseases, or end-stage malignant disease; menopausal women and women receiving contraceptive treatment; women with an alcohol intake exceeding 20g per day and recurrent smokers.

The IISPV Ethics Committee approved this study (CEIm; 23c/2015). Each participant provided informed consent.

Anthropometric parameters and biochemical analyses

From each participant, we obtained anthropometric variables such as weight, height, BMI and waist-hip ratio. Additionally, specialized nurses collected blood samples in ethylenediaminetetraacetic acid tubes using a BD Vacutainer® system, on an empty stomach, and just before bariatric surgery. Blood samples were processed into plasma and serum aliquots. Serum was collected in tubes without anticoagulant and stored at − 80°C. Plasma was separated from EDTA tubes by centrifugation at 3500 rpm for 15 minutes at 4°C.

Biochemical variables measured using a conventional automated analyzer included levels of glucose, insulin, glycosylated hemoglobin A1c (HbA1c), total cholesterol, high density lipoprotein-cholesterol (HDL-C), low density lipoprotein-cholesterol (LDL-C), triglycerides, aspartate aminotransferase (AST), alanine aminotransferase (ALT), gamma glutamyl transferase (GGT), alkaline phosphatase (ALP), lactate dehydrogenase (LDH) and ferritin.

Histopathological analysis of the liver biopsies

During bariatric surgery, we collected liver samples in a tube with RNAlater solution (Qiagen, Hilden, Germany) at 4ºC and then preserved in a formaldehyde solution for histopathological diagnosis. Liver biopsies were obtained whenever liver alteration was suspected. Liver samples were scored and classified by an experienced hepatopathologist through eosin-hematoxylin staining according to Kleiner criteria [18] into NL (n = 44), SS (n = 66) and MASH (n = 64). Due to the availability, a subset of patients had a fragment of the liver biopsy frozen at -96ºC for further proteomic processing: NL (n = 37), SS (n = 40) and MASH (n = 64). All cases of NASH corresponded to grade I-II with mild-moderate inflammation without the existence of hepatic fibrosis.

Serum proteomic analysis

Protein extraction and quantification for serum samples:

Before proteomic analysis, the most abundant plasma proteins (albumin, immunoglobulin (Ig)G, antitrypsin, IgA, transferrin, haptoglobin, fibrinogen, alpha2-macroglobulin, alpha1-acid glycoprotein, IgM, apolipoprotein AI, apolipoprotein AII, complement C3 and transthyretin) were depleted to increase the number of identified/quantified proteins. Ten microliters of each sample were passed twice through the Human-14 Multiple Affinity Removal Spin cartridge (Agilent Technologies, Catalog Number 5188–6560) following the manufacturer’s protocol. Flow-through fractions were concentrated and buffer exchanged to about 100 µl of 6M urea in 50 mM ammonium bicarbonate (ABC) using 5K MWCO spin columns (Agilent 5185–5991).

Protein digestion and peptide 11-plex TMT labelling:

Twenty-five micrograms of total protein (quantified by Bradford’s method) were reduced with 4 mM 1,4-dithiothreitol (DTT) for 1 hour at 37°C and alkylated with 8 mM iodoacetamide (IAA) for 30 minutes at 25°C in the dark. Samples were then digested overnight (pH 8.0, 37°C) with sequencing-grade Trypsin/Lys-C Protease Mix (ThermoFisher Scientific, CA, USA) at an enzyme ratio of 1:50. Digestion was quenched by acidification with 1% (v/v) formic acid, and peptides were desalted on Oasis HLB SPE columns (Waters, Massachusetts, USA) before TMT 11-plex labelling (ThermoFisher Scientific, CA, USA) following the manufacturer’s instructions. Briefly, samples were resuspended with triethyl ammonium bicarbonate (TEAB) 100 mM, then each TMT channel, previously prepared with CAN, was added to each sample.

To normalize all samples across different TMT-multiplexed batches, a pool containing all samples was labelled with the TMT-126 tag and included in each TMT batch. Then, each plex of samples was mixed and the different TMT 11-plex batches were desalted on Oasis HLB SPE columns before nanoLC-MS analysis.

NanoLC-(orbitrap)MS/MS analysis:

Labelled and multiplexed peptides were loaded on a trap nano-column (75 µm I.D.; 1.5 cm length; 3 µm particle diameter, ThermoFisher Scientific, CA, USA) and separated onto a C-18 reversed phase (RP) µPAC™ Neo HPLC column (180 µm bed; 50 cm length; 2.5 x 16 µm pillar diameter, ThermoFisher Scientific, CA, USA) on a Vanquish Neo VN-S10 System (ThermoFisher Scientific). Chromatographic separation was performed with a 180-minute gradient using water LC-MS grade (0.1% formic acid) and acetonitrile (0.1% formic acid) as mobile phases at a flow rate of 300 nL/min.

Mass spectrometry (MS) analyses were performed on an Orbitrap Eclipse from ThermoFisher Scientific by an enhanced FT-resolution MS spectrum (R = 60,000 FHMW) followed by data-dependent FT-MS/MS acquisition (R = 50,000 FHMW, 30% NCE HCD) from the ten most intense parent ions with a charge state acquisition from two to six and dynamic exclusion of 0.7 min.

Protein identification and quantification:

Protein identification and quantification were performed using Proteome Discoverer software v.2.5 (ThermoFisher Scientific, CA, USA) with the Mascot search engine (v2.8, Matrix Science) combining raw data files obtained from each plex. For protein identification, the workflow used the Mascot node combining Homo sapiens and contaminants databases, assuming trypsin digestion. The fragment ion mass tolerance assumed an error of 20 mmu for FT-MS/MS fragmentation mass and 10 ppm for FT-MS precursor ion mass. Oxidation of methionine and acetylation of the N-terminal were set as dynamic modifications, carbamidomethylation as a static modification, and TMT-11plex as a quantitation method. The false discovery rate (FDR) and protein probabilities were calculated by Percolator, with peptide identification set to a maximum of 1%. Peptide quantitation data were retrieved from the ‘Reporter ions quantifier’ node in Proteome Discoverer, using the area of unique and razor peptides and total peptide amount as normalization. Peptide and protein results are expressed in abundance area and are dimensionless.

Liver proteomic analysis

Protein extraction and quantification for liver tissue samples:

Liver tissue samples were homogenized by adding RIPA buffer (Fisher Scientific) and lysate using a BlueBender with 2 freeze-thaw cycles at -80ºC. Then samples were rotated 1h and 30 min at 4ºC. Protein extractions were centrifuged at 16000g for 20 min and the supernatants were collected for protein precipitation with the addition of 10% TCA/acetone. Finally, protein pellets were resuspended in 6M urea/50 mM ammonium bicarbonate (ABC) for digestion.

Protein digestion and peptide 11-plex TMT labelling:

We followed the same methodology described above for serum samples.

High pH reverse phase fractionation and nanoLC-(orbitrap)MS/MS analysis:

Labelled and multiplexed peptides were fractionated by High pH Reversed-Phase Peptide Fractionation Kit (ThermoFisher Scientific, CA, USA) according to manufacturer’s protocol to increase protein identification through orthogonal peptide fractionation of complex peptide samples combined with low pH nanoLC-MS/MS analysis. Briefly, High pH Reversed-phase Peptide Fractionation Kit uses a sequential elution with eight different ratios of 0,1% triethylamine (a high pH buffer) and acetonitrile. Then, the eluted fractions were evaporated to dryness on a SpeedVac and resuspended with water 0.1% formic acid for direct nanoLC-MS/MS analysis.

Fractions of each TMT plex were loaded on a trap nano-column (75 µm I.D.; 1.5 cm length; 3 µm particle diameter, ThermoFisher Scientific, CA, USA) and separated onto a C-18 reversed phase (RP) µPAC™ Neo HPLC column (180 µm Bed; 50 cm length; 2.5 x16 µm pillar diameter, ThermoFisher Scientific, CA, USA) on an Vanquish Neo VN-S10 System (ThermoFisher Sicientific). The chromatographic separation was performed with a 180 min gradient using Milli-Q water (0.1% formic acid) and acetonitrile (0.1% formic acid) as mobile phase at a flow rate of 300 nL/min.

Mass spectrometry analyses were performed following the same methodology described above for serum samples.

Protein identification and quantification:

We followed the same methodology described above for serum samples.

Statistical analysis

The anthropometric and biochemical variables parameters of the patients were analyzed using the SPSS/PC + for Windows statistical package (v27.0; SPSS, Chicago, Illinois, USA). The Kolmogorov-Smirnov test was employed to assess the distribution of variables. Variables were presented as the median and interquartile range because they exhibited a non-normal distribution. Comparative analyses were conducted using a nonparametric Mann-Whitney U test. P-values < 0.05 were considered statistically significant.

Regarding the serum proteomics data, only those proteins identified in > 60% of the samples in at least one of the experimental groups were considered. For these proteins, missing value estimation after data filtering was performed using the KNN algorithm (k-nearest neighbors) from Metaboanalyst v5.0 to carry out univariate statistical analysis. Regarding the liver tissue proteomics data, only those proteins identified in 100% of the samples were considered. For statistics analyses of the proteomics data, a log base 2 data transformation was applied. One-way ANOVA with Tukey HSD Post hoc and Benjamini Hochberg FDR correction was performed between groups, with an adjusted p-value cut-off of < 0.05 applied in each case.

Cytoscape software v3.10.2, with the ClueGO plugin, was used to perform gene enrichment analysis and pathway clustering. STRING was used as the database to obtain information on the included proteins, focusing exclusively on Homo sapiens, and for conducting the gene enrichment analysis. ClueGO served as the plugin to perform functional analysis of the proteins and their interactions. We selected the Reactome database to identify pathways developed by the included proteins, setting a medium network specificity and displaying only pathways with a p-value less than 0.05, with Benjamini-Hochberg p-value correction applied. The remaining parameters were kept at their default settings, including GO term grouping and preferred layout.

Subjects

We first subclassified our subjects according to their hepatic histology into NL (n = 44) and MASLD (n = 130), as shown in Table 1. In this case, subjects were comparable in terms of sex, age, BMI, waist-hip ratio, cholesterol profile and levels of ALP and LDH. Subjects with MASLD presented higher levels of glucose, insulin, HbA1c, triglycerides, hepatic transaminases and ferritin than NL subjects. Then, we subclassified our MASLD cohort into SS (n = 66) and MASH (n = 64), as also shown in Table 1. In this regard, subjects with MASH also presented increased levels of LDH compared to the NL group. Unfortunately, we did not report significant differences between SS and MASH subjects.

Table 1

Anthropometric and basic biochemical parameters of the study cohort with morbid obesity classified into normal liver and metabolic-dysfunction associated steatotic liver disease, and this last one group into simple steatosis and metabolic dysfunction-associated steatohepatitis.
Variables	NL (n = 44)	MASLD (n = 130)	MASLD
Variables	NL (n = 44)	MASLD (n = 130)	Simple Steatosis (n = 66)	MASH (n = 64)
Sex (%M/%W)	0/100	0/100	0/100	0/100
Age (years)	41 (34–49)	47.7 (37–50)	43 (37–50)	44 (35–49)
BMI (kg/m²)	43.9 (41.2–49.3)	45.1 (43.2–49.9)	45.8 (43–51.2)	46.1 (43.3–50.1)
Waist-hip (m) ratio	0.88 (0.84–0.95)	0.92 (0.87–0.97)	0.91 (0.87–0.97)	0.92 (0.87–0.98)
Glucose (mg/dl)	89 (79.5–100.7)	116 (91.7–146.2)*	109 (93–133)*	124.5 (90.2–157.7)*
Insulin (mUI/L)	9.2 (5.7–12.8)	17.9 (10.8–31.2)*	19 (11.1–33)*	16.5 (10.8–25.7)*
HbA1c (%)	5.3 (5–5.7)	6.1 (5.4–7.8)*	6.1 (5.4–7)*	6.1 (5.5–8.2)*
Triglycerides (mg/dl)	103 (76.5–135)	144.5 (112.7–201.7)*	151 (116–197)*	137.5 (110.2–200.5)*
Cholesterol (mg/dl)	164 (139.5–199.5)	166.8 (148–189.6)	164.8 (144–192)	166.9 (151–188.4)
HDL-C (mg/dl)	40 (32.6–50.2)	37 (33–45.2)	36.7 (32–46)	37.8 (33–43.9)
LDL-C (mg/dl)	102.1 (76.6–126.9)	95.3 (80.6–114)	90.9 (76.2–114)	100 (84.6–116)
AST (UI/L)	22.5 (19–41.7)	36 (24–52)*	35 (24.5–52)*	36 (23.7–52.7)*
ALT (UI/L)	23.5 (16–44.5)	35.5 (26.2–57.5)*	35 (29–49)*	35 (26–65)*
GGT (UI/L)	17 (12–25.5)	26 (17–48.2)*	26 (18–45)*	26 (15–70)*
ALP (UI/L)	65 (53–76)	67.5 (57–79)	67.5 (52.7–76)	69.5 (61–80.7)
LDH (UI/L)	382.5 (339.2–423.2)	396 (341–466)	399 (351–433.5)	404.5 (345.5–547.7)*
Ferritin (ng/ml)	37 (22–80.4)	53 (27.9–137.5)*	75.1 (33.4–185.4)*	46 (26.9–110.7)*
Data are expressed as the median and interquartile range. (^#) Significant differences compared to the NL group were considered when p-value < 0.05 using the Mann-Whitney test. NL, normal liver; MASLD, Metabolic Dysfunction-Associated Steatotic Liver Disease; MASH, Metabolic Dysfunction-Associated Steatohepatitis; M, men; W; women; BMI, body mass index; HbA1c, glycosylated hemoglobin A1c; HDL-C, high density lipoprotein – cholesterol; LDL-C, low density lipoprotein – cholesterol; AST, aspartate-amino transferase; ALT, alanine-aminotransferase; GGT, gamma-glutamyl transferase; ALP, alkaline phosphatase; LDH, lactate dehydrogenase.

Proteomics analysis in serum samples

After proteomics analysis in serum samples (n = 240), a total of 537 proteins were identified. Information regarding relative quantification and identification, such as Mascot score, protein FDR confidence and other details, are in Supplementary Table 1. Samples were filtered to include proteins present in > 60% of the samples in at least one experimental group, and KNN estimation was applied, resulting in 257 confident proteins.

We first examined which proteins differed in concentration and pattern between MO subjects with NL and MO subjects with MASLD. Women with MALSD presented significantly increased levels of 9 proteins and significantly decreased levels of 4 proteins compared to NL women (Table 2).

Table 2

Serum proteins significantly upregulated or downregulated in metabolic-dysfunction associated steatotic liver disease (MASLD) compared to normal liver.
UPREGULATED in MASLD vs NL
Coding Name	Log2FC	Adj p-val
Fructose-bisphosphate aldolase (ALDOB)	2,05	< 0,001
Clusterin (CLU)	1,73	0,017
Collectin-10 (COLEC10)	1,62	0,015
Pigment epithelium-derived factor (SERPINF1)	1,50	0,009
Serotransferrin (TF)	1,45	0,040
Collectin-11 (COLEC11)	1,42	0,026
Vitronectin (VTN)	1,39	0,017
Scavenger receptor cysteine-rich type 1 protein M130 (CD163)	1,09	0,027
Vitamin K-dependent protein S (PROS1)	0,90	0,027
DOWNREGULATED in MASLD vs NL
Coding Name	Log2FC	Adj p-val
Complement factor D (CFD)	-0,93	0,027
Gelsolin (GSN)	-1,30	0,026
Sex hormone-binding globulin (SHBG)	-1,64	< 0,001
Adiponectin (ADIPOQ)	-1,67	0,002
Upregulated and downregulated proteins were identified by positive or negative log2 fold changes (log2FC), respectively. Statistical differences were considered significant when the adjusted p-value (Adj P-val) was less than 0.05. One-way ANOVA with Tukey HSD post hoc and Benjamini-Hochberg FDR correction was performed between groups. Detailed information on proteins and further statistical data are provided in Supplementary Table 2.

Next, we evaluated specifically the serum proteome between women with NL and those women with SS. We reported 5 significantly increased proteins and 5 significantly decreased proteins in SS subjects compared to NL participants (Table 3).

Table 3

Serum proteins significantly upregulated or downregulated in simple steatosis compared to normal liver.
UPREGULATED in Simple Steatosis vs NL
Coding Name	Log2FC	Adj p-val
Fructose-bisphosphate aldolase (ALDOB)	2,11	0,006
Pigment epithelium-derived factor (SERPINF1)	1,63	0,017
Serotransferrin (TF)	1,61	0,046
Vitronectin (VTN)	1,53	0,037
Vitamin K-dependent protein S (PROS1)	1,13	0,011
DOWNREGULATED in Simple Steatosis vs NL
Coding Name	Log2FC	Adj p-val
Complement factor D (CFD)	-0,93	0,031
Complement factor C6 (C6)	-1,09	0,040
Sex hormone-binding globulin (SHBG)	-1,78	< 0,001
Adiponectin (ADIPOQ)	-1,84	0,002
Gelsolin (GSN)	-1,86	0,030
Upregulated and downregulated proteins were identified by positive or negative log2 fold changes (log2FC), respectively. Statistical differences were considered significant when the adjusted p-value (Adj P-val) was less than 0.05. One-way ANOVA with Tukey HSD post hoc and Benjamini-Hochberg FDR correction was performed between groups. Detailed information on proteins and further statistical data are provided in Supplementary Table 3.

Then, we compared the differential altered proteins between women with NL and MASH. We found 11 significantly increased proteins and only 2 decreased proteins in the MASH group compared to the NL group (Table 4).

Table 4

Serum proteins significantly upregulated or downregulated in metabolic-dysfunction associated steatohepatitis compared to normal liver.
UPREGULATED in MASH vs NL
Coding Name	Log2FC	Adj p-val
Fructose-bisphosphate aldolase (ALDOB)	1,99	0,003
Clusterin (CLU)	1,93	0,003
Collectin-10 (COLEC10)	1,80	0,003
Collectin-11 (COLEC11)	1,73	0,019
Scavenger receptor cysteine-rich type 1 protein M130 (CD163)	1,54	0,019
Attractin (ATRN)	1,44	0,044
Pigment epithelium-derived factor (SERPINF1)	1,35	0,019
Vitronectin (VTN)	1,23	0,044
Complement factor H (CFH)	1,17	0,035
Thrombospondin-4 (THBS4)	0,94	0,019
Apolipoprotein A-IV (APOA4)	0,73	0,026
DOWNREGULATED in MASH vs NL
Coding Name	Log2FC	Adj p-val
Sex hormone-binding globulin (SHBG)	-1,50	0,003
Adiponectin (ADIPOQ)	-1,52	0,019
Upregulated and downregulated proteins were identified by positive or negative log2 fold changes (log2FC), respectively. Statistical differences were considered significant when the adjusted p-value (Adj P-val) was less than 0.05. One-way ANOVA with Tukey HSD post hoc and Benjamini-Hochberg FDR correction was performed between groups. Detailed information on proteins and further statistical data are provided in Supplementary Table 4.

Finally, we compared the circulating proteome between women with SS and women with MASH, but we did not report significantly differently altered proteins between these groups (data not shown).

Serum Proteomic Profile Comparison Between Groups

A hierarchical clustering analysis of the top 25 proteins revealed distinct profiles of upregulation and downregulation across each group (Fig. 1). In this analysis, it is difficult to define a consistent pattern of protein characteristics for each group. However, we can identify a subset of proteins that appear to be more downregulated in NL and simultaneously overexpressed in MASH subjects, while these two groups do not show relevant differences compared to the SS group. Additionally, this analysis reveals some proteins, such as gelsolin (GSN), adiponectin (ADIPOQ), complement factor D (CFD), glutathione peroxidase 3 (GPX3), sex hormone-binding globulin (SHBG), antithrombin-III (SERPINC1) and complement factor C6 (C6), that are overexpressed in NL and seem to be downregulated in the SS group, but do not present a clear pattern in the MASH group.

Proteomics analysis in liver samples

Subsequently, we performed a proteomics analysis on liver samples from 141 subjects, identifying a total of 4671 proteins. Relative quantification and identification details, including Mascot score and protein FDR confidence are provided in Supplementary Table 5. After filtering proteins present in 100% of the samples, 2081 proteins were confidently identified.

First, we aimed to determine which altered proteins differ in hepatic concentration between NL and MASLD subjects. However, no significant differences were found in this comparison (data not shown). Next, we evaluated the hepatic proteome between women with NL and those with SS. Similar to the previous comparison, we did not find significant differences in protein concentration (data not shown). When comparing the differential altered proteins between women with NL and MASH, once again, we did not find significant differences (data not shown). Finally, we compared the hepatic proteome between women with SS and MASH. This analysis revealed 72 significantly increased proteins and 84 decreased proteins in the MASH group compared to the SS group (Supplementary Table 6). Given the vast amount of information added by this analysis, we have represented the 25 most significant proteins in Fig. 2.

Biological processes enrichment analysis

To identify the pathways (Reactome) associated with the significant proteins in each comparison, we conducted an enrichment analysis of the metabolic pathways, organized by the percentage of proteins involved.

For circulating proteomics analysis, altered proteins significantly different between the NL and MASLD groups correspond completely to pathways related to complement cascade, including regulation and triggering of this process (Fig. 3A). On the other hand, the comparison of proteome between NL and SS groups again correspond entirely to pathways from the complement cascade (Fig. 3B). Then, the differently altered proteins between NL and MASH groups correspond in 50% to complement cascade pathways, and also to integrin cell surface interactions or response to elevated platelet cytosol calcium pathways (Fig. 3C).

For liver proteomics analysis, altered proteins significantly different between the SS and MASH groups correspond primarily to pathways related to the metabolism of amino acids and derivatives, antimicrobial peptides and fatty acid metabolism, among other pathways such as hemostasis, apoptosis, neutrophil degranulation, binding to scavenger receptors, complement cascade, peroxisome, metabolism of vitamins, and amyloid fiber formation pathways (Fig. 4).

In this study, we evaluated the proteomics profile in serum samples and liver biopsies of a homogeneous cohort of women with MO, classified according to liver histopathology.

In the circulating proteome analysis, we identified 257 proteins. The most relevant findings in the different groups were as follows: MASLD subjects presented 14 altered proteins, with increased levels of 9 and decreased levels of 5, compared to NL individuals. After an enrichment analysis. Subjects with SS presented 10 altered proteins, 5 upregulated and 5 downregulated, compared to NL. MASH individuals had 13 altered proteins, 11 upregulated and 2 downregulated, compared to NL, involved in the complement cascade pathway, integrin cell surface interactions and response to elevated platelet cytosolic calcium.

Next, we compared our findings with the literature. In a proteomics study on serum samples of 636 individuals with liver biopsy-confirmed MASH, authors aimed to identify protein signatures corresponding to key histological phenotypes such as steatosis, inflammation, ballooning and fibrosis [12]. Although our studies are not directly comparable, we concur with their finding that low levels of adiponectin are associated with inflammation without fibrosis. Moreover, they suggested that collectin-11 has a positive association with fibrosis, whereas we found increased levels of collectin-11 in subjects with MASH compared to NL. However, in our study, all MASH patients correspond to preliminary stages of MASH without fibrosis.

In other three studies of proteomics in serum, we found coincident results with our findings. In the first, authors conducted a study on the general population, defining two groups: fatty liver (by ultrasound) and liver fibrosis (by transient elastography) [19]. Analyzing their results, we found that thrombospondin-4 is directly associated with both hepatic conditions. We found increased levels of this protein in serum samples of our MASH subjects compared to NL. In the second study, authors suggested that some proteins are associated with MASLD, such as vitronectin, which is associated to extracellular matrix responses to hepatic injury [20]. In this sense, we found increased levels of vitronectin in serum samples of SS and MASH subjects compared to NL. Finally, the third study identified some proteins as candidate biomarkers of MASLD with ELISA validation, including high levels of clusterin, complement factor H, vitronectin and thrombospondin-4 [21]. We also found increased levels of these four proteins in MASH compared to NL. These findings reinforce the relevance of these molecules in MASLD pathogenesis, which are mostly involved in complement cascade [21]. Future research is needed to validate the specific role of complement proteins in MASLD and their potential use as biomarkers for MASH progression. The same article also suggested that apolipoprotein E differentiates between controls and MASLD subjects [21]. Although we did not find differences in serum samples regarding this protein, we observed decreased levels in liver biopsies of MASH subjects compared to SS. This finding is consistent with studies that base the inclusion of apolipoprotein E deficiency as a MASH model in murine fed with high-cholesterol diet [22].

In the liver proteomics analysis, we identified 2081 proteins. We did not report significant differences in the proteome when comparing MASLD to NL groups or when comparing SS or MASH to NL. However, we found 72 proteins with increased levels and 84 proteins with decreased levels in the MASH group compared to the SS group. Enrichment analyses indicated that these proteins are primarily involved in amino acid and derivatives, antimicrobial peptides and fatty acid metabolism pathways. Additionally, these proteins are implicated in lesser proportions in hemostasis, apoptosis, neutrophil degranulation, scavenger receptors, the complement cascade, peroxisomes, metabolism of vitamins and cofactors and amyloid fiber formation processes.

A proteomics study in liver samples of small groups of MASLD subjects found some proteins significantly increased in SS or MASH subjects. They reported increased levels of endoplasmin and peroxiredoxin-4 in subjects with SS compared to controls and high levels of cytochrome b5 subunit a in MASH subjects compared to controls [23]. In comparison, we found increased levels of these three proteins in our MASH subjects compared to SS but did not report significant changes in these proteins in SS or MASH subjects compared to the NL group. Discrepancies can be attributed to their study including small groups of 6 individuals, and a heterogeneous cohort of men and women. In addition, their control group was composed by non-obese subjects. Additionally, they used a different technique (2-DIGE-MALDI-TOF/TOF). In contrast, we evaluated a considerable number (n = 174) and homogeneous (only women) cohort of subjects with MO using the high-throughput technique HPLC-MS/MS.

In a review of mitochondrial molecular pathophysiology of MASLD regarding the liver tissue proteome, authors highlighted potential pathophysiological biomarkers such as cytochrome b5 (linked to apoptosis), hemoglobins (linked to oxidative stress) and cytochromes P2E1 and P4A11 (linked to lipid droplets) [24]. We found these proteins increased in liver samples of our MASH subjects compared to SS, except for hemoglobins, which were decreased in our MASH subjects. However, our enrichment analysis of the liver proteome indicated that these pathways are in the minority but highlighted the fatty acid metabolism associated with lipid droplets.

Our study supported by previous findings, suggests some proteins in serum samples that could be important for the study of possible pathogenic pathways in MASH. Increased levels of clusterin, collectin-11, vitronectin, complement factor H and thrombospondin-4, and decreased levels of adiponectin, are aligned with other published reports about MASH.

Focusing on the altered hepatic proteome between MASH subjects and SS individuals, these proteins are mostly related to molecular pathways such as the metabolism of amino acids and derivatives, antimicrobial peptides and fatty acid metabolism. Other less significant pathways include hemostasis, apoptosis, neutrophil degranulation, binding to scavenger receptors, the complement cascade, peroxisomes, metabolism of vitamins and amyloid fiber formation. It is well-known that amino acid metabolism, especially branched-chain amino acids and aromatic amino acids, is altered in MASLD and contributes to disease progression because it is strongly involved in lipid accumulation processes and inflammatory responses [25]. Additionally, altered fatty acid metabolism is a major factor in MASH, inducing lipotoxicity [26]. However, we cannot establish any relationship between the proteomic findings obtained in serum and liver tissue.

As mentioned, most altered proteins in serum samples between different liver histological groups are related to the complement cascade. This pathway is also altered in liver tissue. This finding is consistent with the role of the complement cascade in lipid metabolism and immune responses [27]. Activation of the innate immune system plays a key role in MASLD pathogenesis when the complement system recognizes danger signals such as tissue injury. Moreover, activation of the complement cascade is involved in disease progression and severity, including the clearance of apoptotic cells, liver inflammation, hepatic fibrosis and liver regeneration [28, 29].

In this proteomics study of a homogeneous cohort of women with MO, we identified a groups of serum proteins composed of 13 altered proteins in MASH subjects compared to NL. Of these, 6 coincide with previous literature. We found increased levels of fructose-bisphosphate aldolase, clusterin, collectin-10, collectin-11, scavenger receptor cysteine-rich type 1 protein M130, attractin, pigment epithelium-derived factor, vitronectin, complement factor H, thrombospondin-4 and apolipoprotein A-IV. Additionally, we found decreased levels of sex hormone-binding globulin and adiponectin. Most of these proteins are related to the complement cascade. Further studies in different population groups are needed to confirm if this proteins profile can be integrated in a panel of biomarkers for the diagnosis or follow-up of MASH.

Rinella ME, Lazarus JV, Ratziu V, Francque SM, Sanyal AJ, Kanwal F et al (2023) A multisociety Delphi consensus statement on new fatty liver disease nomenclature. J Hepatol 79:1542–1556. https://doi.org/10.1016/j.jhep.2023.06.003
Chalasani N, Younossi Z, Lavine JE, Charlton M, Cusi K, Rinella M et al (2018) The diagnosis and management of nonalcoholic fatty liver disease: Practice guidance from the American Association for the Study of Liver Diseases. Hepatology 67:328–357. https://doi.org/10.1002/hep.29367
Lazarus JV, Mark HE, Anstee QM, Arab JP, Batterham RL, Castera L et al (2022) Advancing the global public health agenda for NAFLD: a consensus statement. Nat Rev Gastroenterol Hepatol 19:60–78. https://doi.org/10.1038/s41575-021-00523-4
Shroff H, VanWagner LB (2020) Cardiovascular Disease in Nonalcoholic Steatohepatitis: Screening and Management. Curr Hepatol Rep 19:315–326. https://doi.org/10.1007/s11901-020-00530-0
Friedman SL, Ratziu V, Harrison SA, Abdelmalek MF, Aithal GP, Caballeria J et al (2018) A randomized, placebo-controlled trial of cenicriviroc for treatment of nonalcoholic steatohepatitis with fibrosis. Hepatology 67:1754–1767. https://doi.org/10.1002/hep.29477
Mózes FE, Lee JA, Vali Y, Selvaraj EA, Jayaswal ANA, Boursier J et al (2024) Diagnostic accuracy of non-invasive tests to screen for at‐risk MASH —An individual participant data meta‐analysis. Liver Int liv15914. https://doi.org/10.1111/liv.15914
Frantzi M, Latosinska A, Mischak H (2019) Proteomics in Drug Development: The Dawn of a New Era? Proteom Clin Apps 13:1800087. https://doi.org/10.1002/prca.201800087
Geyer PE, Holdt LM, Teupser D, Mann M (2017) Revisiting biomarker discovery by plasma proteomics. Mol Syst Biol 13:942. https://doi.org/10.15252/msb.20156297
Younossi ZM, Baranova A, Ziegler K, Del Giacco L, Schlauch K, Born TL et al (2005) A genomic and proteomic study of the spectrum of nonalcoholic fatty liver disease. Hepatology 42:665–674. https://doi.org/10.1002/hep.20838
Cheng Q, Yuan X, Lin S, Zhao Y, Wang H, Zhu F et al (2022) Serum proteome profiling reveals differentially expressed proteins between subjects with metabolically healthy obesity and nonalcoholic fatty liver disease. J Proteom 260:104556. https://doi.org/10.1016/j.jprot.2022.104556
Yuan X, Sun Y, Cheng Q, Hu K, Ye J, Zhao Y et al (2020) Proteomic analysis to identify differentially expressed proteins between subjects with metabolic healthy obesity and non-alcoholic fatty liver disease. J Proteom 221:103683. https://doi.org/10.1016/j.jprot.2020.103683
Sanyal AJ, Williams SA, Lavine JE, Neuschwander-Tetri BA, Alexander L, Ostroff R et al (2023) Defining the serum proteomic signature of hepatic steatosis, inflammation, ballooning and fibrosis in non-alcoholic fatty liver disease. J Hepatol 78:693–703. https://doi.org/10.1016/j.jhep.2022.11.029
Ramachandran P, Xu G, Huang HH, Rice R, Zhou B, Lindpaintner K et al (2022) Serum Glycoprotein Markers in Nonalcoholic Steatohepatitis and Hepatocellular Carcinoma. J Proteome Res 21:1083–1094. https://doi.org/10.1021/acs.jproteome.1c00965
Auguet T, Bertran L, Capellades J, Abelló S, Aguilar C, Sabench F et al (2023) LC/MS-Based Untargeted Metabolomics Analysis in Women with Morbid Obesity and Associated Type 2 Diabetes Mellitus. IJMS. ;24:7761. https://doi.org/10.3390/ijms24097761
Bertran L, Capellades J, Abelló S, Durán-Bertran J, Aguilar C, Martinez S et al (2023) LC/MS-Based Untargeted Metabolomics Study in Women with Nonalcoholic Steatohepatitis Associated with Morbid Obesity. IJMS 24:9789. https://doi.org/10.3390/ijms24129789
Geer EB, Shen W (2009) Gender differences in insulin resistance, body composition, and energy balance. Gend Med 6:60–75. https://doi.org/10.1016/j.genm.2009.02.002
Lauretta R, Sansone M, Sansone A, Romanelli F, Appetecchia M (2018) Gender in Endocrine Diseases: Role of Sex Gonadal Hormones. Int J Endocrinol 2018:1–11. https://doi.org/10.1155/2018/4847376
Kleiner DE, Brunt EM, Van Natta M, Behling C, Contos MJ, Cummings OW et al (2005) Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology 41:1313–1321. https://doi.org/10.1002/hep.20701
Abozaid YJ, Ayada I, Van Kleef LA, Vallerga CL, Pan Q, Brouwer WP et al (2023) Plasma proteomic signature of fatty liver disease: The Rotterdam Study. Hepatology 78:284–294. https://doi.org/10.1097/HEP.0000000000000300
Perry AS, Hadad N, Chatterjee E, Ramos MJ, Farber-Eger E, Roshani R et al A prognostic molecular signature of hepatic steatosis is spatially heterogeneous and dynamic in human liver 2024. https://doi.org/10.1101/2024.01.26.24301828
Miller MH, Walsh SV, Atrih A, Huang JT, Ferguson MAJ, Dillon JF (2014) The serum proteome of nonalcoholic fatty liver disease: A multimodal approach to discovery of biomarkers of nonalcoholic steatohepatitis. J Gastro Hepatol 29:1839–1847. https://doi.org/10.1111/jgh.12614
Schierwagen R, Maybüchen L, Zimmer S, Hittatiya K, Bäck C, Klein S et al (2015) Seven weeks of Western diet in apolipoprotein-E-deficient mice induce metabolic syndrome and non-alcoholic steatohepatitis with liver fibrosis. Sci Rep 5:12931. https://doi.org/10.1038/srep12931
Rodríguez-Suárez E, Duce AM, Caballería J, Arrieta FM, Fernández E, Gómara C et al (2010) Non‐alcoholic fatty liver disease proteomics. Proteom Clin Apps 4:362–371. https://doi.org/10.1002/prca.200900119
Nuño-Lámbarri N, Barbero-Becerra V, Uribe M, Chávez-Tapia N (2016) Mitochondrial Molecular Pathophysiology of Nonalcoholic Fatty Liver Disease: A Proteomics Approach. IJMS 17:281. https://doi.org/10.3390/ijms17030281
Tricò D, Biancalana E, Solini A (2021) Protein and amino acids in nonalcoholic fatty liver disease. Curr Opin Clin Nutr Metabolic Care 24:96–101. https://doi.org/10.1097/MCO.0000000000000706
Syed-Abdul MM (2023) Lipid Metabolism in Metabolic-Associated Steatotic Liver Disease (MASLD). Metabolites 14:12. https://doi.org/10.3390/metabo14010012
Guo Z, Fan X, Yao J, Tomlinson S, Yuan G, He S (2022) The role of complement in nonalcoholic fatty liver disease. Front Immunol 13:1017467. https://doi.org/10.3389/fimmu.2022.1017467
Rensen SS, Slaats Y, Driessen A, Peutz-Kootstra CJ, Nijhuis J, Steffensen R et al (2009) Activation of the complement system in human nonalcoholic fatty liver disease. Hepatology 50:1809–1817. https://doi.org/10.1002/hep.23228
Segers FM, Verdam FJ, De Jonge C, Boonen B, Driessen A, Shiri-Sverdlov R et al (2014) Complement Alternative Pathway Activation in Human Nonalcoholic Steatohepatitis. PLoS ONE 9:e110053. https://doi.org/10.1371/journal.pone.0110053

The authors declare no competing interests.

SupplementrayTable1.xlsx
Supplementary Table 1
SupplementaryTable2.xlsx
Supplementary Table 2
SupplementaryTable3.xlsx
Supplementary Table 3
SupplementaryTable4.xlsx
Supplementary Table 4
SupplementaryTable5.xlsx
Supplementary Table 5
SupplementaryTable6.xlsx
Supplementary Table 6

Download PDF

Version 1

posted

You are reading this latest preprint version

Proteomics profile of serum and liver samples in women with morbid obesity and metabolic dysfunction-associated steatohepatitis

Status:

Version 1

Abstract

Figures

Introduction

Materials and methods

Subjects

Anthropometric parameters and biochemical analyses

Histopathological analysis of the liver biopsies

Serum proteomic analysis

Liver proteomic analysis

Statistical analysis

Results

Subjects

Proteomics analysis in serum samples

Serum Proteomic Profile Comparison Between Groups

Proteomics analysis in liver samples

Biological processes enrichment analysis

Discussion

Conclusions

References

Additional Declarations

Supplementary Files

Status:

Version 1