3.1 Routine index analysis
When the body is damaged by heavy metals such as Pb or Cd, the body weight, diet, hair color and mental state will also change (Dai et al. 2021). Therefore, the poisoning situation of the body and the intervention situation of therapeutic substances can be judged to a certain extent through the observation of conventional indicators. In this study, there were no significant differences in body weight, diet, hair color and mental state during adaptive feeding. During the experiment period of 5 weeks, mice in the blank control group had flexible response, good appetite and bright fur. Compared with the blank control group, mice in the Pb-Cd model group showed inappetite, poor spirit, significantly reduced activity, slow weight gain and dark hair color (Fig. 1). The results showed that combined exposure to Pb and Cd caused damage to body weight, mental state and hair in mice.
In addition, when the body is damaged by Pb and Cd complex heavy metals, a series of changes will occur in cell morphology, substance metabolism and so on, leading to changes in organ index. Therefore, the analysis of organ index can reflect the liver injury of mice. In this study, the liver index of each group was measured. As can be seen from Table 1, the liver index of Pb and Cd model group was 4.15 ± 0.33%, which was significantly higher than that of blank control group (P < 0.01), the results showed that combined exposure to Pb and Cd could cause liver enlargement in mice.
Table 1
Changes of liver indexes in mice of each group. (x̄± SD, n = 10)
Group
|
Liver index (%)
|
Blank control group
|
3.51 ± 0.19
|
Pb-Cd group
|
4.15 ± 0.33**
|
Note: * Compared with blank control group, **P<0.01. |
3.2 Hematological index analysis
Long-term exposure to Pb or Cd pollution will lead to damage to blood quality, therefore, hematological indicators are often used to evaluate the degree of harm of heavy metals to the body and the efficacy of therapeutic agents (Rehman et al. 2021). In the present study, hematological index of RBC, WBC, HGB, PLT, PCT, HCT, LYM and MON were monitored. As shown in Table 2, compared with the blank group, PCT in the Pb-Cd group was significantly increased (P < 0.05), WBC, PLT and MON showed an upward trend, while RBC, HGB, HCT and LYM showed a downward trend.
Table 2
Changes of blood index and Biochemical index in mice. (x̄ ± SD, n = 10)
Group
|
Blank control group
|
Pb-Cd group
|
WBC
|
5.71 ± 1.72
|
6.35 ± 0.36
|
RBC
|
7.31 ± 0.27
|
7.07 ± 0.35
|
HGB
|
151.56 ± 4.38
|
149.24 ± 8.18
|
PLT
|
222.60 ± 28.42
|
330.90 ± 47.61*
|
PCT
|
0.25 ± 0.04
|
0.47 ± 0.15**
|
HCT
|
37.33 ± 0.96
|
36.77 ± 2.12
|
LYM
|
2.86 ± 0.51
|
1.99 ± 0.25
|
MON
|
1.24 ± 0.26
|
2.01 ± 0.14*
|
AST(U/L)
|
33.25 ± 1.46
|
38.25 ± 3.89**
|
ALT(U/L)
|
54.20 ± 4.69
|
60.72 ± 4.62**
|
Note: * Compared with blank control group, *P<0.05; **P<0.01. |
3.3 ALT and AST analysis
Liver function examination is the examination item that reflects liver physiology function (Ragab AboZaid et al. 2021). There are many liver function indicators, including ALT, AST, total bilirubin, direct bilirubin, indirect bilirubin, alkaline phosphatase, glutamyl transpeptidase, albumin, globulin, prealbumin, total bile acid, etc. These indicators can reflect whether there is damage to liver cells, whether there is obstruction of the biliary tract, but also reflect the synthetic function of the liver (Zubrzycki et al. 2020). For example, in humans, the normal value of AST is 8–40 U/L, and when ALT is significantly elevated and the AST/ALT ratio is greater than 1, damage to liver parenchyma is indicated. In the present study, ALT and AST indexes of different groups of mice were systematically observed, and the results were shown in Table 2. Compared with the blank control group, the AST and ALT enzyme activities in the Pb-Cd group were significantly increased (P < 0.01), indicating that Pb and Cd complex heavy metals did have significant damage to the liver.
3.4 Pathological analysis
In order to explore the disease process of organs, tissues or cells, some pathomorphological examination method can be used to examine their lesions, discuss the causes, pathogenesis and development process of lesions, and finally make pathological diagnosis (Murakami et al. 2020). Pathological examination has been widely used in clinical work and scientific research (Baraban et al. 2020). In this study, in order to explore the liver injury of mice exposed to Pb and Cd, HE staining was performed to evaluate the histopathological changes of mouse liver.
The prepared sections were observed under a microscope, and the results were shown in Fig. 2. The liver tissue structure of mice in the blank control group was complete, with clear morphology, orderly arrangement of liver cells, normal structure and distinct hierarchy. The liver cells of Pb-Cd group were massive necrosis, uneven arrangement, vacuolar degeneration and congestion.
3.6 Analysis of lipidomics
In recent years, omics technology has been developed vigorously with the development of analytical chemistry technology (Marvasi et al. 2019). Metabolomics is a discipline that reflects the endogenous and exogenous changes of the body and provides further understanding of the disease (Feizi et al. 2021). Lipidomics is an important branch of metabolomics. The changes of lipid metabolism under different physiological and pathological conditions are analyzed and compared to reflect the changes of physiology and pathology (Wu et al. 2021). Therefore, in addition to studying the toxic mechanism of lead and cadmium from the perspective of classical oxidation targets, this study also used non-targeted lipidomics to analyze the changes in liver tissue of mice in each group and to search for possible lipid regulatory pathways.
3.6.1 Quality control of non-targeted lipidomics data
In the process of this study, 5 QC samples were randomly inserted during the detection of liver samples from each group of mice to verify the reliability of the experimental method and the stability of the instrument. Unsupervised PCA analysis was conducted on the pre-processed QC sample data of the five samples, and the results were shown in Fig. 3.
The results showed that the QC samples were all within 2std, indicating that the experimental method was reliable and the instrument had good stability.
3.6.2 Multivariate statistical analysis of lipid data
After the pre-treatment of the original lipid data of each group, Metaboanalyst 5.0 (https://www.metaboanalyst.ca/) was used for data processing such as missing value and normalization of total peak area, and SIMCA 14.1 software was used for multivariate statistical analysis of the processed data.
Firstly, PCA statistical analysis was performed on the lipid data of liver tissue of mice in each group to obtain PCA score under positive and negative ion mode. As can be seen from Fig. 4, the lipid data of liver tissue in the blank group and the Pb-Cd group were well separated. The results showed that the two groups of mice could be distinguished normally, suggesting that Pb-Cd had a significant effect on lipid disorders.
In order to further screen out the different lipids, the mass spectrometry data of the blank control group and Pb-Cd model group were further analyzed by supervised OPLS-DA. As shown in Fig. 5A and 5B, parameter R2x of the OPLS-DA model represented the fitting degree of the model, and parameter Q2 represented the prediction ability of the model. In Fig. 5A, R2X = 0.564 (cum) and Q2 = 0.9 (cum) in the blank control group and Pb-Cd group in the negative ion mode, where R2 was greater than 0.5, the difference between R2X and Q2 was less than 4, and Q2 was close to 1, indicating that it had good fitting and prediction ability in the negative ion mode. Similarly, as shown in Fig. 5B, in the positive ion mode, R2X = 0.515 (cum), Q2 = 0.88 (cum), R2 > 0.5, the difference between R2X and Q2 was less than 3, and Q2 was close to 1, indicating that lipids in liver tissues had good fitting and prediction ability in the positive ion mode. In order to prevent over-fitting of the OPLS-DA model, 200 displacement tests were carried out for the establishment of the OPLS-DA model, as shown in Fig. 5C and 5D. Both the y-intercepts of Q2 in positive and negative ion modes were less than 0, indicating that no over-fitting occurred in the OPLS-DA model and the results were reliable.
Volcano map is a combination of multiple of variation analysis and T test, which can intuitively show the significance of lipid changes between the two groups of samples. Through experimental screening of the FC values in the blank group and Pb-Cd group, a volcano diagram was made for the screened differential lipids, as shown in Fig. 6, Log2 (FC) was the abscissa and -Log10 (P) was the ordinate.
As can be seen from the Fig. 6, there were obvious differences in lipids between the blank group and Pb-Cd group, in which the blue color indicated down-regulation, and the red color indicated up-regulation, indicating that the lipid metabolites in the liver tissues of mice had undergone great changes after the lead and cadmium complex heavy metals were poisoned.
3.6.3 Screening of differential lipids
In order to further screen differential lipids, VIP > 1, P < 0.05, FC > 2 or FC < 0.5 were used as parameters for screening differential metabolites. According to LIPID MAPS, HMDB, Massbank of North America (MONA) and other online sites, the LIPID ratio and secondary mass spectrometry were compared. A total of 24 different lipids were identified, and the specific substances were shown in Table 4.
Table 4
Differential lipid table of Blank group and Pb-Cd group.
Compound name
|
Formula
|
Class
|
RT
|
m/z
|
VIP
|
FC
|
P-value
|
Stearoyl-L-Carnitine
|
C25H49NO4
|
FA
|
6.56
|
426.342
|
1.022
|
8.800
|
0.016
|
Pentaethylene glycol
|
C23H45NO4
|
FA
|
5.96
|
426.342
|
1.307
|
7.872
|
0.017
|
6-Aminohexanoate
|
C6H13NO2
|
FA
|
15.62
|
392.316
|
1.504
|
2.633
|
0.002
|
Linoleic acid
|
C18H32O2
|
FA
|
11.60
|
456.332
|
1.210
|
3.703
|
0.003
|
LPE 20:5
|
C25H42NO7P
|
GP
|
5.64
|
500.276
|
4.172
|
0.443
|
0.002
|
PE (18:1/22:6)
|
C45H76NO8P
|
GP
|
23.69
|
790.536
|
1.014
|
0.387
|
0.004
|
LPE 18:0
|
C23H48NO7P
|
GP
|
9.87
|
480.308
|
7.000
|
0.484
|
0.005
|
PC(P-18:1/P-18:1)
|
C44H86NO7P
|
GP
|
21.01
|
340.356
|
1.373
|
2.310
|
0.006
|
PC(16:0/18:1)
|
C42H82NO8P
|
GP
|
19.78
|
703.468
|
1.130
|
6.477
|
0.045
|
LPE(20:3)
|
C25H46NO7P
|
GP
|
11.56
|
580.361
|
1.538
|
0.394
|
0.000
|
PG(22:6/18:0)
|
C46H79O10P
|
GP
|
23.76
|
738.507
|
1.795
|
0.487
|
0.020
|
PC(16:0/20:5)
|
C44H78NO8P
|
GP
|
24.40
|
764.548
|
2.263
|
2.889
|
0.025
|
LPE(18:2)
|
C23H44NO7P
|
GP
|
23.17
|
712.491
|
1.027
|
0.478
|
0.042
|
LPE 14:1
|
C19H38NO7P
|
GP
|
13.06
|
342.336
|
5.239
|
3.660
|
0.000
|
LPE 16:1
|
C21H42NO7P
|
GP
|
15.93
|
370.367
|
2.846
|
2.170
|
0.001
|
PE (16:0/20:5)
|
C41H72NO8P
|
GP
|
23.36
|
738.505
|
5.256
|
0.314
|
0.001
|
PE (20:5/22:6)
|
C47H72NO8P
|
GP
|
20.67
|
808.491
|
1.677
|
0.450
|
0.001
|
Bexarotene
|
C24H28O2
|
PR
|
23.45
|
762.507
|
4.048
|
0.371
|
0.000
|
agnuside
|
C22H26O11
|
PR
|
19.75
|
465.304
|
5.365
|
0.416
|
0.000
|
Ginkgolide B
|
C20H24O10
|
PR
|
17.86
|
572.480
|
3.202
|
2.321
|
0.000
|
Abietic acid
|
C20H30O2
|
PR
|
5.62
|
498.262
|
4.345
|
0.426
|
0.001
|
SM(d18:1/18:0)
|
C41H83N2O6P
|
SP
|
9.09
|
762.556
|
1.362
|
17.399
|
0.014
|
SM(d18:0/16:0)
|
C39H81N2O6P
|
SP
|
22.04
|
618.476
|
1.006
|
0.498
|
0.003
|
Sphingosine-1-Phosphate
|
C18H38NO5P
|
SP
|
11.57
|
366.336
|
4.498
|
2.834
|
0.002
|
Cluster analysis was performed on the 24 different lipid metabolites screened, which could more comprehensively and intuitively display the relationship between the samples and the expression patterns of different groups of lipids. As shown in Fig. 7, there were significant differences in metabolites between the Pb-Cd group and the blank group. The experimental results showed that Pb-Cd complex heavy metals could cause a significant lipid metabolism disorder.
3.4.4 Pathway analysis and biological interpretation
Among the 24 differential lipid metabolites screened, 12 differential lipids were down-regulated in the Pb-Cd model group compared with the blank control group, including PC (P-18:1/P-18:1), PC (16:0–18:1), SM (d18:1/18:0), Ginkgolide B, PC (16:0/20:5), Stearoyl-L-Carnitine, Pentaethylene glycol, Aminocaproic acid, LPE (16:1), LPE (14:1), Sphingosine-1-Phosphate and Linoleic acid. While PG (22:6/18:0), LPE (18:2), LPE (18:0), agnuside, PE (20:5/22:6), PE (18:1/22:6), PE (16:0/20:5), Bexarotene, SM (d18:0–16:0), LPE (20:3), LPE (20:5) and Abietic acid were up-regulated. The 24 selected differential lipids included four major categories of fatty acyl, glycerol phospholipids, enol esters and sphingolipids, and 11 subcategories of fatty acid esters, fatty acid conjugates, linear acid derivatives, glycerol phospholipid ethanolamine, glycerol phospholipid choline, glycerol phosphate glyceride, vitamin A-like, terpene glycoside, terpene lactone, diterpene and sphingolipids. According to the HMDB ID number of differential lipids, the pathways of 24 differential lipids were analyzed using Metaboanalyst 5.0 online site.
The results were shown in Fig. 8, where the color red represented a low P value, and the circle size represented the path influence value, and the larger the influence value is, the higher the influence value is. A total of 8 pathways were identified, including Linoleic acid metabolism, sphingolipid metabolism, Glycerophospholipid metabolism, alpha-Linolenic acid metabolism, arachidonic acid metabolism, glycosylphosphatidylinositol-anchor biosynthesis, biosynthesis of unsaturated fatty acids and fatty acid degradation. Among them, linoleic acid metabolism was the most influential pathway, followed by sphingolipid metabolism and glycerol phospholipid metabolism. The results showed that Pb-Cd could cause lipid disorders, possibly by regulating linoleic acid metabolism, sphingolipid metabolism and glycerolipid metabolism.