MALAT1 and MEG3 genes expression in non-alcoholic fatty liver disease in type 2 diabetes mellitus patients: a case-control study

ABSTRACT

Aim of the study

This study aims to investigate the nonalcoholic fatty liver disease (NAFLD) serum molecular profile among individuals diagnosed with type 2 diabetes (T2DM).

Methods

In this observational case-control investigation, 120 participants were included and subdivided forming 4 different groups as follows: group І: NAFLD non diabetic individuals, group ІІ: T2DM without NAFLD patients, group III: NAFLD diabetic patients and group IV: healthy control. MALAT1 and MEG3 expression in serum from all groups was measured.

Results

Expression of MALAT1 was upregulated in NAFLD and diabetic patients (p = 0.037 and 0.033 respectively). The cutoff value was determined for MALAT1 expression in NAFLD and diabetic by the ROC curve and was > 0.54 and > 0.67 respectively. By multivariate analysis, the only reliable indicator for MALAT1 expression in NAFLD and diabetics was determined to be ESR. Furthermore, we found that NAFLD patients showed greater MEG3 expression than those with T2DM (p = 0.033).

Conclusion

Expression of MALAT1 was upregulated in NAFLD and T2DM indicating that it might be an early diagnostic marker for both diseases and helps in the development of novel therapeutic agents. Moreover, MEG3 expression was higher among NAFLD patients than those NAFLD patients with T2DM, which suggests the feasibility of decreased MEG3 expression could be a viable predictive biomarker for early T2DM detection among NAFLD-diagnosed patients.

KEYWORDS:

• Non-alcoholic fatty liver disease (NAFLD)
• type 2 diabetes mellitus (T2DM)
• quantitative real-time-polymerase chain reaction (qPCR)

1. Introduction

NAFLD prevalence, which is considered one of the most frequent liver disorders, is becoming a world wide public health concern. When fat accounts for more than 5 to 10% of the liver’s weight, the disease arises.It may cause cirrhosis, hepatocellular cancer (HCC), and hepatic fibrosis, which would increase mortality and morbidity. Therefore, indicators for the diagnosis of NAFLD are urgent needs [1]. According to age, sex, and ethnicity, NAFLD prevalence may differ. Actually, it appears that aging is connected with the rise in both the NAFLD prevalence and liver disease stage [2]. In addition, men have a two times greater frequency of NAFLD than women [3,4].

Diabetes mellitus (DM) has become a serious health issue worldwide. It is characterized by a chronic hyperglycemia state and disturbance in the metabolism of carbohydrates, fat, and protein. It is related to the deficiency of absolute or relative insulin secretion and/or insulin action, which is controlled by a variety of factors, including inherited and environmental influences.It affects the vascular system and leads to micro and macroangiopathy [5].

People with type 2 DM (T2DM) have a very high NAFLD prevalence. In fact, according to some research, NAFLD affects between one-third and two-thirds of diabetic individuals [6,7]. Estimates from several research have revealed that T2DM patients who are asymptomatic have a high burden of advanced fibrosis, ranging from 5% to 7% [6,7]. Patients with NAFLD frequently have high blood levels of triglycerides (TG) and low serum high-density lipoprotein (HDL) levels. It is believed that 50% of people with dyslipidemia who visit lipid clinics develop NAFLD [8,9].

Moreover, there is a clear link between NAFLD and the risk of DM. If someone has NAFLD, their chance of acquiring DM increases by around 5-fold [10]. Additionally, it has been shown that improved NAFLD lowers the risk of acquiring DM.[10]Till now; it is unpredictable which people with NAFLD will acquire DM [11].

To better understand the molecular basis of NAFLD, gene expression profiling has long been used [12,13]. The existence of NAFLD has been linked to DNA methylation modification and gene expression alteration [14].

The study’s purpose was to examine NAFLD serum molecular profile in T2DM patients and controls using RT-qPCR to identify a disease-serum molecular profile that may be validated for early NAFLD diagnosis in T2DM patients while also introducing novel screening and diagnostic methods.

2. Materials and methods

After giving their signed informed consent; this case-control study was including 120 participants subdivided forming four different groups at Alexandria University’s Internal Medicine Department; group І: 30 adult NAFLD nondiabetic patients, group ІІ: 30 adult T2DM patients without NAFLD, group III: 30 adult NAFLD with T2DM and group IV: 30 healthy subjects.

Exclusion criteria included chronic liver disease, a recent traumatic or surgical history, hematological disorders or malignancy, chronic inflammatory or autoimmune related diseases, a recent severe significant infection history at the study admission time, CKD with eGFR 60 ml/min per 1.73 m2, and also alcoholics.

All participants had their height and weight assessed in order to determine their body mass index (BMI). The superior iliac crest and the lower rib border were used to estimate the waist circumference. A mercury sphygmomanometer was also used for systolic and diastolic blood pressure measuring.

This study adhered to our institution’s human subject’s ethical guidelines as well as the Helsinki Declaration.The Alexandria University Faculty of Medicines Ethics Committee is responsible for offering ethical approval with reference number IRB NO: 00012098 – NAW NO: 00018699.

2.1. Laboratory investigations

A Biosystem (BTS350) analyzer was used to measure each participant’s fasting blood glucose and insulin level, complete profile of lipids, tests of liver function, C reactive protein (CRP), and serum-involved uric acid.The Modification of Diet in Renal Disease (MDRD) equation was employed to obtain the estimated GFR. Individuals’ insulin resistance was measured using the Homeostatic Model Assessment for Insulin Resistance (HOMA-IR).Also, complete blood count (CBC)was measured by MINDRAY BC-3000 plus analyzer, glycated hemoglobin (HbA1C) was measured by Mispa-i2 analyzer, and erythrocyte sedimentation rate (ESR).

2.2. Genomic study

According to the manufacturer’s recommendations, the RNeasy Mini kit was used to extract RNA from serum. (catalog # 217004, Qiagen, Germany). The samples of RNA were kept at −80°C before being used. Total RNA Reverse transcription to single-stranded cDNA was performed by the use of a thermal cycler (Simpli Amp, Applied Biosystems, Thermo Scientific, USA) configured according to the manufacturer’s instructions using the High-Capacity Transcription kit of cDNA Reverse transcription. (catalog # 4375575, Applied Biosystems, USA). Using a (RT-qPCR), the MALAT1 and MEG3 gene expressions were determined.Quantification and analysis were carried out on the STRATAGENE MX3005P real time QPCR system (Sigma-Aldrich, USA) according to manufacturer instructions. To determine relative expression, the target gene expression was normalized first to that of the G6PD mRNA house keeping gene and then determined using the 2−ΔΔCT method. Each response assay plate also underwent non-template control (NTC) procedures.

3. Results

Regarding demographic data and risk factors of the four studied groups (Table 1), there was a statistically significant difference among the four study groups concerning age (p < 0.001*). In terms of the risk factors prevalence, there was a difference with statistical significance between the four included study groups concerning BMI (p = 0.13) and hypertension (p = 0.032). However, a difference with no statistical significance was found among the four groups in terms of gender, and smoking.

Table 1. Comparison between the different studied groups according to demographic data and prevalence of risk factors.

	Group I (n = 30)		Group II (n = 30)		Group III (n = 30)		Group IV (n = 30)		Test of Sig.	p
	No.	%	No.	%	No.	%	No.	%
Gender
Male	9	30.0	12	40.0	9	30.0	5	16.7	χ^2=3.993	0.262
Female	21	70.0	18	60.0	21	70.0	25	83.3
Age (years)
Min. – Max.	28.0–65.0		29.0–71.0		35.0–63.0		14.0–60.0		F=16.439*	<0.001*
Mean ± SD.	44.50 ± 9.71		49.10 ± 10.08		51.43 ± 7.34		35.10 ± 11.44
P1	0.002*		<0.001*		<0.001*
BMI (kg/m2)
Min. – Max.	25.80–48.80		21.40–52.70		22.40–61.0		19.53–48.80		χ^2=3.768	0.013*
Mean ± SD.	35.61 ± 6.43		34.24 ± 6.65		36.34 ± 7.70		30.60 ± 7.88
P1	0.040*		0.209		0.013*
Smoking	3		10.0		6		20.0		χ^2 = 1.867	MCp=0.643
Hypertension	3		10.0		8		26.7		χ^2 = 1.867	MCp=0.032

IQR: Inter quartile rangeSD: Standard deviationχ²: Chi square test MC: Monte Carlo.

F: F for ANOVA test. p: p value for comparing between the studied groups.

Pairwise comparison between each two groups was done using Post Hoc Test (Tukey)..

p1:p value for comparing between Group IV and each other group *: Statistically significant at p ≤ 0.05.

Diabetic patients, without NAFLD (group II) or with NAFLD, (groups III(showed significantly increased ESR, FBS, and HbA1c than patients with NAFLD only (group I).Also, they showed significantly increased GFR, fasting insulin, and HOMA IRthan controls (group IV). (Table 2)

Table 2. Comparison between the different studied groups according to ESR, CRP, FBS, HbA1c, Fasting insulin, HOMA-IR, Total cholesterol, HDL, LDL, TG, AST, ALT, serum uric acid, and eGFR.

	Group I (n = 30)	Group II (n = 30)	Group III (n = 30)	Group IV (n = 30)	p
ESR	14.63 ± 7.55	22.27 ± 9.04	25.43 ± 16.85	19.80 ± 13.61	0.001*
	p1 = 0.001*,p2 <0.001*,p3 = 0.695
p4	0.128	0.087	0.036*
CRP	7.75 ± 21.71	4.81 ± 3.24	6.52 ± 7.61	4.55 ± 4.24	0.407
FBS	81.81 ± 18.18	150.93 ± 57.42	172.77 ± 75.20	83.23 ± 7.89	<0.001*
	p1 <0.001*,p2 <0.001*,p3 = 0.570
p4	0.818	<0.001*	<0.001*
HbA1c	5.54 ± 0.64	7.80 ± 1.81	8.02 ± 1.71	5.46 ± 0.53	<0.001*
	p1 <0.001*,p2 <0.001*,p3 = 0.915
p4	0.995	<0.001*	<0.001*
Fasting insulin	14.01 ± 7.48	20.48 ± 15.68	16.21 ± 12.45	9.93 ± 2.21	0.008*
	p1 = 0.539,p2 = 0.942,p3 = 0.588
p4	0.012*	0.002*	0.010*
HOMA-IR	2.32 ± 0.82	8.89 ± 9.87	5.78 ± 6.48	2.66 ± 4.38	<0.001*
	p1 = 0.002*,p2 = 0.151,p3 = 0.096
p4	0.277	<0.001*	0.012*
Total cholesterol	187.7 ± 29.54	180.2 ± 32.68	191.9 ± 46.79	177.9 ± 45.41	0.484
HDL	46.27 ± 7.92	43.13 ± 4.46	45.57 ± 6.94	43.90 ± 7.83	0.273
LDL	117.8 ± 21.99	111.7 ± 18.77	114.1 ± 40.93	111.6 ± 35.75	0.848
TG	123.4 ± 32.84	131.1 ± 56.99	145.7 ± 49.14	120.8 ± 48.38	0.183
AST	26.60 ± 12.87	25.15 ± 9.84	23.97 ± 8.83	37.63 ± 62.33	0.698
ALT	24.23 ± 13.19	25.60 ± 11.69	23.78 ± 12.61	43.30 ± 84.20	0.489
GGT	32.72 ± 6.77	32.07 ± 9.86	33.22 ± 10.94	28.54 ± 7.35	0.170
Serum uric acid	4.84 ± 1.30	5.57 ± 1.23	5.55 ± 1.45	5.29 ± 1.92	0.212
eGFR	103.1 ± 20.98	98.20 ± 22.65	92.62 ± 19.21	115.7 ± 20.93	<0.001*
	p1 = 0.805,p2 = 0.221,p3 = 0.732
p4	0.097	0.009*	<0.001*

F: F for ANOVA test, Pairwise comparison bet. each two groups was done using Post Hoc Test (Tukey).

p: p value for comparing between the studied groups *: Statistically significant at p ≤ 0.05.

p1:p value for comparing between group I and group II. p2: p value for comparing between group I and group III.

Table 3. Comparison between the different studied groups according to MALAT1 and MEG3 genes expression.

	Group I (n = 30)	Group II (n = 30)	Group III (n = 30)	Group IV (n = 30)	H	p
MALAT1 gene expression
	0.08–27.28	0.06–95.67	0.01–120.3	0.10–34.06	6.623	0.085
Mean ± SD.	4.27 ± 6.29	7.17 ± 18.74	6.78 ± 22.07	2.55 ± 6.39
Median (IQR)	1.53 (0.46–6.11)	2.11 (0.52–3.73)	1.24 (0.34–3.41)	0.51 (0.25–1.48)
MEG3 gene expression
Min. – Max.	0.25–60.97	0.16–34.78	0.12–42.22	0.15–8.22	12.016*	0.007*
Mean ± SD.	7.72 ± 14.11	5.13 ± 7.76	4.65 ± 10.55	1.41 ± 1.69
Median (IQR)	2.30 (1.24–5.39)	1.10 (0.63–7.36)	1.13 (0.64–1.95)	0.96 (0.41–1.32)
Sig. bet. grps.	p1 = 0.177,p2 = 0.033*,p3 = 0.432
p4	0.001*	0.043*	0.214

IQR: Inter quartile range SD: Standard deviationH: H for Kruskal Wallis test..

Pairwise comparison bet. each two groups was done using Post Hoc Test (Dunn’s for multiple comparisons test).

p: p value for comparing between the studied groups *: Statistically significant at p ≤ 0.05.

p₁: p value for comparing between group I and group II. p₂: p value for comparing between group I and group III.

p3:p value for comparing between group II and group III. p4: p value for comparing between group IV and each other group.

The expression of the MALAT1 and MEG3 genes was assessed among the research-involved groups (Table 3). A difference with no statistical significance was found as regards MALAT1 expression among the four studied groups (p = 0.085). However, a statistically significant difference was observed regarding MEG3 expression among the four groups (p = 0.007). Groups I (NAFLD) and II (T2DM) showed higher MEG3 gene expression than group IV (controls). Also, MEG3 gene expression was higher significantly among group I (NAFLD) than in group III (NAFLD with diabetes) (p = 0.033).

Table 4. univariate and multivariate Linear regression analysis for the parameters affecting MALAT gene expression (n = 60) for NAFLD patient group (I + III).

	Univariate		^#Multivariate
	p	B (95%C.I)	p	B (95%C.I)
Female	0.706	1.737 (−7.430–10.903)
Age (years)	0.718	−0.083 (−0.543–0.376)
Smoker	0.675	−2.944 (−16.942–11.053)
Hypertension	0.572	−3.194 (−14.448–8.060)
Weight (kg)	1.000	−0.00005 (−0.213–0.212)
Height (cm)	0.731	−0.087 (−0.588–0.415)
BMI (kg/m^2)	0.812	0.072 (−0.530–0.673)
Waist circumference (cm)	0.804	0.036 (−0.251–0.322)
Hemoglobin	0.650	0.783 (−2.658–4.224)
1^st ESR	<0.001*	0.704 (0.465–0.943)	<0.001*	2.143 (1.418–2.867)
2^nd ESR	<0.001*	0.305 (0.143–0.467)	<0.001*	−0.938 (−1.362 – −0.515)
CRP	0.037*	0.270 (0.017–0.523)	0.470	0.063 (−0.110–0.235)
ANA	0.528	−5.328 (−22.130–11.474)
FBS	0.017*	0.070 (0.013–0.127)	0.763	0.007 (−0.037–0.050)
HbA1c	0.172	1.612 (−0.720–3.943)
Fasting insulin	0.552	−0.123 (−0.536–0.289)
HOMA-IR	0.595	−0.231 (−1.094–0.632)
Total cholesterol	0.125	0.083 (−0.024–0.190)
HDL	0.236	0.339 (−0.228–0.906)
LDL	0.990	0.001 (−0.129–0.131)
TG	0.931	0.004 (−0.094–0.103)
AST	0.325	−0.189 (−0.571–0.192)
ALT	0.652	−0.075 (−0.406–0.256)
GGT	0.001*	0.721 (0.291–1.152)	0.064	0.305 (−0.018–0.627)
Serum uric acid	0.942	0.109 (−2.890–3.108)
eGFR	0.613	0.052 (−0.153–0.257)
Liver size	0.603	1.337 (−3.776–6.451)
Echogenicity	0.514	4.287 (−8.766–17.339)
Size of spleen	0.888	−0.796 (−12.079–10.487)

Note: B: Unstandardized Coefficients.

C.I: Confidence intervalLL: Lower limitUL: Upper Limit.

^#All variables with p < 0.05 was included in the multivariate.

*:Statistically significant at p ≤ 0.05.

Table 5. Univariate and multivariate Linear regression analysis for the parameters affecting MALAT gene expression (n = 60) for diabetic patient group (II + III).

	Univariate		^#Multivariate
	p	B (95%C.I)	p	B (95%C.I)
Female	0.743	−1.828 (−12.910–9.255)
Age (years)	0.129	−0.456 (−1.049–0.137)
Smoker	0.489	−5.135 (−19.890–9.621)
Hypertension	0.776	−1.741 (−13.951–10.470)
Weight (kg)	0.960	−0.008 (−0.304–0.289)
Height (cm)	0.311	0.341 (−0.327–1.009)
BMI (kg/m^2)	0.832	−0.079 (−0.818–0.661)
Waist circumference (cm)	0.937	0.014 (−0.332–0.360)
Hemoglobin	0.193	2.359 (−1.223–5.941)
1^st ESR	<0.001*	0.706 (0.357–1.055)	0.006*	1.883 (0.557–3.209)
2^nd ESR	0.009*	0.305 (0.080–0.530)	0.017*	−0.840 (−1.525 – −0.155)
CRP	<0.001*	1.702 (0.909–2.494)	0.466	0.441 (−0.764–1.647)
ANA	0.517	−6.883 (−28.016–14.251)
FBS	0.055	0.075 (−0.002–0.152)
HbA1c	0.254	1.734 (−1.280–4.747)
Fasting insulin	0.891	0.026 (−0.350–0.401)
HOMA-IR	0.919	0.032 (−0.601–0.666)
Total cholesterol	0.268	0.073 (−0.058–0.203)
HDL	0.173	0.612 (−0.275–1.500)
LDL	0.899	0.011 (−0.158–0.180)
TG	0.642	−0.023 (−0.123–0.077)
AST	0.093	−0.479 (−1.039–0.082)
ALT	0.298	−0.229 (−0.666–0.208)
GGT	0.211	0.321 (−0.188–0.830)
Serum uric acid	0.380	−1.754 (−5.722–2.213)
eGFR	0.253	0.145 (−0.106–0.396)
Liver size	0.927	0.303 (−6.259–6.865)
Echogenicity	0.504	−4.296 (−17.089–8.497)
Size of spleen	0.871	−1.562 (−20.700–17.577)

B: Unstandardized Coefficients.

C.I: Confidence interva lLL: Lower limit UL: Upper Limit.

^#:All variables with p < 0.05 was included in the multivariate.

*:Statistically significant at p ≤ 0.05.

Table 6. Univariate and multivariate Linear regression analysis for the parameters affecting MEG gene expression (n = 60) for NAFLD patient group (I + III).

	Univariate		^#Multivariate
	p	B (95%C.I)	p	B (95%C.I)
Female	0.180	−4.729 (−11.697–2.239)
Age (years)	0.922	0.017 (−0.337–0.372)
Smoker	0.860	0.958 (−9.852–11.767)
Hypertension	0.186	5.736 (−2.837–14.308)
Weight (kg)	0.921	−0.008 (−0.172–0.156)
Height (cm)	0.753	−0.061 (−0.448–0.325)
BMI (kg/m^2)	0.906	0.027 (−0.437–0.492)
Waist circumference (cm)	0.460	−0.082 (−0.302–0.138)
Hemoglobin	0.022*	2.987 (0.447–5.527)	0.039*	2.633 (0.142–5.124)
1^st ESR	0.139	−0.171 (−0.400–0.057)
2^nd ESR	0.122	−0.107 (−0.244–0.029)
CRP	0.321	0.101 (−0.100–0.301)
ANA	0.383	−5.677 (−18.595–7.241)
FBS	0.611	−0.012 (−0.058–0.034)
HbA1c	0.238	−1.076 (−2.882–0.730)
Fasting insulin	0.102	0.259 (−0.053–0.571)
HOMA-IR	0.382	−0.291 (−0.954–0.371)
Total cholesterol	0.682	−0.017 (−0.101–0.067)
HDL	0.047*	0.434 (0.006–0.862)	0.179	0.288 (−0.136–0.713)
LDL	0.482	−0.035 (−0.135–0.065)
TG	0.831	0.008 (−0.068–0.084)
AST	0.917	0.016 (−0.281–0.312)
ALT	0.563	0.074 (−0.181–0.329)
GGT	0.587	0.099 (−0.263–0.460)
Serum uric acid	0.945	−0.079 (−2.392–2.234)
eGFR	0.194	0.103 (−0.054–0.259)
Liver size	0.819	0.765 (−5.903–7.434)
Echogenicity	0.327	−4.947 (−14.968–5.073)
Size of spleen	0.545	2.639 (−6.037–11.316)

B: Unstandardized Coefficients.

C.I: Confidence interva lLL: Lower limit UL: Upper Limit.

^#:All variables with p < 0.05 was included in the multivariate.

*:Statistically significant at p ≤ 0.05.

Table 7. Univariate and multivariate Linear regression analysis for the parameters affecting MEG gene expression (n = 60) for diabetic patient group (II + III).

	Univariate		^#Multivariate
	p	B (95%C.I)	p	B (95%C.I)
Female	0.223	−3.047 (−8.001–1.908)
Age (years)	0.033*	−0.288 (−0.551 – −0.025)	0.065	−0.243 (−0.501–0.016)
Smoker	0.519	−2.164 (−8.845–4.516)
Hypertension	0.750	−0.885 (−6.409–4.639)
Weight (kg)	0.826	0.015 (−0.120–0.149)
Height (cm)	0.624	0.075 (−0.229–0.379)
BMI (kg/m^2)	0.883	0.025 (−0.310–0.359)
Waist circumference (cm)	0.739	−0.026 (−0.183–0.130)
Hemoglobin	0.466	0.600 (−1.038–2.237)
1^st ESR	0.286	−0.095 (−0.272–0.082)
2^nd ESR	0.337	−0.052 (−0.159–0.055)
CRP	0.322	−0.204 (−0.612–0.205)
ANA	0.371	−4.290 (−13.821–5.241)
FBS	0.981	0.0004 (−0.035–0.036)
HbA1c	0.409	−0.570 (−1.941–0.801)
Fasting insulin	0.871	−0.014 (−0.184–0.156)
HOMA-IR	0.798	−0.037 (−0.323–0.249)
Total cholesterol	0.021*	−0.067 (−0.124 – −0.010)	0.761	−0.016 (−0.124–0.091)
HDL	0.468	0.148 (−0.258–0.555)
LDL	0.011*	−0.095 (−0.167 – −0.023)	0.336	−0.067 (−0.205–0.071)
TG	0.150	0.032 (−0.012–0.077)
AST	0.574	0.073 (−0.186–0.333)
ALT	0.292	0.105 (−0.093–0.303)
GGT	0.140	−0.171 (−0.400–0.058)
Serum uric acid	0.660	0.399 (−1.406–2.203)
eGFR	0.138	0.085 (−0.028–0.197)
Liver size	0.327	−2.558 (−7.739–2.622)
Echogenicity	0.270	−3.196 (−8.946–2.554)
Size of spleen	0.766	−1.295 (−9.950–7.361)

B: Unstandardized Coefficients.

C.I: Confidence interva lLL: Lower limit UL: Upper Limit.

^#:All variables with p < 0.05 was included in the multivariate.

*:Statistically significant at p ≤ 0.

Furthermore, NAFLD patients (groups I and III together) displayed significantly higher MALAT1 and MEG3 gene expression when compared with the study-involved controls (p = 0.037 and 0.007, respectively). Additionally, compared to the control, only MALAT1 expression, but not MEG3 gene expression, was significantly higher among diabetic patients (groups II and III together) (p = 0.033). (Supplementary data)

3.1. Validity of MALAT1 and MEG 3 genes expression to discriminate between patients and controls

In Figure 1, the receiver operating curve (ROC) shows the cutoff value for MALAT1 expression to diagnose NAFLD patients was > 0.54 with an 0.635 detected area under the curve (AUC),71.67% sensitivity, and 53.33% specificity. The cutoff value for MEG 3 expression to diagnose NAFLD patients was > 1.197 with an 0.676 AUC and 61.67% sensitivity and 73.33% specificity.The cutoff value for MALAT1 expression to distinguish between diabetic patients from the control group was also determined by the ROCcurve and the cutoff was > 0.67 with 0.639AUC, 70% sensitivity,and 56.67% specificity.

Figure 1. Right figure shows ROC curve for MALAT1 and MEG3 genes expression to discriminate diabetic patient (groups II & III) from controls while left figure shows ROC curve for MALAT1 and MEG3 genes expression to discriminate NAFLD patient (groups I & III) from controls.

3.2. Linear regression analysis

In univariate analysis, ESR and CRP were significant predictors of MALAT1 gene expression in both NAFLD (group I & III), (Table 4) and diabetic (group II & III), (Table 5) patients, and FBS, and GGTin NAFLD patients. In multivariate analysis, ESR only was associated with MALAT1 gene expression and it was the only reliable indicator of MALAT1 gene expression in patients with NAFLD and diabetes. Also, in univariate analysis, HDL and Hb were significant predictors of MEG3 gene expression in NAFLD (group I & III), (Table 6) and age, total cholesterol, and LDL in diabetic (group II & III), (Table 7) patients. In multivariate analysis, Hb only was associated with MEG3 gene expression in NAFLD patients.

3.3. Correlation study

A detailed correlation analysis of MALAT1 and MEG3 expression was done with numerous clinical and biochemical parameters of the study participants and was summarized in the supplementary Tables . In diabetic individuals (groups II & III), MALAT1 gene expression was correlated positively with eGFR and inversely correlated with ESR. Additionally, in controls (group IV), MALAT1 expression had a positive correlation with HbA1 and a negative correlation with AST.The MEG3 expression was inversely correlated with age in diabetic patients (groups II & III), ESR in NAFLD patients (groups I & III), and total cholesterol and LDL in diabetic without NAFLD patients (group II). (Tables 1:7)

4. Discussion

NAFLD affects up to 20% of the general population and may be up to 90% of those adults with obesity. Steatosis, hepatitis, and fibrosis are among the vast range of lesions that NAFLD may cause different degrees of these three types of lesions might develop in the same patient. Fibrosis may proceed to cirrhosis and HCC in 20% and 10% of cases, respectively [15]. NAFLD is most prevalent in the Middle East and South America, where as it is least prevalent in Africa. The following: obesity (51.34%), T2DM (22.51%), dyslipidemia (69.16%), hypertension (39.34), and metabolic syndrome (42.54%) were all among the metabolic comorbidities linked with NAFLD [16].

There is a clear association between both NAFLD and diabetes risk. If a person has NAFLD, their chances of acquiring diabetes rise by around 5-fold. The NAFLD and T2DM association may be described by insulin resistance, dyslipidemia, and hepatic TG buildup in NAFLD, as well as a dysfunctional B-cell among those with T2DM. People with T2DM are more likely than healthy people to develop significant liver diseases such as cirrhosis, fibrosis, and HCC [17].

The major study purpose was to evaluate the NAFLD-related serum molecular profile of T2DM patients and healthy controls using (RT-qPCR) and to identify a disease-specific pattern that can be validated for early detection of NAFLD in T2DM, thereby providing a potential screening and diagnostic tool.

In the current investigation, all diabetes-diagnosed patients (groups II and III) had substantially higher BMI and waist (p = 0.006 and < 0.001 respectively). According to Prasad and Gupte S. et al, [18] who studied 30 patients with T2DM and 32 controls, patients diagnosed with diabetes had significantly greater BMI and waist circumference than controls (p = 0.015 and p= <0.001, respectively).

In this study, there were no statistically significant differences in lipid profile, functions of the liver, and uric acid across all groups. In contrast to our findings, Shrestha N, and Bhatt NP et al. [19] studied 100 patients, 62 (62%) male, and 38 (38%) female. Based on their ALT levels, the patients were subdivided forming two groups: those with normal ALT levels and others with elevated levels of ALT, and it was discovered that NAFLD patients reported significantly increased levels of ALT (p = 0.001) but no significant rise in AST or GGTlevels. The difference between their study and ours may be attributed to different inclusion criteria and ethnic populations.

A lot of studies show that diabetics have considerably higher levels of transaminases than non-diabetics, especially with NAFLD [20,21]. In contrast to our findings, Shrestha N and Bhatt NP et al [19] discovered a significant increase statistically in both serum ALT and GGT among diabetic patients in comparison to the involved controls in their study which include 105 patients newly diagnosed with T2DM without a liver disease history, with p = 0.003 and p = 0.001, respectively. The variation between their and our investigations might be attributed to differences in inclusion criteria and ethnic populations.

Long non-coding RNAs (lncRNAs) are increasingly becoming recognized as contributors to biological processes that influence disease pathophysiology in humans. For many years, researchers have been studying the lncRNAs’ role in obesity, T2DM, and related comorbidities such as NAFLD development [22]. In the current investigation, there was a relation of statistical significance among MALAT1, MEG3 gene, and T2DM. Expression of the MALAT1 gene was substantially greater among patients diagnosed with diabetes (groups II&III) than among those control-involved participants (p = 0.033). Also, group II (diabetic patients only) showed higher MEG3 gene expression than controls (p = 0.043). In agreement with our study, Prasad and Gupte S. et al. [18] discovered that patients with diabetes had considerably higher MALAT1 and MEG3 levels of gene expression than controls. (p < 0.05).

The expression of MALAT1 gene in this investigation was increased significantly among all patients diagnosed with NAFLD (group I&III) than among the controls (p = 0.037) and the expression of MEG3 was significantly increased among the NAFLD diagnosed groups (group I&III) than those considered as controls (p = 0.037). Also, the MEG3 expression was significantly increased among those involved in group I (NAFLD) than in group III (NAFLD with diabetes) (p = 0.033).

In agreement with our study, Yu F and Lu Z et al. [23] revealed that MALAT1 gene expression was upregulated in both liver fibrotic tissues and hepatic stellate cells (HSCs) which are activated. There was also a negative loop of feedback between levels of MALAT1 and also miR-101b. Different inclusion criteria may account for the disparity between their study and ours. Kim Ya and Park KK et al. [24] showed that MALAT1 improves hepatic steatosis and insulin resistance by enhancing the SREBP-1c protein nuclear stability. Also, Leti F and Legendre C et al. [25] found that MALAT1 expression upregulation may contribute to NAFLD development via processes involving inflammatory chemokines. In contrast to our study, Khalifa O and Errafi K et al. [26] revealed that MEG3 expression was decreased in correlation with the NAFLD progression. The difference between their study and ours may be attributed to different inclusion criteria, ethnic populations, and sample size.

5. Conclusion

We concluded that MALAT1 expression was upregulated in NAFLD and diabetic patients which suggests the possibility to be an early diagnostic marker for both diseases that helps in the development of novel therapeutic agents. In NAFLD and diabetic individuals, ESR was revealed to be the sole significant predictor of MALAT1 gene expression. Furthermore, the expression of MEG3 was greater among NAFLD patients than those having NAFLD with T2DM, implying that decreasing MEG3 expression might be considered a possible predictive biomarker for early T2DM diagnosis in NAFLD patients.

Biographical state

Please include a brief biography for each author. This might be a brief (no more than 200 words) copy of your department’s website or an academic networking page.

Disclosure statement

The authors declare that there are no stated conflicts of interest.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/20905068.2023.2251221

Additional information

Notes on contributors

Eman Yousef Moursi

Eman Yousef Moursi Professor at Internal Medicine department, Faculty of Medicine, Alexandria University. Dr Eman has experience and multiple publications in the field of Internal medicine and Diabetes.

Pacint Elsayed Moez

Pacint ElSayed Moez Professor of Clinical Pathology Faculty of Medicine Alexandria University. Professor Moez is also the Director of Medical Research Centre and Executive Director of Proteomics Lab.

Ehab Mostafa Hassouna

Ehab Mostafa Hassouna Professor of Internal Medicine, Faculty of Medicine, Alexandria University, Egypt. Prof Ehab has experience and multiple puplications in the field of Internal Medicine and Hepatobiliary diseases.

Eman Zakareya Elkemary

Eman Zakareya ElKemary Lecturer at Clinical and Chemical Pathology department, Faculty of Medicine, Alex University. Dr. Eman has experience and multiple publications in the field of clinical chemistry and molecular biology.

Geilan Mahmoud Marei

Geilan Mahmoud Marei Assistant Colleague of Internal Medicine department and Specialist in Clinical Nutrition, Faculty of Medicine, Alexandria University, Egypt. Dr Geilan has experience in the field of Internal Medicine and Clinical Nutrition.

Mohammed Said Sayed Shater

Mohammed Said Sayed Shater Lecturer of Internal Medicine, Faculty of Medicine, Alexandria University, Egypt. Dr Mohammed has experience and multiple puplications in the field of Internal Medicine and Hepatobiliary diseases.

Mai Hesham Mohammed Badrah

Mai Hesham Mohammed Badrah Assistant Professor at Internal Medicine department, Faculty of Medicine, Alexandria University. Dr Mai has experience and multiple publications in the field of Internal medicine and Diabetes.