This article is mentioned in:

Introduction

Recently Homo sapiens longevity assurance homolog 2 of yeast LAG (Lass2) has attracted the interest of researchers since a large amount of evidence has demonstrated that Lass2 is a potential tumor-suppressor gene. Deficiency of Lass2 is involved in the tumorigenesis of various types of tumors, especially hepatocellular carcinoma (HCC) (1–4). It has been reported that two of three non-specific Lass2-deleted mice presented liver cancer spontaneously when they were about nine months old (3). In our recently published study, using a hepatocellular-specific Lass2-knockout (KO) animal model, we found that Lass2-KO mice were more susceptible to the carcinogen DEN, i.e., DEN caused the Lass2-KO mice to develop liver tumors earlier and the tumors developed more rapidly (5).

To explore the biological functions of Lass2 and the related mechanisms it employs to suppress HCC, the structures and functions of the livers of the Lass2-KO mice and wild-type (WT) mice were compared. Meanwhile, microarrays of mRNAs and miRNAs of the livers from the two genotypes were performed and analyzed.

Materials and methods

Animals

The hepatocyte-specific Lass2-KO mice used in the present study were generated by the crossing of mice (C57BL/6J) carrying floxed the second exon of Lass2 and Albumin-Cre transgenic mice (C57BL/6J), as previously reported (5). All protocols for animal care and use were approved by the Regulations for the Administration of Affairs Concerning Experimental Animals (The State Science and Technology Commission of P.R. China, 1988), and the Animal Experimental Center of Jiangsu University was licensed for animal experiments. All of the mice were housed in pathogen-free (SPF) animal facilities under a standard 12-h-light/12-h-dark cycle. Animals received free access to water and commercial mouse chow throughout the present study. One-month-old male Lass2-KO and -WT mice were sacrificed by cervical dislocation. The mice for the experiments that followed were n=10/group. Liver weight/body weight were measured.

Biochemical estimation

The mice were sacrificed by cervical dislocation, and blood harvested from the inferior vena cava was centrifuged at 1,600 × g for 10 min at room temperature to obtain serum for hepatic biochemical estimation. The activities of aspartate aminotransferase (AST), alanine aminotransferase (ALT) and lactate dehydrogenase (LDH) in serum were estimated using an AutoAnalyzer (Hitachi 7600, Japan) at the Affiliated Hospital of Jiangsu University.

Histological sections and staining

Liver tissues were immediately removed from the sacrificed mice, partly fixed in AAF (100% alcohol 85 ml, acetic acid 5 ml, formalin 10 ml) for morphological examination, and partly stored at −80°C for further use. The tissues were paraffin-embedded and sectioned (5-μm thick). Periodic acid-Schiff (PAS) staining was performed using a commercially available kit (cat. no. 0609A14; Shellfish Gamma Biotechnology Co., Ltd., Nanjing, China). Briefly, the sections were incubated in 0.5% periodic acid solution for 15 min, rinsed with distilled water, and exposed to Schiff’s reagent for 20 min followed by two 3-min exposures to 0.6% sodium metabisulfite.

Western blotting

Tissues were lysed in RIPA lysis buffer containing the protease inhibitor phenylmethanesulfonyl fluoride (cat. nos. P0013C and ST506; both from Beyotime Institute of Biotechnology). The cell extracts were centrifuged at 12,000 × g for 20 min at 4°C in a Beckman Avanti-30 centrifuge, and the supernatants were used for the experiments. The protein concentrations were determined with the BCA assay kit (cat. no. P0009; Beyotime Institute of Biotechnology). The equivalent tissue proteins (10 μg/lane) were subjected to electrophoresis on a Mini-Protean Tetra Electrophoresis System (165–8001; Bio-Rad, Hercules, CA, USA) and transferred onto PVDF membranes (Millipore, Bedford, MA, USA) via a semi-dry transfer system (Bio-Rad). The membranes were blocked with 5% non-fat milk for 1 h at room temperature in TBST [50 mM (pH 7.5) Tris, 0.9% NaCl and 0.1% Tween-20] and then incubated with a 1:200 dilution of rabbit anti-mouse/human albumin (cat. no. BS6520; Bioworld Technology, Inc., Dublin, OH, USA) overnight at 4°C. The membranes were washed and then incubated with peroxidase goat anti-rabbit antibody (cat. no. XR-9920; ProSci Inc., Poway, CA USA) for 1 h at room temperature and developed using the BeyoEcl Plus kit (cat. no. P0018; Beyotime Institute of Biotechnology) for 1 min, and then scanned by ChampChemi professional (SG2011; Beijing Sage Creation Science Co., Ltd., Beijing, China).

RNA isolation and qPCR analysis

Total RNA including miRNA from liver tissues was extracted by TRIzol (cat. no. 15596026; Invitrogen Corp., Carlsbad, CA, USA) or the miRNeasy Mini kit (cat. no. 217004; Qiagen, Hilden, Germany) according to the manufacturer’s suggestions. For mRNA qPCR, RNA was transcribed into cDNA using the QuantiTect reverse transcription kit and QuantiTect SYBR-Green PCR kits (cat. no. 205311 and no. 204243; both from Qiagen) according to the manufacturer’s protocols. For miRNA qPCR, RNA was transcribed into cDNA using the miScript Reverse II transcription kit (cat. no. 218161; Qiagen). The reaction component consisted of total RNA 1 μg, miScript HiSpec buffer 4 μl, Nucleics Mix 2 μl, miScript reverse transcriptase mix 2 μl, RNase-free H2O up to 20 μl. The reaction was carried out at 37°C for 60 min and 95°C for 5 min on the ABI PCR 9700 system (Applied Biosystems, Foster City, CA, USA). cDNA was diluted in 80 μl nuclease-free H2O for further application by LightCycler 480 SYBR-Green I master (Roche, Switzerland; cat. no. 04887352001). The reaction system for qPCR consisted of: LightCycler 480 SYBR-Green I Master 5 μl, forward primer 0.2 μl, reverse primer 0.2 μl, cDNA 1 μl, nuclease-free H2O 3.6 μl. PCR was run on ABI 7500 Fast (Applied Biosystems) and normalized against the expression of GAPDH or U6. The program was performed at 95°C for 10 min, 95°C for 10 sec plus 60°C for 30 sec for 40 cycles. For the melting curve evaluation, the temperature was slowly increased from 60°C to 97°C, and 5 acquisitions per °C were performed continuously. All samples were analyzed in triplicate. Relative expression was calculated using the comparative threshold cycle (Ct) method and was indicated as n-fold change = −(ΔCtKO − ΔCtWT). The primers for Tnfaip3, NF-κB and GAPDH were: for Tnfaip3, 5′-CAGCACCTAAG CCAACGAGT-3′ and 5′-TGGACCTGTCAATGTGTTCG-3′; for NF-κB, 5′-AGCTTATGCCGAACTTCTCG-3′ and 5′-GAC TCCGGGATGGAATGTAA-3′; for GAPDH, 5′-GCAAGG TCATCCCAGAG-3′ and 5′-AAGTCGCAGGAGACAAC-3′; for miR-694 and U6, the miRNA specific primers were respectively: 5′-CTGAAAATGTTGCCTGAAG-3′ and 5′-CAAGG ATGACACGCAAATTCG-3′, whereas the reverse primer was the manufacturer-provided miScript universal primer.

Microarrays of mRNA and miRNA

Liver tissues from 2 one-month male Lass2-KO mice and 2 one-month male WT mice were respectively subjected to the experiments of Mouse OneArray (MOA 2.0 ver.) and Mouse and Rat miRNA OneArray (MRmiOA 4.0 ver.) using Phalanx microarray platform by Oebiotech Co. (Shanghai, China). Briefly, RNA was extracted from liver tissues with TRIzol reagent (Invitrogen), whose quantity and purity were assessed using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies Inc., Wilmington, DE, USA). The purity and integrity of each extracted RNA met the requirements: A260/A280 >1.6, A260/A230 >1 and RNA integrity number (RIN) value >5. Small RNA fraction indicated the abundance of RNA <200 nt compared with the overall RNA area from Agilent RNA 6000 Nano assay and was acceptable for the miRNA assay. Possibility of genomic DNA contamination was excluded by gel electrophoresis. Two micrograms of RNA from each group was respectively converted into cyanine-5 labeled target cRNA, hybridized to either Mouse OneArray or Mouse and Rat miRNA OneArray by the Affymetrix GeneChip fluidics station 450, and scanned with an Affymetrix GeneChip scanner 3000 7G. After normalization, differentially expressed mRNAs or miRNAs were established at log2 |Fold change| >1 and P<0.05.

For advanced data analysis, all biological replicates were pooled and calculated to identify differentially expressed genes based on the threshold of fold change and the P-value. The correlation of expression profiles between biological replicates and treatment conditions was demonstrated by unsupervised hierarchical clustering analysis. A subset of genes was selected for clustering analysis. An intensity filter was used to select genes where the difference between the maximum and minimum intensity values exceeded 4,000 among all microarrays. For this microarray project, the number of genes clustered was 272 genes. According to previously selected differentially expressed gene lists, Gene Ontology (GO) analysis was performed by Oebiotech Co. Targetscan 5.1 was utilized to predict miR-targeting mRNA. mRNA-miRNA integration analysis demonstrated potential mRNA targets with inverse expression alterations as their regulatory miRs displayed in the mRNA microarray or miRNA microarray.

Statistical analysis

Data were analyzed by the Student’s t-test. P<0.05 was considered to indicate a statistically significant difference, and P≤0.01 and P≤0.001 are indicated by relevant symbols in the figures and legends. qPCR and western blot analyses were repeated three times.

Results

Structural alterations in the liver tissues from hepatocyte-specific Lass2-KO mice were noted in the PAS-stained sections

The average ratio of liver weight/body weight of the Lass2-KO mice was higher than the ratio in the control WT mice (Fig. 1A). Compared to the liver tissues from the WT mice, the hepatocytes of the Lass2-KO displayed abundant vesicles (Fig. 1B).

Figure 1 Representative image of structural alterations in the liver tissues from a hepatocyte-specific Lass2-KO mouse. (A) The average ratio of the liver weight/body weight of the Lass2-KO mice was higher than this ratio in the control WT mice. (B) Compared to the liver tissues from the WT mice, hepatocytes of the Lass2-KO displayed abundant vesicles. Scale bar, 20 μm. WT, wild-type; KO knockout. ***P≤0.001.

Liver functions and the expression of ALB are altered in Lass2-KO mice compared with the WT control

The production of ALB in the liver was attenuated in the Lass2-KO mice (Fig. 2A). The levels of ALT, AST and LDH in serum were respectively elevated in the Lass2-KO mice, indicating abnormal liver function in the Lass2-KO mice (Fig. 2B–D), in accordance with the structural injury of the Lass2-KO hepatocytes.

Figure 2 Alterations in the liver function of hepatocyte-specific Lass2-KO mice. (A) Expression of ALB was attenuated in the Lass2-KO mice. Levels of (B) ALT, (C) AST and (D) LDH in serum were respectively elevated in the Lass2-KO mice. *P<0.05, **P≤0.01 and ***P≤0.001.

Profiles of mRNAs and miRNAs in the liver tissues are respectively reprogrammed in Lass2-KO mice vs. WT control

mRNA and miRNA microarray results showed that over 600 mRNA were markedly upregulated and over 700 genes were downregulated; whereas 13 miRNAs were markedly upregulated and 31 miRNAs were downregulated (Table I). According to the GO analysis, ‘response to wounding’ and ‘inflammatory response’ were among the top-10 altered pathways (Table II). miRNA-mRNA integrated analysis identified 4 upregulated and 3 downregulated miRNAs and their respective negatively controlled target genes (Table III).

Table I Altered miRNAs in the Lass2-KO mouse liver tissues compared to the WT mouse liver tissues.

Table I

Altered miRNAs in the Lass2-KO mouse liver tissues compared to the WT mouse liver tissues.

Upregulated miRNAs	Downregulated miRNAs
mmu-miR-1198-5p	mmu-miR-1192	mmu-miR-467d-3p
mmu-miR-125b-5p	mmu-miR-1224-5p	mmu-miR-467f
mmu-miR-142-3p	mmu-miR-1247-3p	mmu-miR-483-5p
mmu-miR-199a-3p	mmu-miR-149-3p	mmu-miR-494-3p
mmu-miR-199b-3p	mmu-miR-1894-3p	mmu-miR-5109
mmu-miR-199a-5p	mmu-miR-1931	mmu-miR-5136
mmu-miR-199b-5p	mmu-miR-1934-3p	mmu-miR-669a-3p
mmu-miR-2137	mmu-miR-30c-1-3p	mmu-miR-669o-3p
mmu-miR-2861	mmu-miR-320-3p	mmu-miR-669c-3p
mmu-miR-326-5p	mmu-miR-346-3p	mmu-miR-669p-3p
mmu-miR-34a-5p	mmu-miR-466f-3p	mmu-miR-694
mmu-miR-491-3p	mmu-miR-466h-3p	mmu-miR-711
mmu-miR-5130	mmu-miR-466i-3p	mmu-miR-744-5p
	mmu-miR-466q	mmu-miR-762
	mmu-miR-467b-3p	mmu-miR-882
		mmu-miR-92a-2-5p

Table II Top 10 enrichment pathway terms from the Gene Ontology (GO) analysis (biological process).

Table II

Top 10 enrichment pathway terms from the Gene Ontology (GO) analysis (biological process).

Gene set name	Genes in overlap	P-value
Response to wounding	10	0.0407
Humoral immune response	3	0.0588
Homeostasis of number of cells	2	0.112
Viral genome replication	2	0.122
Locomotory behavior	5	0.122
Inflammatory response	6	0.148
Development of primary sexual characteristics	2	0.172
Response to external stimulus	12	0.172
Jak-Stat cascade	2	0.225
Viral infectious cycle	2	0.235

Table III miRNA-mRNA integrated analysis of the Lass2-KO mouse liver tissues.

Table III

miRNA-mRNA integrated analysis of the Lass2-KO mouse liver tissues.

A, Upregulated miRNAs and their downregulated predicted target genes
mmu-miR-2861	mmu-miR-5130	mmu-miR-142-3p	mmu-miR-125b-5p
Lrrc41	Osbpl7	Atp2a2	Fam116a
Srm		Baz1a	Ier3ip1
		Rras	Map2k7
		Mastl	Fam78b
		Arl15	Slc17a7
		Lifr	Rasal2
		Tgfb2
		Myst2
B, Downregulated miRNAs and their upregulated predicted target genes
mmu-miR-1192	mmu-miR-466f-3p		mmu-miR-694
Tcf4	Pgm2l1	Col4a1	Mef2c
Ap3m1	Tm6sf1	Junb	Ncoa7
Sgk1	Angptl2	Cttnbp2nl	Ska1
Scd2	Prc1	2700081O15Rik	Gng2
Adam23	Smad7	Ptgfrn	Abi2
Elavl4	Klf6	Ncoa7	Nipal1
Rbm28	Trp53inp1	Zfp532	Tnfaip3
Phf15	Tmem65	Samd4	Fam120c
Arrdc4	Dgkd	Prrg3	Srebf1
Slc16a5	Fam120c	Zmat3	Man1c1
Arhgef17	Gpm6b	Ptp4a1	Vldlr
Zfp532	Nr4a1	Adam23	Fbln5
Nrxn2	Eif4enif1	Tcf4	Rai2
Ccna2	Mef2c	Jun	Prrg3
	Fam46a	Pde4d	Fam46a
	Elavl4	Tnfaip8	Plekha6
	Odz3		E130203B14Rik
			Nup153
			Trib1
			Cd63
			Pde5a
			Serpinh1 (PAI-1)
			Sgms1
			Elavl4
			Pde4d

qPCR confirms that miR-694 is markedly downregulated

The level of miR-694 was downregulated in the Lass2-KO liver tissue, as confirmed by qPCR (Fig. 3).

Figure 3 qPCR confirmed that miR-694 was markedly downregulated. The level of miR-694 was downregulated in the Lass2-KO liver tissues, as confirmed by qPCR. WT, wild-type; KO, knockout. **P≤0.01.

Tnfaip3 is markedly upregulated whereas NF-κB is downregulated

Tnfaip3, one of the markedly upregulated mRNAs in the Lass2-KO liver tissues, which is also a putative target gene of miR-694, was confirmed by qPCR (Fig. 4A). Its negatively controlled NF-κB was found to be downregulated (Fig. 4B).

Figure 4 Tnfaip3 was markedly upregulated and NF-κB was downregulated in the Lass2-KO liver tissues. (A) qPCR assay showed that the mRNA level of Tnfaip3 was markedly upregulated. (B) mRNA level of NF-κB, which is commonly considered to be negatively regulated by Tnfaip3, was found to be downregulated. **P≤0.01 and ***P≤0.001.

Discussion

Lass2 is a member of the Lass family, which is conserved among eukaryotes and is abundantly distributed in the liver, kidney and brain. The major function of Lass2 is synthesizing long-chain ceramide-C:24–26 (6). Ceramide serves as the precursor of a series of more complex sphingolipids. Short-chain ceramides function as a second messenger in a variety of cellular events, including apoptosis and differentiation (7,8), and regulate various cellular processes linked to cancer development, progression, metastasis and resistance to therapy (9,10). By comparison, long-chain ceramide, as the important element in constructing membranous structures, may regulate the cellular behavior via influencing the property of membranes. For example, ablation of Lass2 causes morphological alterations of the property of membranes (11,12). In the present study, numerous vesicles were noted in the hepatocytes from young Lass2-KO mice (Fig. 1), in accordance with previous reports from other researchers. Based on the attenuated production of albumin and increased hepatic biochemical indices in the Lass2-KO mice, (Fig. 2), it appears that the hepatocellular-specific Lass2-KO mice underwent hepatocellular injury even at an early age.

The microarrays of mRNAs and miRNAs of the Lass2-KO mice liver tissues vs. control demonstrated that miR-694 was upregulated, which was confirmed by qPCR. Moreover, the predicted target gene Tnfaip3 was found to be upregulated, as shown in the results of either the microarray of mRNA or qPCR. NF-κB, which is usually commonly negatively controlled by Tnfaip3 was found downregulated in the Lass2-KO mouse liver tissues. In our previous study (5), another target gene Serpinh1 (PAI-1) of miR-694 was found to be upregulated. The data strongly suggest that Lass2 deletion influences the expression level of miR-694.

However, the function of miR-694 is uncertain. According to miRNA-mRNA integrated analysis in this study, the down-regulated Lass2-related miR-694 elevated 25 mRNAs including Tnfaip3. Tnfaip3 is commonly considered as an inflammation suppressor (13) by inactivation of lymphocytes of suppressing NF-κB (14,15). Although it is commonly considered a tumor suppressor, overexpression of Tnfaip3 has also been reported in several non-lymphoma solid cancers, including HCC (16–18). Our data suggest that Tnfaip3/NF-κB might play a role in inhibiting the inflammation and protecting injured hepatocytes caused by deletion of Lass2. In another report, Lass2-KO mice displayed resistance to LPS-induced liver injury (19), which might also be explained by the inhibitory immunity mediated by Tnfaip3/NF-κB.

The functions of other predicted target genes of miR-694 are listed in Table IV. According to published reports, miR-694 target genes are involved it the regulation of various important cellular events, including transcription, cell signal transduction, metabolism of lipid and glucose and trafficking, suggesting a diversity of regulatory functions of miR-694. miRNAs act as highly effective regulators of intracellular events. Moreover, via endocytosis or exocytosis of miRNA-containing vesicles, miRNAs modulate the extracellular milieu including stromal cells and extracellular matrix (20). For example, altered Tnfaip3/NF-κB may influence the immunocytes in the liver, or PAI-1 which was found to be upregulated, may regulate the synthesis of ECM in the liver (5), either of which is a predicted target gene of miR-694. The actual functions of miR-694 may likely be beyond what is listed in Table IV.

Table IV Functions of the miR-694 predicted target genes.

Table IV

Functions of the miR-694 predicted target genes.

Function	miR-694 target gene (ref.)
Transcription	Mef2c (21)
	Ncoa7 (22)
	Tnfaip3 (13,14)
	Elavl4 (23,24)
Cell cycle	Ska1 (25)
	Nup153 (26)
Cell signal transduction	Gng2 (27)
	Abi2 (28,29)
	Pde5a (30)
	Pde4d (31)
Metabolism regulator	Trib1 (32–34)
	Sgms1 (35–37)
	Srebf1 (38)
	Vldlr (39)
Trafficking	CD63 (40,41)
	Man1c1 (42)
	Nup153 (26)
	Sgms1 (43)
ECM deposition	Fbln5 (44,45)
	PAI-1 (46)
Immunity regulation	Trib1 (47,48)
	Tnfaip3 (13,14)

Overall, the present study first reports the attenuated expression level of miR-694 in Lass2-KO mouse liver tissue and its alteration of the Tnfaip3/NF-κB pathway. Our data strongly suggest that miR-694 functions in maintaining homeostasis of the liver and provide the basis to explore the functions of Lass2-related microRNAs.

Acknowledgements

This study was supported by the Grant from the State Key Laboratory of Oncogenes and Related Genes (no. 90-10-02, to X.L.) and Clinical Medicine Science & Technology Project of Jiangsu Province of China (no. BL2013024).

Abbreviations:

References