Effect of a common missense variant in LIPA gene on fatty liver disease and lipid phenotype: New perspectives from a single‐center observational study

Abstract Lysosomal acid lipase deficiency (LAL‐D) is an autosomal recessive disease characterized by hypoalphalipoproteinemia, mixed hyperlipemia, and fatty liver (FL) due to mutations in LIPAse A, lysosomal acid type (LIPA) gene. The rs1051338 single‐nucleotide polymorphism (SNP) in LIPA gene, in vitro, could adversely affect the LAL activity (LAL‐A). Nonalcoholic fatty liver disease (NAFLD) is often associated with metabolic syndrome, and the diagnosis requires the exclusion of excess of alcohol intake and other causes of hepatic disease. The aim of the study was to evaluate the impact of rs1051338 rare allele on lipid phenotype, severity of FL, and LAL‐A in patients suffering from dyslipidemia associated with NAFLD. We selected 74 subjects with hypoalphalipoproteinemia or mixed hyperlipemia and evaluated transaminases, liver assessment with controlled attenuation parameter (CAP), LAL‐A, rs1051338 SNP genotype. The presence of rare allele caused higher levels of triglycerides and hepatic transaminase and lower levels of high‐density lipoprotein cholesterol (HDL‐C). Multivariate analysis highlighted independent association between rare allele and FL severity in subjects with NAFLD. The rs1051338 SNP may modulate FL severity and atherogenic dyslipidemia in patients suffering from NAFLD.


| INTRODUC TI ON
Lipase A, lysosomal acid type (LIPA) gene encodes for lysosomal acid LIPAse (LAL), an enzyme that is able to catalyze triglycerides (TG) and cholesterol esters hydrolysis in lysosomes. Lysosomal acid LIPAse deficiency (LAL-D) is an autosomal recessive disease characterized by hypoalphalipoproteinemia, mixed hyperlipemia and hepatic steatosis. 1,2 The rs1051338 polymorphism is a common variant of the LIPA gene with a global frequency of the minor allele (GMAF) of 0.25010.
The variant is located at the exon 2 level and involves the replacement of adenine with a cytosine in position 46 (ACC>CCC). A recent in silico and ex-vivo study 3 has reported that this nucleobases substitution causes an aminoacid change at a translational level: 16-threonine (Thr, T) with 16-proline (Pro, P) at the level of the signal peptide of the protein. Consequently, the signal peptide of LAL-Pro is reduced of two aminoacids compared with that of LAL-Thr, causing an early truncation of the central core and a displacement of two amino acids at the cleavage site. As a result, the change in the signal peptide of LAL disrupts the normal sorting and transport of LAL to the lysosome. The polymorphic LAL is susceptible to cytosolic proteasomal degradation, thus reducing lysosomal LAL protein and activity (LAL-A). 3 Moreover, the rs1051338 polymorphism in LIPA gene has been previously associated by disequilibrium linkage to increased risk of atherogenic dyslipidemia, metabolic syndrome, obesity, and cardiovascular disease. 3,4 Non-alcoholic fatty liver disease (NAFLD) is the presence of fat in the liver (FL) after the exclusion of alcohol excess or other causes of liver disease: NAFLD pathogenesis is multifactorial and influenced by presence of metabolic syndrome or genetic polymorphisms. 5 LAL-A reduction was observed in patients with NAFLD 6 and, in the metabolic-molecular prospective, LAL impairment results in reduced hepatocyte effectiveness in processing lipids, which, consequently, are accumulated in the liver. 1 The rare allele (c. 46C, p.16P) of rs1051338 has been previously associated with LAL-A slight deficiency and could be linked to NAFLD pathogenesis. Therefore, the LAL-A reduction causes the activation of the transcription factors of the sterol regulatory element-binding protein (SREBP) family with consequent activation of lipogenesis which results in an alteration of the release of cholesterol and free fatty acids in the cytosol. 7 This mechanism supports the increase in intracellular fatty acids causing the development and progression of NAFLD. 8 On the other hand, the cytosolic accumulation of fatty acids and triglycerides typical of NAFLD can cause a reduction of LAL-A by a negative feedback regulation. 9 The clinical characterization of the rs1051338 polymorphism can provide useful prognostic data as well as a better diagnostic classification. The rs1051338 polymorphism may be consider as other polymorphisms whose association with NAFLD was already established. 10,11 The study of genetic background of fatty liver and NAFLD could lead to better therapeutic strategies soon.
We suggest the hypothesis that the presence of rs1051338 polymorphism may exacerbate the pathological phenotype of patients suffering from dyslipidemia and hepatic steatosis associated with NAFLD. Thus, the aim of the study was to evaluate the impact of rs1051338 rare allele on lipid phenotype, severity of FL, and LAL-A in patients suffering from dyslipidemia associated with NAFLD.

| Study design
The present is an observational study designed to test the influence of the rs1051338 polymorphism in LIPA gene on lipid phenotype, liver assessment, and LAL-A.
Inclusion criteria were established to screen low-moderate LAL-D and were obtained by an adaptation of a model described in the literature, 12 which were originally designed to identify geneticdetermined LAL-D. Low-density lipoprotein cholesterol levels were not considered as inclusion criteria to avoid the risk of underestimation due to the presence of high TG level typical of mixed LAL-D-like dislipidemia. Conversely, low HDL-C levels were used as a main inclusion criterion because they are strongly associated with LAL-D. 2,[13][14][15][16][17] Thus, patients included had high-density lipoprotein cholesterol (HDL-C) <50 mg/dl and, at least, one of the following characteristics: alanine aminotransferase (ALT) or aspartate aminotransferase (AST) >1.5 maximum laboratory value or TG >150 mg/dl or hepatomegaly or hepatic steatosis in the context of NAFLD.
Exclusion criteria were age <18 years and >80 years, acute or chronic artery or heart disease, chronic and acute liver disease with different diagnosis than NAFLD, acute and chronic nephropathies,

| Genetic analysis
Genomic DNA was extracted from peripheral blood leukocytes with the standard techniques. 18 Genetic analysis was performed by restriction fragment length polymorphism analysis for rs1051338 of LIPA gene which was previously described. 19

| LAL activity determination
Blood LAL activity (nmol/spot/h) was measured with DBS extracts using the inhibitors Lalistat2. Peripheral blood was collected with ethylene-diamine-tetra acetic acid blood, spotted on to filter paper, and allowed to dry overnight at room temperature. Samples were stored double-bagged with desiccant at −20°C and analyzed within 2 weeks of storage. Uninhibited and inhibited with Lalistat 2 (Chemical Tools, South Bend, IN, USA) activities were dosed. LAL activity was determined by subtracting activity in the inhibited reaction from uninhibited reaction (total lipase) and expressed as nmol/ spot/h of methylumbelliferone. 20 Inter and intra-assay variations were 2.4 and 2.3%, respectively. measures were up to 10 and acceptability criteria were IQR/median <30% and success rate >60%. 25 According with the literature, [26][27][28] in Table S1, CAP, VCTE, and SWE are reported the cutoff values used for steatosis and fibrosis staging.

| Liver assessment and results ranking
Patients were also divided depending on early versus advanced stages of steatosis as follow: S0, S1, and S2 were considered early stage, patients with S3 were considered at severe stage of disease. 29  Table S2) influence on outcomes using classic and stepwise methods. Thus, ordinal multivariate analysis was performed on steatosis severity groups (ranked as 0 for S0, 1 for S1, 2 for S2, and 3 for S3) and in the former analysis, the covariates were age, sex, TG, body mass index (BMI), and rs1051338 genotypes. The covariates of ordinal multivariate analysis for fibrosis stages were steatosis severity, age, sex, TG, BMI, and rs1051338 genotypes.

| Statistical analysis
Multivariate logistic regression was also performed to test predictors of severe stage of disease. Data analyzed were set as follow: 0 for presence of S0, S1 and S2 and 1 for presence of S3, 1 for male and 0 for female sex. BMI, TG levels and age were ranked in quartiles (Table S2).

| RE SULTS
At inclusion visit, a total of 74 subjects met inclusion criteria and were enrolled in the study (57 males and aged between 30 and 77 years). Table 1 gives demographic, anthropometric, hematologic,  Table S3. Furthermore, significant differences emerged in steatosis severity (S0, S1, S2, S3) within genotypic groups: the rare allele (c.46C) was more frequent in patients with more severe steatosis and vice versa ( Table 2). No other significant differences were found in other considered variables between c.46C and c.46AA genotypes ( Figure 1 and Table 2).
Higher severity fibrosis groups (VCTE measured) were associated with higher TG, AST, and BMI levels, and lower HDL-C levels than healthier patients (Table S5 and S6). Moreover, rare allele (c.46C) was borderline (p = 0.062) more frequently associated with severe fibrosis at VCTE (Table S6). No statistically significant differences emerged within fibrosis severity groups (F) when assessed with SWE, except for the age of patients, see Table S5.
Steatosis severity was independently associated with the rare allele (c.46C) and increasing of BMI in ordinal regression analysis ( Figure 2A). The analysis performed for hepatic fibrosis measured with VCTE showed that higher stages of fibrosis were associated with higher TG and BMI levels and to male gender ( Figure S1A). No statistically independent variables were correlated to fibrosis staged with SWE ( Figure S1B). Figure 2B shows statistically significant association between the risk allele (c.46C) and the presence of severe steatosis (S3) in contrast to early stages of disease (S0, S1, and S2). Additionally, the higher BMI was independently correlated with the severe stage (S3) of steatosis.

| DISCUSS ION
In this study, the association between the rs1051338 SNP in the LIPA gene and hepatic steatosis was evaluated in patients with atherogenic dyslipidemia and NAFLD. Inclusion criteria used for patients' enrollment and designed for identification of mild-moderate LAL-D were effective in producing homogeneous sample and limited variability range of age, BMI, and lipid and glycemic profile.
The prevalence of rare allele (c.46C) rs1051338 SNP was observed in 46% of cases, 39.2% HZ and 6.8% OZ, with MAF (minor allele frequency) of 26.4%: this distribution respects the Hardy-Weinberg equilibrium and agreed with Global MAF of 25.010% reported in the literature. NAFLD was diagnosed in 87.8% of patients (55.4% S3 or severe steatosis) which represented a very high prevalence, although the literature reports that NAFLD prevalence ranged between 50 and 95% in subjects with metabolic risk factors and dyslipidemia, such as our sample. [31][32][33] Lower HDL-C and borderline higher TG levels than wild type were observed in the rare allele carriers. This findings agreed with the literature, but such differences have been observed only in linkage of disequilibrium study. 4   In conclusion, the rs1051338 polymorphism of the LIPA gene influences the lipid profile contributing to worse a form of dyslipidemia LAL-D-like and was significantly associated to a worse hepatic steatosis in subjects with NAFLD. These results should be confirmed by further studies, which could finally define rs1051338 functional nature and the existence of "mild" forms of LAL-D that could play an important role in NAFLD pathogenesis. Furthermore, the elevation in transaminase levels observed in patients carrying the rare allele may suggest an increased risk of nonalcoholic steatohepatitis.

CO N S ENT S TATEM ENT
The study (2020-383 id.10748) was approved by the Institutional Review Board of IRCCS Ospedale Policlinico San Martino (Genoa, Italy) and conducted in accordance with the guidelines of the Declaration of Helsinki. All patients signed an informed consent form before inclusion in the present study.

ACK N OWLED G M ENT
The authors received no financial support for the research.

CO N FLI C T O F I NTE R E S T
The authors declare that there is no conflict of interest regarding the publication of this paper.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.