Markedly Low Prevalence of Fatty Liver despite Obesity in Prader-Willi Syndrome: A Search for Protective Genetic Markers

doi:10.21203/rs.3.rs-3142002/v1

Background & Aims.

Non-alcoholic fatty liver disease (NAFLD) is the most common liver disease in all ages that may cause significant morbidity and mortality. The pathogenesis of the disease is not fully elucidated but genetic factors have a major role in the development of NAFLD. Prader-Willi syndrome (PWS) is a neurogenetic, multisystemic disorder in which the main symptom is lack of satiety with uncontrolled eating and severe obesity. Despite obesity, NAFLD is relatively rare in PWS. The aim of this study was to assess whether known NAFLD-associated small nucleotide variants (SNVs) play a role in the protection from NAFLD in PWS.

Approach & Results.

Using targeted amplicon next generation sequencing, we studied DNA from patients with PWS and genotyped 13 SNVs that were previously associated with high risk for NAFLD. The study population included 142 (69 females) individuals with genetically confirmed PWS. Median age was 17.5 years, BMI z-score was 2.13 ± 1.9 and mean ALT and AST were 22 ± 20 units/L and 29 ± 17 units/L, respectively. Five of the 13 SNVs showed significantly lower frequency of the risk allele in our cohort compared to healthy population frequencies. Cumulative risk score for all 13 SNVs was also significantly lower in our cohort of PWS patients compared to the healthy population (adjusted p-value, 1.85E-5). Furthermore, it was found that Ashkenazi Jews have lower frequency of the risk alleles of NAFLD.

Conclusions.

Our results show that genetic factors may protect patients with PWS from developing NAFLD. Larger scale studies should be performed to confirm our findings.

Health sciences/Gastroenterology

Health sciences/Risk factors

Biological sciences/Genetics/Medical genetics

Biological sciences/Biological techniques/Sequencing/Next generation sequencing

Prader-Willi Syndrome

Fatty Liver

NASH

Next Generation Sequencing

Nonalcoholic fatty liver disease (NAFLD), currently regarded now as metabolic (dysfunction) associated fatty liver disease (MAFLD), is seen worldwide and is the most common liver disorder in western industrialized countries and worldwide ^1,2. The prevalence of NAFLD is estimated to be 32·4% worldwide and increased significantly over the past two decades^2,3. The complete mechanism of NAFLD's pathogenesis has not been fully elucidated. Insulin resistance has a key role in the development of hepatic steatosis and, potentially, steatohepatitis ^4,5. However, despite the strong association, not all patients with NAFLD exhibit insulin resistance, suggesting that NAFLD may be a heterogeneous syndrome with more than one cause. Among other factors, genetic factors may play a major role in the etiology of NAFLD. Indeed, at least four genetic variants in four different genes were reported to be responsible for encoding hepatic lipid metabolism regulatory proteins associated with the development and progression of NAFLD^6–8. Furthermore, genetic polymorphisms involved in insulin signaling are associated with hepatic fibrosis in NAFLD⁹.

Prader-Willi syndrome (PWS) is a neurogenetic, multisystemic condition resulting from the absence or reduced expression of specific paternal genes within the 15q11-q13 region of chromosome 15. Its incidence is estimated to be between 1 in 10,000 to 1 in 30,000 live births. The primary characteristic of this syndrome is the inability to experience satiety, leading to uncontrolled eating and significant obesity. Other features include neonatal hypotonia, characteristic facial features, growth hormone deficiency, hypogonadism, scoliosis, developmental delay, cognitive impairment, and severe behavioral problems^10,11. Interestingly, despite obesity, NAFLD is relatively rare in PWS. The observation that individuals with PWS have relative hypo-insulinemia and a higher insulin sensitivity than non-PWS obese subjects may be one of the explanations of the low incidence of NAFLD in these patients ¹². However, studies that described the low incidence of NAFLD in PWS showed that it was not related to the increased insulin sensitivity^13,14. These findings led us to hypothesize that genetic variance in patients with PWS may protect them from developing NAFLD.

Patients

The national multidisciplinary PWS clinic at Shaare Zedek Medical Center treats nearly all individuals with a genetic diagnosis of PWS in Israel. There are 177 individuals 94M/83F ages 0.5m-55.2y actively followed in the center (35 ages 0.5m-5y, 64 from 5y-15y, 54 from 15y-30y and 24 above 30 years, Table 1). DNA samples were obtained from 142 patients above age 5 years in this cohort. Of them 48% were females with a median age of 17.5 years. Eighty four percent were Jewish (55% Ashkenazi Jews), 9% Muslim Arab and 1.4% Druze. Blood measurements of 30 patients with PWS were compared to that of 28 healthy age- gender-and BMI-matched controls. Table 2 summarizes the clinical characteristics of the study population. Genetic analysis for PWS showed that 37% had uniparental disomy (UPD) and 61% had paternal deletion of 15q11-13. Table 3 presents the differences in Age, Sex and BMI between control and affected participants with clinical measurements.

Ethics Declaration

Ethical approval for this study was obtained from the Shaare Zedek Medical Center institutional review board. DNA and whole blood samples in this study were donated to the Shaare Zedek Medical Genetics Institute for research with informed consent according to Shaare Zedek institutional review board guidelines and as set forth in the Declaration of Helsinki. Informed consent was obtained from all participants.

Blood biochemistry

Serum levels of cholesterol, triglycerides, high-density lipoprotein (HDL), low-density lipoprotein (LDL), alanine aminotransferase (ALT), and aspartate aminotransferase (AST) were determined using the Cobas C-111 chemistry analyzer (Roche, Switzerland).

Multiplex amplicon sequencing

Amplicons flanking the variants targeted in Table S1, were amplified from each DNA sample with indexed multiplex PCR followed by pooling and sequencing on an Illumina MiSeq platform.

Bioinformatics analysis

Raw sequencing reads were mapped to the human reference genome (hg38) using bwa (version 0.7.17). The resulting alignment files were used by HaplotypeCaller from the GATK toolset (version 3.8) for variant calling. Polygenic risk scores were calculated based on the GWAS catalog (https://www.ebi.ac.uk/gwas/home)(4). Statistical analysis and visualizations were performed in R. Raw unweighted Risk Score (RS) was calculated as:

$$RS = {\sum }_{1}^{13}RA$$

Where RA is the occurrences of risk-associated alleles for each of the 13 single nucleotide variants (SNVs).

[Methods should include a description and/or reference for the assay methods used in measuring liver enzymes, cholesterol, TG, etc.]

Ethics

This study was approved by the local institutional Helsinki committee.

Clinical data

The study population included 142 patients with PWS (Table 1). In order to assess the prevalence of elevated liver enzymes and dyslipidemia in our cohort, we compared the laboratory results of the cohort to those of normal controls. We measured ALT, AST, triglycerides, total cholesterol, HDL and LDL levels in 30 adults with PWS and 28 age- and weight-matched healthy controls. For the cohort, Z-score BMI was 2.13 ± 1.9 and mean ALT and AST was 22 ± 20 units/L and 29 ± 17 units/L, respectively. Table 2 depicts the clinical characteristics showing that the study and the control groups are comparable in age, gender, and BMI. Analysis of blood tests showed that ALT (Fig. 1A), AST (Fig. 1B), triglycerides (Fig. 1C), and total cholesterol (Fig. 1D) levels were significantly lower in the PWS group, but there were no differences in circulating HDL (Fig. 1E) and LDL (Fig. 1F). These results are in line with previous reports and suggest that the prevalence of liver abnormalities in our cohort is low¹⁴.

Genotyping

In order to look for genetic alleles that protect these individuals from developing NAFLD/NASH, we designed an amplicon sequencing panel targeted to cover 13 common SNVs, which were previously shown by GWAS to be associated with NAFLD, fibrosis or NASH⁶. For six of these variants, the reference allele was associated with increased risk for the disease phenotypes, while for the other seven the alternative allele was associated with increased risk. The list of the variants covered by our sequencing panel is described in Table S1. Using this panel, we determined the genotype of 13 variants for all 142 patients in our cohort (Tables 4 and S1).

The allele frequency within the cohort was between 0.05 and 0.65 for each of the 13 variants (Table 4). Comparing the allele frequency of each variant between the study group and 1000 genome project frequency data (phase 3, N = 2504)¹⁵ revealed a set of five variants in which the frequency of the risk-associated allele was significantly lower within our cohort (Fisher’s exact test, p < 0.05) (Fig. 2). This group included two variants in GCKR and variants in LPIN1, MERTK and ENPP1. A single variant, in SOD2, displayed the opposite trend, i.e., the frequency of the risk-associated allele was significantly higher in our cohort.

Four of the five variants with lower frequency of the risk-associated allele (GCKR*2, LPIN1 and ENPP1) had significantly lower rates of heterozygosity in the cohort compared to that of the 1000 genomes project, while there were no significant differences in risk-associated allele homozygosity rates. The variant in MERTK, however, had a significantly higher homozygosity rate of the risk-associated allele vs. 1000 genomes with no difference in heterozygosity rate.

Polygenic risk score

As with many other genotype-phenotype correlations, the GWAS data indicate that the genetic effect determining the risk for the NAFLD/NASH phenotype may result from a combination of a number of variants in multiple genes⁶. Accordingly, we calculated an unweighted raw risk-score (RS) for each individual in our cohort by summing the occurrences of risk-associated alleles (see Methods). We compared this RS with an RS calculated from the allele frequencies of each variant in the 1000 genomes project global population (red dashed line) and population of Ashkenazi Jews (blue dashed line) (Fig. 3). For all ethnic groups in our cohort (except Bedouin patients) the median RS was significantly lower than the population RS.

Genetic determinants of PWS

Deletions in the PWS region in chromosome 15 and UPD are the two most common causes of PWS. Of the 142 patients in our cohort, 87 were diagnosed with paternal deletions, 50 with UPD and the rest with imprinting center defects (IMP) or yet undetermined (Table 1). Comparing the allele frequencies and genotype occurrences between the group of patients with chromosome 15 deletions or UDP revealed several differences (Table S1). However, none of the differences between paternal deletion and UPD patients were significant among the 13 SNVs.

Patients of Ashkenazi Jewish ancestry

One hundred twenty of the 142 patients in our cohort were Jewish. Among them, 54 were non-Ashkenazi Jews, 36 had one parent with Ashkenazi (AJ) ancestry and 30 had 2 parents with AJ ancestry. We compared the allele frequencies of the 66 patients with AJ ancestry to the allele frequency of global AJ data from GnomAD (https://gnomad.broadinstitute.org/) (Fig. 4, Table S1). Out of the five variants that showed significantly lower frequency of the risk-associated allele, only for the variant in LPIN1, the allele frequency (AF) in our cohort supported less risk for NASH than in all AJs. For the other four variants, the frequency of the risk-associated allele was equal to or higher in AJs in our cohort compared to GnomAD AJ allele frequencies. Similar to what we observed for several individual variants, the RS for the AJ population was lower than that of the global GnomAD population (Fig. 4).

Obesity and genetic risk factors

BMI of each patient was standardized according to the relevant age group range (Fig. S1A). We then compared the raw RS of all variants (Fig. S1B) or serum level of liver markers (Fig. S2) for the patients within 1, 2 or 3 standard deviations from the mean. Median raw RS were lower than the global population for all BMI groups, except for patients with the highest BMI’s (SD > 3)(Fig. S1B). There were no significant differences in liver enzyme levels between the BMI groups, indicating that even patients with the highest obesity levels had no evidence of NASH or NAFLD (Fig. S2)

NAFLD is the most common liver disease in all ages that may cause significant morbidity and mortality. To date, the pathogenesis of the disease is not fully elucidated, but it is well-established that genetic factors play a major role in the pathogenesis of the disease^2,3. The rising rates of obesity and metabolic syndrome together with genetic factors leads to the dramatic increase of NAFLD and resulting complications. One cannot predict which individuals with metabolic syndrome will progress to NAFLD, NASH and significant liver disease. In the current study, we studied a cohort of patients with PWS who are known to develop obesity and metabolic syndrome in early life but surprisingly do not have fatty liver. We hypothesized that genetic factors contribute to the low incidence of NASH despite obesity in PWS. Indeed, five genes, known to have a role in NAFLD, were found to be different in this population and may explain the low incidence of NAFLD. These include glucokinase regulator (GCKR), ectoenzyme nucleotide pyrophosphate phosphodiesterase 1 (ENPP1), superoxide dismutase 2 (SOD2), MER protocol-oncogene, tyrosine kinase (MERTK) and LPIN1.

One of the hallmarks of PWS is obesity from an early age. Thus, it is intriguing why the incidence of fatty liver is low in these patients. Indeed, several investigators studied this question. The higher insulin sensitivity might protect PWS subjects from developing metabolic syndrome and NAFLD. Fintini and colleagues¹⁴ used a validated method of ultrasound grading (range from G0 to G3) for the diagnosis of NAFLD and found a reduced risk of NAFLD in PWS children. Bedugni et al ¹³ reported that NAFLD is less frequent in PWS than in non-PWS women. In both studies the lower rate of NASH was not related to the increased insulin sensitivity in these patients. In the current study we compared liver enzymes and other blood results of a group of patients with PWS to age, sex and BMI matched healthy controls and showed that liver enzymes were significantly lower in PWS patients. As this was a retrospective study, most of our cohort did not perform liver US but based on previous studies it is plausible that fatty liver is also low in our cohort.

In the past decade, GWAS have contributed to the identification of numerous genes that are involved in the pathogenesis of NAFLD and associated with prognosis of the disease. The most rigorous studies were done on PNPLA3 which is associated with NAFLD at all stages, across various geographical regions and ethnicities. PNPLA3 is a multifunctional enzyme that plays a role in lipid regulation and hepatic fibrogenesis. In the current study, we screened for 13 SNVs that were found to be the most significant in NAFLD in various studies. Of these variants we found five that had significantly different allele frequencies in our cohort vs. 1000 genome controls. These alleles are involved in lipogenesis, insulin resistance and hepatic fibrosis. One of them is GCKR, which regulates de novo lipogenesis by controlling the influx of glucose in hepatocytes and has been associated with children ¹⁶ and adults ¹⁷ with NAFLD. GCKR increases the lipogenic pathway by providing substrates for fatty acid biosynthesis and thus increases the risk of NAFLD. In our study both markers of the GCKR gene were significantly lower in the PWS cohort than in the control group, suggesting a lower risk for NAFLD. Another gene that is commonly associated with NAFLD is ENPP1. This gene has a role in insulin resistance and was shown to be correlated with increased risk for hepatic fibrosis in patients with NAFLD ⁹. In our cohort the risk-associated allele frequency was significantly lower than in the control group, suggesting lower risk for fibrosis. In contrast, the A16V variant in SOD2, previously associated with increased risk to fibrosis, was significantly enriched in our cohort. This variant occurs in the mitochondrial targeting sequence, and was shown to affect SOD2 protein import into the mitochondria, mRNA stability and MnSOD activity¹⁸. This variant was previously associated with breast, prostate and brain cancers, diabetes and cardiovascular disease¹⁹. In studies related to NAFLD and fibrosis, it was mostly associated with the severity of NAFLD and progression of the disease to fibrosis ¹⁸. According to his notion, and the fact that the allele frequency of this SOD2 variant is high in our cohort of non-NAFLD patients, suggest that it is not pertinent for assessing the risk to develop NAFLD. MERTK is also linked to hepatic fibrosis but has been shown to be associated with reduced risk for fibrosis in patients with NAFLD and chronic hepatitis C infection²⁰. Indeed, in our cohort the risk protective allele was significantly higher than control suggesting that it plays a role in the reduced liver damage in PWS. Lastly, LPIN1, a phosphatidate phosphatase, is also linked to reduced risk for NAFLD (15). LPIN1 is highly expressed in the liver and adipose tissue and involved in the synthesis of phospholipids and triglycerides. Its increased expression was shown to be protective from NAFLD ²¹. Interestingly, the risk protective allele at this gene target was also enriched in our study cohort.

In recent years polygenic risk score (PRS) has emerged as a powerful tool to assess an individual's relative risk to a disease. This approach has shown promising results by predicting risk for coronary artery, type 2 diabetes mellitus, various cancer types and Alzheimer disease²². Since no single genetic variant can adequately predict the genetic risk for NAFLD/NASH, the PRS approach seems attractive to these disorders as well. Indeed, promising results were shown when combining different variants, mostly from PNPLA3, TM6SF2, GCKR loci^21,23−25. Our own results showing that patients with clinical risk factors, such as high BMI, and low PRS values did not develop NASH, support the use of this approach.

Contemporary Ashkenazi Jews (AJ) are descendants of several hundred founder individuals from central and Eastern Europe. This population maintained genetic isolation, separated from its neighbors by religious and cultural practices as well as consanguinity²⁶. Due to the genetic bottleneck and founder effects in AJ, the allele frequencies in large parts of the human genome are different when comparing people of AJ origin to other geographically close populations. As a consequence, there was a notable increase in the occurrence of autosomal-recessive disorders and a relatively elevated occurrence of gene variants that increase the susceptibility to prevalent conditions like breast and ovarian cancer, inflammatory bowel disease, and Parkinson's disease ²⁷. By contrast, longevity studies have also identified various genetic lifespan increasing determinants which are enriched in AJ vs. other ethnicities, providing protection against hypertension, cardiovascular disease, and metabolic syndromes ^28,29. Similarly, out of the 13 variants associated with higher risk for NASH/NAFLD/fibrosis, 5 had significantly lower frequency of the risk-associated allele in AJ compared to the global population. This suggests that at least in terms of genetic risk factors, the AJ population is at lower risk to develop fatty liver diseases in general.

There are several limitations to this study. As this was a retrospective study some of the laboratory results were not available for a few of the study participants. Liver ultrasounds to assess for fatty liver were not performed as part of this study. However, in some of our patients who performed abdominal ultrasound due to various indications the liver appearance was normal. Indeed, normal ultrasound in our cohort would support previous reports that showed the lack of fatty liver in PWS but these studies consistently show this finding and the significantly low liver enzymes in our cohort corroborates the lack of fatty liver.

In conclusion, the current study suggests that the importance of genetics in a cohort of obese patients who do not have NAFLD. We showed specific genes that significantly differ from the healthy population and unweighted Risk Score for all 13 genes that were assessed were significantly lower in our cohort. This study further supports the notion that NAFLD is multifactorial disease and genetics plays a significant role in its pathogenesis.

Author contribution

OM, VGT, DAZ, GA and ES designed the study. OM, DAZ and ES wrote the main manuscript. OM, VGT, TM, DAZ, SB, JT, HJH and ES contributed to the data collection and investigation. OM and DAZ preformed the bioinformatics analysis. All the authors read and approved the final manuscript.

Data availability

The sequencing data generated during the current study are available at the NCBI SRA with accession number PRJNA995227.

Competing interests

All authors declare no competing interests.

Younossi, Z. M. et al. Changes in the prevalence of the most common causes of chronic liver diseases in the United States from 1988 to 2008. Clin Gastroenterol Hepatol 9, 524-530 e521; quiz e560, doi:10.1016/j.cgh.2011.03.020 (2011).
Riazi, K. et al. The prevalence and incidence of NAFLD worldwide: a systematic review and meta-analysis. Lancet Gastroenterol Hepatol 7, 851-861, doi:10.1016/S2468-1253(22)00165-0 (2022).
Vernon, G., Baranova, A. & Younossi, Z. M. Systematic review: the epidemiology and natural history of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis in adults. Aliment Pharmacol Ther 34, 274-285, doi:10.1111/j.1365-2036.2011.04724.x (2011).
Sanyal, A. J. et al. Nonalcoholic steatohepatitis: association of insulin resistance and mitochondrial abnormalities. Gastroenterology 120, 1183-1192, doi:10.1053/gast.2001.23256 (2001).
Cariou, B., Byrne, C. D., Loomba, R. & Sanyal, A. J. Nonalcoholic fatty liver disease as a metabolic disease in humans: A literature review. Diabetes Obes Metab 23, 1069-1083, doi:10.1111/dom.14322 (2021).
Eslam, M., Valenti, L. & Romeo, S. Genetics and epigenetics of NAFLD and NASH: Clinical impact. J Hepatol 68, 268-279, doi:10.1016/j.jhep.2017.09.003 (2018).
Dongiovanni, P., Romeo, S. & Valenti, L. Genetic Factors in the Pathogenesis of Nonalcoholic Fatty Liver and Steatohepatitis. Biomed Res Int 2015, 460190, doi:10.1155/2015/460190 (2015).
Albhaisi, S. & Sanyal, A. J. Gene-Environmental Interactions as Metabolic Drivers of Nonalcoholic Steatohepatitis. Front Endocrinol (Lausanne) 12, 665987, doi:10.3389/fendo.2021.665987 (2021).
Dongiovanni, P. et al. Genetic variants regulating insulin receptor signalling are associated with the severity of liver damage in patients with non-alcoholic fatty liver disease. Gut 59, 267-273, doi:10.1136/gut.2009.190801 (2010).
Cassidy, S. B., Schwartz, S., Miller, J. L. & Driscoll, D. J. Prader-Willi syndrome. Genet Med 14, 10-26, doi:10.1038/gim.0b013e31822bead0 (2012).
Grugni, G. et al. The Italian National Survey for Prader-Willi syndrome: an epidemiologic study. Am J Med Genet A 146A, 861-872, doi:10.1002/ajmg.a.32133 (2008).
Hirsch, H. J., Gross, I., Pollak, Y., Eldar-Geva, T. & Gross-Tsur, V. Irisin and the Metabolic Phenotype of Adults with Prader-Willi Syndrome. PLoS One 10, e0136864, doi:10.1371/journal.pone.0136864 (2015).
Bedogni, G. et al. Is non-alcoholic fatty liver disease less frequent among women with Prader-Willi syndrome? Obes Facts 7, 71-76, doi:10.1159/000358570 (2014).
Fintini, D. et al. Non-Alcoholic Fatty Liver Disease (NAFLD) in children and adolescents with Prader-Willi Syndrome (PWS). Pediatr Obes 11, 235-238, doi:10.1111/ijpo.12052 (2016).
Genomes Project, C. et al. A global reference for human genetic variation. Nature 526, 68-74, doi:10.1038/nature15393 (2015).
Santoro, N. et al. Variant in the glucokinase regulatory protein (GCKR) gene is associated with fatty liver in obese children and adolescents. Hepatology 55, 781-789, doi:10.1002/hep.24806 (2012).
Petta, S. et al. Glucokinase regulatory protein gene polymorphism affects liver fibrosis in non-alcoholic fatty liver disease. PLoS One 9, e87523, doi:10.1371/journal.pone.0087523 (2014).
Al-Serri, A. et al. The SOD2 C47T polymorphism influences NAFLD fibrosis severity: evidence from case-control and intra-familial allele association studies. J Hepatol 56, 448-454, doi:10.1016/j.jhep.2011.05.029 (2012).
Palma, F. R. et al. Mitochondrial Superoxide Dismutase: What the Established, the Intriguing, and the Novel Reveal About a Key Cellular Redox Switch. Antioxid Redox Signal 32, 701-714, doi:10.1089/ars.2019.7962 (2020).
Rueger, S. et al. Impact of common risk factors of fibrosis progression in chronic hepatitis C. Gut 64, 1605-1615, doi:10.1136/gutjnl-2014-306997 (2015).
Valenti, L. et al. LPIN1 rs13412852 polymorphism in pediatric nonalcoholic fatty liver disease. J Pediatr Gastroenterol Nutr 54, 588-593, doi:10.1097/MPG.0b013e3182442a55 (2012).
Torkamani, A., Wineinger, N. E. & Topol, E. J. The personal and clinical utility of polygenic risk scores. Nat Rev Genet 19, 581-590, doi:10.1038/s41576-018-0018-x (2018).
Trepo, E. & Valenti, L. Update on NAFLD genetics: From new variants to the clinic. J Hepatol 72, 1196-1209, doi:10.1016/j.jhep.2020.02.020 (2020).
Gao, F. et al. Individualized Polygenic Risk Score Identifies NASH in the Eastern Asia Region: A Derivation and Validation Study. Clin Transl Gastroenterol 12, e00321, doi:10.14309/ctg.0000000000000321 (2021).
Kawaguchi, T. et al. Risk estimation model for nonalcoholic fatty liver disease in the Japanese using multiple genetic markers. PLoS One 13, e0185490, doi:10.1371/journal.pone.0185490 (2018).
Zlotogora, J., Patrinos, G. P. & Meiner, V. Ashkenazi Jewish genomic variants: integrating data from the Israeli National Genetic Database and gnomAD. Genet Med 20, 867-871, doi:10.1038/gim.2017.193 (2018).
Ostrer, H. & Skorecki, K. The population genetics of the Jewish people. Hum Genet 132, 119-127, doi:10.1007/s00439-012-1235-6 (2013).
Barzilai, N. et al. Unique lipoprotein phenotype and genotype associated with exceptional longevity. JAMA 290, 2030-2040, doi:10.1001/jama.290.15.2030 (2003).
Newman, A. B. & Murabito, J. M. The epidemiology of longevity and exceptional survival. Epidemiol Rev 35, 181-197, doi:10.1093/epirev/mxs013 (2013).

Table 1. Demographic characteristics of the study population.

Parameter	Value	Count (N)	Percent (%)
Total	N	142
Sex	Female	69	48.6
	Male	73	51.4
Age	Median	17.5
	Range	5.6-57
Ethnicity	Jew	120	84.5
	Muslim arab	13	9
	Bedouin arab	6	4.2
	Druze	2	1.4
	Samaritan	1	0.7
Subethnicity among the Jews	Non Ashkenazi	54	45
	1 Ashkenazi parent	36	30
	2 Ashkenazi parents	30	25
Genetic diagnosis	Uniparental disomy (UPD)	50	35.2
	Paternal deletion 15q11-13 (Del)	87	61.3
	Other/Not determined	5	3.5

Table 2. Clinical characteristics of the study population.

Measurements	Sample size	Mean	SD	95% Confidence interval
BMI, SDS	116	2.13	1.91	1.78 - 2.48
ALT GPT	118	22.95	20.40	19.23 - 26.67
AST GOT	109	29.16	17.07	25.92 - 32.4
GGT	49	27.64	36.76	17.09 - 38.2
Alkaline phosphatase	110	188.68	127.43	164.6 - 212.76
Albumin	75	4.65	3.12	3.93 - 5.36
Triglycerides	96	86.57	36.40	79.2 - 93.95
Cholesterol	97	170.76	42.81	162.13 - 179.39
HDL	91	50.22	15.79	46.93 - 53.51
LDL	90	106.35	37.62	98.47 - 114.23
HbA1C	67	5.74	1.29	5.43 - 6.06
Glucose	112	87.58	30.14	81.94 - 93.22

Table 3. Comparison between subjects with and without PWS

	Control	PWS	*P value*
N	28	30	-
Age (Year)	29.6 ± 6.2	29.9 ± 6.2	n.s.
DEL/UPD	-	17/13	-
Sex (% Male)	57%	60%	n.s.
BMI (Kg/M²)	28.2 ± 5.3	28.4 ± 5.1	n.s.

Deletion (DEL); Uniparental Disomy (UPD). Data are expressed as mean ± SD. Not significant (n.s.).

Table 4. Variants analyzed in this study, their molecular effect on the protein, the phenotype associated with the alternative variant and their overall allele frequencies (AF) in the study cohort and in the 1000 genome database. Shaded are the 5 variants showing significant lower frequency for the risk-associated allele.

Gene	SNP ID	Function	Effect	Impact	Phenotype	AF cohort	AF 1000G
FNDC5/ Irisin	rs3480	HSCs activation	Non-coding; reduced expression	?	↓ fibrosis	0.521	0.583
LYPLAL1	rs12137855	TGL catabolism	?	?	↑ NAFLD	0.137	0.164
LPIN1	rs13412852	Regulation of lipid metabolism	intronic	?	↓ NASH, fibrosis	0.338	0.205
GCKR	rs1260326	Regulation of de novo lipogenesis	p.P446L	loss	↑ NAFLD, NASH, fibrosis	0.493	0.707
GCKR	rs780094	Regulation of de novo lipogenesis	intronic	?	↑ NAFLD, NASH, fibrosis	0.504	0.698
MERTK	rs4374383	Innate immunity, HSCs activation	Non-coding; Linked with sequence regulated by IRF1	loss	↓ fibrosis	0.655	0.521
IRS1	rs1801278	Insulin signaling	p.G972R	loss	↑ fibrosis	0.060	0.053
ENPP1	rs1044498	Insulin signaling inhibitor	p.K121Q	gain	↑ fibrosis	0.250	0.342
SOD2	rs4880	Mitochondrial antioxidant	p.A16V	loss	↑ fibrosis	0.521	0.411
KLF6	rs3750861	Regulation of de novolipogenesis; fibrogenesis	Splice variant IVS1-27G	loss	↓ fibrosis	0.056	0.067
IFNL4/ IL28B	rs12979860	Innate immunity	Alternative IFNL3/4 transcription	loss	↓ fibrosis	0.398	0.356
LARGE1/UCP2	rs695366	Mitochondrial lipid metabolism OxPhos	−866 promoter variant	gain	↓ NASH	0.254	0.264
PNPLA3	rs738409	Lipid droplets remodeling	p.I148M	loss	↑ NAFLD, NASH, fibrosis, HCC	0.225	0.262

No competing interests reported.

Markedly Low Prevalence of Fatty Liver despite Obesity in Prader-Willi Syndrome: A Search for Protective Genetic Markers

Status:

Version 1

Abstract

Background & Aims.

Approach & Results.

Conclusions.

Figures

Introduction

Methods

Patients

Ethics Declaration

Blood biochemistry

Multiplex amplicon sequencing

Bioinformatics analysis

Ethics

Results

Clinical data

Genotyping

Polygenic risk score

Genetic determinants of PWS

Patients of Ashkenazi Jewish ancestry

Obesity and genetic risk factors

Discussion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1