Acute pancreatitis (AP) is a common critical disease, has poor outcomes and a high mortality rate, and requires complex clinical management (Hazra and Gulliford, 2014; Lee and Papachristou, 2019). The etiology of AP is diverse (AGA, 2007; IAP/APA, 2013). In Western countries, gallstones and alcohol consumption represent the most common causes (Sekimoto et al., 2006; Yadav and Lowenfels, 2013) whereas in China, hypertriglyceridemia (HTG) rather than alcohol consumption is the second leading cause (Yin et al., 2015; Jin et al., 2019). Hypertriglyceridemia-induced acute pancreatitis (HTG-AP) has been defined as pancreatitis with a triglyceride (TG) level of up to 1000 mg/dL (11.3 mmol/L) alone, or 500 mg/dL (5.65 mmol/L) accompanied by lipemic or lactescent (milky in appearance) blood, after excluding other etiologies (Zafrir et al., 2018). By comparison with biliary acute pancreatitis, HTG-AP occurs earlier, is more severe, has a higher recurrence rate, and affects more males than females (Yin et al., 2015; Li et al., 2018; Jin et al., 2019).

HTG can be broadly divided into primary and secondary categories in terms of etiology (Yuan et al., 2007; Rygiel, 2018). Secondary HTG is primarily related to lifestyle factors such as heavy alcohol consumption and smoking, physiological factors such as pregnancy, or comorbid pathological conditions such as obesity, diabetes mellitus, hypothyroidism, and renal dysfunction (Black and Sprecher, 1993; Merkel et al., 2002). By contrast, primary HTG has been found to be mainly due to genetic defects in five lipid metabolism-related genes (Nawawi et al., 2020) namely LPL (lipoprotein lipase, OMIM #609708), LMF1 (lipase maturation factor 1, OMIM #611761), GPIHBP1 (glycosylphosphatidylinositol-anchored high density lipoprotein-binding protein 1, OMIM #612757), APOA5 (apolipoprotein A-V, OMIM #606368), and APOC2 (apolipoprotein C-II, OMIM #608083). LPL is the key enzyme catabolizing triglyceride (Olivecrona, 2016). LMF1 functions in the maturation of LPL (Peterfy et al., 2007). GPIHBP1 mediates the transmembrane transport and binding of LPL (Beigneux et al., 2007). APOA5 and APOC2 act as basic activators of LPL activity (Breckenridge et al., 1978; Pennacchio et al., 2001).

The abovementioned dichotomous classification of HTG provides a useful framework for understanding the etiology of HTG. However, in most cases, the etiology of HTG is complex, and often involves primary-primary (i.e., gene-gene) and primary-secondary (i.e., gene-environment) interactions (Wang et al., 2008; Hegele et al., 2014; Cole et al., 2015; Serveaux Dancer et al., 2018; Chen et al., 2019; Dron et al., 2019; Pu et al., 2020). Herein, we describe the clinical, genetic and functional analysis of a patient with alcohol-triggered HTG-AP, which serves to provide us with several novel insights into the complex etiology of HTG.

In the present study, we report the detailed clinical, biochemical and family analysis of a patient with alcohol-triggered HTG-AP as well as the in vitro functional analysis of the rare LPL p.Ile252Met variant. We obtained long-term data on TG levels and post-heparin LPL mass and activity for the patient and performed in vitro functional analysis of the missense variant of interest in both the “heterozygous” and “homozygous” state and under different experimental conditions (with and without heparin treatment). Integration of these findings provides several novel insights into the complex etiology of HTG and HTG-AP.

First, the patient was found to carry a known common risk single nucleotide polymorphism (SNP) (i.e., APOA5 c.553G > T (p.Gly185Cys); rs2075291) and a rare variant [i.e., LPL c.756T > G (p.Ile252Met)]. APOA5 c.553G > T has an allele frequency of 0.06765 in East Asians (according to gnomAD); and in the Asian population, c.553T carriers displayed an increased risk of HTG as compared with c.553 GG carriers, with the overall random effects odds ratio (OR) being 3.55 [95% confidence interval (CI) 2.46 to 5.13] in the context of a dominant genetic model (He et al., 2016). By reference to the definitions employed by Manolio et al. (2009), the APOA5 c.553G > T variant may be best described as a common genetic variant with moderate genetic effect (on HTG) in the Asian population. By contrast, the LPL c.756T > G (p.Ile252Met) variant is extremely rare in normal populations (allele frequency of 0.000003978 according to gnomAD) and has not previously been reported in HTG patients. The p. Ile252 site is conserved across the classes Mammalia and Aves (Figure 3C), suggesting its functional importance for protein structure and function. Consistent with the conservation of Ile252 over 300 million years of evolution, p.Ile252Met was predicted to be pathogenic by 9 of the 13 in silico algorithms employed by VarSome (Kopanos et al., 2019). Further, post-heparin LPL activity in the patient is approximately 70% of normal (Table 1 and Figure 1). It should be noted here that the post-heparin LPL activity in the patient was, in fact, a reflection of the net result of the complex interaction between the LPL gene and its regulating genes (such as APOC2, APOA5, GPIHBP1, and LMF1) as well as environmental and lifestyle factors. Thus, to decipher the direct effect of the p.Ile252Met variant on LPL protein and function, we further performed an in vitro transfection study, which demonstrated that the mutant p.Ile252Met allele caused a 40% loss of enzyme activity as compared to the wild-type allele (Figure 6). Based upon these observations, we believe that it is reasonable to define the LPL p.Ile252Met variant as a rare genetic variant that confers a rather moderate effect on LPL function (and presumably on HTG). Taken together, the patient was determined to have a digenic basis for his previous mild HTG and the current severe HTG-AP. To the best of our knowledge, this is the first time that trans heterozygosity for a common variant and a rare variant, both of which have a moderate effect on HTG, has been described in this disease. In this regard, it is pertinent to mention that in the context of TG-related genes, common variants are generally thought to have a small genetic effect whereas rare variants are thought to have a strong genetic effect (Johansen and Hegele, 2011; Dron and Hegele, 2020). Additionally, it should also be appreciated that the APOA5 p.Gly185Cys variant appeared to amplify the effect of the LPL p.Ile252Met variant on reducing LPL activity, as the reduction of the plasma LPL activity in the patient (30% lower than normal) was 10% higher than the reduction of the LPL activity conferred by the “heterozygous” mutant LPL variant in vitro (20% lower than wild-type). Although care should always be exercised when comparing in vivo data directly with in vitro data, such a difference has a strong biological basis. The APOA5 p.Gly185Cys impairs the function of APOA5; since APOA5 is an activator of LPL, this loss-of-function leads in turn to less activated LPL (Dorfmeister et al., 2008; Huang et al., 2012; Sharma et al., 2014; Chang et al., 2018).

Second, alcohol abuse is the most important environmental factor causing HTG owing to its impact on very-low-density lipoprotein secretion, lipolysis and free fatty acid fluxes from adipose tissue to the liver (Van de Wiel, 2012; Klop et al., 2013; Zemánková et al., 2015). Herein, having performed long-term follow-up of the TG levels in the patient, we were able to report a previously undescribed scenario: whereby the consumption of 150 g alcohol apparently triggered the attack of severe HTG-AP, the daily consumption of ∼100 g alcohol over 20 y is likely to have contributed to his pre-existing mild-HTG. This latter postulate was based on the observation that the TG levels in the patient remained normal, or at most borderline high, during the nearly 3-year follow-up period, during which the patient had completely abstained from drinking alcohol. Further follow-up studies are, however, needed to draw a firm conclusion.

Third, most of the HTG-associated rare LPL missense variants reported to date result in a complete or >90% functional loss of the affected allele, with this functional loss often being linked to upstream effects on protein synthesis, transport and secretion rather than to an isolated effect on enzyme activity (Rouis et al., 1996; Murano et al., 2005; Yu et al., 2006; Liu et al., 2016; Shi et al., 2020). To date, we are aware of only one rare LPL missense variant in the literature (i.e., a homozygous p.Ala203Thr variant found in a LPL-deficient patient) that specifically disrupted enzyme activity (Beg et al., 1990). However, both in vivo and in vitro findings (Beg et al., 1990) pointed to the p.Ala203Thr variant experiencing a complete loss of enzymatic activity. By contrast, the LPL p.Ile252Met mutant currently described here retained 60% enzyme activity as compared to its wild-type counterpart. Interestingly, p.Ile252Met occurred within a region of LPL, encompassing amino acid positions 243 to 266, that plays a crucial role in determining lipase substrate specificity (Dugi et al., 1995).

Lastly, we observed that whereas LPL mass was significantly increased in the patient during the acute phase of HTG-AP, LPL activity remained essentially unchanged at 70% normal (Figure 1). We surmised that this may be due to a dominant-negative effect of the mutant LPL p.Ile252Met allele, given that LPL is active mainly as a homodimer (Griffon et al., 2009; Arora et al., 2019; Beigneux et al., 2019). To explore this postulate, we mimicked the heterozygous state of the variant by mixing an equal amount of wild-type and mutant expression vectors in the transfection experiments. As shown in Figure 6, the impact of the “homozygous” variant was approximately double that of the “heterozygous” variant in terms of reducing LPL activity, thereby excluding such a postulate. It might also be that under stress of severe HTG, a positive feedback cycle stimulated LPL synthesis. However, a significant fraction of rapidly synthesized LPL proteins may not fold correctly and hence might not function normally. In this regard, it did not escape our attention that in the aforementioned LPL-deficient patient carrying the LPL p.Ala203Thr variant, higher than normal LPL mass concentrations were present in both pre- and post-heparin samples (Beg et al., 1990). It should, however, be noted that no information was available as to whether or not the patient had AP when the analysis was performed. Moreover, unlike our patient who had retained 70% normal LPL activity, the homozygous p.Ala203Thr carrier had no LPL activity at all. Consequently, we surmise that in this latter patient, the increased LPL mass may more likely be due to a feedback stimulus resulting from the absence of LPL activity. Irrespective of the underlying reasons, the increased LPL mass in both patients is consistent with the in vitro evidence that the respective LPL missense variants did not affect either protein synthesis or secretion.

The above notwithstanding, it should be emphasized that this study focused on a single case with HTG-AP. Further studies of more cases are required for us to obtain a better understanding of the genotype-phenotype relationships underlying this rare entity. Moreover, although loss-of-function APOA5 variants have previously been shown to lead to less activated LPL by several studies (Dorfmeister et al., 2008; Huang et al., 2012; Sharma et al., 2014; Chang et al., 2018), co-transfection experiments of APOA5 p.Gly185Cys and LPL p.Ile252Met expression vectors would be expected to yield confirmatory evidence for the gene-gene interaction between the two variants.

We provide concrete evidence that gene-gene and gene-environment interactions are responsible for causing mild HTG and HTG-AP in this patient. The novel insights generated from this study not only improve our understanding of the complex etiology of HTG, but also have important implications for variant interpretation as well as disease prevention. For example, the study of the novel LPL p.Ile252Met variant clearly demonstrates that rare variants can confer a small or modest effect in the same way as common variants. Moreover, whereas the evaluation of evolutionary conservation and in silico pathogenicity prediction are both helpful in assessing the pathogenic relevance of missense variants, the precise underlying biological mechanisms can only be addressed by in vitro functional analysis. Finally, subjects found to carry any genetic risk factors for HTG should be well advised to be abstemious with regard to their alcohol consumption.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.

The studies involving human participants were reviewed and approved by the Ethics Committee of Jinling Hospital. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

QY: conceptualization, methodology, project administration, and writing-original draft. NP: data curation, investigation, methodology, and writing-original draft. X-YL: investigation, methodology, and software. X-LS and W-WC: validation and visualization. G-FZ: software and visualization. Y-PH: investigation and software. JZ: data curation and funding acquisition. F-XC: project administration and validation. B-QL: project administration and resources. Z-HT: formal analysis and funding acquisition. CF: supervision and writing – review and editing. DC: formal analysis and writing – review and editing. J-MC: conceptualization, formal analysis, supervision, writing-original draft, and writing – review and editing. W-QL: funding acquisition, project administration, and resources. All authors contributed to revision of the manuscript and approved the final manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.