Serum lipids, leptin, and soluble leptin receptor in alcohol dependence: A cross-sectional and longitudinal study

Highlights

• Interactions between lipid and leptin metabolism and alcohol dependence.

• Sex differences in alcohol dependence with regard to lipid and leptin activities.

• Higher HDL-Cholesterol and lower LDL/HDL in male patients than in controls.

• Higher concentrations of soluble leptin receptor in male and female patients.

Abstract

Objective

Alcohol dependence affects metabolic processes. Further research is needed to apply this knowledge clinically. In this study, possible differences in serum lipids and/or leptin activities between alcohol-dependent in-patients and healthy controls and possible associations with alcohol-related blood parameters and with prospective outcomes in alcohol dependence were assessed sex-specifically.

Method

We measured and compared (median) serum lipids (triglycerides and total, HDL, and LDL cholesterol) and leptin activities (leptin, soluble leptin receptor [ObRe], and free leptin index) in 200 (males 56.5 %) early-abstinent alcohol-dependent in-patients and 240 (males 55.4 %) healthy controls and assessed alcohol-related readmissions during a 24 -month post-inclusion period.

Results

Male patients showed higher HDL cholesterol (61 versus 48 mg/dl), lower LDL/HDL ratios (2.06 versus 3.04), and lower free leptin index (0.30 versus 0.59) at study inclusion compared to healthy controls. In patients, ObRe levels were higher than in controls and decreased from inclusion to the second study-visit (at median 5 days later; males: 16.7–13.8 versus 11.0 ng/ml; females: 17.0–13.4 versus 12.1 ng/ml). The free leptin index increased between the two time points in females (0.80 versus 1.20). Lipids and leptin activities correlated with carbohydrate-deficient transferrin levels and liver enzyme activities. None of the serum parameters were significantly associated with alcohol-related readmissions.

Conclusion

Our data support that serum lipid levels and leptin activities are involved in alcohol dependence. The parameters appear as possible indirect biomarkers for alcohol dependence.

Introduction

Alcohol dependence is a demanding worldwide health-related challenge. Current epidemiological reviews report that harmful use of alcohol accounts for 5.1 % of all deaths and 5.3 % of all disability-adjusted life years (World Health Organisation, 2018). A better understanding of the etiopathogenetic mechanisms and the development and improvement of preventive, diagnostic, and therapeutic strategies are urgently needed.

Various direct and indirect biomarkers for chronic alcohol consumption have been reported and were tested as diagnostic tools in the treatment of alcohol dependence. Carbohydrate-deficient transferrin (CDT) and liver enzymes, such as glutamic-oxaloacetic transaminase (GOT), glutamic-pyruvic transaminase (GPT), gamma-glutamyl transferase (GGT), and mean corpuscular volume (MCV), are parameters with limited sensitivity and low specificity. Nevertheless, these parameters are often used in the clinical setting as indicators for chronic alcohol consumption (Andresen-Streichert et al., 2018). Serum lipoproteins are standard laboratory parameters, which also have been associated with alcohol consumption. Thus, they may serve as indirect markers for chronic alcohol intake.

A role of metabolic pathways in alcohol dependence is well-established (Ehlers and Wilhelmsen, 2007; Landgren et al., 2008; French et al., 2010; Wang et al., 2012; Lichenstein et al., 2014). Recently, we found that the body mass index (BMI) is related to alcohol dependence and also predicts alcohol-related readmission following in-patient withdrawal treatment (Lenz et al., 2017; Weinland et al., 2019a, 2019b). The field may provide a solid basis to develop novel and effective prevention and treatment strategies.

The most prominent and replicated finding is an increase in high-density lipoprotein (HDL) cholesterol in alcohol dependence (mainly male alcohol-dependent patients), and also in other excessive drinking patterns (Cauley et al., 1987; Szegedi et al., 2000; Brien et al., 2011; De Bruyne et al., 2017; Rosoff et al., 2019). A systematic review and meta-analysis of 63 studies conducted by Brien et al. (2011) reported consistently increased levels of HDL cholesterol in alcohol-drinking subjects of both sexes. In support of this, positive correlations of HDL cholesterol with established indirect markers of chronic alcohol consumption including CDT, GOT, GPT, GGT, and MCV have been reported for alcohol-dependent patients (Szegedi et al., 2000). Contrary to these findings, Bell et al. (1985) found no significant associations of HDL cholesterol and the number of drinks in patients with alcohol dependence, but a significant decline of HDL cholesterol during two weeks of abstinence.

Other lipid biomarkers, such as total cholesterol, low density lipoprotein (LDL) cholesterol, and triglycerides have also been investigated with respect to alcohol abuse. In contrast to HDL cholesterol, no consistent differences in serum concentrations of total and LDL cholesterol have been observed between alcohol-dependent patients and control subjects (De Bruyne et al., 2017). However, a recent case-control study reported decreased odds for the diagnosis of alcohol dependence in patients with higher LDL cholesterol (Goodyear et al., 2017). For triglycerides, both decreased (De Bruyne et al., 2017) and increased (Goodyear et al., 2017) levels have been found. To our knowledge, no specific analysis with respect to the HDL/LDL ratio in alcohol-dependent patients has been published.

The appetite regulating peptide leptin secreted by adipocytes of the subcutaneous white fat tissue has been subject to numerous investigations in the context of alcohol dependence and craving (Hube et al., 1996). Leptin modulates appetite via hypothalamic receptors and leads to a reduced production of appetite stimulating neurohormones (Zhang et al., 1994). However, overweight and adiposity were associated with higher levels of leptin, which were attributed to resistance of leptin receptors under overweight conditions (El-Haschimi et al., 2000; Friedman, 2002).

Due to its possible role in appetite-regulating processes and in dopaminergic, mesolimbic pathways (Inui, 1999; Fulton et al., 2000; Palmiter, 2007), many studies have been conducted and revealed that leptin serum levels relate to alcohol consumption, alcohol dependence, and craving (Wurst et al., 2007; Addolorato et al., 2009; Lenz et al., 2010; Hillemacher et al., 2015). Increased leptin serum levels have repeatedly been associated with higher alcohol consumption in mice (Kiefer et al., 2001a, 2001b) and men (Nicolás et al., 2001; Wurst et al., 2007; Addolorato et al., 2009). Moreover, a stronger craving for alcohol has been linked to higher leptin concentrations (Kiefer et al., 2001a, 2001b; Kraus et al., 2004; Hillemacher et al., 2007a). Increasing leptin levels during 12 weeks of alcohol withdrawal therapy have been reported to predict relapse (Kiefer et al., 2005). Patients with higher leptin levels at the beginning of alcohol detoxification treatment showed a significant decline during withdrawal (Kiefer et al., 2001a, 2001b). Furthermore, treatment with acamprosate and naltrexone with the aim of maintaining abstinence leads to lower leptin serum levels with extended abstinence (Kiefer et al., 2005).

The soluble leptin receptor (ObRe) is one of the six known leptin receptor isoforms named ObRa to ObRf. The single membrane-spanning leptin receptor, which is mainly expressed in the hypothalamus (a brain area important to regulation of body weight), belongs to the cytokine receptor family (Tartaglia et al., 1995). The ObRb represents the long form with an intracellular domain of 302 amino acids, while the ObRa, ObRc, ObRd, and ObRf belong to the four short forms with cytoplasmic tails of 30–40 amino acids with unique C-termini (Wauman et al., 2017). Until now, ObRe has only been detected in a soluble form. It is a circulating part of the longer isoform, ObRb (Mantzoros et al., 2011). Tu et al. (2008) suggested that the ObRe might act as an antagonist in leptin surface binding and endocytosis. Besides, ObRe is the leading binding protein for leptin in human blood, it affects bioavailability of leptin (Schaab and Kratzsch, 2015). In rodents (Uotani et al., 1999; Wang et al., 1999) and in human studies (Chan et al., 2002), leptin suppressed ObRe and ObRe levels. ObRe levels were shown to be decreased in obese humans (van Dielen et al., 2002) and low concentrations of ObRe have been linked to insulin resistance (Sandhofer et al., 2003). Relevance of ObRe with respect to various diseases has been demonstrated. ObRe correlates negatively with hepatic steatosis in patients with type 2 diabetes (Cernea et al., 2018). Moreover, ObRe was shown to be linked to a higher risk of mild cognitive impairment (MCI) in diabetic type 2 patients (Yin et al., 2018). The leptin/ObRe ratio, the free leptin index (FLI), serves as a marker for leptin bioavailability with higher ratios indicating higher leptin activity (Cernea et al., 2018). Furthermore, the FLI and the ObRe appear to be linked to breast cancer (Rodrigo et al., 2017).

The published clinical research is limited in that primarily male cohorts were studied, and sex-specific analyses have been neglected. In addition, the study cohorts consisted only partially of alcohol-dependent patients and also included other alcohol consuming phenotypes beyond the diagnosis of alcohol dependence. Moreover, even though ObRe is a relevant modulator of leptin bioavailability, the associations of ObRe and the FLI with alcohol dependence have not yet been investigated. Furthermore, to our knowledge, the potential of lipid serum levels, ObRe, and FLI for use in predicting alcohol-related readmissions has not been studied.

In this study, we compared serum levels of triglycerides, total, HDL, and LDL cholesterol, the HDL/LDL ratio, leptin, ObRe, and FLI between alcohol-dependent in-patients and healthy control subjects. We investigated the relationships of these parameters with body mass index (BMI), CDT, GOT, GPT, GGT, and MCV. We tested whether serum lipid and leptin parameters may predict outcome. Because serum lipids and leptin activities are known to depend on the BMI and because previous studies used leptin/BMI ratios (Sandhofer et al., 2003; Kraus et al., 2004; Hillemacher et al., 2007a; Lenz et al., 2012b; Rodrigo et al., 2017), we conducted these analyses again using lipid parameters and leptin activities divided by BMI. Because of the importance of sex differences in science (Tannenbaum et al., 2019) and in particular in alcohol dependence (Lenz et al., 2012a; Clayton and Collins, 2014), we aimed at consistently performing sex-specific analyses.