Sugar-sweetened beverages (SSBs), such as sodas, fruit-flavored drinks, and sports drinks, are a significant source of dietary added sugars and a major contributor to excess energy intake (1). Global averages of SSB consumption range up to one 8-ounce serving/day (2), and contribute between 3 and 10% of daily energy consumption (3–6). Observational data suggest that higher SSB consumption is linked to a host of chronic diseases, including cardiovascular disease (CVD), type 2 diabetes (T2D), obesity, non-alcoholic fatty liver disease (NAFLD), and gout (7–11). Consequently, dietary guidance consistently recommends limiting added sugar consumption, particularly from SSBs (1, 12, 13). Although secular trends in dietary behavior suggest a decline in SSB consumption in recent years, national surveys suggest that >50% of US and European youth and adults continue to consume at least one serving of SSB daily (3, 5, 6, 14–16). Thus, SSB consumption continues to be a major public health concern globally.

Observational and experimental evidence linking SSB consumption to heritable metabolic risk factors and disease risk, i.e., those with underlying genetic predispositions, have paved the way for gene–SSB interaction studies (17–19). These gene–diet interaction studies may provide insight into the molecular mechanisms by which SSB consumption influences disease risk. From a public health perspective, this knowledge could be used to develop personalized dietary recommendations for the primary prevention and treatments of chronic diseases, and may provide motivation for patients to adhere to lifestyle guidance (20). The purpose of this review is to summarize current gene–SSB interaction studies on chronic disease risk factors and disease outcomes.

We conducted a comprehensive PubMed search for literature on SSB consumption and genetic interactions for publications through July 2017. Our search strategy combined terms for SSB consumption [“sugar-sweetened beverage(s),” “soda(s),” “sugar,” “beverage(s),” “drink(s)”], diet (“diet,” “dietary,” “food”), genetics [keywords: “gene(s),” “genome,” “genetic”], and/or interactions [“interaction(s),” “modify(ies),” “modification(s)”]. We also cross-referenced recovered studies and relevant reviews to identify additional studies. We identified nine published studies linking SSB to a variety of health outcomes with consideration for underlying genetic predisposition (Table 1). Studies included cross-sectional and longitudinal analyses of population-based cohort studies, and meta-analytic data from population-based cohort studies. The genetic variants investigated were either limited to single loci or expanded multiple loci (Table 2).

In this review, we have summarized the limited body of evidence describing how genes implicated in various diseases may interact with SSB consumption to modify cardiometabolic health and chronic disease risk. To date, the strongest evidence for interaction between genes and SSB consumption is for obesity in three independent studies that consistently support a link between a genetic risk for obesity and SSB consumption on changes in adiposity (26–28). The majority of current evidence suggests that the adverse effects of SSB consumption on health may be strongest among subgroups with a genetic predisposition for the adverse metabolic outcomes, with some exceptions. In two studies, one related to changes in body weight (28) and one related to gout (67), the genetic predisposition seems to attenuate or possibly mask the association between SSB intake and the outcome.

The public health implications of these findings should be carefully interpreted both in context of the effect size of the interaction and in the frequency of the effect variant. For impactful personalized nutrition, the effect of the interaction should be clinically meaningful and the frequency of the effect variant common among the population. Such an interaction may be helpful in serving as a motivational tool to encourage compliance with guidance on lifestyle modification for select individuals (20). In the current review, effect sizes of interactions ranged from larger interactions with likely impactful effect sizes (26, 51) to smaller effect sizes (28, 37). While interactions with smaller effect sizes may have unclear clinical and translational impact, they may provide unique biological insight and help frame subsequent research questions.

Inferring potential mechanisms by which genetics and SSB consumption may influence disease onset is more feasible in interaction studies involving genetic variants with known functions. Batt and colleagues present an example of a gene by SSB consumption interaction study with the potential to provide insight into the molecular origin of increased risk of gout for some individuals (67). It is known that SLC2A9 encodes the GLUT9 transporter, a renal uric acid transporter (59, 65, 68), and genetic variants of the transporter associate with blood levels of uric acid and risk of gout (66, 69–71). SSB are composed of 50% fructose, and a major end product of fructose metabolism is uric acid. Genetic variants associated with a reduced ability to clear the increased uric acid produced during fructose metabolism may be one mechanism by which increased SSB consumption could lead to increased blood levels of uric acid in some individuals (59, 72). Thus, it is possible that individuals with a SLC2A9 mutation that leads to a defective GLUT9 transporter may have decreased ability to eliminate high uric acid loads produced from high-fructose metabolism following high SSB consumers. Thus, these individuals may generally have higher blood uric acid levels, so even a small increase in mean uric acid levels may lead to an increase in gout risk. Potential for mechanistic insights such as this cannot be determined from studies of genes with limited knowledge of their biological function or a host of pleiotropic genes.

There are general differences in the methods used to select genetic variants among the studies described here. Single genetic variants previously known to be strongly associated with the outcomes of interest from genome-wide association studies were selected for T2D, CVD, NAFLD, and gout, while an aggregate score, also related to the outcome, was selected for obesity and MetS. In our research with respect to glycemic traits (37), we have selected genetic variants related to both the exposure and outcome. In the case of the adiposity genetic risk score analyses, as the functionality of several BMI-related loci remain largely unknown, it is difficult to pinpoint mechanisms by which the score aggregating several of these variants interacts with SSB consumption. Thus, individual variant interaction tests may be insightful as single variants can be more readily mapped to a biological function. Both Qi and colleagues (26) and Brunkwall and colleagues (27) conducted follow-up analyses and provided evidence for modest effects of each individual variant by SSB consumption interaction on BMI. While nominal significance was observed for a few individual SNPs [Qi and colleagues: rs543874 in SEC16B (p = 0.0003) and Brunkwall and colleagues: rs1555543 in PTBP2 (p = 0.02)], these findings do not show significance when accounting for multiple testing. On the contrary, Olsen and colleagues provide evidence that GPRC5B rs12444979 individual interaction with SSB consumption for change in waist circumference, which warrants further follow-up. Current knowledge regarding genetic variants related to variability in SSB consumption is limited. A genome-wide association analysis of macronutrient intake has identified one locus that was associated with carbohydrate intake (73). However, efforts to disclose genetic variants related to single foods and beverages are ongoing and are made possible by large biobanks with genome-wide genetic data and dietary intake data (https://biobankengine.stanford.edu/coding/INI1309).

Direct-to-consumer genetic testing companies that offer nutrition recommendations currently do not offer personalized recommendations related to SSB consumption as they do not account for the interactions related to SSB consumption described in this Review. While they may not include these recommendations simply because of the greater risk of chronic diseases with higher SSB consumption in the general population, it may also, in part, be driven by various limitations in the current evidence. With the exception of obesity, the described studies are of limited sample sizes and have not been replicated. For example, the studies related to NAFLD and MetS had modest sample sizes of 200 and 828, respectively. In the case of obesity, effort has been placed on replicating interaction findings first reported by Qi et al. in non-U.S. populations and despite minor differences in findings, the consistency of the results across longitudinal studies with multiple sampling of diet increased confidence in the finding. Attempting replication of other recent interaction findings in larger and more diverse populations, with prospective designs, is warranted to corroborate initial findings in this field as has been recently conducted for T2D (74). For the purpose of generalizability of findings, replication attempts should include populations of different age groups to account for variability in SSB consumption with age (i.e., higher consumption among younger individuals) (26, 67), and cohorts from various countries to account for differences in SSB formulation (i.e., Europe SSB are sweetened with sucrose, which is composed of 50% fructose and 50% glucose, whereas a higher proportion of SSB in the U.S. are sweetened with high-fructose corn syrup composed of 55% fructose and 45% glucose). Other considerations for future research include differences in the exposure, SSB consumption, both in terms of dietary assessment methodology (i.e., semi-quantitative food-frequency questionnaire or 7-day food records) and in SSB consumption definition (i.e., inclusion of fruit juice). Including fruit juice in SSB definition may contribute to different results (51).

In summary, the detrimental effects of SSB consumption on disease risk is of public health concern, regardless of genetic predisposition. Imamura and colleagues estimated that SSB consumption could contribute to 1.8 million T2D events in the US over 10 years, and to 2.6 million events in the UK, even after accounting for obesity status (32). The gene by SSB consumption interaction data to-date are interesting and may suggest that some individuals who consume greater SSB may be more susceptible to greater risk of adiposity; however, further studies are needed on the effect of sugars in those with genetic predisposition for diabetes or CVD. In the meantime, there is a continued need to develop public health interventions that reduce the consumption of SSB globally.

HD contributed to the conception and design of the research; and DH and HD contributed to the acquisition, analysis, and interpretation of the data. All authors drafted the manuscript, critically revised the manuscript, agreed to be fully accountable for ensuring the integrity and accuracy of the work, and read 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. The handling editor declared a shared affiliation, though no other collaboration, with one of the authors HD.