Gene Variants of the SLC2A5 Gene Encoding GLUT5, the Major Fructose Transporter, Do Not Contribute to Clinical Presentation of Acquired Fructose Malabsorption

Irina Taneva Medical Laboratory for Clinical Chemistry, Microbiology, Infectious Diseases and Genetics “Prof.Schenk/Dr.Ansorge & Colleagues" Dorothee Grumann Medical Laboratory for Clinical Chemistry, Microbiology, Infectious Diseases and Genetics “Prof.Schenk/Dr.Ansorge & Colleagues" Dietmar Schmidt Medical O ce Internal Medicine & Gastroenterology Elina Taneva Medical Laboratory for Clinical Chemistry, Microbiology, Infectious Diseases and Genetics “Prof.Schenk/Dr.Ansorge & Colleagues" Ulrike von Arnim Department of Gastroenterology, Hepatology and Infectious Diseases, Otto-von-Guericke University Thomas Ansorge Medical Laboratory for Clinical Chemistry, Microbiology, Infectious Diseases and Genetics “Prof.Schenk/Dr.Ansorge & Colleagues" Thomas Wex (  t.wex@schenk-ansorge.de ) Medical Laboratory for Clinical Chemistry, Microbiology, Infectious Diseases and Genetics “Prof.Schenk/Dr.Ansorge & Colleagues"


Introduction
Gastrointestinal symptoms are frequent causes for medical check-up in medical o ce and clinics as well.
Pathophysiological mechanisms are numerous including infectious causes (viral, bacterial, parasites), immunologically-related disorders (e.g. re ux disease, gastritis, coeliac disease, and chronic in ammatory bowel disease), intolerance towards nutrients or components (e.g. food-related allergies) and malabsorption syndromes of nutrients such as lactose (1) and fructose (2) as most common ones.
Malabsorption syndromes have rarely monogenic causes. Pathogenic mutations strongly affecting function of lactase (1,3) or aldolase B (2,4) as key enzymes for metabolizing lactose and fructose, respectively, are present in very few subjects. The overall majority of patients suffering from lactose-and fructose-related symptoms are caused multifactorally by hereditary, environmental, sociology-economic factors;in particular diet plays are major role. Lactose-related disorders had been widely studied for decades and have been comprehensively understood concerning its multi-factorial pathogenesis including genetic polymorphisms (5)(6)(7). In contrast, fructose-related disorders are not well understood.
Notably, fructose-associated disorders have sharply increased over the last 2 decades strongly associated with increasing demand of fructose consumption in a variety of food (8-10). Incomplete intestinal absorption of fructose can lead to various symptoms in vulnerable subjects such as atulence, diarrhea, bloating, nausea and pain (10)(11)(12). Routinely, diagnosis of fructose malabsorption syndrome is mainly performed by hydrogen breath test after ingestion of de ned amount of fructose (12). Several studies demonstrated pathophysiological role of fructose transporters, in particular GLUT5, for the development of fructose-associated symptoms, although divergent results were reported. Mice experiments (either knockout or feeding models) highlighted a predominant role of slc2a5 gene (glut5) as major intestinal transporter for fructose and signaling molecule for the induction of down-stream acting genes encoding fructolytic and gluconeogenic enzymes (13)(14)(15). Furthermore, slc2a5 -mRNA was found to be regulated by age and fructose in rat model (16,17). In human intestinal cell line models fructose was shown positively regulate SLC2A5/GLUT5 expression by transcriptional and posttranscriptional regulation (18, 19), however ex vivo studies of human samples did not show differences in SLC2A5-mRNA or GLUT5 protein content in intestinal biopsies from patients with fructose intolerance (20). In addition to GLUT5 in intestine, the glucose-transporter GLUT2 contributes to the hepatic uptake of fructose since GLUT5 is weakly expressed in liver (21). Very recently, several studies revealed potential regulatory role of GLUT5 for malignancy and proliferation of various tumor cells driven by high consumption of fructose as carbon source (22,23). Based on these recent observations, diagnostic and therapeutic potential of targeting this metabolic pathway has been started to be investigated (24)(25)(26)(27). The only study addressing the role of gene variants of SLC2A5/GLUT5 for the development of fructose malabsorption studied 8 patients suffering from isolated fructose malabsorption without nding any functionally relevant gene variant. Mutational analysis was performed by single strand conformation polymorphism analysis and in one index case by Sanger sequencing (28).
In order to analyze potential role of SLC2A5/GLUT5 gene variants in a larger cohort, a genomic region comprising of complete coding region, adjacent introns and 700bp of proximal promoter region of 35 subjects with acquired fructose malabsorption were comprehensively analyzed by Sanger sequencing.

Study design and patients
Study design was composed of retrospective and prospective parts ( Figure 1

Statistical analysis
Categorical data are expressed as absolute numbers with percentages. Age is shown as a mean with standard deviation. Frequencies of identi ed gene variants were compared towards public databases such as "thousand genome project". Due to the low numbers of most gene variants identi ed, data are presented descriptively in most cases. If possible, allele frequencies were analyzed by χ2 test. Differences were considered statistically signi cant when p < 0.05.

Results
Characterization of study group As summarized in Table 1, the majority of patients was female (29/35) and the mean age was about 38 years. Retrospective evaluation of clinical records revealed abdominal pain and diarrhea as leading symptoms. Data concerning the duration of symptomatic disease was available for 40% of the study group. Furthermore it was shown that 34 out of the 35 patients analyzed had negative serology for coeliac disease; either anti-deamidated gliadin IgG and IgA antibodies (n=30) or anti-tissue transglutaminase IgG/IgA (n=4) ( Table 2).

Molecular analysis of SLC2A5 (GLUT5)-gene
Sequence analysis of SLC2A5 gene of 35 patients was successfully performed for all 35 patients. In total, 10 gene variants were identi ed from those 5 were frequent and ve were identi ed just once (Table 3). One heterozygous synonymous gene variant, leaving amino acid p.Leu278 unchanged, was localized in exon 8 of the SLC2A5 gene ( Table 3). The ve frequent gene variants are located in promoter and adjacent intronic regions and demonstrated frequencies between 10.0 and 47.1%. Comparison of identi ed allele frequencies between public databases and own data demonstrate similar ranges for frequent variants ( Table 3). The ve rare variants (including the missense variant) were identi ed in individual patients only at heterozygous state. Based on the criteria of the guidelines of the "American College of Medical Genetics" (ACMG) none of the gene variants are considered having pathological relevance (Table 3).

Discussion
Here, we demonstrated that gene variants of SLC2A5 gene encoding the fructose transporter GLUT5 are not generally involved in the pathogenesis of acquired fructose malabsorption. Based on the number of subjects studied (n=35) we cannot exclude a rare role of such gen variants in every individual patient suffering from fructose malabsorption, but the rate of these cases having "pathogenic GLUT5 variants" among all these patients are likely to be < 3-5%.
A major strength of the study is the clinically based de nition of the study cohort prior to molecular analysis. Related gastrointestinal disorders such as hereditary lactose-intolerance and acquired lactose malabsorption, both leading to pathological lactose breath test, were excluded. Hereditary ALDOB de ciency was practically ruled out by analyzing the three major mutations of ALDOB leading to this disease. Coeliac disease, also considered as chimera among gastrointestinal diseases, was kept out of the study group by two approaches. First, since secondary lactose malabsorption is a well-known leading symptom for coeliac disease, these patients were excluded by abnormal lactose breath test. Second, for 34 out of the 35 individuals, corresponding serological parameters (IgA/IgG tissue anti-transglutaminase or deamidated anti-gliadin antibodies) were found to be negative.
The symptoms reported by our patients are rather unspeci c and in line with those reported in similar studies with patients (I) suffering from fructose malabsorption tested by breath test (29)(30)(31), or (II) classi ed as having irritable bowel disease (IBS) (32,33). However it is notable that symptoms in our studies were assessed retrospectively only by analyzing patients' medical records, and no structured interview or assessment of questionnaire in context to e.g. IBS-related Rome criteria (34) was performed.
In summary, the 35 subjects included in the molecular analysis of the SLC2A5/GLUT5 gene presented (I) clinically relevant symptoms that are consistent with acquired fructose-malabsorption, (II) demonstrated abnormal fructose-induced breath test and (III) relevant other related diagnoses (e.g. coeliac disease, hereditary fructose intolerance) were basically excluded. Taken together, we are con dent that overall the great majority of these 35 subjects are patients suffering from acquired fructose malabsorption. It has been shown that the individual ability of metabolizing fructose for subjects without any side effects differs widely from 5 -50 g (reviewed in 21) supporting the multi-factorial etiology of acquired fructose malabsorption.
The aim of the study, molecular analysis of SLC2A5/GLUT5 gene concerning gene variants associated to fructose malabsorption, was based on other studies showing the role of SNPs/mutations affecting the uptake/metabolism of related sugars. Variants including partial de ciency of sucrose-isomaltase were shown to be associated with IBS (35). Analysis of UK biobank data revealed that gene variants in human ketohexokinase gene are associated with loss of function and resulting in the rare benign condition of fructosuria (36). In vitro mutation analysis in rats between GLUT5 and its closest related transporter (GLUT7) revealed that single amino acids (e.g. p.Gln166Glu) are responsible for the speci c transport of fructose, and mutation of this residue to p.166Glu results in the uptake of glucose, whereas other variants and chimera between GLUT5 and GLUT7 demonstrated strong reduction or even complete lack of fructose uptake (37).
The fact that the allele frequencies of the 10 gene variants between ours and those reported in database were very similar strongly implies that none of these variants have a relevant role for the clinical manifestation of acquired fructose malabsorption. Notably, 5 of the 10 gene variants were singular ndings that do not allow nal any general conclusion due to study size. But taken into account the very low frequencies reported in databases, the potential relevance for the very frequent fructose-absorption syndrome seems to be very limited.
While this study demonstrate that SLC2A5/GLUT5-related gene variants do not playing a relevant role in the pathogenesis of acquired fructose malabsorption, other pathogenic factors have been recently identi ed to be associated with this disorder. Trelis and co-workers identi ed a frequent association of the disease with the infection of parasites, in particular Giardia intestinalis (38). Several animal studies identi ed speci c changes in the gut microbiome in context to genetic host factors (39) and the intake of fructose (39,40) showing that Akkermansia spec. seems to play an important role in the prevention fructose-induced metabolic dysregulation. Over-expression of slc2a5 in slc2a5/glut5-knock out mice let to profound increase of fructose utilization and subsequent higher levels of Clostridium and Enterococcus spec. (41). Overall most related studies demonstrate that higher intestinal luminal levels of fructose caused by changes in fructose consumption or absorption will likely affect bacterial load and composition of the microbiome (reviewed in 42).
In humans, several studies highlighted the role of the transcription factor ChREBP encoded by the MLXIPL gene for the predisposition concerning fructose intolerance malabsorption (21,43) and diarrheapredominant IBS patients with impaired intestinal fructose transport (44). The association between ChREBP and fructose malabsorption was further supported by animal models (45,46). Nuclear receptor LXR (lxralpha, NR1H3) is another transcriptional regulator of GLUT5 expression identi ed in mice and human that is thought to be a potential pharmaceutical target for selective modulation of GLUT5 expression in context to cancer and metabolic disease (47). Notably, authors identi ed a functional LXR responsive element in the human SLC2A5/GLUT5 promoter region located at position -385 based on transcriptional start site, but none of our 35 patients showed a variation at this position.
Overall, these different ndings strengthen the hypothesis that fructose-related malabsorption syndrome associated with different pathological conditions has multi-factorial etiology. Different transcriptional regulatory patterns affecting the SLC2A5/GLUT5 gene expression contribute to the pathology, whereas gene variants of SLC2A5/GLUT5 including the promoter region, which was the focus of this study, do not play a relevant role.

Declarations
Ethics approval and consent to participate Study scheme and experimental protocols were approved by the Ethic Committee of the Land Sachsen-Anhalt (Vote No. 19/15). The study was performed in accordance with the Helsinki Declaration. All patients who participated in the study provided signed informed consent.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analysed during this study are included in this published article.

Competing interests
The authors declare that they have no competing interests.

Not applicable
Authors' contributions I.T. included patients, acquired data, analyzed primary data, and was involved in writing of the manuscript. D.G. included patients, acquired data, analyzed primary data, and was involved in writing of the manuscript. D.S. included patients and provided retrospective data of study cohort. E.T. was involved in providing funding, planning of study, providing data of breath tests. U.v.A. was involved in study design, assessment of serological data. T.A. was involved in providing funding, planning of study, providing data of breath tests. T.W. was involved in study design, inclusion of patients, primary data analysis, and writing of the manuscript. All authors critically reviewed and approved the manuscript.  Multiple symptoms were possible. Note that data were retrospectively recorded from medical records of patients; no structured interview was performed.  Study scheme