Dominant-negative effect of lactase missense variants: hetero-complex assembly with the wild-type enzyme impairs intracellular trafficking and digestive function

Recent papers in Gut have highlighted the importance of the genetic makeup and the influence of the gut microbiome on the amount of lactose in the gut in lactose-intolerant individuals.^{1 2} Here, we present biochemical data strongly suggesting that heterozygous carriers of certain lactase gene (LCT) variants could also be lactose-intolerant.

It is widely known that lactose malabsorption in adult-type hypolactasia (ATH) is caused by lactase downregulation beyond infancy.³ Another form of lactose intolerance is congenital lactase deficiency (CLD), a rare disease caused by biallelic LCT variants in the coding region that manifests neonatally with severe diarrhoea and dehydration after lactose ingestion with food.⁴ LCT encodes lactase-phlorizin hydrolase (LPH), a β-galactosidase, exclusively expressed in enterocytes and localised at the brush border membrane where it digests luminal lactose.⁴ Pathogenic forms of LPH elicited by LCT variants are characterised by intracellular block in the ER and complete loss of digestive activity.⁵ Since dimerisation of LPH monomers in the ER is an essential biosynthetic step,⁶ we asked whether an interaction between wild type (LPH ^wt ) and mutant monomers (LPH ^mut ) may occur in heterozygotes.

We therefore investigated the interaction of FLAG-tagged LPH ^wt with myc-tagged missense (p.G1363S-myc, p.S1124L-myc, p.S688P-myc, p.R1587H-myc) and truncated (p.E1612*, p.S1150Pfs*19, p.Y1390*) LPH mutants (see online supplemental appendix 1). All are pathogenic and severely affect function and trafficking.^{5 7} The potential interaction was investigated by coexpression of LPH ^wt and LPH ^mut in COS-1 cells, followed by reciprocal immunoprecipitation with anti-myc, and Western blotting with anti-FLAG or anti-LPH antibodies. As already shown, COS-1 cells, although neither intestinal nor polarised, can be used for meaningful studies of function and trafficking competence.^{6 8} Our data demonstrate an interaction between LPH ^wt and each of the full-length missense mutants (figure 1A). These LPH ^mut dominantly impacted trafficking with a complete block in the ER, assessed by endoglycosidase H-sensitivity indicating a mannose-rich type of glycosylation (figure 1A) and by confocal microscopy (figure 1C). Strikingly, interaction of these mutants resulted in complete loss of the digestive function of the heterodimer towards lactose (figure 1D, upper panel). In contrast, all three truncated mutants neither interacted with myc-tagged LPH ^wt (figure 1B) nor impacted enzymatic activity (figure 1D, lower panel).

Figure 1.

Interaction between wild type (LPH ^wt ) and mutant monomers (LPH ^mut ) and its effects on intracellular trafficking and digestive function. (A) Immunoprecipitation of coexpressed FLAG-tagged wild type LPH (LPH ^wt-FLAG ) and myc-tagged mutants of LPH (LPH ^mut-myc ) in COS-1 cells was carried out using anti-myc antibody. The immunoprecipitates were digested with endoglycosidase H (endo H) and further analysed by Western blotting with anti-FLAG antibody. The blots demonstrate unequivocally that the wild type LPH and all full length LPH mutants interact avidly and the resulting heterodimeric complex (LPH ^wt /LPH ^mut ) is fully endo H-sensitive indicative of an intracellular block in the ER/early secretory pathway. (B) Immunoprecipitation in coexpressed LPH ^wt-myc and the truncated LPH mutants (LPH ^mut ) was carried out with anti-myc. The immunoprecipitates were treated with endo H and further analysed by Western blotting with anti-LPH antibodies. The blot does not reveal protein bands corresponding to truncated LPH ^mut indicating that an interaction between wild type LPH ^wt and truncated LPH ^mut mutants did not occur. (C) Immunofluorescence images of coexpressed LPH ^wt-FLAG (blue) with LPH ^mut-myc (green) were identified by confocal laser microscopy. KDEL-ds Red (red) was used as a marker of the ER. White bars=20 µm. (D) LPH ^wt-FLAG was coexpressed with the various myc-tagged LPH ^mut-myc in COS-1 cells followed by immunoprecipitation of the cellular lysates with anti-myc antibody (upper panel). Pro-LPH ^wt-myc was coexpressed with the truncated pro-LPH ^mut in Cos-1 cells followed by immunoprecipitation of the cellular lysates with anti-LPH antibodies (lower panel). The samples were split in two parts and either assayed for enzymatic activity using lactose as substrate or analysed by Western blotting. The protein band intensities were quantified and the enzymatic activities of the LPH proteins were related to these intensities. Relative specific activities are measured in comparison to the homozygous control (LPH ^wt-myc +LPH ^wt or LPH ^wt-myc +LPH ^wt-FLAG ) that was set to 100%. SEM, n≥3, paired Student’s t-test, *p<0.05, **p<0.005, and ***p<0.0005. LPH, lactase-phlorizin hydrolase.

This study unequivocally demonstrates that full-length LPH ^mut interact efficiently with LPH ^wt and abolish the digestive function of resulting heterodimers, with the transmembrane domain being essential for proper dimerisation.

The question remains as to the relevance of this observation with respect to actual enzyme activity of heterozygotes. A heterozygous carrier of an LCT variant can usually be expected to have 25% of unaffected homodimers and ~25% of normal activity. Depending on effects on allosteric regulation or degradation, enzyme activities can also be significantly lower. The number of families with CLD and thus the number of known obligate heterozygotes is extremely small with 14 different LCT variants described, including only six missense variants,^{5 9} which may explain the current lack of information about any clinical symptoms and diet in heterozygotes.

For LPH, the clinical picture could even be complicated, if a heterozygote carried an ATH allele, a genetic variant upstream of LCT,³ on the other chromosome. Despite the high frequency of ATH worldwide, calculations on such compound heterozygosity are difficult due to large regional differences in the frequency of both ATH¹⁰ and CLD.⁹ Nevertheless, it should be considered in unusual forms of lactose intolerance and investigated accordingly.

In summary, we present a novel concept of a dominant-negative effect of LCT missense variants suggesting that clinical effects occur in heterozygous carriers of such variants and that clinical differentiation from ATH could be difficult and should be considered and investigated in future studies.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

Not applicable.