T2D: Unclear role of unclassified genesJanuary 6, 2020
Diabetes is a heterogeneous group of metabolic disorders whose associated complications cause significant morbidity and mortality. Precision medicine involves tailoring the treatment to the patient. There is increasing interest in applying the concept to diabetes. The precision medicine approach is widely used in oncology and is rapidly becoming an important tool in the medical field, especially due to the emergence of new technologies such as next generation sequencing (NGS). NGS-enabled precision medicine is also used to treat rare monogenic subtypes of diabetes, but there are challenges in applying the same to T1D and T2D. The first step in developing targeted treatments, particularly in common polygenic subtypes of diabetes, is to understand the often-subtle differences in pathophysiology and factors which may influence treatment responses.
Gene sequencing, especially of the candidate genes for a disease in question, are usually done using Sanger sequencing, which is accurate. However, it is relatively slow and expensive, as single genes need to be sequenced sequentially. This makes it more difficult to be used in disorders where there are significant overlaps in the phenotype, both within and between different genetic aetiologies, and where the genetic causes are yet to be understood. NGS is a new technology that allows the parallel sequencing of many genes at a reasonable cost.
Differential treatments can be given after NGS analysis of HNF1A/4A mutations in case of monogenic diabetes, like in maturity-onset diabetes of the young (MODY), neonatal diabetes due to potassium channel gene mutations and mild fasting hyperglycemia due to glucokinase mutations. Furthermore, early detection of diabetes caused by a single gene allows early prediction of other clinical aspects associated with that specific gene, thus enabling the possibility of interventions soon after diagnosis, especially within the first six months of life, in the case of neonatal diabetes.
To date, there are 24 genetic causes of neonatal diabetes identified and mutations in KCNJ11 gene are the commonest, occurring in almost one-third of all the cases. The KCNJ11 gene encodes for Kir6.2 subunit of the pancreatic ATP-dependent potassium channel that links glucose to insulin secretion. Mutations in this gene results in KATP channel being unresponsive to metabolically generated ATP. Thus, the babies show insulin deficiency and are treated with insulin injections.
Several experiments suggested the possibility of using sulphonylureas, typically used in T2D, to bind and close the KATP channel in permanent neonatal diabetes (PNDM). Being a good example of precision medicine, this cannot be applied generally due to complexities and other gene mutations like that on the insulin (INS) gene. Heterozygous dominant-negative INS gene mutations cause abnormal preproinsulin and proinsulin. Therefore, a permanent insulin treatment will be needed. This kind of genetic heterogeneity is common for all subtypes of diabetes, including the common polygenic forms (T1D and T2D), and helps in defining discrete sub-groups for potential drug repurposing or new drug discovery.
Several genome-wide association studies (GWAS) to understand polygenic diabetes resulted in over 100 T2D susceptibility loci being discovered, but the determining functions of these associated gene variants and identifying their causal links are very difficult. Despite these difficulties and complexities, these genetic risk variants could be useful in assisting diagnosis and for selecting the correct treatment. For example, in T1D, a genetic risk score based on 30 T1D-associated risk variants — each weighted according to individual risk contribution — can reliably differentiate T1D from T2D and T1D from monogenic diabetes (Oram R.A et al 2016, Diabetes Care, Vol. 39 No.3, pp 337-344 & Oatel KA et al 2016, Diabetes Vol 65 No.7, pp 2094-2099). Such platforms have significant impact in terms of making an early and correct clinical diagnosis and starting suitable interventions. It is also important to note that the genes associated with monogenic diabetes may also be implicated in polygenic disease. For example, common E23K mutations of KCNJ11 gene have been shown to be associated with T2D susceptibility (54,100). However, small-size effects of genetic risk variants become an important limiting factor for the translation of such data into clinical practice.
Influence of non-glycaemic pathways
Glycated haemoglobin (HbA1c) is an accepted diagnostic test for T2D and several studies have shown more than 60 different genetic loci altering its levels in glycaemic pathways (Franklin CS et al Annals Hum Genetics, 74 (2010), pp. 471-478, Pare G et al PLoS Genetics, 4 (2008), p. e1000312, Soranzo N et al Diabetes, 59 (2010), pp. 3229, Chen P et al Diabetes, 63 (2014), pp. 2551-2562, Ryu J et al Hum Mutat, 33 (2012), pp. 655-659) and non-glycaemic pathways (Panzer S et al Blood, 59 (1982), pp. 1348-1350, Coban E et al Acta Haematol, 112 (2004), pp. 126-128). As HbA1C is a diagnostic test for diabetes, genetic variation acting through non-glycaemic pathways that falsely lower HbA1c can result in missed T2D cases (Byr L et al Clin Chem, 47 (2001), pp. 153-163 & Venkataraman K et al Diab Med J Br Diab Assoc, 29 (2012), pp. 911-917). Conversely, raised HbA1c can result in over-diagnoses.
The role of unclassified genetic variants influencing HbA1c are unclear and may act through glycaemic or erythrocytic pathways, but more studies are needed to evaluate these findings. In case of the FN3K locus, it was identified in GWAS to be associated with HbA1c, but not through glycaemic or erythrocytic indices. The FN3K gene encodes for fructosamine 3-kinase, which is involved in the deglycation of glycated proteins, a nonglycaemic nonerythrocytic mechanism that likely influences HbA1c.
One of the major drawbacks
of the discovery and use of genetic markers is that they are limited by the low power of the studies as well as
by a lack of availability of cohorts of patients with specific diseases.
There is a need for clinicians and scientists to work in close cooporation for translational research.