L3 Genetics Genetics of diabetes and obesity

Overview

The lecture frames obesity and type 2 diabetes as an evolutionary mismatch: a genome selected for caloric scarcity now operating in a post-1980s environment of caloric surplus and sedentary behaviour. It builds from the evolutionary origins of metabolic vulnerability (thrifty-gene selection and genetic drift), through the genetic architecture of obesity (monogenic versus polygenic) and the epigenetic layer through which environment modifies gene expression across generations, to two worked variant examples, FTO in Europeans and CREBRF in Pacifica populations, whose mechanisms and diabetes risk differ sharply. It closes with the polygenic architecture of T2D, polygenic risk scores as an aggregate predictor, and how genetic pathophysiology identifies drug targets.

Lecture objectives as stated: be aware of ideas of genetic predisposition (epigenetics and selection); give examples of how common variants contribute to common diseases (FTO, CREBRF, MC4R); give examples of how genetics has contributed to understanding disease pathophysiology. The lecture links to Tutorial 9 and the Metabolism module.

The evolutionary mismatch premise

• Biology is adapted to a world that no longer exists. The ancient genome was selected for caloric scarcity and high energy expenditure; the modern genome is still adapted for scarcity but sits in an environment of caloric surplus and sedentary behaviour.
• The transition point is the obesogenic environment, dated post-1980s.
• Causal chain: past evolutionary history + early developmental environment → genotype / epigenotype → later environment.
• Core claim: current mismatches between past developmental environments and later environments are the primary drivers of modern metabolic disease.

Evolutionary origins of metabolic traits

Two explanations are contrasted, selection versus chance.

Thrifty gene theory (selection)

• Mechanism: directed natural selection.
• Function: specific genes increased metabolic efficiency to extract and store energy.
• Context: highly advantageous for hunter-gatherers during periods of unpredictable food supply and famine (Neel, 1962).

Genetic drift (chance)

• Mechanism: random fluctuations in gene variant frequencies.
• Function: variants become more or less common purely due to chance.
• Context: heavily influenced by population bottlenecks and founder effects during human migration.

Architecture of obesity: monogenic vs polygenic

• Heritability: genes account for 40 to 70% of the variation in weight in a population.
• 70 to 100 genes identified in GWAS analyses.
• Gene-environment interactions are exacerbated by evolutionary thrift.
• Neuroendocrine control of appetite is covered in a later lecture.

Polygenic obesity (complex)

• Nature: common, continuous trait.
• Mechanism: additive genetic variation; hundreds of alleles each exert a small cumulative effect on the phenotype.

Monogenic obesity

• Nature: rare, severe, early-onset.
• Mechanism: severe disruption of specific hunger/satiation neurological pathways.
• Examples: mutations in leptin pathways and the melanocortin 4 receptor (MC4R).
• MC4R phenotype as given: disrupted “fullness” signals, obsessive food-seeking behaviour, obesity by 10 years.
• Appetite-pathway detail from the inset diagram: a neuron releases AgRP, which acts on and inhibits the MC4R receptor in the membrane of a target cell (AgRP antagonism at MC4R).

Environment shifts the threshold, not the genome

• A population distribution of body weight/BMI sits well left of the obesity threshold (BMI ≥ 30) pre-1980s. The obesogenic environment/lifestyle shifts the whole distribution rightward, so the upper tail of the post-1980s curve now crosses the threshold.
• Core takeaway: the genome has not changed in 40 years. Environmental conditions shift the distribution curve, massively increasing the penetrance of pre-existing genetic variants.

Genes mandate vulnerability, not disease. The same genotypes cross into disease because the distribution moved, not because the sequence changed.

Epigenetics: environmental modification of DNA

• Hardware: the underlying static sequence of DNA base pairs.
• Software: DNA methylation marks specifically at CpG / CG sites. These chemical marks repress or express genes without altering the underlying sequence.
• Environment, lifestyle and the microbiome create epigenetic “marks”; these structural modifications govern whether a gene is actively transcribed or silenced.
• Structural context from the diagram: condensed chromosome → nucleosomes/histone-wrapped DNA → unwound double helix carrying methyl tags.
• Cross-reference given on the slide: ELM2 Genetics Bokor lectures.

Intergenerational transmission of metabolic risk

Three generations are affected simultaneously by one exposure in pregnancy:

1. F0 (grandmother/mother): maternal environment and lifestyle directly affect her own somatic cells.
2. F1 (the fetus): the intrauterine environment directly shapes the developmental epigenetics of the child.
3. F2 (the germ cells): the reproductive cells that will become the next generation are also marked by the F0 environment.

Synthesis: a single environmental exposure during pregnancy can simultaneously alter the epigenetic signatures of three generations.

Epigenetic biomarkers in clinical practice

• Epigenetic changes are highly dynamic, detectable, and persist across a lifetime.
• Workflow: input = patient factors (e.g. obesity, smoking) → detection = DNA methylation scoring from whole blood → output = the AHRR gene.
• AHRR: toxin metabolism, hypomethylation, serves as a persistent quantifiable biomarker for prior exposures such as cigarette smoking.
• EWAS: similar to GWAS but looking at methylation markers.

The FTO variant: the European “hunger gene”

• Gene: fat mass and obesity-associated gene.
• Prevalence: found in 40% of European chromosomes; 16% of people of European descent carry two copies (AA alleles). Highly prevalent (52%) in Yoruban populations, but rare (14%) in Chinese/Japanese populations.
• Phenotypic impact: individuals with two copies (AA) carry an average of +3 kg of body weight compared with non-carriers. Despite being the largest known contributor to European obesity, it accounts for only 1% of total variation.
• Mechanism in order: FTO genetic variant → gene regulatory and epigenetic pathways activated → elevated ghrelin (the hunger hormone) → increased caloric intake.
• Citation given: Schwartz and Morton, Nature 418:595-597.
• Appetite-regulation figure elements listed: neurons, food intake and energy expenditure, the arcuate nucleus, NPY/AgRP and melanocortin signalling, ghrelin from the stomach, leptin from fat tissue, insulin from the pancreas, PYY; key entries were melanocortin receptor (MC4R) blocked by AgRP, ghrelin receptor, NPY/PYY receptor, melanocortin receptor (MC3R), and leptin receptor of insulin receptor.

Flag carried from the transcript: on this slide the inset appetite-regulation figure is reproduced at very small size, its individual labels are only partly legible, and the key entries above are transcribed at the limit of legibility. The "two copies / AA alleles" annotation appears as a small separate annotation in the original text layer.

Obesity-associated genes differ by population

A Venn diagram of obesity-associated genes by population (PubMed 28405013):

• Shared by East Asians, Africans and Europeans: FTO, MC4R, POMC.
• East Asians only: ALDH2, CDKAL1, ITIH4, KCNQ1, KLF9, NT5C2.
• Africans only: BRE, GALNT10.
• East Asians ∩ Europeans: BDNF, GIPR, GP2, MAP2K5, PCSK1, SEC16B.
• Africans ∩ Europeans: MIR148A/NFE2L3.
• Europeans only: ADAM23, ADCY9, AGBL4, AKAP6, ASB4, ATP2B1, BDNF, C9orf93, CADM1, CADM2, CALCR, CBLN1, CBLN4, CLIP1, COBLL1, CREB1, DDC, DMXL2, EHBP1, ELAVL4, ELP3, EPB41L4B, ERBB4, ETS2, ETV5, FAIM2, FAM120AOS, FANCL, FLJ35779, FHIT, FIGN, FOXO3, GBE1, GNAT2, GNPDA2, GPR61, GPR120, GPRC5B, GRID1, GRP, HHIP, HIF1AN, HIP1, HMGA1, HNF4G, HOXB5, HS6ST3, HSD17B12, IFNGR1, INO80E, KAT8, KCNK3, KCNK9, KCNMA1, KCTD15, LEPR, LMX1B, LOC100287559, LOC284260, LOC285762, LPIN2, LRP1B, LRRN6C, MAF, MIR548A2, MIR548X2, MRPS33P4, MTCH2, MTIF3, NAV1, NEGR1, NLRC3, NPC1, NRXN3, NT5C2, NTRK2, OLFM4, PACS1, PARK2, PGPEP1, PMS2L11, PRKCH, PRKD1, PTBP2, RAB27B, RARB, RABEP1, RALYL, RASA2, RMST, RPL27A, RPTOR, SBK1, SCARB2, SDCCAG8, SLC2A10, SLC22A3, SLC39A8, SH2B1, STXBP6, TAL1, TCF7L2, TDRG1, TFAP2B, TLR4, TMEM18, TMEM160, TNKS, TNNI3K, TOMM40, TUB, UBE2E3, USP37, ZBTB10, ZNF608, ZZZ3.
• Native North Americans (separate, non-overlapping circle): MAP2K3.
• Samoans (separate, non-overlapping circle): CREBRF.

The Native North American and Samoan circles are drawn outside and not overlapping the other three, indicating population-specific associations.

The CREBRF paradox: a Pacifica founder effect

• Gene/variant: CREBRF (rs373863828 / p.Arg457Gln).
• Prevalence: 5 to 27% frequency in Pacifica populations. 7% of Samoans carry two copies; 38% carry one copy, which demonstrates a dominant effect.
• Phenotypic impact: carriers of two copies are on average +8 kg heavier. It represents 2% of the total 70% heritability for obesity in Samoa.
• Mechanism in order: CREBRF genetic variant → regulates gene expression → promotes rapid conversion of glucose into fat storage.
• The evolutionary paradox: while it drives severe fat mass accumulation, it paradoxically protects against type 2 diabetes (odds ratio 0.59). A pure “thrifty” survival advantage in the past that has become a liability today.

Replication in NZ Māori and Pacifica

• Yes: risk for obesity, but protective for gestational diabetes and T2D.
• OR 0.59 (95% CI 0.47 to 0.73, $p=1.9\times 10^{-6}$) for type 2 diabetes.
• Source study (Diabetologia 2018, 61:1603-1613): “Discordant association of the CREBRF rs373863828 A allele with increased BMI and protection from type 2 diabetes in Māori and Pacific (Polynesian) people living in Aotearoa/New Zealand”.
• Graphical abstract content, in order: (1) CREBRF is a negative nuclear regulatory factor of the CREB3 translation factor, encoded on chromosome 5q; (2) the missense variant rs373863828 minor A allele is carried by 28% of pregnant Polynesian women with obesity; (3) pregnant women with obesity carrying the minor A allele are 8x less likely to develop gestational diabetes, a negative predictive value of 90%.
• Related paper (2020): “The Pacific-specific CREBRF rs373863828 allele protects against gestational diabetes mellitus in Māori and Pacific women with obesity” (Krishnan, Murphy, Okesene-Gafa, Ji, Thompson, Taylor, Merriman, McCowan, McKinlay).
• The Otago “CREBRF Women’s Study: Can your genes protect you against diabetes?” is referenced.

Synthesising the mismatch: FTO vs CREBRF

	FTO variant	CREBRF variant
Primary population	European descent (16% homozygous)	Pacifica / Samoan (7% homozygous)
Evolutionary driver	Broad population drift	Severe founder effect / selection (voyaging)
Phenotypic weight gain	+3 kg	+8 kg
Mechanism of action	Neurological (elevated ghrelin drives hunger)	Metabolic (cellular pathway rapidly converts glucose to fat)
Type 2 diabetes risk	Increases parallel to weight	Decreases (protective, OR 0.59)

Conclusion: fat mass is a shared outcome, but the biological architecture driving it is fundamentally distinct across populations.

Genetic architecture of type 2 diabetes

As in obesity, many gene variants each with modest effect (odds ratio). Variants associated with T2D at or near genome-wide significance, ordered by chromosome (Florez, J Clin Endo Metab 93:4633, 2008):

Marker	Chr	Description	Gene region	Function	Risk allele	OR	P value
rs10923931	1	Intronic	NOTCH2	Transmembrane receptor implicated in pancreatic organogenesis	T	1.13	4.1 x 10-8
rs7578597	2	Missense: T1187A	THADA	Thyroid adenoma; associates with PPARγ	T	1.15	1.1 x 10-9
rs4607103	3	38 kb upstream	ADAMTS9	Secreted metalloprotease expressed in muscle and pancreas	C	1.09	1.2 x 10-8
rs4402960	3	Intronic	IGF2BP2	Growth factor binding protein; pancreatic development	T	1.14	8.9 x 10-16
rs1801282	3	Missense: P12A	PPARG	Transcription factor involved in adipocyte development	C	1.19	1.5 x 10-7
rs10010131	4	Intron-exon junction	WFS1	Endoplasmic reticulum transmembrane protein	G	1.15	4.5 x 10-5
rs7754840	6	Intronic	CDKAL1	Homologous to CDK5RAP1, CDK5 inhibitor; islet glucotoxicity sensor	C	1.12	4.1 x 10-11
rs864745	7	Intronic	JAZF1	Transcriptional repressor; associated with prostate cancer	T	1.10	5.0 x 10-14
rs13266634	8	Missense: R325W	SLC30A8	β-cell zinc transporter ZnT8; insulin storage and secretion	C	1.12	5.3 x 10-8
rs10811661	9	125 kb upstream	CDKN2A/B	Cyclin-dependent kinase inhibitor and p15 tumour suppressor; islet development	T	1.20	7.8 x 10-15
rs12779790	10	Intergenic region	CDC123-CAMK1D	Cell cycle/protein kinase	G	1.11	1.2 x 10-10
rs7903146	10	Intronic	TCF7L2	Transcription factor; transactivates proglucagon and insulin genes	T	1.37	1.0 x 10-48
rs1111875	10	7.7 kb downstream	HHEX	Transcription factor involved in pancreatic development	C	1.13	5.7 x 10-10
rs5219	11	Missense: E23K	KCNJ11	Kir6.2 potassium channel; risk allele impairs insulin secretion	T	1.14	6.7 x 10-11
rs7961581	12	Intronic	TSPAN8-LGR5	Cell surface glycoprotein implicated in GI cancers	C	1.09	1.1 x 10-9
rs8050136	16	Intronic	FTO	Alters BMI in general population	A	1.17	1 x 10-12
rs757210	17	Intronic	HNF1B	Transcription factor involved in pancreatic development	A	1.12	5 x 10-6
rs2237892	11	Intronic	KCNQ1	Voltage-gated potassium channel	C	1.40	2.5 x 10-40

Key points to take from the table:

• All odds ratios are modest; the odds ratio column was highlighted on the slide to make exactly this point.
• The largest effects shown are TCF7L2 (OR 1.37) and the appended KCNQ1 row (OR 1.40).
• Three functions were circled on the slide as notable: THADA (associates with PPARγ), PPARG (adipocyte development transcription factor) and FTO (alters BMI in the general population).
• Most genes cluster around pancreatic development/islet biology and insulin secretion (NOTCH2, ADAMTS9, IGF2BP2, CDKAL1, SLC30A8, CDKN2A/B, HHEX, KCNJ11, HNF1B) rather than adiposity.

Flag carried from the transcript: this table is a low-resolution reproduction; the red highlight box over the odds ratio column partly overlays the appended KCNQ1 row's p value, and that row's reference column is blank/absent.

Polygenic risk scores

• Input: genome-wide polygenic scores aggregating 2.1 million markers (summing weight-raising and weight-lowering alleles).
• Output: can predict weight and obesity trajectories from birth to middle age.
• Weight difference (kg) between high-score and low-score individuals: birth 0.06, age 8 3.5, age 18 12.3, middle age 13.
• The high-score tail carries increased risk for: extreme obesity, bariatric surgery, coronary disease, heart failure, mortality.
• Source: A. V. Khera et al., Cell 2019.
• Consultative insight: avoid genetic determinism (“my genes make me fat”). Individual variants offer poor risk prediction, but aggregate polygenic risk scores accurately map an individual’s lifelong susceptibility to environmental triggers.

From pathophysiology to targeted pharmacology

Core concept: understanding the precise genetic pathophysiology identifies exact therapeutic targets where modulation yields dramatic results. Two gene → drug → outcome chains, grouped by the defect addressed:

Peripheral tissue resistance

• PPARG → glitazones → dramatically improves insulin sensitivity in peripheral tissues for type 2 diabetes.

Pancreatic secretion failure

• KCNJ11 → sulphonylureas → directly triggers insulin release, serving as a targeted therapy for MODY (maturity-onset diabetes of the young).

Re-evaluating the metabolic landscape

1. The inherited blueprint: obesity and T2D are complex, polygenic traits (~60% heritability) shaped by evolutionary mechanisms like genetic drift and “thrifty” selection.
2. The environmental trigger: genes do not mandate disease; they mandate vulnerability. The post-1980s obesogenic environment acts via epigenetic marks to push genetically susceptible populations past the disease threshold.
3. The clinical future: by abandoning generic models of obesity, we leverage polygenic risk scores for early prediction and use targeted pharmacology (e.g. PPARG/glitazones) to correct precise pathophysiological failures.

Pharmacogenetics of GLP-1 agonists

• Wegovy and Ozempic are GLP-1 agonists.
• Responsiveness is dependent on genetic variants, which is why Wegovy may not work in some people.
• A 23andMe report was set for the following Friday.

Self-test

1. State the premise of the evolutionary mismatch model and name the causal chain it proposes from evolutionary history to disease.
2. Distinguish the thrifty gene theory from genetic drift as explanations for metabolic traits, giving the mechanism and context of each.
3. What proportion of population variation in weight is attributable to genes, and how many obesity genes have been identified in GWAS analyses?
4. Distinguish polygenic from monogenic obesity in terms of nature and mechanism.
5. Describe the clinical phenotype of MC4R-related obesity as given in the lecture, and state how AgRP interacts with MC4R.
6. Explain how the post-1980s environment increased the number of people with obesity even though the genome has not changed.
7. Distinguish the “hardware” and “software” of the epigenome, and state where the methylation marks sit.
8. Describe how a single environmental exposure during pregnancy can affect three generations, naming F0, F1 and F2.
9. Describe the input, detection step and output of the epigenetic biomarker workflow, and explain what AHRR hypomethylation reports.
10. What is EWAS and how does it differ from GWAS?
11. List the prevalence figures for the FTO variant across European, Yoruban and Chinese/Japanese populations.
12. Describe the mechanistic chain by which the FTO variant increases body weight, and state its average weight effect and its contribution to total variation.
13. Which three obesity-associated genes are shared by East Asians, Africans and Europeans, and which two genes appear in circles that do not overlap the other populations at all?
14. Describe the CREBRF variant: its identifier, its allele frequencies in Samoans, and its average weight effect in homozygotes.
15. Explain why CREBRF is described as a paradox, and give the odds ratio that supports this.
16. Describe the finding replicated in NZ Māori and Pacifica populations for gestational diabetes in pregnant women with obesity who carry the minor A allele.
17. Compare FTO and CREBRF across evolutionary driver, mechanism of action and effect on type 2 diabetes risk.
18. Describe the general genetic architecture of type 2 diabetes shown in the Florez table, and name the two variants with the largest odds ratios.
19. Which biological process do most of the T2D-associated genes in the table relate to? Give three examples with their function.
20. Explain how much weight difference separates high and low polygenic score individuals at birth, age 18 and middle age, and what this trajectory shows.
21. Explain why aggregate polygenic risk scores are useful when individual variants are not, and state the caution the lecturer attaches to them.
22. For each of PPARG and KCNJ11, name the drug class targeting it and the clinical outcome.
23. A patient of European descent has a high BMI and reports persistent hunger and high caloric intake. A patient of Samoan descent has a higher BMI still but a lower than expected risk of type 2 diabetes. Explain the distinct genetic architecture likely to underlie each presentation.
24. A patient asks why a GLP-1 agonist has not worked for them. What explanation does the lecture offer?
25. Integrative: using the thrifty gene concept, penetrance and epigenetics, explain how a variant that was advantageous in the past becomes a driver of metabolic disease today, using CREBRF as the worked example.

Answers

Reveal answers

1. Biology is perfectly adapted to a world that no longer exists: the ancient genome was selected for caloric scarcity and high energy expenditure, while the modern genome, unchanged, sits in caloric surplus and sedentary behaviour. Chain: past evolutionary history + early developmental environment → genotype/epigenotype → later environment; mismatch between the two is the primary driver of modern metabolic disease.
2. Thrifty gene: mechanism is directed natural selection; specific genes increased metabolic efficiency to extract and store energy; advantageous for hunter-gatherers facing unpredictable food supply and famine (Neel, 1962). Genetic drift: mechanism is random fluctuation in variant frequencies; variants become more or less common purely by chance; heavily influenced by population bottlenecks and founder effects during human migration.
3. Genes account for 40 to 70% of the variation in weight in a population; 70 to 100 genes have been identified in GWAS analyses.
4. Polygenic: common, continuous trait; additive genetic variation with hundreds of alleles each exerting a small cumulative effect. Monogenic: rare, severe, early-onset; severe disruption of specific hunger/satiation neurological pathways (leptin pathways, MC4R).
5. Disrupted “fullness” signals, obsessive food-seeking behaviour, and obesity by 10 years. A neuron releases AgRP, which acts on and inhibits (antagonises) the MC4R receptor in the target cell membrane.
6. The genome has not changed in 40 years. The obesogenic environment/lifestyle shifts the whole population BMI distribution rightward, so the upper tail crosses the obesity threshold (BMI ≥ 30). This massively increases the penetrance of pre-existing genetic variants rather than creating new ones.
7. Hardware is the underlying static sequence of DNA base pairs; software is DNA methylation marks. The marks sit specifically at CpG / CG sites and repress or express genes without altering the underlying sequence.
8. F0 is the grandmother/mother, whose environment and lifestyle affect her own somatic cells. F1 is the fetus, whose developmental epigenetics are shaped directly by the intrauterine environment. F2 is the fetus’s own germ cells, which will become the next generation and are also marked by the F0 environment. One exposure therefore alters the epigenetic signatures of three generations simultaneously.
9. Input is patient factors such as obesity and smoking; detection is DNA methylation scoring from whole blood; output is the AHRR gene. AHRR is involved in toxin metabolism, and its hypomethylation is a persistent, quantifiable biomarker of prior exposure such as cigarette smoking. Epigenetic changes are dynamic, detectable, and persist across a lifetime.
10. EWAS is an epigenome-wide association study, similar in design to GWAS but looking at methylation markers rather than sequence variants.
11. Found in 40% of European chromosomes, with 16% of people of European descent carrying two copies; 52% in Yoruban populations; 14% in Chinese/Japanese populations.
12. FTO genetic variant → gene regulatory and epigenetic pathways activated → elevated ghrelin (hunger hormone) → increased caloric intake. Homozygotes (AA) average +3 kg versus non-carriers. Despite being the largest known contributor to European obesity, it accounts for only 1% of total variation.
13. Shared by all three: FTO, MC4R, POMC. Non-overlapping circles: MAP2K3 in Native North Americans and CREBRF in Samoans, indicating population-specific associations.
14. CREBRF, variant rs373863828 / p.Arg457Gln. Frequency 5 to 27% across Pacifica populations; in Samoans 7% carry two copies and 38% carry one copy, demonstrating a dominant effect. Homozygotes are on average +8 kg heavier, and the variant represents 2% of the total 70% heritability for obesity in Samoa.
15. It drives severe fat mass accumulation by promoting rapid conversion of glucose into fat storage, yet it paradoxically protects against type 2 diabetes, odds ratio 0.59. It was a pure thrifty survival advantage in the past that has become a liability today.
16. The effect replicates: increased risk for obesity but protection against gestational diabetes and T2D (OR 0.59, 95% CI 0.47 to 0.73, p = 1.9 x 10-6). Pregnant women with obesity carrying the rs373863828 minor A allele are 8x less likely to develop gestational diabetes, giving a negative predictive value of 90%; 28% of pregnant Polynesian women with obesity carry the allele.
17. Evolutionary driver: broad population drift (FTO) versus severe founder effect/selection through voyaging (CREBRF). Mechanism: neurological, elevated ghrelin driving hunger (FTO) versus metabolic, cellular conversion of glucose to fat (CREBRF). T2D risk: increases parallel to weight (FTO) versus decreases, protective with OR 0.59 (CREBRF). Fat mass is a shared outcome but the underlying architecture differs by population.
18. Many gene variants each with modest effect, mirroring obesity. Nearly all odds ratios in the table are between about 1.09 and 1.20; the largest are TCF7L2 (rs7903146, OR 1.37) and KCNQ1 (rs2237892, OR 1.40).
19. Most relate to pancreatic development, islet biology and insulin secretion rather than adiposity. Examples: NOTCH2, a transmembrane receptor implicated in pancreatic organogenesis; SLC30A8, the β-cell zinc transporter ZnT8 involved in insulin storage and secretion; KCNJ11, the Kir6.2 potassium channel whose risk allele impairs insulin secretion. (HHEX, HNF1B and IGF2BP2 in pancreatic development are also acceptable.)
20. Birth 0.06 kg, age 8 3.5 kg, age 18 12.3 kg, middle age 13 kg. The difference is negligible at birth and widens across development, so a genome-wide polygenic score can predict weight and obesity trajectories from birth to middle age; the high-score tail carries increased risk of extreme obesity, bariatric surgery, coronary disease, heart failure and mortality.
21. Individual variants offer poor risk prediction because each has a small effect, but aggregating 2.1 million markers into a score accurately maps an individual’s lifelong susceptibility to environmental triggers. The caution is to avoid genetic determinism: genes mandate vulnerability, not disease.
22. PPARG is targeted by glitazones, which dramatically improve insulin sensitivity in peripheral tissues in type 2 diabetes, addressing peripheral tissue resistance. KCNJ11 is targeted by sulphonylureas, which directly trigger insulin release and serve as a targeted therapy for MODY, addressing pancreatic secretion failure.
23. The European patient’s presentation fits FTO: a common European variant acting through gene-regulatory and epigenetic pathways to raise ghrelin, increasing hunger and caloric intake, with about +3 kg in homozygotes and T2D risk rising in parallel with weight. The Samoan patient’s fits CREBRF: a founder-effect variant regulating gene expression to convert glucose rapidly into stored fat, about +8 kg in homozygotes, but protective against T2D with OR 0.59.
24. Wegovy and Ozempic are GLP-1 agonists, and responsiveness to them depends on genetic variants, so some individuals do not respond.
25. The thrifty gene concept holds that variants increasing metabolic efficiency for extracting and storing energy were selected because they aided survival through unpredictable food supply and famine. CREBRF does exactly this, promoting rapid conversion of glucose to fat storage, and reached high frequency in Pacifica populations through a severe founder effect during voyaging, while also protecting against T2D. The variant itself has not changed, but the post-1980s obesogenic environment shifts the population weight distribution rightward and epigenetic marks laid down by environment, lifestyle and the microbiome alter gene expression, so the same genotype now crosses the obesity threshold. Penetrance rises without any change in sequence: an advantage under scarcity becomes a liability under surplus.