rs3787348 — PTPN1 PTPN1 intronic variant

PTP1B's Brake on Weight Loss — The PTPN1 rs3787348 Variant

When you try to lose weight, two hormones drive most of the work: insulin directs your cells to take up glucose, and leptin tells your hypothalamus to suppress appetite and ramp up energy expenditure. Both signals are time-limited — an enzyme called PTP1B, encoded by the PTPN1 gene on chromosome 20, dephosphorylates and inactivates their receptors, ending each signaling pulse. When PTP1B is more active or more abundant, those windows close faster, and both insulin efficiency and leptin responsiveness decline.

rs3787348 sits within an intron of PTPN1, inside the same ~100-kb haplotype block¹¹ ~100-kb haplotype block
A region of DNA inherited together because it rarely recombines; all associated PTPN1 variants fall within this block, identified by Bento et al. 2004 that harbors every PTPN1 variant linked to insulin resistance and type 2 diabetes in human genetics studies. The T allele tags the risk haplotype. The signal at rs3787348 is intronic — the variant itself does not change the PTP1B protein, but it marks a regulatory haplotype associated with higher PTPN1 transcript levels through a functional 3'-UTR insertion²² 3'-UTR insertion
A variant in the untranslated region at the end of the PTPN1 mRNA that stabilizes the transcript, increasing the amount of PTP1B protein produced in skeletal muscle in the same block.

The Mechanism

PTP1B terminates insulin signaling by removing phosphate groups from activated tyrosine residues on the insulin receptor kinase domain — the same step that initiates glucose uptake in muscle and fat cells. In the hypothalamus, PTP1B dephosphorylates JAK2³³ JAK2
Janus kinase 2, the first intracellular enzyme activated when leptin binds its receptor in the hypothalamus; its inactivation by PTP1B blunts leptin's appetite-suppressing and thermogenic signals, terminating the leptin signal. When the PTPN1 risk haplotype drives higher PTP1B expression, both dephosphorylation events happen faster and more completely — producing a state of simultaneously reduced insulin sensitivity and reduced leptin sensitivity. In whole-body PTP1B knockout mice this dual pathway effect manifests as insulin hypersensitivity and resistance to diet-induced obesity; in the context of the human risk haplotype, the converse obtains.

The additive dose-response of rs3787348 — with each copy of the T allele increasing the weight-loss blunting effect — is consistent with a codominant regulatory mechanism where the risk haplotype progressively elevates PTPN1 expression.

The Evidence

The primary functional evidence for rs3787348 specifically comes from Yamakage et al. 2021⁴⁴ Yamakage et al. 2021
Yamakage H et al. Association of protein tyrosine phosphatase 1B gene polymorphism with the effects of weight reduction therapy on bodyweight and glycolipid profiles in obese patients. J Diabetes Investig, 2021, a prospective cohort study of 447 obese Japanese patients who underwent a standardised 3-month lifestyle intervention (caloric restriction to 25 kcal/kg ideal body weight, 60% carbohydrate, more than 30 minutes of moderate exercise at least 3 days per week). Genotyping rs3787348 and a second PTPN1 tag SNP (rs6067484) revealed a striking genotype-dependent gradient in weight loss outcomes: GG homozygotes lost 5.1 ± 0.5 kg on average, GT heterozygotes lost 4.1 ± 0.2 kg, and TT homozygotes lost only 3.1 ± 0.3 kg (p = 0.001). BMI reductions followed the same pattern (−1.9, −1.5, and −1.2 kg/m², respectively; p = 0.001), as did waist circumference reductions and leptin reductions during therapy. The T allele was also associated with higher baseline BMI (p = 0.041), consistent with a chronic dampening of leptin-mediated energy balance.

The broader haplotype context was established by Bento et al. 2004⁵⁵ Bento et al. 2004
Bento JL et al. Association of protein tyrosine phosphatase 1B gene polymorphisms with type 2 diabetes. Diabetes, 2004, who identified the 100-kb LD block and estimated an odds ratio of approximately 1.3 for type 2 diabetes from the risk haplotype — with a population-attributable risk of 17-20%, among the highest for any noncoding T2D locus. Florez et al. 2004⁶⁶ Florez et al. 2004
Florez JC et al. Association of PTPN1 polymorphisms with measures of glucose homeostasis in Hispanic Americans: the IRAS Family Study. Diabetes, 2004 confirmed these findings in 811 Hispanic Americans, with all 20 common haplotype-block SNPs associating with the insulin sensitivity index and fasting glucose.

Notably, a 2025 meta-analysis of 75,595 individuals across multiple cohorts found no statistically significant association between rs3787348 and type 2 diabetes risk (OR = 1.02, 95% CI 0.95–1.08), confirming that the T2D risk at this locus is likely distributed across multiple correlated variants in the LD block rather than residing uniquely at rs3787348 itself. The clearest individual signal for this SNP is in treatment response, not incident T2D.

Practical Actions

The weight-loss blunting effect of the T allele has a direct clinical implication: carriers who invest in a calorie-restriction and exercise program should expect a more modest BMI response than the population average, not a failure of adherence. Concretely, at the TT genotype, a 3-month structured program produces roughly 40% less weight loss than the GG reference — a difference that often goes undiagnosed and leads to premature discontinuation of a working intervention.

Exercise is specifically valuable because moderate-intensity aerobic activity transiently reduces PTP1B protein activity in skeletal muscle, extending the effective insulin signaling window — a direct pharmacological bypass of the PTP1B-mediated brake. Higher exercise volumes and more frequent sessions maintain this suppression for more hours per week. Monitoring leptin and adiponectin levels during a weight loss program provides objective data on whether the hormonal response to weight loss is proportional — T allele carriers may see blunted leptin reductions even as weight falls.

Interactions

The most relevant interaction is with rs6067484, the closely linked PTPN1 tag SNP that co-tags the same 100-kb risk haplotype. In the Yamakage 2021 study, rs6067484 showed only nominal association with waist circumference response, while rs3787348 showed the stronger and broader effect on BMI, bodyweight, and leptin reduction. Carriers of the T allele at rs3787348 who also carry the G allele at rs6067484 represent the densest haplotype coverage of the PTPN1 risk block.

PTP1B also interacts with the leptin receptor (LEPR) pathway — animal studies demonstrate that PTP1B × leptin receptor interactions compound the degree of hypothalamic leptin resistance. Carriers of rs3787348 T allele and risk variants in LEPR may experience additive impairment of energy balance regulation.