Module 2 Study Guide

Genomic Medicine Training Program

Module 2: Personalizing Nutrient Requirements & Resolving Nutritional Paradoxes Using Genomic Medicine

Overview

9 lessons cover the following topics:

• Monogenic vs Polygenic approach • Key Nutrient Pathways: Vitamins A, C, D, E; B vitamins (including folate/homocysteine cycles); CoQ10; Omega fatty acids; Selenium; Calcium • Micronutrients, Genes and Examples of Nutrigenetic and Nutrigenomic Interventions • Gene-Function Matrix • Case Studies

Lesson 1: Introduction and Omega Fatty Acids

• Introduction to personalizing nutrient requirements • Monogenic and polygenic examples • Introduction to Chrissi: Case Report • Lipid dysregulation and chronic disease • Omega fatty acid metabolism • Role of FADS1 and FADS2 gene SNPs • Clinical examples

©GENOMA INTERNATIONAL | genomainternational.com

Lesson 2: CoQ10

• CoQ10 metabolism • Roles of CoQ10 • Role of NQO1 gene SNPs • Nutritional paradox • Chrissi: Case Study • CoQ10 and statins

Lesson 3: Carotenoids and Vitamin A

• Carotenoid and Vitamin A metabolism • Role of carotenoids • Digestion, absorption and transport of carotenoids and Vitamin A • Roles of Vitamin A and carotenoids • Role of BCMO1 gene SNPs • Vitamin A transport genes • Clinical example: vegan • Chrissi: Case Study • Genes and macular degeneration research

Lesson 4: Vitamin E

• Tocopherol, tocotrienols • Factors impacting bioavailability • Roles of Vitamin E • Genes regulated by Vitamin E • Functions of gamma tocopherol • Vitamin E metabolism • Genes regulating Vitamin E metabolism, role of SNPs • Research study • Chrissi: Case Study

©GENOMA INTERNATIONAL | genomainternational.com

Lesson 5: Vitamin C

• Vitamin C biochemistry and metabolism • Ascorbic acid vs ascorbate • Roles of Vitamin C • Digestion, absorption, transport, distribution and excretion • Genes impacting Vitamin C requirements, role of SNPs • SLC23A1 research study • Chrissi: Case Study • Nutritional paradox

Lesson 6: Selenium

• Selenium basics • Roles of selenoproteins: GPx • Role of GPx gene SNPs • Role of thioredoxin • Role of selenoproteins: DIO • Role of selenoproteins in thyroid hormone synthesis • Role of DIO2 gene SNPs • Role of selenoproteins: redox, immune, more • Selenium requirements • Nutritional paradox • Chrissi: Case Study

©GENOMA INTERNATIONAL | genomainternational.com

Lesson 7: Antioxidants

• Polygenic Model • Dietary and endogenous antioxidants • Role of antioxidants in estrogen metabolism • Role of genes and gene SNPs in antioxidant cascade • Chrissi: Case Study • Ted: Case Study • Margaret: Case Study • Ali: Case Study • Introduction to Gene-Function Matrix

Lesson 8: Calcium and Vitamin D

• Calcium metabolism • Roles of calcium • Roles of vitamin D • Calcium homeostasis and vitamin D • Role of CaSR and gene SNPs • Genes linked to calcium and vitamin D, role of gene SNPs • Gene-function matrix: vitamin D • Chrissi: Case Study • Gene-function matrix: Chrissi • VDR • Nutritional paradox • Bone metabolism • Gene function matrix: Chrissi

©GENOMA INTERNATIONAL | genomainternational.com

Lesson 9: B vitamins

• Polygenic approach to B vitamin requirements • Roles of B vitamins • Digestion, absorption of vitamin B12 • Role of B vitamins in Transmethylation/Transsulfuration (TM/TS) pathway • Methylation biochemistry • TM/TS: Simple to complex • Methyl donors • Chrissi: Case Study • Role of gene SNPs • Gene-function matrix • TM/TS and glutathione, free radical quenching • Role in detoxification, estrogen metabolism • Ted: Case Study • Gene function matrix • Margaret: Case Study • Gene function matrix • Ali: Case Study • Gene function matrix

©GENOMA INTERNATIONAL | genomainternational.com Module 2 Lesson 1

Time Stamp Title/Slide Subject 0:00 Genomic Medicine Training Program 0:08 Module 2 Session 1 1:00 Overview of Module 2 Session 1 1:33 Personalized Nutrition: Each is Different 2:48 Personalized Nutrition in Era of Genomic Medicine 3:51 Personalized Nutrition and Nutritional Paradox 4:51 Nutritional Paradox Defined 5:25 Understanding 4 Science Areas for Genomic Medicine 6:33 Monogenic Examples in Module 2 8:24 Polygenic Examples in Module 2 9:37 Chrissi: Case Report 9:43 Details for Case Report 10:33 Health Issuess for Case Report 11:29 Details for Case Report 12:18 Health Issues for Case Report 13:38 Lab Testing for Case Report 14:50 Omega 3 FA Requirements in Era of Genomic Medicine 15:33 Details for Case Report 15:55 Lipid Dysregulation and Chronic Disease 17:29 Dietary Lipids and Chronic Disease 18:27 Heart Disease and Dietary Lipids in Literature 19:52 Toll Like Receptor Graphic 21:34 Breast Cancer, Obesity and Lipids Research Articles 21:56 Background for Review Artilce 23:00 Breast Cancer, Obesity, Lipids and Inflammation 25:44:00 Inflamed Adipose Tissue Illustration 27:48:00 Plant based Omega 3 FA 28:23:00 Genes and PUFA Metabolism 31:37:00 Genotype with Major and Minor Alleles FADS1 32:07:00 FADS1 Genotype and PUFA Metabolism Graphic 32:50:00 Nutritional Impact of FADS1 SNP Omega 3 34:22:00 Nutritional Impact of FADS1 SNP Omega 6 34:29:00 Genotype and PUFA Metabolism Graphic 35:24:00 Chrissi Genomic Test Results 36:12:00 Culinary Translation 36:41:00 FADS1 Genotype and PUFA Metabolism Graphic 36:58:00 Inflamed Adipose Tissue Illustration 37:27:00 Take‐away, Summary and Insights 42:44:00 End Genomic Medicine Training Program

Module 2: Personalizing Nutrient Requirements and Resolving Nutritional Paradoxes Using Genomic Medicine

Joe Veltmann PhD FAAIM DCCN Roberta L. Kline MD FACOG

Overview of Module 2

• Nutrient Requirements in the Era of Genomic Medicine • Gene Variants Influence Nutrient Utilization and Requirements • Genomic Testing Personalizes Nutrient Requirements and Resolves Nutritional Paradoxes – Monogenic and Polygenic Examples – Case Reports • Summary

1 Personalized Nutrition: Each is Different

• Each person metabolizes a nutrient differently depending on his or her genetic code. Personalized • Adequacy defined by RDA may need to be adjusted based on gene variants associated Nutrition with: – digestion absorption In the Era of – transport metabolism Genomic – utilization. • Gene variants or single nucleotide Medicine polymorphisms (SNPs) influence gene expression, metabolic processes and nutritional requirements.

Personalized Nutrition In the Era Genomic Medicine

• Nutrients and bioactives “speak” to genes, influencing a nutrient’s functional requirement. • Meeting the RDA for a pro‐nutrient may not represent nutrient requirement of the active molecule in a biochemical or metabolic pathway (Nutritional Paradox).

2 The Nutritional Paradox

• Relying on lab test to determine a nutrient’s requirement can be misleading when genes are involved in the biotransformation of pro‐nutrient to active molecule. • A Nutritional Paradox can alter disease initiation, development or progression.

• Fundamental Understanding in These 4 Personalized Scientific Areas: Nutrition – Cellular Biology (Cell Signaling Pathways). – Nutritional Biochemistry (Substrates or in the Era of Cofactors). – Metabolic Pathways (Absorption, Transport, Genomic Excretion, etc). – Interconnectedness Between Biological Medicine Systems.

Nutrient Utilization: Personalizing Nutrient Requirement Based on DNA Monogenic FADS1: Omega 3 and Omega 6 PUFA Requirements NQO1: Ubiquinone/Ubiquinol Requirement (Statin Therapy) BCMO1: β‐carotene metabolism & Vitamin A requirement α-TTP: Alpha Tocopherol Homeostasis and Requirement SLC23A1: Vitamin C Homeostasis and Requirement GPx: Selenium Requirement

3 Nutrient Utilization: Personalizing Nutrient Requirement Based on DNA Polygenic

• Antioxidant Cascade: Quenching Free Radicals • Calcium & Vitamin D Requirements • Transmethylation/Transsulfuration: – Nutrient Requirements for B2, B6, B9, B12, Choline – Nutrient Requirements for Downstream Substrates Associated with Other Biological Systems

Case Report: Chrissi

Female, 45 years of age, lives in Mid‐ West; Polish Heritage Occupation: Student/Full‐time Mom (2 children) Case Report Height: 5’ 9 Weight: 190 BMI: 28 Details: Chrissi Diet: Ham, Cheese, Bread, Chicken, Salad, Veggies, Yogurt, Beef Exercise: Biking, Running (1 hr/day); Runs ½ Marathons

4 Health Issues: – Rheumatoid Arthritis – Yo‐yo Dieting—Weight Gain – Ruptured Achilles Tendon – Poor Recovery from Exercise – Osteopenia Case Report – Bi‐polar Disorder Details: Chrissi – Extreme Fatigue – Stomach Issues (Inflammation) – Sick Often – Insomnia Meds: Lexapro (10 mg); Vyvance (20 mg) Nutritional Supplements: None

Health Issues: – Rheumatoid Arthritis – Yo‐yo Dieting—Weight Gain – Ruptured Achilles Tendon – Poor Recovery from Exercise – Osteopenia Case Report – Bi‐polar Disorder Details: Chrissi – Extreme Fatigue – Stomach Issues (Inflammation) – Sick Often – Insomnia Meds: Lexapro (10 mg); Vyvance (20 mg) Nutritional Supplements: None

5 Case Report: Chrissi Laboratory Testing: Ultimate Wellness Genomic Panel • Bone Health • Inflammation • Detox • Redox • Weight Management • Exercise and Risk of Injury • Nutrient Utilization • Transmethylation/Transsulfuration • Emotional Health • Eating Behaviors

Omega 3 and 6 Fatty Acid Requirements in the Era of Genomic Medicine

6 Obesity, Type 2 Diabetes, Cardiovascular Disease Lipid Dysregulation Lipids have diverse functions: ‐cell signaling agents and Chronic ‐cell membrane trafficking and sorting ‐part of cell morphogenesis and proliferation Diseases cell signaling agents cell membrane trafficking and sorting Precursors to pro‐inflammatory and part of cell morphogenesis and anti‐inflammatory eicosanoids derived proliferation from PUFA (polyunsaturated fatty acids)

Dietary Lipids and Chronic Diseases

Differences between Traditional American Diet (TAD) versus Mediterranean Diet (MD): • TAD: high in saturated fat and n‐6 PUFAs, low in n‐3 PUFAs • MD: high in n‐3 PUFAs and monounsaturated fats, fruits, veggies, whole grains, wine and poultry

Mediterranean diet lowers incidence of CAD, hypertension, diabetes, arthritis, inflammatory and autoimmune disorders.

Omega 3 Articles in the Literature‐Heart Disease

The Journal of Nutritional Biochemistry Volume 53, March 2018, Pages 9-19 Research Article Flaxseed oil rich in omega-3 protects aorta against inflammation and endoplasmic reticulum stress partially mediated by GPR120 receptor in obese, diabetic and dyslipidemic mice models.

AlexandreMoura-AssisaMilessa SilvaAfonsobVanessade

OliveiraaJoseaneMoraricGustavo Aparecidodos SantosdMarciaKoikebAna MariaLottenbergbRodrigoRamos

CatharinodLicio AugustoVellosocAdelinoSanchez Ramos da SilvaeLPde MourafEduardo

RocheteRopellefJosé RodrigoPaulifDennys Esper CorrêaCintraag

7 From Metabolic Products, Toll‐like Receptors, Cell Signaling to Inflammatory Response

Graphic source: Haynes BF, Soderberg KA, et al., (2015). https://clinicalgate.com/introduction‐to‐the‐immune‐system‐2/

Omega 3 Articles in the Literature‐Breast Cancer

Background and Objectives for Review Article

• Obesity: risk factor for developing breast cancer (postmenopausal women). • Adipose tissue in the breast under obese conditions induces inflammation. • n‐3 polyunsaturated fatty acids (n‐3 PUFA): potent anti‐inflammatory.

Objectives: 1. Review current status of breast cancer and its relationship with obesity. 2. Review current research and knowledge on the role of n‐3 PUFA in reducing/preventing breast cancer cell growth in vitro, in vivo and in human studies 3. How n‐3 PUFA may modulate signaling pathways mitigating their effects on breast cancer development.

8 Adipose Tissue, Obesity & Inflammation • Obesity: hypertrophy and hyperplasia of adipose tissue • Shift in macrophage subtypes from M2:M1 triggers inflammation and differential expression of T1 and T2 cells. – Upregulation of pro‐inflammatory cytokines—IL‐6, IL‐8, MCP‐1 – Downregulation of anti‐inflammatory cytokines—IL‐4, IL‐10 • Adipose tissue: secretes exosomes‐‐ mediating fat cell communication with other cells including breast cells

Plant‐based Omega‐3 Fatty Acids and EPA/DHA

• Sources of plant‐based omega‐3 fatty acids – Flaxseed, Chia, Hemp, Algae, Walnuts, Canola • Conversion of ALA to EPA dependent on several delta‐desaturase enzymes encoded by the FADS1 and FADS2 genes. • Most important is FADS1: converts eicosatetraenoic acid (ETA, 20:4n‐3) to eicosapentaenoic acid (EPA, 20:5n‐3).

9 Genes Involved in PUFA Metabolism

Linoleic Acid (LA) α‐Linolenic Acid (ALA) (18:2 n=6) (18:3 n=3) Pro‐inflammatory FADS2 Anti‐ Eicosanoids D6D Y‐Linoleic Acid (GLA) Stearadonic Acid (SDA) inflammatory (18:3 n=6) (18:4 n=3) Eicosanoids

Series 1 Prostaglandins

Di‐HomoY‐Linoleic Acid (DGLA) Eicosatetranoic Acid (ETA) (20:3 n=6) (20:4 n=4) FADS1 Series 2 Prostaglandins D5D Series 3 Prostaglandins Arachidonic Acid (AA) Eicosapentanoic Acid (EPA) (20:4 n=6) (20:5 n=3) Series 4 Leukotrienes Series 5 Leukotrienes

Adrenic Acid (DTA) Docosapentanoic Acid (DPA) (22:4 n=6) (22:5 n=3) FADS2 D6D Docosahexanoic Acid (DHA) (22:6 n=3) Adapted from: Merino DM, Ma DWL, et al. (2010). Lipids in Health and Dis 9:63. © Genoma International All rights reserved. 28

FADS1 Genotypes on Omega 3 and Omega 6 PUFA Gene Function Alleles Omega 3 and 6 PUFA Phenotype FADS1 encodes for Major Little or No Impact delta 5 desaturase M/m Moderate Impact enzyme minor High Impact

Minor allele associated with decreased ability to convert: Omega 6 PUFA: DGLA to AA Omega 3 PUFA: ETA to EPA and DHA

FADS1 Gene Variants in PUFA Metabolism

Series 1 Prostaglandins

10 Impact of Gene SNP on FADS1: Omega 3 FA • Nutritional Significance of SNP on FADS1: – May not generate sufficient EPA Levels: • Individuals consuming plant‐based omega 3‐fatty acid (Vegetarians). • Individuals not consuming sufficient quantity of EPA‐rich seafood. – EPA is precursor to DHA; insufficient DHA. – EPA precursor to PG3 and LT5; insufficient anti‐inflammatory molecules. – EPA down‐regulates pro‐inflammatory cytokines (IL‐1B, IL‐6, TNF‐ alpha).

Impact of Gene SNP on FADS1: Omega 6 FA

Nutritional Significance of SNP on FADS1: – DGLA converted to PG1 (pro‐ inflammatory). – Less PG2 and LT4 generated‐‐much less inflammation, unless individual consumes meat, dairy and egg yolks, rich sources of AA.

FADS1 Gene Variants in PUFA Metabolism

Series 1 Prostaglandins

11 Genomic Test Results: Chrissi

Gene Hi Impact Mod Impact Nutritional Recommendation FADS1 X Cold Water Fish or EPA/DHA Supplementation. Limit intake of Meat, Dairy Products and Egg Yolks.

Culinary Translation

FADS1 Gene Variants in PUFA Metabolism

Series 1 Prostaglandins

12 Image source: Al‐Jawadi, A et al. Protective properties of n‐3 fatty acids and implications in obesity‐associated breast cancer. 2018 JNutrBio (53): 1‐8. © Genoma International All rights reserved. 37

Take‐Aways: FADS1 Genotype

• Impacts Nutrient Requirements: – Omega 3: EPA and DHA – Omega 6: AA • Impacts Functional Biomarkers: – Omega 3: ALA and SDA, EPA and DHA – Omega 6: DGLA and AA • Disrupts Homeostasis/ Biological Systems: – Increases inflammation and risk of chronic disease. – Decreases cognitive development (infants). – Increases neurodegenerative disease risk (Alzheimer’s, Parkinson’s).

13 Module 2 Lesson 2

Time stamp Topic / Slide Subject 0:00 Genomic Medicine Training Program 0:02 Module 2 Session 2 Personalized Nutrition Program 0:10 CoQ10 Requirement in Era of Genomic Medicine 0:45 CoQ10 Requirement and Roles in Body 1:47 Loss of CoQ10 Over Times in Organs 3:56 Synthesis of CoQ10 7:42 CoQ10 and Electron Transport Chain 9:06 Antioxidant Role of CoQ10 11:18 Electron Transport Chain and CoQ10 11:59 NQO1 Gene and CoQ10 12:17 Major and Minor Alleles for NQO1 13:02 Ubiquinone versus Ubiquinol 14:15 Lab test for CoQ10 15:18 Health Issues for Chrissi 15:40 Synthesis of CoQ10 17:15 Culinary NQO1 Gene and CoQ10 Limits 1 17:34 Culinary NQO1 Gene and CoQ10 Limits 2 18:03 Culinary NQO1 Gene and CoQ10 Limits 3 18:50 Statins and CoQ10 Synthesis 20:21 CoQ10 Synthesis and Statin Therapy Research 23:54 Take Aways 26:29:00 Biological Systems Integrated Graphic 27:50:00 Summary 31:24:00 End Genomic Medicine Training Program

Module 2: Personalizing Nutrient Requirements and Resolving Nutritional Paradoxes Using Genomic Medicine

Joe Veltmann PhD FAAIM DCCN Roberta L. Kline MD FACOG

Nutrient Utilization: Personalizing Nutrient Requirement Based on DNA Monogenic FADS1: Omega 3 and Omega 6 PUFA Requirements √ NQO1: Ubiquinone/Ubiquinol Requirement (Statin Therapy) BCMO1: β‐carotene metabolism & Vitamin A requirement α-TTP: Alpha Tocopherol Homeostasis and Requirement SLC23A1: Vitamin C Homeostasis and Requirement GPx: Selenium Requirement

1 CoQ10 Requirement in the Era of Genomic Medicine.

CoQ10 and Its Roles in the Body

• Electron Transport Chain • Can function as Antioxidant: – Quenches Superoxide Free Radicals – Participates in Antioxidant Cascade with Vitamin E Radical • 200 mg/day CoQ10 for 2 weeks improved recovery time following exercise training.

CoQ10 (Ubiquinone) Found in every cell of the body • Highest concentrations: heart, liver, lungs and kidney. • Concentration decreases with age. • Body capable of synthesizing if diet is rich in fish, meats, and micronutrients.

Image source: Kalen, A et al. Lipids 1989. (24) 579.

2 CoQ10 Synthesis Glycolysis/ Beta‐oxidation

Protein Rich Foods

Transmethylation/ Transsulfuration

Image adapted from: dos Santos, GC et al. Coenzyme Q10 and its effects in the treatment of neurodegenerative diseases. Braz. J. Pharm. Sci. vol.45 no.4 São Paulo Oct./Dec. 2009. © Genoma International All rights reserved. 7

Function of CoQ10: Electron Transport Chain Participates in chemical reactions associated with energy (ATP) production in mitochondria.

Reduced CoQ10

Graphic source: http://users.humboldt.edu/rpaselk/C438.S12/C438Notes/C438nLec25.htm

Function of CoQ10: Antioxidant Partner

Fat‐soluble antioxidant in cellular bilayer

CoQ10 = Ubiquinone CoQH2 = Ubiquinol

3 NQO1 Gene and CoQ10 Biotransformation

NQO1 = NAD(P)H quinone oxidoreductase • NQO1 SNP decreases enzymatic activity: – Ubiquinone to ubiquinol conversion reduced. • NQO1 SNPs linked to higher risk: – Bladder cancer – Gastro‐intestinal cancer – Lung cancer – Benzene toxicity – Breast cancer © Genoma International All rights reserved. 10

NQO1 Genotypes: Decreased CoQ10 Utilization Gene Function Alleles CoQ10 Phenotype

NQO1 encodes for the Major Little or No Impact NAD(P)H quinone M/m Moderate Impact oxidoreductase; Minor High Impact catalyzes conversion of Minor allele associated with ubiquinone (pro‐ decreased ability to convert CoQ10 nutrient) to ubiquinol to ubiquinol; increases risk of (active molecule). ubiquinol deficiency in susceptible individuals.

Ubiquinone Versus Ubiquinol

Characteristic Ubiquinone Ubiquinol Form of CoQ10 Oxidized Reduced

Percent in Body Low: 10 to 25 % High: 80‐90 %

Active Form NO YES

Absorption POOR GREAT (300 % Better) Antioxidant Effect NO YES

Measureable YES (Stable) NO (Unstable)

Cost $ $$

4 Laboratory Results for CoQ10

Genomic Test Results: Nutrigenomic Intervention Gene High Impact Mod Impact Nutritional Action Step

NQO1 X Ubiquinol Supplementation

CoQ10 Synthesis Glycolysis/ Beta‐oxidation

Protein Rich Foods

Transmethylation/ Transsulfuration

5 Culinary Limitations: NQ01

NQ01 SNP impacts ubiquinol production: – Outcome: reduced conversion ubiquinone to ubiquinol. – Ubiquinol cannot be directly obtained from the diet.

Culinary Limitations: NQ01

NQ01 SNP impacts ubiquinol production: – Outcome: reduced conversion ubiquinone to ubiquinol. – Ubiquinol cannot be directly obtained from the diet. NQ01 gene may be functional, however poor substrate availability (TYR, MET), plus marginal deficiencies in micronutrient cofactors may impede synthesis of CoQ10.

Culinary Limitations: NQ01

6 Statins and CoQ10 Supplementation

https://bjcardio.co.uk/2015/10/coenzyme‐q10‐and‐cardiovascular‐disease‐an‐overview/

Take‐Aways: CoQ10 & NQO1

• Impacts Nutrient Requirements: – Ubiquinone and Ubiquinol • Impacts Usefulness of Laboratory Biomarker Ubiquinone: Nutritional Paradox • Disrupts Homeostasis/ Biological Systems: – Electron Transport Chain (Energy) – Antioxidant Cascade (Quenching Free Radicals) – Statin Therapy + NQO1 Genotypes: Ubiquinol Deficiency

7 Integrate genomic test results into a comprehensive health and wellness strategy. Biological Systems Are Interconnected!

Genomic testing and interpretation requires basic understanding of nutritional biochemistry and metabolic pathways.

Understand the important biochemical and metabolic pathways associated with each gene SNP to help direct clinical nutritional interventions.

Summary Overcome nutritional paradoxes by integrating gene SNP results with nutrigenomic, lifestyle or dietary interventions.

Optimizing a person’s nutrient requirements requires an understanding of his/her genotype.

8 Module 2 Lesson 3

Time Stamp Topic / Slide Subject 0:00 Genomic Medicine Training Program 0:05 Module 2 Session 3 Nutritional Requirements in Era of Genomic Medicine 0:37 Carotenoids and Vitamin A Requirements Graphic 1:22 Learning Objectives 3:39 Carotenoids Precursor to Vitamin A 3:48 Carotenoids and Vitamin A Requirements Graphic 3:59 Carotenoids Precursor to Vitamin A 5:22 Provitamin A Carotenoids Flow Chart Graphic 7:05 Lycopene 7:36 Carotenoids in Biological Systems 8:53 Digestion and Absorption of Carotenoids‐Review 10:07 Digestion and Absorption B‐carotene‐Review 11:34 Conversion of B‐carotene to Vitamin A ‐Review 12:22 Flow Chart Conversion of B‐carotene to Vitamin A Graphic 12:51 Conversion of B‐carotene to Vitamin A Transport 13:39 Role of Vitamin A in Biological Systems 14:12 Genomic Medicine B‐carotene Metabolism 14:49 Genes Associated with Apical Side of Enterocyte 15:58 BCMO1 in Enterocyte 16:37 Flow Chart of BCMO1 17:59 Genes Associated with Basal Side of Enterocyte 18:48 Vitamin A Transport Genes 19:33 Graphic Digestion/Absorption/Transport/Excretion Vitamin A 19:59 Graphic Digestion 21:55 Graphic Lymphatic System and Blood Circulation 23:26 Transporters within Small Intestine 24:42:00 Retinyl Esters to Chylomicrons 24:05:00 Graphic Extraheaptic Target Tissue and Vitamin A 28:16:00 Graphic Liver and Vitamin A 29:42:00 Graphic Excretion of Retinol 31:00:00 Genes Transport, Lipoproteins, Enzymes, Receptor Sites Associated with B‐carotene 31:26:00 BCMO1 Genotype Matrix 32:33:00 Nutritional Paradox BCMO1 34:28:00 Vegan with BCMO1 Genotype 35:29:00 Chrissi Case History‐Health Issues, Diet, Meds, etc 36:28:00 Biomarkers Vitamin A/Carotenoids 36:52:00 Chrissi Genomic Test Results and Nutrition Action Steps 38:05:00 Dietary Sources 38:28:00 Modification of Chrissi's Diet 38:50:00 Lutein Non‐Provitamin A Carotenoid Graphic 38:58:00 Age Related Macular Degeneration and Lutein 39:03:00 BCMO1 and ARMD Research Article 40:12:00 BCMO1 and CD36 and ARMD Research Results 43:01:00 MPOD and BCMO1 and CD36 Research Results 43:44:00 Carotenoid Supplement Study 44:13:00 Oil Filled Capsule Study Conclusions 45:01:00 Take‐aways BCMO1 46:18:00 Take‐aways Nutritional Paradox 46:53:00 Take‐aways Genomic Medicine 47:43:00 Biological Systems Integrated 50:36:00 End Genomic Medicine Training Program

Module 2: Personalizing Nutrient Requirements and Resolving Nutritional Paradoxes Using Genomic Medicine

Joe Veltmann PhD FAAIM DCCN Roberta L. Kline MD FACOG

Carotenoids and Vitamin A Requirements in the Era of Genomic Medicine

1 Learning Objectives

• Carotenoids and Vitamin A Metabolism. • BCMO1 Genotype, Beta Carotene, and Vitamin A Status. • BCMO1 Genotype, Carotenoids, and Macular Degeneration. • Nutritional Paradox. • Case History (Chrissi): • Nutrigenomic Interventions; and • Culinary Genomics. • Key Take‐Aways.

Carotenoids: Precursor to Vitamin A and Much More

• More than 600 carotenoids found in nature: • only 60 can be converted to retinal (intermediate vitamin A molecule). • Most studied pro-vitamin A carotenoids: • alpha carotene, beta-carotene and beta cryptoxanthin. • Most studied carotenoids but not provitamin A precursors: • lycopene, lutein and zeaxanthin.

Carotenoids and Vitamin A Requirements in the Era of Genomic Medicine

2 Carotenoids: Precursor to Vitamin A and Much More

Image source: Mein JR, Lian F, et al., (2008). Nutr. Rev. 66:667-683.

3 Role of Carotenoids In Biological Systems

• Carotenoids have diverse functions: – anti‐cancer agents – antioxidants – heart disease – decrease risk of cataracts, age‐ related macular degeneration Carotenoids act as cell signaling agents (recent discovery).

Graphic Source: Bonet ML, Canas JA, et al. (2015) . Arch Biochem Biophys 572: 112‐125

Digestion and Absorption of Carotenoids: A Brief Review • Carotenoids: fat‐soluble nutrients linked to dietary fat digestion and lipoprotein transport. • Digestion: dietary fat, bile acids, pancreatic enzymes, gastric lipase needed to hydrolyze esters of carotenoids: – only free form carotenoids absorbed.

Digestion and Absorption of Beta-Carotene

• Duodenum: beta-carotene incorporated into mixed micelles and transported to brush border membrane of enterocyte.

• Old paradigm: assumed beta-carotene absorbed by passive diffusion.

• New paradigm: specialized proteins involved in absorption and transport of beta-carotene/ Vitamin A across the enterocyte.

4 Conversion of B‐carotene to Vitamin A: A Brief Review

• In enterocyte, 40 to 60% of beta‐carotene converted to vitamin A intermediate (retinal). • Beta‐carotene: 2 X vitamin A activity than other carotenoids. • Activity regulated by cytoplasmic enzyme: beta carotene 15’, 15’ mono‐oxygenase (BCMO1) encoded by BCMO1 gene.

BCMO1 Pro- Vitamin A Molecule

Active Vitamin A Molecule

Retinyl Ester

Conversion of Retinal to Retinol (Active Vitamin A) and Transported to Tissues: A Brief Review • Retinal converted to retinol (active vitamin A molecule) and then retinyl esters in the enterocyte. • Retinyl esters packaged into chylomicrons: – Chylomicrons transported by lymphatics and released into blood circulation; – Chylomicrons picked up by liver, de‐esterified into retinol; and – Retinol transported to target tissues by retinol binding protein (RBP).

5 Role of Vitamin A In Biological Systems

• Vitamin A:

Ultraviolet – Vision Light – Embryonic development. Atmospheric Pollution – Cell differentiation

– Membrane and skin Stress protection – Cell membrane Poor Diet

antioxidant Image source: http://buahmerahmix.ph/beta-carotene.html

Genomic Medicine: Beta‐Carotene Metabolism 1. Beta‐Carotene Digestion and Absorption • Gene Variants Associated with Esterase, Pancreatic Lipase, Bile Acids. • Level of Dietary Fat.

Genomic Medicine: Beta‐Carotene Metabolism 1. Beta‐Carotene Digestion and Absorption • Gene Variants Associated with Esterase, Pancreatic Lipase, Bile Acids. • Level of Dietary Fat. 2. Gene Variants Associated with Genes Encoding for Specialized Beta‐ Carotene Transport Proteins (Apical Side of Enterocyte): • Scavenger Receptor Class B Member 1 (SR‐B1). • Neimann‐Pick C1‐like 1 (NPC1L1). • Scavenger Receptor Class B Member 3 (SCARB3 or CD36).

6 Genomic Medicine: Beta‐Carotene Metabolism

3. Beta‐Carotene Metabolism within Enterocyte: • Gene variants on BCMO1 can reduce beta‐carotene conversion to retinal by 80%. • Retinoic acid synthesis reduced.

BCMO1 Pro- Vitamin A Molecule

Active Vitamin A Molecule

Genomic Medicine: Beta‐Carotene Metabolism

3. Beta‐Carotene Metabolism within Enterocyte: • Gene variants on BCMO1 can reduce beta‐carotene conversion to retinal by 80%. • Retinoic acid synthesis reduced.

4. Basal Side of Enterocyte: Vitamin A transport genes: • ATP‐binding Cassette Transporter 1 (ABCA1). • Microsomal (Triglyceride) Transfer Protein (MTP).

7 Genomic Medicine: Vitamin A (Retinyl Esters) Transport and Excretion • Vitamin A Carriers Outside GI Tract: • Lymph: ApoB, ApoE, ApoA5, ApoC3. • Blood: HDL, LDL, VLDL, ApoA1, LPL, ApoBR, CETP, PLTP, ApoB. • Liver: SR‐B1, LPR, LDLR. • Tissues: ABCA1, LDLR.

• Vitamin A Excretion: CYP26A1, CYP26B1, CYP26C1.

Location of Vitamin A Transporter Proteins (SR‐B1, CD36, NPC1‐L1) Within Regions of the Small Intestine

From Retinyl Esters to Chylomicron

http://ib.bioninja.com.au/standard‐level/topic‐6‐human‐physiology/61‐digestion‐and‐absorption/lipid‐absorption.html

Chylomicron Chylomicron

HDL HDL

HDL

10 Genes Associated with Beta Carotene Metabolism & Vitamin A Utilization • Transporter Proteins (8): – ABCA1, N‐P C1L1, CETP, CD36, MTP, PLTP, RBP, SR‐B1, • Lipoproteins (8): – ApoA1, ApoA5, ApoB, ApoE, ApoC3, HDL‐C, LDL‐C, VLDL‐C • Enzymes (5): – BCMO1, LPL, CYP26A1, CYP26B1, CYP26C1 • Receptor Sites (5): – ApoBR, LPR, LDLR, RAR, RXR

BCMO1 Genotype: Susceptible to Vitamin A Deficiency Gene Function Alleles Vitamin A Phenotype

Gene encodes for beta‐ Major Little or No Impact carotene 15,15'‐ M/m Moderate Impact monooxygenase; found minor High Impact in enterocyte cytoplasm, converting Minor allele associated with beta‐carotene to decreased ability to convert beta‐ retinal; gene SNPs alter carotene to retinal, increasing risk of efficiency of enzyme as vitamin A deficiency in susceptible much as 80%. populations.

Two Nutritional Paradoxes Associated with BCMO1 Genotype • Strict Vegans: – Blood Concentration of Beta‐Carotene‐‐High. – Blood Concentration of Vitamin A (Retinol)‐‐Low. • Individuals consume diet replete with fat‐soluble nutrients: – But blood concentrations of Vitamins A, E, D‐‐Low. – Carriers for gene variants associated with lipid uptake and intracellular transport proteins (SR‐B1, CD36, ABCA1 and NPC1L1), the BCMO1 genotype and extrahepatic transport proteins.

11 Vegan with BCMO1 Genotype

BCMO1 Genotype

Adapted from: Henning, P et al. Retinoid receptors in bone and their role in bone remodeling. Front. Endocrinol., 11 March 2015.

Case History: Chrissi • Health Issues (9): Rheumatoid Arthritis Yo‐yo Dieting—Weight Gain Ruptured Achilles Tendon Poor Recovery from Exercise Osteopenia Bi‐polar Disorder Extreme Fatigue Susceptible to Colds/Flu Insomnia • Diet: Ham, Cheese, Bread, Chicken, Salad, Veggies, Yogurt, Beef (No fish in diet: does not like the taste) • Exercise: Biking, Running (1 hr/day); Runs ½ Marathons. • Meds: Lexapro (10 mg); Vyvance (20 mg). • Nutritional Supplements: None.

Nutritional Biomarkers: Vitamin A/Carotenoids

12 Case History: Test Results for BCMO1 Gene SNP (Chrissi)

Gene High Moderate Nutritional Action Step Impact Impact BCMO1 X Increase colorful fruits and vegetables: 3‐5 servings per day Multi‐vitamin as insurance policy: 50% Beta‐carotene; 50% vitamin A palmitate Evaluate vitamin A biomarkers in 90 days

Dietary Sources

Vitamin A (retinol) Beta Carotene • Grass fed meats • Sweet potatoes (cooked) • Cold water fish • Winter squash (cooked) • Cod liver oil • Kale, collard, turnip greens (cooked) • Mollusks • Carrots (cooked) • Cheese (goat, cheddar, • Swiss chard, spinach, romaine (raw) camembert) • Mango, cantaloupe, red grapefruit • Eggs • Basil, thyme, cilantro, chives • Red chili, cayenne, paprika

Modification to Chrissi’s Diet & Nutritional Supplements Based on DNA Data Before DNA testing After DNA Testing Diet Ham Limit to only 1 serving per month Cheese Eliminate White Bread Eliminate and substitute with gluten free bread; limit to 3 to 4 times per week Eggs Egg white only Chicken White meat, without the skin Salad Colorful and fresh, 4‐6 servings per day Veggies Colorful and fresh, 4‐6 servings per day Yogurt Eliminate Beef Lean beef only Other proteins Beans, Protein Powders, Tofu, Nuts and Seeds Fruits Colorful, fresh, low sugar fruits Nutritional NONE EPA/DHA, Ubiquinol, Multiple Vitamin‐Mineral with 50% Supplements Vitamin A Ester and 50% Carotenoids

13 Image source: Mein JR, Lian F, et al., (2008). Nutr. Rev. 66:667-683.

Age‐Related Macular Degeneration (ARMD) and Lutein • Factors Associated with ARMD: – Age – Sun Exposure – Poor Diet – Family History – Smoking – Alcohol – Obesity – Hypertension/ Heart Disease Image source: https://www.webmd.com/eye‐health/macular‐ – Medications degeneration/understand‐amd‐17/video‐age‐related‐macular‐ degeneration‐explained

BCMO1 and Macular Degeneration

Genetic variants in BCMO1 and CD36 are associated with plasma lutein concentrations and macular pigment optical density (MPOD) in humans. Abstract: • Lutein is recovered at high concentration in the human macula lutea. Micronutrient implicated in prevention of age‐related macular degeneration. • Objective. To identify genes which affect blood and retina lutein concentrations among candidate genes (intestinal sterol transporters (CD36) and carotenoid oxygenases(BCMO1). • Design. A comparative plus an observational study. • Participants. Twenty‐nine healthy subjects for the comparative study and 622 subjects for the observational study.

14 Results: BCMO1, CD36 & Macular Degeneration

Results. Six SNPs linked to ABCG8, BCMO1, CD36, and NPC1L1 explained 25% and 38% of the plasma and MPOD variance, respectively.

Plasma lutein levels and genotype: – Carriers with two minor alleles for BCMO1 had lower (P < 0.05) plasma lutein compared to individuals with a major and minor allele. – Carrier with two major alleles for CD36 had lower (P < 0.05) plasma lutein compared to individuals with a minor allele. – Carriers of the minor allele for CD36 had higher plasma lutein; this was confirmed in the cohort of 622 subjects.

Results: BCMO1, CD36 & Macular Degeneration

Results. Six SNPs linked to ABCG8, BCMO1, CD36, and NPC1L1 explained 25% and 38% of the plasma and MPOD variance, respectively. MPOD and genotype: – Carriers with two minor alleles for BCMO1 had higher (P < 0.05) MPOD compared to individuals with a major allele. – Carrier with two major alleles for CD36 had higher (P < 0.05) MPOD compared to individuals with a minor allele. Conclusions: – BCMO1 and CD36 are implicated in plasma and retina concentrations of lutein. – Gene variants in these genes can modulate blood and retina concentrations of lutein.

15 Conclusions: 1. Commercially available carotenoid food supplements often do not achieve their label claim. 2. Oil filled soft gel capsules are best way to provide a stable carotenoid supplement. 3. Use carotenoid interventions in oil based formulas rather than powder filled formulas based on DNA results of individual.

Take‐Aways : Carotenoids & BCMO1

BCMO1 gene regulates conversion of beta‐carotene to its intermediate (retinal) and then to retinol (active vitamin A molecule). Nutritional Paradox: • Strict vegans consume a carotenoid‐rich diet but low serum vitamin A (retinol) concentrations linked to BCMO1 genotype. • Omnivores: diet replete with all fat soluble vitamins but serum levels are low. – Carriers for gene variants associated intracellular transport proteins (SR‐B1, CD36, ABCA1 and NPC1L1), along with BCMO1 genotype can affect vitamin A and all fat soluble vitamin requirements.

Take‐Aways: Carotenoids & BCMO1

BCMO1 gene regulates absorption of lutein and retinal storage of lutein in retina (macular degeneration).

Nutritional Paradox: • BCMO1 genotype can influence the incidence and severity of macular degeneration.

16 Take‐Aways: Carotenoids & BCMO1

Genomic test results and biomarker testing can help clinicians: • Affirm Current Dietary Choices; • Recommend Culinary Genomic Food Maps; and • Provide Nutritional Supplement to Meet Nutrient Requirement for Vitamin A.

Integrate genomic test results into a comprehensive health and wellness strategy. Biological Systems Are Interconnected!

17 Module 2 Lesson 4

Time Stamp Title / Slide Subject 0:00:00 Genomic Medicine Training Program 0:02:00 Module 2 Session 4 0:10:00 Vitamin E Requirement in the Era of Genomic Medicine 0:24:00 Learning Objectives for Module 2 Session 4 2:05:00 Personalizing Nutrient Requirements ‐ Clinical Relevance 4:12:00 Vitamin E Section 1 4:23:00 Vitamin E Tocopherol Vitamers 5:13:00 Difference between Tocopherol and Tocotrienol Vitamers 5:35:00 Vitamin E Activity between Types of Tocopherols and Tocotrienols 7:06:00 Dietary Sources of Vitamin E 7:32:00 Bioavailability Factors for Vitamin E #1 Gender and Age 8:23:00 Bioavailability Factors for Vitamin E #2 Lifestyle 8:41:00 Bioavailability Factors for Vitamin E #3 Diet 9:31:00 Bioavailability Factors for Vitamin E #4 Drugs and GIT Disease 10:45:00 Bioavailability Factors for Vitamin E 5 Inherited Gene Mutations 12:34:00 Bioavailability Factors for Vitamin E # 6 Gene Variants 13:06:00 Function of Vitamin E Antioxidant Electron Transport Chain 14:14:00 Antioxidant Cascade 16:20:00 Function of Vitamin E Antioxidant #2 Lipid BiLayer 17:44:00 Functions of Vitamin E Non Antioxidant Regulate Gene Expression 19:51:00 Genes Regulated by Vitamin E 20:11:00 Functions of Vitamin E Non Antioxidant Inhibit Smooth Muscle 20:41:00 Functions of Vitamin E Non Antioxidant‐Platelet Addhesion 21:25:00 Alpha and Gamma Tocopherols in Foods 22:26:00 Function of Gamma Tocopherol #1 Endothelial Lining 23:12:00 Function of Gamma Tocopherol #2 Cncer Cells 24:38:00 Function of Gamma Tocoopherol #3 Intracellular Antioxidant/ Cytokines 25:05:00 Take‐Aways Section 1 26:20:00 Section 2 Vitamin E Metabolism 26:42:00 Vitamin E Overview Absorption, Transport, Excretion Graphic 27:06:00 Vitamin E Metabolism Digestion‐ Esterase, Lipase, Bile Acids 27:43:00 Vitamin E Metabolism Proteins and Genes Associated with Transport 28:36:00 Summary Genes in Vitamin E Metabolism 29:03:00 Graphic Total Vitamin E Metabolism 29:17:00 Graphic Enterocyte 29:39:00 Genes in Entercoyte 29:45:00 Location Vitamin E transporters 30:20:00 Graphic Vitamin E Lymph and Blood 30:49:00 Diet to Chylomicron 31:19:00 Graphic Extra Hepatic Tissue Vitamin E Metabolism 32:38:00 Graphic Vitamin E Liver 35:54:00 Graphic Vitamin E Metabolism Complete 37:01:00 Fate of Vitamin E Vitamers 37:32:00 Vitamin E Biomarker #1 Serum Vitamin E 37:39:00 Vitamin E Biomarker #2 Inter‐individual Variation High 37:43:00 Gene Variants Explain High Variation with Serum Vitamin E Levels 38:10:00 Take‐Aways Section 2 38:59:00 Section 3 Alpha TTP Gene and Vitamin E Requirement 39:09:00 Learning Objectives for Section 3 39:19:00 Experimental Design #1 40:06:00 Experimental Design #2 41:05:00 Results of Experiment 41:49:00 Conclusion of Experiment 42:40:00 Alpha TTP Genotype and Vitamin E Concentration 43:34:00 Chrissi Case Report 43:38:00 Chrissi Dietary History, Health Issues, etc 44:38:00 Genomic Test Results for Chrissi 46:20:00 Modification of Chrissi Diet Based on Genomic Test Results 47:20:00 Dietary Sources of Alpha Tocopherol 48:57:00 Chrissi Genes for Vitamin E Transport in Enterocyte 50:43:00 Chrissi Genes for Vitamin E in Lymph 51:07:00 Chrissi Genes for Blood 52:01:00 Chissi Genes for Vitamin E in Liver 52:29:00 Chrissi Genes for Vitamin E in Tissues 53:01:00 Chrissi Genes for Vitamin E Excretion 53:39:00 Take‐Aways for Module 2 Session 4 and Questions 58:46:00 End Genomic Medicine Training Program

Module 2: Personalizing Nutrient Requirements and Resolving Nutritional Paradoxes Using Genomic Medicine

Joe Veltmann PhD FAAIM DCCN Roberta L. Kline MD FACOG

Part 3: Vitamin E Requirements in the Era of Genomic Medicine

1 Learning Objectives

• Section 1: Functions of Vitamin E • Section 2: Vitamin E Metabolism – Digestion, Absorption, Transport/Distribution and Excretion • Section 3: Alpha Tocopherol Transfer Protein Genotype and Serum Alpha Tocopherol Concentration: – Research Example – Case History (Chrissi) – Nutrigenomic Interventions • Nutritional Paradox • Take‐Aways

Clinical Relevance of Personalizing Micronutrient Requirements

Diet Name Eat to Live Fat Eat, Drink Vegan‐ Metabolism and Be Aggressive Diet Healthy WT Loss

Features Hypocaloric, Hi animal P; WT vegan Low Carb Maintenance Def Nutrients B12, B3, D, B1, D, E, Ca, D, Ca, K w/o energy E, Ca, Se Mg, K intake and Zn adjusted

Def Nutrient B12, D Ca, D D @ 2000 kcal/day

Section 1: Functions of Vitamin E

2 Similarities and Differences Between Structural Analogs (Vitamers) of Vitamin E Vitamin E • Tocopherols: – alpha tocopherol – beta tocopherol – gamma tocopherol – delta tocopherol

Similarities and Differences Between Structural Analogs (Vitamers) of Vitamin E Vitamin E • Tocopherols: – alpha tocopherol – beta tocopherol – gamma tocopherol – delta tocopherol • Tocotrienols: – alpha tocotrienol – beta tocotrienol – gamma tocotrienol – delta tocotrienol

3 Dietary Sources of Vitamin E

Alpha‐tocopherol • Corn • Wheat germ • Sunflower Oil Gamma Tocopherol • Walnuts • Soybeans • Palm Oil

Common Factors Affecting Vitamin E Bioavailability

• Gender and Age: – Vitamin E serum levels decrease with age. – Females have higher serum levels of vitamin E.

Common Factors Affecting Vitamin E Bioavailability

• Gender and Age: – Vitamin E serum levels decrease with age. – Females have higher serum levels of vitamin E.

• Lifestyle Choices: – Smoking – Alcohol

4 Common Factors Affecting Vitamin E Bioavailability

• Gender and Age: – Vitamin E serum levels decrease with age. – Females have higher serum levels of vitamin E. • Lifestyle Choices: – Smoking – Alcohol • Dietary Variables: – High Fat Diet – Competing Nutrients: • Vitamin K, EPA, plant sterols, retinoic acid, dietary fiber – Amount of natural or synthetic Vitamin E consumed daily

Additional Factors Affecting Vitamin E Bioavailability

• Drugs: – Inducers or Inhibitors of Cytochrome P450 Enzymes • GI Tract Diseases: – Malabsorption Syndrome, Cystic Fibrosis, Crohn’s Disease

Genetic Factors Affecting Vitamin E Bioavailability • Inherited Mutations Leading to Reduced Vit E Status: – Abetalipoproteinemia (Microsomal Triglyceride Transfer Protein Mutation) – Hypobetalipoprotein (ApoB gene mutation) – Vitamin E Transport Mutation (α‐tocopherol transfer protein)

5 Genetic Factors Affecting Vitamin E Bioavailability • Inherited Mutations Leading to Reduced Vit E Status: – Abetalipoproteinemia (Microsomal Triglyceride Transfer Protein Mutation). – Hypobetalipoprotein (ApoB gene mutation). – Vitamin E Transport Mutation (α‐tocopherol transfer protein). • Gene Variants (Polymorphisms) Resulting in Decreased Plasma Vitamin E Concentrations: – Genes Associated with Digestion, Absorption, Transport/Distribution, Utilization, Excretion of Vitamin E.

Functions of Vitamin E: Antioxidant

• First line of defense against lipid peroxidation. • Prevents auto‐oxidation of PUFAs associated with plasma membrane phospholipids and lipoprotein particles: – Ability to reversibly interconvert between alpha‐tocopherol and alpha‐tocopherolquinone forms. – Cofactors: Vitamin C, Selenium, GSH, Polyphenols, CoQ10

Antioxidant Cascade

CoQ10 recycles β‐Carotene, Polyphenols Lipoic Acid Polyphenols Vit E recycles Vit C recycles Vit C recycles GSH

Disarms Free Radicals Recycles Vit E Recycles Vit C

6 Functions of Vitamin E: Antioxidant

• Alpha‐tocopherol limits production of superoxide by inhibiting NADPH oxidase activity. • Inhibits phospholipase A2 and cyclooxygenase activities decreasing free radical production.

Image adapted: http://www.pnas.org/content/103/52/19908/F1.large.jpg

Functions of Vitamin E: Non‐Antioxidant

• Regulates Gene Expression: – Up‐regulates: • α‐TTP (alpha tocopherol transfer protein) • CYP3A4 (Cytochrome P450 3A4) • CYP3A5 (Cytochrome P450 3A5) • CTG (Connective Tissue Growth Factor) – Down‐regulates: – CD36 (Scavenger Receptor Class B Member 3) – MMP‐1 (Collagenase) – COL1A1 (Collagen 1 alpha 1)

Genes Regulated by Vitamin E

7 Functions of Vitamin E: Non‐Antioxidant

• Inhibits Smooth Muscle Cell Proliferation. • Supports RBC Maturation. • Promotes Development of CNS and Peripheral Neurons. • Signal Transduction (Cell Signaling Agent).

Functions of Vitamin E: Non‐Antioxidant

• Fosters Monocyte‐endothelial Adhesion. • Stimulates Cytokine Release When Monocytes Release Reactive Oxygen Species. • Modify Platelet Adhesion and Aggregation.

Gamma Tocopherol: Vitamin E Vitamer

Image source: Cook‐Mills JM, Marchese ME, et al., (2010) Antioxi Redox Signal 15: 1607‐1638

8 Functions of Gamma Tocopherol

Reduces Reactive Oxygen Species and Inflammation by Effectively Trapping Reactive Nitrogen Oxides (Peroxynitrite) in Endothelial Lining.

Image source: Cook‐Mills JM, Marchese ME, et al., (2010) Antioxi Redox Signal 15: 1607‐1638

Functions of Gamma Tocopherol

Regulates Factors that Guard Against Certain Cancers (CRC and Prostate) and Degenerative Diseases: – Protects pancreatic cells from IL‐1B. – Reduces TNF‐alpha levels. – Inhibits formation of superoxide radicals. – Reduces several pro‐thrombotic factors.

Functions of Gamma Tocopherol

Activates Genes Involved in Protecting Against Alzheimer’s Disease: – Boosts Anti‐inflammatory Cytokines and Intracellular Antioxidant Defenses.

9 Take‐Aways Section 1: Vitamin E Functions

• Vitamin E is a Generic Term for 8 Chemically Related. Compounds: 4 with a Tocopherol Backbone, 4 with a Tocotrienol Backbone. • Alpha Tocopherol Only Vitamin E Vitamer Concentrated and Retained from Food Sources by Human Body. • Multiple Factors Influence Vitamin E Status: Mutations/ SNPs. • Vitamin E’s Role in the Human Body: Antioxidant Versus Non‐ Antioxidant. • Vitamin E Deficiency: peripheral neuropathy, ataxia, muscle weakness, retinopathy, increase risk of CVD, prostate CA, cataracts. © Genoma International All rights reserved. 28

Section 2: Vitamin E Metabolism

Overview: Vitamin E Digestion, Absorption, Transport/ Distribution and Excretion

Vitamin E in Food Supplemented in Water

Lymph

10 Vitamin E Metabolism: Proteins and Genes

• Vitamin E Digestion and Absorption: – Esterase, Pancreatic Lipase, Bile Acids

Vitamin E Metabolism: Proteins and Genes

• Vitamin E Digestion and Absorption: – Esterase, Pancreatic Lipase, Bile Acids • Vitamin E Transport and Distribution: – Enterocyte (5): –Scavenger Receptor Class B Member 1 (SR‐B1) –Neimann‐Pick C1‐like 1 (NPC1L1) –Scavenger Receptor Class B Member 3 (SCARB3 or CD36) –ATP‐binding cassette transporter 1 (ABCA1) –Microsomal (Triglyceride) Transfer Protein (MTP)

Vitamin E Metabolism: Proteins and Genes • Vitamin E Digestion and Absorption • Vitamin E Transport and Distribution – Enterocyte – Lymph: ApoB, ApoE, ApoA5, ApoC3 – Blood: HDL, LDL, VLDL, ApoA1, LPL, ApoBR, CETP, PLTP, ApoB – Liver: SR‐B1, LPR, LDLR, Alpha‐tocopherol Transfer Protein (α‐TTP) – Tissues: ABCA1, LDLR

• Vitamin E Excretion: CYP4F2, CYP3A4, CYP3A5

Vitamin E Absorption and Intracellular Vitamin E Transport Receptor‐Mediated Lipid Transport Proteins: • Apical side of Enterocyte – Scavenger Receptor Class B Member 1 (SR‐B1) – Neimann‐Pick C1‐like 1 (NPC1L1) – Scavenger Receptor Class B Member 3 (SCARB3): • AKA Cluster of Differentiation 36 CD36) • Basal side of Enterocyte – ATP‐Binding Cassette Transporter 1 (ABCA1) – Microsomal (Triglyceride) Transfer Protein (MTP)

12 Location of Vitamin E Transporter Proteins Within Regions of the Small Intestine

Vitamin E: From Diet to Chylomicron

http://ib.bioninja.com.au/standard‐level/topic‐6‐human‐physiology/61‐digestion‐and‐absorption/lipid‐absorption.html

14 Metabolic Fate of Vitamin E Vitamers

• Phase I Detoxification: – CYP4F2 and CYP3A4/ CYP3A5 • Phase II Detoxification: – Sulfation and Glucuronidation • Phase III Detoxification: – Metabolites Found in Bile, Feces and Urine

Vitamin E Biomarker and Individual Variability

• Serum or Plasma Alpha Tocopherol Concentration Used as Vitamin E Biomarker.

Vitamin E Biomarker and Individual Variability

• Serum or Plasma Alpha Tocopherol Concentration Used as Vitamin E Biomarker. • Inter‐Individual Variability High: 20 to 80%!

15 Vitamin E Biomarker and Individual Variability

• Serum or Plasma Alpha Tocopherol Concentration Used as Vitamin E Biomarker. • Inter‐Individual Variability High: 20 to 80%! • Gene Variants Associated with Alpha Tocopherol Transfer Protein Contribute Most to Plasma or Serum Alpha Tocopherol Concentrations. Could Explain Conflicting Conclusions about Vitamin E Status and Health Outcomes.

Take‐Aways: Section 2 Vitamin E

• Absorption, transport and distribution of Vitamin E vitamers from food requires an orchestrated process among different organs, lipoproteins, membrane receptors and transport proteins, and enzymes. • Genes encode for many of the proteins involved in these processes. • Gene variants associated with alpha tocopherol transfer protein significantly alter concentration of serum alpha tocopherol concentrations.

Section 3: Gene Variant for Alpha TTP Gene and Vitamin E Requirement

16 Learning Objectives: Section 3

• Alpha Tocopherol Transfer Protein (α‐TTP) Genotype: – Impact on Circulating Levels of Alpha Tocopherol. • Research to support hypothesis • Case History (Chrissi): – Nutrigenomic Interventions • Take‐Aways

α‐TTP Gene SNPs and Vitamin E Status

Experimental Design: • Determine Whether Gene Variants associated with α‐TTP gene affect Serum Vitamin E Concentrations in Finnish Men With and Without Prostate Cancer: – Supplemented with 50 mg/day Alpha Tocopherol Acetate for 3 Years.

α‐TTP Gene SNPs and Vitamin E Status

Experimental Design: • 982 cancer cases and 851 controls. • All smokers: 5 or more cigarettes/ day); 50‐69 years of age. • 50 mg/ day alpha tocopherol was supplemented for both controls and cancer cases. • Serum alpha tocopherol determined at the start of the study and then 3 years later by HPLC. • Compare genotype x alpha tocopherol supplementation on serum vitamin E levels.

17 α‐TTP Gene SNPs and Vitamin E Status

Results: • Carriers of both copies of the minor alleles had approximately 3% lower baseline alpha tocopherol concentrations than men with both copies of the major alleles. α ‐TTP Genotype Subjects α‐tocopherol (mg/L) % Difference (95 % CI) Major Alleles 294 13.0 (12.7‐13.4) Referent Major/ Minor Alleles 373 12.8 (12.5‐13.2) ‐1.5 (0.27) Minor Alleles 153 12.6 (12.2‐13.0) ‐3.1 (0.03)

α‐TTP Gene SNPs and Vitamin E Status Conclusions: • Study confirms vitamin E status in blood linked to alpha tocopherol transfer protein gene variants. • Caucasian men (smokers) with the alpha‐TTP genotype (2 copies of the minor allele) associated with: – lower baseline serum alpha tocopherol concentrations; – lower serum alpha tocopherol response to 50 mg/day alpha tocopherol acetate supplementation; and – may require higher vitamin E intake to meet alpha tocopherol requirement and maintain homeostasis.

α-Tocopherol Transfer Protein Genotype: Plasma Vitamin E Concentrations Gene Function Alleles Vitamin E Phenotype

Gene encodes for minor High Impact tocopherol transfer M/m Moderate Impact protein; little affinity Major Low or No Impact for vitamin E vitamers Minor allele associated with other than alpha decreased plasma alpha tocopherol tocopherol; genotype concentrations; may need to determines vitamin E supplement depending on intake of concentration in blood. type and amount of dietary Vitamin E‐rich foods. © Genoma International All rights reserved. 54

18 Case History: Chrissi

Genomic Test Results for α‐TPP: Before and After Biomarker Test (Chrissi) Gene High Moderate Nutritional Intervention/ Impact Impact Lifestyle Choices α-TTP X Increase intake of foods rich in vitamin E Supplement with mixed tocopherols Inhibitors of Vitamin E Bioavailability?

α-TTP Serum α- Serum α- Reference Nutritional Genotype tocopherol tocopherol 90 Range Intervention Before days After (ug/mL) Intervention Intervention Minor 5.1 ug/mL 11.4 ug/mL 5.5 to 17 400 IU mixed Alleles ug/mL tocopherol

19 Modification to Chrissi’s Diet & Nutritional Supplements Based on DNA Data Before DNA testing After DNA Testing Diet Ham Limit to only 1 serving per month Cheese Eliminate White Bread Eliminate and substitute with gluten free bread; limit to 3 to 4 times per week Eggs Egg white only Chicken White meat, without the skin Salad Colorful and fresh, 4‐6 servings per day Veggies Colorful and fresh, 4‐6 servings per day Yogurt Eliminate Beef Lean beef only Other proteins Beans, Protein Powders, Tofu, Nuts and Seeds Fruits Colorful and fresh, low sugar fruits Nutritional NONE EPA/DHA, Ubiquinol, Multiple Vitamin‐Mineral with 50% Supplements Vitamin A Ester and 50% Carotenoids, Mixed Tocopherols © Genoma International All rights reserved. 58

Dietary Sources of Alpha Tocopherol

Oils Nuts Herbs & Spices Fruits & Vegetables Wheat germ Almonds Dried: Cooked tomatoes Hazelnut Hazelnuts Basil Dried apricots Olive Pine Nuts Oregano Spinach Sunflower Brazil Nuts Sage Avocado Almond Peanuts Thyme Swiss chard Safflower Chili powder Red peppers Grapeseed Cayenne Dandelion greens Seeds Canola Turnip greens Sunflower Fresh: Other Cilantro Spirulina Egg Eel © Genoma International All rights reserved. 59

Chrissi: Genotype Matrix for Lipoproteins and Transporter Proteins Location Gene Protein High Moderate Low / No Impact Impact Impact

Enterocyte‐ SR‐B1 Scavenger Receptor Class B X apical Member 1 (SR‐B1 or SCARB1) NPC1L1 Neimann‐Pick C1‐like1 CD36 Scavenger Receptor Class B Member 3 (SCARB3) Enterocyte‐ MTP Microsomal Transfer Protein basal ABCA1 ATP‐binding Cassette X Transporter

20 Chrissi: Genotype Matrix for Lipoproteins and Transport Proteins Location Gene Protein High Moderate Low / No Impact Impact Impact

Lymph Apo A5 Apolipoprotein A5 X Apo B Apolipoprotein B Apo C3 Apolipoprotein C3 X Apo E Apolipoprotein E X

Chrissi: Genotype Matrix for Lipoproteins and Transporter Proteins Location Gene Protein High Moderate Low / No Impact Impact Impact

Blood HDL High Density Lipoprotein‐C LDL Low Density Lipoprotein‐C VLDL Very Low Density Lipoprotein Apo A1 Apolipoprotein A1 X Apo B Apolipoprotein B Apo BR Apolipoprotein B Receptor CETP Cholesterol Ester Transfer Protein X LPL Lipoprotein Lipase X PLTP Phospholipid Transfer Protein

Chrissi: Genotype Matrix for Lipoproteins and Transporter Proteins Location Gene Protein High Moderate Low / No Impact Impact Impact

Liver α‐TTP Alpha‐tocopherol Transfer X Protein SR‐B1 Scavenger Receptor Class B X Member 1 (SCARB1) LDLR Low Density Lipoprotein‐C X Receptor LPR Lipoprotein Receptor

21 Chrissi: Genotype Matrix for Lipoproteins and Transporter Proteins Location Gene Protein High Moderate Low /No Impact Impact Impact

Tissues ABCA1 ATP‐binding Cassette X Transporter SR‐B1 Scavenger Receptor Class B X Member 1 (SCARB1) LDLR Low Density Lipoprotein‐C X Receptor LPR Lipoprotein Receptor

Chrissi: Genotype Matrix for Detoxification Location Gene Protein High Moderate Low / No Impact Impact Impact

Liver‐Phase I CYP4F2 Cytochrome P450 4F2 CYP3A4 Cytochrome P450 3A4* CYP3A5 Cytochrome P450 3A5 Liver‐Phase II SULT1A1 Sulfotransferase 1A1 X UGT1A1 UDP‐glucuronosyltransferase 1A1

UGT1A3 UDP‐glucuronosyltransferase 1A3

UGT1A4 UDP‐glucuronosyltransferase 1A4 Bile‐Phase III MDR3/ ABCB4 Multidrug resistance protein 3

Take‐Aways Section 3 Vitamin E: α‐TTP Genotype and Vitamin E Status

• Digestion and absorption of fat soluble vitamins (carotenoids, vitamin A, and vitamin D) use similar proteins and processes outlined in this section for Vitamin E. • Lipoproteins are inexorably tied to the transport and distribution of vitamin E in blood and tissues (Polygenic versus Monogenic). • Serum level of alpha tocopherol is regulated by α-TTP. • Vitamin E requirement may need to be adjusted based on genomic test results (Nutritional Paradox); use culinary genomics roadmap.

22 Module 2 Lesson 5

Time Stamp Title / Slide Subject 0:00 Genomic Medicine Training Program 0:09 Module 2 Session 5 Vitamin C 0:48 Vitamin C Requirement in Era of Genomic Medicine 1:36 Learning Objectives 2:25 Vitamin C Biochemistry in Plants and Animals 3:13 Vitamin C Inborn Error of Metabolism in Humans # 1 3:17 Vitamin C Inborn Error of Metabolism in Humans # 2 4:02 Glucose, Ascorbic Acid, and Dehydroascorbic Acid 5:39 Ascorbic Acid Versus Ascorbate: ph Dependent 6:27 Dietary Sources of Ascorbic Acid 7:16 Vitamin C Deficiency: Scurvy 8:53 Functions of Vitamin C 10:30 Vitamin C in Collagen Synthesis 12:55 Vitamin C in Hydrolyation of Proline and Lysine in Collagen Synthesis 14:32 Vitamin C in the Biosynthesis of Carnitine 16:09 Vitamin C in the Biosynthesis of Catecholamine Enzyme‐Dopamine Beta‐Hydroxylase 18:34 Vitamin C Functions‐Others 19:23 Iron Absorption, Transport and Utilization: Vitamin C as Co‐factor 27:28:00 Vitamin C in the Electron Transport Chain 28:16:00 Vitamin C uptake and recycling in Mitochondria 29:53:00 Vitamin C in Bile Acid Synthesis 31:13:00 Functions of Vitamin C in Stimulating Immune System 31:30:00 Vitamin C and the Common Cold or Flu 32:19:00 Vitamin C in Immune Function 36:21:00 Vitamin C in the Prevention and Treatment of Infections 37:19:00 Functions of Vitamin C‐Anioxidant Cascade 37:50:00 Vitamin C in Antioxidant Cascade 38:46:00 Ascorbate in Lipid Bilayer 40:23:00 Traditional and Non‐tradiational Roles of Vitamin C 42:53:00 Vitamin C Absorption‐Passive Diffusion and Active Transport SVCT1 and SVCT2 43:37:00 Vitamin C Digestion, Absorption/Transport/ Distribution and Excretion‐SVCT1 45:14:00 Vitamin C Digestion, Absorption/Transport/ Distribution and Excretion‐SVCT2 46:03:00 Genes Encoding for Protein in Ascorbate Absorption and Intracellular Transport 49:17:00 Genes Encoding for Protein in Ascorbate Absorption and Intracellular Transport‐GLUT1 and GLUT3 50:05:00 Efflux of Vitamin C from Intestinal and Renal Epithelium‐Acorbate 51:12:00 Efflux of Vitamin C from Intestinal and Renal Epithelium‐Dehydroascorbic Acid 52:30:00 Vitamin C Absorption Capacity 54:07:00 Tissue Distribution of Ascorbate 57:21:00 Levels of Ascorbic Acid Linked to SLC23A1 Genotype‐Study 58:58:00 Experimental Design and Results of Study 59:47:00 Conclusions of Study 1:02:12 Genotype Alleles for SLC23A1 1:02:52 Case History Chrissi 1:03:01 Case History Details‐Chrissi 1:04:38 Genomic Test Results SLC23A1 for Chrissi 1:05:26 Nutritional Paradox for SLC23A1 Genotype 1:06:15 Genomic Test Results SLC23A1 for Chrissi Along with Vitamin C Biomarker 1:07:52 Vitamin C Requirement, Genotype and Health Issues 1:09:28 Vitamin C Requirement, Genotype and Estrogen Metabolism Genotypes 1:10:17 Vitamin C and Estrogen Metabolites 1:12:42 Modification of Chrissi's Diet and Supplements Based on DNA Data 1:14:19 Foods Rich in Vitamin C 1:15:56 Take Aways SLC23A1 and Vitamin C Status # 1 1:16:45 Take Aways SLC23A1 and Vitamin C Status # 2 1:20:47 End Genomic Medicine Training Program

Module 2: Personalizing Nutrient Requirements and Resolving Nutritional Paradoxes Using Genomic Medicine

Joe Veltmann PhD FAAIM DCCN Roberta L. Kline MD FACOG

Vitamin C Requirements in the Era of Genomic Medicine

1 Learning Objectives

• Vitamin C Metabolism. • Vitamin C Transport Genotype and Vitamin C Requirement • Research Study • Case History (Chrissi): – Nutrigenomic Interventions • Nutritional Paradox • Take‐Aways

Vitamin C Biochemistry: Plants and Animals

• Abundant molecule found in plants and animals • Exception: humans, primates, guinea pigs, some birds and fish • Gene mutation precludes its synthesis.

Vitamin C: Inborn Error of Metabolism in Humans

• Mutation in gene GULO (encodes for L‐gulonolactone oxidase) leads to inborn error of metabolism and L‐ascorbic acid deficiency.

Image adapted from: https://answersingenesis.org/genetics/human‐gulo‐pseudogene‐evidence‐evolutionary‐discontinuity‐and‐genetic‐entropy/

2 Vitamin C

• Mutation in gene GULO (encodes for L‐gulonolactone oxidase) leads to in born error of metabolism and L‐ascorbic acid deficiency. • Vitamin C: Not part of any metabolic pathway; functions as a cofactor in enzymatic reactions; antioxidant.

Image adapted from: https://answersingenesis.org/genetics/human‐gulo‐pseudogene‐evidence‐evolutionary‐discontinuity‐and‐genetic‐entropy/

Glucose, Ascorbic Acid and Dehydroascorbic Acid

• Ascorbic Acid (Reduced Form): Active Molecule • Dehydroascorbic Acid (Oxidized Form): Inactive Molecule

Image source: Roomi MW et al. Vitamin C in Health: Scientific focus on its anti‐cancer efficacy. Journal Cell Med & Nat Health. June 2016

Ascorbic Acid and Ascorbate: pH Dependent

Ascorbate radical

Graphic source: Figueroa‐Mendez R and S Rivas‐Aranchibia (2015). Frontiers in Phys 6:397

3 Dietary Sources of Ascorbic Acid • Fresh Fruits: – Guava Cranberries – Star Fruit Strawberries – Raspberries Blueberries – Watermelon Kiwi • Fresh Vegetables: – Red Peppers – Winter Squash – Tomatoes – Brussels sprouts

Vitamin C Deficiency: Scurvy

• Muscle Pain • Whole body: fatigue, fever, loss of appetite, or malaise • Mouth: bleeding gums or tooth loss • Skin: rashes or red spots • Also common: bruising, coiled hair, failure to thrive, irritability, muscle weakness, swollen gums, or weight loss; anemia

Functions of Vitamin C

• Reducing Agent in Different Reactions: – Iron‐dependent hydroxylase enzymes • Lysine and proline hydroxylases (collagen synthesis/ wound healing/ bone health) • Carnitine biosynthesis – Copper‐dependent hydroxylase enzyme • Catecholamine synthesis (Dopamine Beta‐hydroxylase) – Support and structure of neurons (differentiation, maturation, survival)

4 Image Source: Nakarani K (2018). Biochemistry at Surat Municipal Institute of Medical Education and Research. Collagen Biochemistry.

Vitamin C: Cofactor in Hydroxylation of Proline and Lysine in Collagen Synthesis

Hydroxylation

Glycosylation

Xu, Y et al. iHyd‐PseAAC: Predicting Hydroxyproline and Hydroxylysine in Proteins by Incorporating Dipeptide Position‐Specific Propensity into Pseudo Amino Acid Toxqui, L et al. Chronic Iron Deficiency as an Emerging Risk Factor for Composition. Int. J. Mol. Sci. 2014, 15(5), 7594‐7610. Osteoporosis: A Hypothesis. 2015 Nutrients 7(4):2324‐2344 © Genoma International All rights reserved. 14

Biosynthesis of Carnitine

Trimethyllysine Hydroxytrimethyllysine

Deoxoycarnitine Deoxoycarnitine aldehyde

Image source: https://commons.wikimedia.org/wiki/File:Biosynthesis_L‐carnitine.png

Image source: http://slideplayer.com/slide/7840442/

5 Vitamin C: Catecholamine Synthesis (Dopamine Beta‐hydroxylase)

Functions of Vitamin C

• Reducing Agent in Different Reactions: – Iron‐dependent hydroxylase enzymes – Copper dependent mono‐oxygenases – Support and structure of neurons (differentiation, maturation, survival) – Non‐heme absorption in small intestine/regulates iron homeostasis – Electron Transport System (Protects mitochondrial function and membrane) – Synthesis of bile acids

SLC11A2 or GPx

Image adapted: https://themedicalbiochemistrypage.org/iron‐copper.php

6 Electron Transport Chain & Vitamin C

Participates in chemical reactions associated with energy (ATP) production in mitochondria.

Reduced CoQ10

Image source: http://users.humboldt.edu/rpaselk/C438.S12/C438Notes/C438nLec25.htm

Vitamin C Uptake and Recycling in Mitochondria: Prevents Mitochondrial Dysregulation

GPx

Image source: Mandl J, Szarka A, et al., (2009). Vitamin C update on physiology and pharmacology. Br J Pharmacol 157:1097‐1110.

Image source: Botham KM, Mayes PA, et al., (2015) Chapter 26: Cholesterol Synthesis, Transport, & Excretion, Harper's Illustrated Biochemistry, Thirtieth Edition, The McGraw‐Hill Education, NY

7 Functions of Vitamin C

• Reducing Agent in Different Reactions: – Iron‐dependent hydroxylase enzymes – Copper dependent mono‐oxygenases – Support and structure of neurons (differentiation, maturation, survival) – Non‐heme absorption in small intestine/regulates iron homeostasis – Electron transport system (protects mitochondrial function and membrane) – Synthesis of bile acids – Stimulates immune system

Vitamin C and the Common Cold • Vitamin C reduces the duration of colds by anywhere from 5 to 21%. • Vitamin C significantly reduce the severity of cold symptoms. • Vitamin C appreciably decreases the incidence of common cold.

Role of Vitamin C in Immune System

• Accumulates in neutrophils, enhances chemotaxis (a), phagocytosis (b), generation of reactive oxygen species c), and ultimately microbial killing (d). • Facilitates apoptosis and removal of neutrophils from sites of infection by macrophages, decreasing necrosis and potential tissue damage. • Enhances differentiation and proliferation of B‐ and T‐cells, increasing antibody production. • Inflammatory mediator: modulates cytokine production and decreases histamine levels. • Supports epithelial barrier function against pathogens and environmental oxidative stress.

8 Vitamin C in the Prevention & Treatment of Infections • Prevention of infections: – Dietary vitamin C intakes (100–200 mg ) needed to saturate plasma levels and optimize cell and tissue levels. • Treatment of established infections: – Requires significantly higher (gram) doses of the vitamin to compensate for the increased inflammatory response and metabolic demand.

Functions of Vitamin C

• Reducing Agent in Different Reactions – Iron‐dependent hydroxylase enzymes – Copper dependent hydroxylase enzyme – Support and structure of neurons (differentiation, maturation, survival) – Non‐heme absorption in small intestine/regulates iron homeostasis – Electron Transport System (Protects mitochondrial function and membrane) – Synthesis of bile acids – Stimulates immune system • Anti‐oxidant (scavenges free radicals): help recycles vitamin E

Antioxidant Cascade

CoQ10 recycles β‐Carotene, Polyphenols Lipoic Acid Polyphenols Vit E recycles Vit C recycles Vit C recycles GSH

Disarms Free Radicals Recycles Vit E Recycles Vit C

9 Function of Ascorbate: Antioxidant Partner

Extracellular, water soluble antioxidant

CoQ10 = Ubiquinone CoQH2 = Ubiquinol

Traditional and Non Traditional Roles of Vitamin C

• Traditional – Collagen Carnitine – Catecholamine Synthesis – Iron Homeostasis – Bile – Immune Function – Antioxidant • Non‐Traditional – Hormonal regulation (vasopressin) – Gene transcription (HIF1α) – Epigenetic modification • Methylated DNA and histones

Vitamin C: Digestion, Absorption, Transport/Distribution and Excretion • Uptake of ascorbate from lumen and transport into cells by: – Passive Diffusion – Active Transport: two sodium dependent vitamin C co‐transporters (SVCT1 and SVCT2)

10 Vitamin C: Digestion, Absorption, Transport/Distribution and Excretion • Uptake of ascorbate from lumen and transport into cells by – Passive Diffusion – Active Transport: two sodium dependent vitamin C co‐transporters (SVCT1 and SVCT2) • SVCT1: predominately in intestinal enterocyte and renal tubular cell (located in apical plasma membrane); – Permits absorption of vitamin C in excess of cellular needs. – Regulates circulating concentrations of ascorbate. – Has greater affinity for ascorbate. – Absorbs ascorbate 9 X faster than SVCT2.

Vitamin C: Digestion, Absorption, Transport/Distribution and Excretion

• SVCT2: found primarily in enterocytes (basal lateral membrane): – Also in brain, eye, skeletal muscle where ascorbate concentration tightly controlled. – Responsible for normal brain function and homeostasis.

Genes Encoding for Proteins in Ascorbate Absorption and Intracellular Transport • SLC23A1 encodes for SVCT1. • SLC23A2 encodes for SVCT2. Enterocyte – SVCT 1 and 2: Ascorbic Acid – Ileum and duodenum Efflux (?) GSH

Image source: Bohndiek, SE et al. Hyperpolarized [1‐13C]‐ascorbic and dehydroascorbic acid: vitamin C as a probe for imaging redox status in vivo. J. Am. Chem. Soc. (2011). © Genoma International All rights reserved. 33

11 Genes Encoding for Proteins in Ascorbate Absorption and Intracellular Transport • SLC23A1 encodes for SVCT1. • SLC23A2 encodes for SVCT2. Enterocyte – SVCT 1 and 2: Ascorbic Acid – Ileum and duodenum Efflux (?) • GLUT1, GLUT3, and GLUT4 (Glucose GSH Transporters): – GLUT1 and GLUT3‐‐dehydroascorbic acid found in foods or oxidation of ascorbic acid intestinal tract. – Jejunum

Image source: Lindblad M, et al. Nutrients 2013, 5(8), 2860‐2879; doi:10.3390/nu5082860

Efflux of Vitamin C from Intestinal and Renal Epithelium • Ascorbate: – Volume Sensitive Anion Channels (membrane proteins mediating the passive transport of organic anions in response to changes in osmolarity) – Exocytosis – Gap junctions • DHA Reduced to Ascorbate: – Extracellular spaces; picked up by capillary supply that supply the mucosa of the small intestine.

Image source: Lindblad M, et al. Nutrients 2013, 5(8), 2860‐2879; doi:10.3390/nu5082860

12 Vitamin C Absorption Capacity

• Low intake of vitamin C (less than 20 mg): 98% efficiency. • Moderate intake of Vitamin C (30 to 180 mg): 70 to 90% efficiency. • High intake of Vitamin C (> 1 g): < 50 % efficiency. – Body absorbs what it needs in 2 hours. – Within 3 to 4 hours any unused portion out of blood stream.

Tissue Distribution of Ascorbate

Tissue mg/kg Transporter Significance Adrenals 550 SVCT2 Catecholamine (Medulla) Synthesis Brain 140 SVCT2 Neuron Development and Function Liver 125 Mostly Bile Acid SVCT1 but Synthesis; Redox some SVCT2 Muscle 35 SVCT2 Redox

(Skeletal) Image source: Lindblad M, et al. Nutrients 2013, 5(8), 2860‐2879; doi:10.3390/nu5082860

Are Circulating Levels of L‐Ascorbic Acid Linked to SLC23A1 Genotype? • Experimental Design (British Women’s Heart and Health Study): – Determine association between L‐ascorbic acid concentrations and SLC23A1 genotypes. – 3,425 women of European ancestry; ages 60 to 79 years.

Timpson, NJ et al. Genetic variation at the SLC23A1 locus is associated with circulating levels of L‐ascorbic acid (Vitamin C). Am J Clin Nutr. 2010 August ; 92(2): 375–382. doi:10.3945/ajcn.2010.29438.

13 Are Circulating Levels of L‐Ascorbic Acid Linked to SLC23A1 Genotype? • Experimental Design (British Women’s Heart and Health Study): – Determine association between L‐ascorbic acid concentrations and SLC23A1 genotypes; 3,425 Women of European ancestry; 60 to 79 yrs. • Results: Carriers of the minor allele for SLC23A1 had a significant reduction in circulating levels of L‐ascorbic acid (‐4.15 μmol/L per minor allele). Genotype SLC23A1 Major M/m mm Per Allele Effect L‐Ascorbic Acid (μmol/L) 43.77 38.63 32.61 ‐4.15

Circulating Levels of L‐Ascorbic Acid Linked to SLC23A1 Genotype • Experimental Design: Meta analysis (totaling 15,087 people, 5 studies). • Results: Pooled analysis confirmed each additional rare allele was associated with a reduction in circulating levels of L‐ascorbic acid of −8.31 μmol/L per minor allele. Genotype SLC23A1 MM Mm mm Effect/minor allele L‐Ascorbic Acid (μmol/L) 56.66 48.21 43.38 ‐8.31

Conclusion: Gene SNP in SLC23A1 (vitamin C active transporter locus) associated with decreased circulating levels of L‐ascorbic acid in European population.

SLC23A1 Genotype: Lower Plasma Ascorbic Acid Concentration Gene Function Alleles Vitamin C Phenotype

Gene encodes for MM Low or No Impact SVCT1, a transporter Mm Moderate Impact protein, associated mm High Impact with the uptake of Minor allele associated with decreased L‐ ascorbic acid from the ascorbic acid concentrations; may need to lumen of the GI tract increase dietary vitamin C‐rich foods or and intracellular supplement, depending on redox state of transport. body, dose used, tissue metabolism and metabolic demand.

14 Case History: Chrissi

Case History: Chrissi • Health Issues (9): Rheumatoid Arthritis Yo‐yo Dieting—Weight Gain Ruptured Achilles Tendon Poor Recovery from Exercise Osteopenia Bi‐polar Disorder Extreme Fatigue Susceptible to Colds/Flu Insomnia • Diet: Ham, Cheese, Bread, Chicken, Salad, Veggies, Yogurt, Beef (No fish in diet: does not like the taste) • Exercise: Biking, Running (1 hr/day); Runs ½ Marathons • Meds: Lexapro (10 mg); Vyvance (20 mg) • Nutritional Supplements: None

Case Report: Chrissi and SLC23A1 Genotype

Gene High Moderate Nutritional Intervention/ Impact Impact Lifestyle Choices SLC23A1 X Increase intake of foods rich in Vitamin C Supplement with Vitamin C

15 Nutritional Paradox: SLC23A1 Genotype

• Chrissi’s diet contained few fruits rich in Vitamin C. • Ascorbic acid involved in many biological processes including maintaining an optimal immune response.

Nutritional Paradox: SLC23A1 Genotype

• Chrissi’s diet contains few fruits rich in Vitamin C. • Ascorbic acid involved in many biological processes including maintaining an optimal immune response. • Genotype: moderate impact on SLC23A1 may have contributed to low concentration of vitamin C before nutrigenomic intervention (500 mg, 2x per day). Before Intervention 90‐days After Ref Range Plasma Vitamin C 14 68 23 to 114 (μmol/L)

Vitamin C Requirement, Genotype, & Health Issues

• Vitamin C Requirements: – Absorption Genotype: SLC23A1 – Tissue Demands (Adrenal Gland and Stress) – Frequency of Cold/ Flu

16 Vitamin C Requirement, Genotype & Estrogen Metabolism Genotypes • Vitamin C Requirements: – Absorption Genotype: SLC23A1 – Tissue Demands (Adrenal Gland and Stress) – Frequency of Cold/ Flu – Metabolic Load, Free Radicals and Genotypes (Estrogen Metabolism)

Vitamin C & Estrogen Metabolism Genotypes • Chrissi’s Estrogen Metabolism Genotypes Detox Gene High Moderate Impact Impact

Phase I CYP1B1 X

Phase II COMT X

GSTM1 X

CAT X

GPx X

SOD2 X

Modification to Chrissi’s Diet & Nutritional Supplements Based on DNA Data Before DNA testing After DNA Testing Diet Ham Limit to only 1 serving per month Cheese Eliminate White Bread Eliminate and substitute with gluten free bread; limit to 3 to 4 times per week Eggs Egg white only Chicken White meat, without the skin Salad Colorful and fresh, 4‐6 servings per day Veggies Colorful and fresh, 4‐6 servings per day Yogurt Eliminate Beef Lean beef only Other proteins Beans, Protein Powders, Tofu, Nuts and Seeds Fruits Colorful and fresh, low sugar fruits Nutritional NONE EPA/DHA, Ubiquinol, Multiple Vitamin‐Mineral with 50% Supplements Vitamin A Ester and 50% Carotenoids, Mixed Tocopherols, Vitamin C © Genoma International All rights reserved. 51

17 Foods Rich in Vitamin C

Kakadu plum Pineapple Thyme Acerola cherries Oranges Parsley Rose hips Kiwi Chives Chili peppers Red grapefruit Wasabi Guava Kale Garlic Black currants Broccoli Natto Lemons Brussels sprouts Papayas Yellow or red Lychee peppers Mustard greens Strawberries

Take‐Aways: SLC23A1 and Vitamin C Status

• Uptake of Vitamin C from the small intestine and circulating blood concentrations of Vitamin C linked to SVCT1 transport protein. • SVCT1 encoded by SLC23A1. • Gene variants associated with SLC23A1 can result in 40 to 50 % reduction in Vitamin C absorption.

Take‐Aways: SLC23A1 and Vitamin C Status

• Carriers of two minor alleles for SLC23A1, who consume few Vitamin C rich foods, could be Vitamin C deficient, impacting redox balance and multiple biological systems. • Culinary strategies can help boost Vitamin C status with different fruits, vegetables and spices in the diet. • Nutritional Paradox: although diet might be replete in vitamin C, gene variants for SLC23A1 may preclude adequate vitamin C absorption and utilization in tissues.

18 Module 2 Lesson 6

Time Stamp Title / Slide Subject 0:00 Genomic Medicine Training Program 0:13 Module 2 Session 6 0:57 Selenium Requirement in the Age of Genomic Medicine 1:54 Learning Objectives 3:48 Selenium Basic Facts 5:41 Low Dietary Selenium and Chronic Disease 6:39 Selenium in Selenium Dependent Enzymes‐Selenoproteins 8:33 Functions of Selenoproteins # 1 9:31 Functions of Selenoproteins # 2 10:28 Functions of Selenoproteins 1‐GPx 13:19 Antioxidant Cascade and Selenoprotein GPx 14:16 GPx (cytosolic) and Iron Absorption 15:13 Selenoprotein in Testes (Spermatogenesis) 17:43 Ubiquitous Nature of Glutathione Peroxidase Enzymes 18:05 Glutathione Peroxidase and Cancer Tumor Progression Study 21:55 Functions of Selenoproteins 2‐Thiroredoxin Antioxidant System 23:47 Thioredoxin Reductase, FAD, NADPH and Selenocyteine 24:44:00 Sites Where Thioredoxin Plays a Role 26:39:00 Thioredoxin Reductase and Selenoprotein 27:36:00 Antioxidant Cascade and Selenoprotein Thioredoxin 29:30:00 Functions of Selenoprotein Enzymes 3‐DIO1, DIO2, DIO3 30:27:00 Selenoprotein and Thyroid Hormone Synthesis 32:21:00 Synthesis of T4 and T3 from Tyrosine TPO Enzyme Graphic 34:15:00 Graphic Steps 1 and 2 of T4 and T3 Synthesis in Thyroid Gland 37:30:00 Step 3 in Synthesis of T4 and T3 in Thyroid Gland 37:42:00 Step 4 in Synthesis of T4 and T3 in Thyroid Gland 37:54:00 Step 5 in Synthesis of T4 and T3 in Thyroid Gland 38:04:00 Step 6 in Synthesis of T4 and T3 in Thyroid Gland 38:18:00 Step 7 in Synthesis of T4 anf T3 in Thyroid Gland 39:01:00 Homeostasis of T3 and T4 Graphic 39:58:00 Factors Impacting Thyroid Hormone Levels and Homeostasis 41:52:00 DIO2 Polymorphism and Hypothyroidism‐Novel Treatment 44:44:00 DIO2 Genotype Description 45:41:00 DIO2 SNP and Neurodegenerative Disease 47:35:00 Graphic Steps 1 to 8 for Synthesis of T4 and T3 in Thyroid Gland 52:21:00 Functions of Selenoproteins 4‐Methionine‐R‐Sulfoxide Reductase 54:15:00 Distribution of Methionine‐R‐Sulfoxide Reductase 55:12:00 Function of Selenoprotein‐5 Selenium and Immune System 57:06:00 Functions of Selenoproteins‐6 Selenium Transport Protein SEPP1 59:57:00 Selenium Transporter Protein and Redox SEPW/ SELW 1:01:52 Function of Selenoprotein‐7 Selenoprotein, ER and Redox 1:02:47 Function of Selenoprotein‐7 SEPS1, SELS, VCP Membrane Seleoprotein 1:04:43 Gene SNP on SEPS1 increased Risk of Auto‐immunity 1:06:37 Function of Selenoprotein #8 DNA Architecture and Epigenetics 1:09:29 Selenium Requirement Using RDA Definition 1:12:20 Selenium Requirement for GPx1 1:13:17 GPx1 Genotypes 1:14:14 ROS versus Antioxidant Fulcrum 1:16:08 GPx1 Gene Variants Study and Selenium Supplement 1:18:03 Graphic Genotype GPx1 and Selenium Supplementation Results 1:22:48 GPx1 Genotype and Selenium Supplementation Conclusions 1:23:45 GPx1 Genotype and Selenium Supplementation MoreConclusions 1:24:42 Nutritional Paradox: GPx1 and Dietary Selenium Requirement 1:24:50 Nutritional Paradox: GPx1 and Dietary Selenium Biomarker Testing 1:25:26 Case History Chrissi 1:25:40 Case History Details Chrissi 1:26:37 Modification of Chrissi Diet and Nutritional Support Based on DNA 1:27:35 Sources of Selenium Rich Foods 1:27:59 Modification of Chrissi Diet and Nutritional Support Based on DNA 1:28:31 Case History Genotype Results for GPx1, DIO2 and GSH 1:31:22 Modification of Chrissi Diet and Nutritional Support Based on DNA 1:32:09 Take‐Aways Gene Variants and Selenium Nutrient Requirement 1:32:30 Take‐Aways Gene Variants and Selenium Nutritional Paradoxes 1:33:16 Take‐Aways Gene Variants and Selenium DNA Directed Interventions 1:34:13 Genomic Testing Benefits for Clinican and Patient 1:35:11 End Genomic Medicine Training Program

Module 2: Personalizing Nutrient Requirements and Resolving Nutritional Paradoxes Using Genomic Medicine

Joe Veltmann PhD FAAIM DCCN Roberta L. Kline MD FACOG

Selenium Requirements in the Era of Genomic Medicine

1 Learning Objectives

• Biological Functions and Clinical Significance of Selenoproteins – Enzymes (4), Se carriers (2), Redox balancers (3), Epigenetics • Research Study: – GPx1 genotype Influences Selenium Requirement • Case History (Chrissi): – Nutrigenomic Interventions • Nutritional Paradox • Summary

Selenium (Se): Basic Facts

• Essential Nutrient • Se Requirement: • 50 to 100 micrograms per day (adults) • 20 to 25 micrograms per day (< 18 yrs) • Serum Levels: • Adults: 60 to 160 micrograms/L • Children (< 2 yrs): 16‐71 micrograms/L

Low Dietary Selenium & Chronic Disease

• Parkinson’s disease • Alzheimer’s disease • Atherosclerosis • Myocardial infarction • Heart failure • Sickle cell disease • Carcinogenesis (Lung, Prostate, Breast, CRC, etc.) • COPD • Autism

2 Selenium

• Micronutrient required for selenium‐dependent enzymes (selenoproteins). – About 25 selenoproteins in humans. – During translation, selenocysteine inserted into proteins at very specific locations in the amino acid sequence to form functional selenoproteins.

Functions of Selenoprotein Enzymes

1. Component of glutathione peroxidase enzyme‐ converts hydrogen peroxide to water. 2. Component of thioredoxin reductase—antioxidant cascade. 3. Component of DIO2, a deiodinase 2 enzyme converts T4 to T3. 4. Component of methionine‐R sulfoxide reductase B1 (Selenoprotein R) protects methionine residues against ROS

Functions of Selenoprotein Enzymes

5. Boosts neutrophil “killing” ability. 6. Se carrier proteins Selenoprotein P (plasma) Selenoprotein W (muscle, brain) 7. Endoplasmic Reticulum & Redox Selenoprotein 15 Selenoprotein S1 8. Epigenetics

3 Functions of Selenoprotein Enzymes: 1. Glutathione Peroxidases • Glutathione peroxidases (8 isozymes): – GPx1 (cytosol and mitochondria) – GPx2 (GI tract and liver) – Gpx3 (kidneys and thyroid) – GPx4 (testes) – GPx5 (epididymis/reproductive tract) – GPx6 (olfactory epithelium) – GPx7 (endoplasmic reticulum) – Gpx8 (lungs) • All considered antioxidant enzymes

Antioxidant Cascade

CoQ10 recycles β‐Carotene, Polyphenols Lipoic Acid Polyphenols Vit E recycles Vit C recycles Vit C recycles GSH

Disarms Free Radicals Recycles Vit E Recycles Vit C

Glutathione Peroxidase

• GPx1 (cytosolic) – Facilitates the recycling

of Vitamin C by coupling SLC11A2 or reduction/oxidation with GPx1 GSH in iron absorption.

4 Selenoproteins in Testes (Spermatogenesis)

• GPx4: – Reduces phospholipid hydroperoxides in testes, protecting spermatozoa against oxidative stress. – Major structural protein capsule associated with mature sperm mitochondria helix‐‐sperm motility. • SEPP1: (Selenoprotein P) – Supplies selenium to testes.

Ubiquitous Nature of Glutathione Peroxidases

• Glutathione peroxidases (8 isozymes): – GPx1 (cytosol and mitochondria) – GPx2 (GI tract and liver) – Gpx3 (kidneys) – GPx4 (testes) – GPx5 (epididymis/reproductive tract) – GPx6 (olfactory epithelium) – GPx7 (endoplasmic reticulum) – Gpx8 (lungs) • All considered antioxidant enzymes

Glutathione Peroxidases Involved in Cancer Tumor Progression

• Mitochondria and NADPH oxidases: intracellular Extracellular sources of H2O2 and other hydroperoxides (ROOH). NADPH oxidases • H2O2 and/or ROOH activates cyclooxygenase‐2 (COX‐2), catalyzing the formation of pro‐ Lipoxygenases inﬂammatory prostaglandin PGE2: – Plays a role in the acute inﬂammatory response, leading to tumor cell proliferation and invasion. – Inhibits apoptosis by the activation of pro‐ survival intracellular pathways (PI3K/Akt, Ras‐ Intracellular MAPK/ERK pathways). – Supports cancer cell migration, invasion and angiogenesis.

– Induces CSCs through the Wnt cell signaling Image source: Jiao Y, Wang Y, et al. (2017). Glutathione peroxidases as pathway. oncotargets. Oncotarget 8:80093‐80102.

5 Functions of Selenoprotein Enzymes: 2. Thioredoxin Antioxidant System Thioredoxin (Trx) NADPH Thioredoxin reductase (TrxR) – Cytosol (TrxR1) – Mitochondrial (TrxR3) – Testes (TGR)

Image source: Whayne TF et al. Thioredoxins in Cardiovascular Disease. May 2015 Canadian Journal of Physiology and Pharmacology 93(11):150521143545001

Functions of Selenoprotein Enzymes: 2. Thioredoxin Andioxidant System Thioredoxin (Trx) NADPH Thioredoxin reductase (TrxR) – Cytosol (TrxR1) – Mitochondrial (TrxR3) – Testes (TGR) • Functions: major antioxidant enzyme system and redox regulator using FAD, NADPH and selenocysteine.

Functions of Selenoprotein Enzymes: 2. Thioredoxin Antioxidant System • Functions: major antioxidant enzyme systems and redox regulator using FAD, NADPH and selenocysteine. – DNA synthesis – Defends against ROS – Apoptosis – Electron donor – Cell growth & survival – Catalyze substrates – Restore antioxidant enzymes – Redox regulator (cell signaling) – Regeneration Vitamin C, alpha lipoic acid, vitamin E, CoQ10 (Antioxidant Cascade)

6 Functions of Selenoprotein Enzymes: 2. Thioredoxin Antioxidant System

Thioredoxin reductase Thioredoxin

Image source: Holmgren, Arne and Jun Lu. Thioredoxin and thioredoxin reductase: Current research with special reference to human disease. Biochemical and Biophysical Research Communications Volume 396, Issue 1, 21 May 2010, Pages 120‐124.

Antioxidant Cascade

CoQ10 recycles β‐Carotene, Polyphenols Lipoic Acid Polyphenols Vit E recycles Vit C recycles Vit C recycles GSH

Disarms Free Radicals Recycles Vit E Recycles Vit C

Functions of Selenoprotein Enzymes: 3. Iodothyronine Deiodinases (DIO1, DIO2, DIO3) Regulation of Thyroid Hormones

7 Selenoprotein: Thyroid Hormone Deiodinases

• Thyroid hormone deiodinases High stress, deficiencies : Se, Zn, (3): B‐vitamins, folate, vitamins A & C • DIO1 and DIO2 catalyze the conversion of inactive T4 Se, Zn, B‐vitamins, (thyroxine) to active T3 in the folate, vitamins A & C peripheral tissues. • DIO1 and DIO3 converts T4 to reverse T3 (rT3). • DIO1/ DIO3 and DIO1/ DIO2 converts T3 and rT3, respectively to inactive thyronine molecule.

Synthesis of T4 and T3 from Tyrosine

Thyroid peroxidase (TPO)

Image source: https://www.slideshare.net/amirthaqiff/20130417124022809

Figure 18‐11a The Thyroid Follicles

Follicle cavity

2. Two iodide atoms Thyroglobulin FOLLICLE CAVITY oxidized to form iodine (contains T3 and T4)

(thyroperoxidase) Endocytosis GPx3 Thyroglobulin Iodide + Lysosomal (I ) Other amino acids digestion

Tyrosine T4 T3

Diffusion

Diffusion TSH‐ MCT 8/10 sensitive ion pump 1. Iodide dependent on FOLLICLE CELL Na/I co‐transport system (TSH stimulated) CAPILLARY

Iodide (I–) TBG, transthryretin, or albumin T4 & T3 Image adapted from : http://leaThe synthesis, storage, and secretion of thyroid hormones.rning.hccs.edu/faculty/michael.phelps/biol2402/lecture‐presentations/ch.‐18‐endocrine

8 Figure 18‐11a The Thyroid Follicles 3. Iodinated Follicle cavity thyroglobulin complex (TSH)

2. Two iodide atoms Thyroglobulin FOLLICLE CAVITY oxidized to form iodine (contains T3 and T4)

(thyroperoxidase) Endocytosis GPx3 Thyroglobulin Iodide + Lysosomal (I ) Other amino acids digestion

Tyrosine T4 T3

Diffusion

Diffusion TSH‐ MCT 8/10 sensitive ion pump 1. Iodide dependent on FOLLICLE CELL Na/I co‐transport system (TSH stimulated) CAPILLARY

Figure 18‐11a The Thyroid Follicles 3. Iodinated Follicle cavity thyroglobulin complex (TSH) 4. Iodinated thyroglobulin complex 2. Two iodide atoms Thyroglobulin FOLLICLE CAVITY forms colloidal droplet oxidized to form iodine (contains T3 and T4) (TSH stimulated) (thyroperoxidase) Endocytosis GPx3 Thyroglobulin Iodide + Lysosomal (I ) Other amino acids digestion

Tyrosine T4 T3

Diffusion

Diffusion TSH‐ MCT 8/10 sensitive ion pump 1. Iodide dependent on FOLLICLE CELL Na/I co‐transport system (TSH stimulated) CAPILLARY

Iodide Lysosomal 5. Lysosomal digestion of (I+) Other amino acids digestion droplet releases T3 and T4

Tyrosine T4 T3

Diffusion

Diffusion TSH‐ MCT 8/10 sensitive ion pump 1. Iodide dependent on FOLLICLE CELL Na/I co‐transport system (TSH stimulated) CAPILLARY

9 Figure 18‐11a The Thyroid Follicles 3. Iodinated Follicle cavity thyroglobulin complex (TSH) 4. Iodinated thyroglobulin complex 2. Two iodide atoms Thyroglobulin FOLLICLE CAVITY forms colloidal droplet oxidized to form iodine (contains T3 and T4) (TSH stimulated) (thyroperoxidase) Endocytosis GPx3 Thyroglobulin

Iodide Lysosomal 5. Lysosomal digestion of (I+) Other amino acids digestion droplet releases T3 and T4

Tyrosine T4 T3

Diffusion 6. T3 and T4 efflux into Diffusion TSH‐ MCT 8/10 sensitive capillary by MCT 8/10 ion pump 1. Iodide dependent on FOLLICLE CELL Na/I co‐transport system (TSH stimulated) CAPILLARY

Iodide Lysosomal 5. Lysosomal digestion of (I+) Other amino acids digestion droplet releases T3 and T4

Tyrosine T4 T3

Diffusion 6. T3 and T4 efflux into Diffusion TSH‐ MCT 8/10 sensitive capillary by MCT 8/10 ion pump 1. Iodide dependent on FOLLICLE CELL Na/I co‐transport system 7. T3 and T4 bound to (TSH stimulated) CAPILLARY either albumin, pre‐

Iodide (I–) albumin or TGB TBG, transthryretin, or albumin T4 & T3 Image adapted from : http://leaThe synthesis, storage, and secretion of thyroid hormones.rning.hccs.edu/faculty/michael.phelps/biol2402/lecture‐presentations/ch.‐18‐endocrine

Figure 18‐11b The Thyroid Follicles Hypothalamus releases TRH Homeostasis Disturbed

Decreased T3 and TRH T4 concentrations in blood or low body temperature Anterior lobe Pituitary gland HOMEOSTASIS

Normal T3 and T4 Anterior concentrations, lobe normal body temperature TSH

Homeostasis Restored

Increased T3 and T concentrations 4 Thyroid in blood gland

Thyroid follicles release T3 and T4

10 Factors Impacting Thyroid Hormone Levels and Thyroid Homeostasis 1. TRH/TSH/Thyroid Axis aka Hypothalamus‐Pituitary‐Thyroid Axis (HPT) 2. Dietary a) Iodine deficiency b) Selenium deficiency (Deiodinase Enzymes) c) Protein/Amino Acid Deficiency (Phe and Tyr) d) Carbohydrates e) Co‐factor deficiencies 3. Cold Stress 4. Conditions with High Energy Requirements: Pregnancy Trauma Fight/Flight Exercise Puberty Injuries/Illness 5. Gene Polymorphisms (GPx3, DIO2)

DIO2 Polymorphism and Hypothyroidism: Novel Treatment

DIO2 Gene Variants Linked to Reduced Enzymatic Activity & T4 to T3 Conversion Gene Function Alleles T3 Phenotype

Gene encodes for a MM Little or No Impact selenium‐dependent Mm Moderate Impact enzyme, deiodinase 2, mm High Impact which catalyzes the Minor allele associated with decreased conversion of T4 to T3. enzyme activity and less T3 converted from T4. Prescribing more T4 (thyroxine) exacerbates hypothyroid symptoms; treat with combination of T4 and T3. © Genoma International All rights reserved. 33

11 DIO2 Polymorphism: Neurodegenerative Disease

Researchers did not evaluate GPx polymorphisms in subjects

Iodide Lysosomal 5. Lysosomal digestion of (I+) Other amino acids digestion droplet releases T3 and T4

Tyrosine T4 T3

Functions of Selenoprotein Enzymes: 4. Methionine‐R‐sulfoxide reductase B1 • Selenoprotein R (Methionine‐R‐sulfoxide reductase B1) – Protects proteins with methionine residues against oxidative stress, which can impair their function.

Image source: Nebraska Redox Biology Center Educational Portal. http://genomics.unl.edu/RBC_EDU/msr.html

12 Distribution of Methionine‐R‐sulfoxide reductase B1 (Selenoprotein R)

Image source: Nebraska Redox Biology Center Educational Portal. http://genomics.unl.edu/RBC_EDU/msr.html

Function of Selenoproteins 5. Selenium and Immune Function • Neutrophils – chemotaxis (a) – phagocytosis (b) – generation of reactive oxygen species c) – microbial killing (d) using H2O2 • Intracellular GPx—selenium dependent enzyme – Converts H2O2 to water and oxygen. – Minimizes tissue necrosis.

Function of Selenoproteins: 6. Selenium Transport Protein and Redox

• Selenoprotein P (SEPP1): ‐ Selenium transport protein—from liver to peripheral tissues. ‐ Provides adequate supply of selenium for thioredoxin reductase and GSH peroxidases. ‐ Critical for selenium homeostasis in brain and testes. ‐ ApoE2 receptor facilitates uptake of SEPP1. ‐ Check ApoE2 receptor genotype—may impact SEPP1 uptake.

13 Selenium Transport Protein and Redox

• Selenoprotein W (SEPW or SelW): – Selenium transport protein—found in heart, skeletal muscle, brain, breast and prostate. – Binds to GSH molecule supporting redox balance. – Protects against oxidative stress‐induced neuronal cell death.

Function of Selenoproteins: 7. Selenoproteins, Endoplasmic Reticulum & Redox

• Selenoprotein 15 (SEP15): – Interacts with endoplasmic reticulum. – Involved in maintaining quality control (redox balance) associated with glycoprotein folding (prostate, kidney, testes, liver and brain).

Function of Selenoproteins: 7. Selenoproteins, Endoplasmic Reticulum & Redox

14 Function of Selenoproteins: 7. Selenoproteins, Endoplasmic Reticulum & Redox

• Selenoprotein S (SEPS1, SelS or VCP‐interacting membrane selenoprotein): – ER‐bound protein involved in cellular response to ER stress activated by detection of misfolded proteins. – Removes and transfers misfolded proteins from ER lumen to cytosol where proteins are tagged with ubiquitin before degradation. Polymorphism in SEPS1 associated with increased plasma pro‐ inflammatory cytokines; predisposition to Hashimoto thyroiditis, preclampsia, CAD, gastrointestinal cancers. SEPS1 may regulate inflammation and autoimmune responses.

Selenium: DNA Architecture and Epigenetics

• Selenium and Genomic Stability: Protects against DNA damage‐‐ formation of DNA adducts, DNA or chromosome breaks, telomere length and function. • Selenium and Gene Expression: Modulation of DNA methylation/inhibition of histone deacetylation (epigenetics).

Image source: Bernstein, CA et al. DNA Damage, DNA Repair and Cancer. Semantic Scholar 2013.

Selenium Requirement: GPx

• RDA for Selenium based on amount of selenium need to maximize glutathione peroxidase activity (GPx) in plasma. – Form of Selenium: – Organic (selenomethionine) versus Inorganic (sodium selenite/selenate). – Selenomethionine increased blood selenium concentrations compared to sodium selenite and sodium selenate. • Absorption rates: Sodium selenite 100 %, selenomethione about 90, sodium selenate 50% • Inorganic forms increased plasma GPx more effectively than organic forms.

15 Selenium Requirement: GPx1

• RDA for Selenium based on amount of selenium need to maximize glutathione peroxidase activity (GPx) in plasma. – GPx Genotype: • Polymorphisms can reduce GPx1 enzymatic activity; linked to increased risk of cancer: Lung Breast Prostate Bladder CRC Gastric • Metabolic syndrome • Heart Disease • Aging/Longevity • Heavy Metal Toxicity • Generalized Oxidative Stress

GPx1 Genotype: Lower Plasma Selenium Concentration Gene Function Alleles GPX1 Phenotype

Gene encodes for a MM Little or No Impact selenium‐dependent Mm Moderate Impact enzyme, glutathione mm High Impact peroxidase, catalyzing Minor allele associated with the reduction of decreased plasma selenium hydrogen peroxide (a concentrations; may need to free radical) to water supplement depending on intake of in the antioxidant type and amount of dietary cascade. selenium‐rich foods. © Genoma International All rights reserved. 47

16 Study: Impact of GPx1 Gene Variants, Dietary Selenium Supplementation on Selenium Status • Experimental Design: – 37 morbidly obese, selenium (Se) deficient women – Consumed one Brazil nut (about 290 ug Se) for 8 weeks – Measurements compared at baseline and 8 weeks across the 3 GPx1 genotypes (MM, Mm, mm): • Blood Se concentrations • RBC GPx activity • DNA damage levels

Source: Cominetti C, de Bortoli MC, et al (2011). Associations between glutathione peroxidase‐1 Pro198Leu polymorphism, selenium status and DNA damage levels in obese women after consumption of Brazil nuts. Nutrition 27: 891–896 © Genoma International All rights reserved. 49

Genotype GPx1 and Selenium Supplementation Parameter Genotype Baseline Se Supplement % Change (290 ug Se) RBC Se MM 60.8 200.8 230 (ug/L) Mm 65.0 207.3 219 mm 59.7 220 269 RBC GPx MM 38.5 57.4 49 Activity Mm 33.0 51.7 56 (u/g Hb) mm 31.4 45.2 43.9 DNA MM 80.8 64.2 ‐20% Breakage Mm 67.6 68.5 1.3 (Comet Test) mm 92.1 114 56.3 Source: Cominetti C, de Bortoli MC, et al (2011). Associations between glutathione peroxidase‐1 Pro198Leu polymorphism, selenium status and DNA damage levels in obese women after consumption of Brazil nuts. Nutrition 27: 891–896 © Genoma International All rights reserved. 50

GPx1 Genotype and Selenium Supplementation

17 GPx1 Genotype and Selenium Supplementation

• In obese populations (oxidative stress), dietary selenium (Se) better incorporated into RBCs for carriers of the GPx1 MM and mm genotypes. • For obese carriers of the GPx1 mm genotype, dietary intake of 290 ug selenium resulted in lower RBC GPx activity and more DNA breakage. • Obese carriers of the MM genotype when supplemented with 290 ug Se had moderate increase in RBC Se and RBC GPx activity and significantly less DNA breakage. • Selenium requirement varies depending on GPx1 genotype in an obese population.

Nutritional Paradox: GPx1 and Dietary Selenium Requirement • RDA for selenium based on amount of selenium needed to maximize glutathione peroxidase activity (GPx) in plasma. • Polymorphisms associated with GPx1 can alter selenium requirement.

Nutritional Paradox: GPx and Dietary Selenium Requirement • RDA for Selenium based on amount of selenium needed to maximize glutathione peroxidase activity (GPx) in plasma. • Polymorphisms associated with GPx1 can alter selenium requirement. • Personalizing selenium requirement requires knowledge of a person’s GPx1 genotype. • Using glutathione peroxidase activity (GPx) activity in plasma may not be the best biomarker.

18 Case History: Chrissi

Case History: Chrissi • Health Issues (9): Rheumatoid Arthritis Yo‐yo Dieting—Weight Gain Ruptured Achilles Tendon Poor Recovery from Exercise Osteopenia Bi‐polar Disorder Extreme Fatigue Susceptible to Colds/Flu Insomnia • Diet: Ham, Cheese, Bread, Chicken, Salad, Veggies, Yogurt, Beef (No fish in diet: does not like the taste) • Exercise: Biking, Running (1 hr/day); Runs ½ Marathons • Meds: Lexapro (10 mg); Vyvance (20 mg) • Nutritional Supplements: None

Modification to Chrissi’s Diet & Nutritional Supplements Based on DNA Data Before DNA testing After DNA Testing Diet Ham Limit to only 1 serving per month Cheese Eliminate White Bread Eliminate and substitute with gluten free bread; limit to 3 to 4 times per week Eggs Egg white only Chicken White meat, without the skin Salad Colorful and fresh, 4‐6 servings per day Veggies Colorful and fresh, 4‐6 servings per day Yogurt Eliminate Beef Lean beef only Other proteins Beans, Protein Powders, Tofu, Nuts and Seeds Fruits Colorful and fresh, low sugar fruits Nutritional NONE EPA/DHA, Ubiquinol, Multiple Vitamin‐Mineral (50% Supplements Vitamin A Ester and 50% Carotenoids, alpha Tocopherols, Vitamin C) © Genoma International All rights reserved. 57

19 Sources of Selenium‐Rich Foods Food Amount (Serving) Selenium (ug) Brazil Nuts (Selenium rich soil) 1 oz (6 kernels) 543.5 Tuna (Yellowfin, cooked, dry heat) 3 oz 92.0 Noodles, egg, cooked, enriched 1 cup 38.2 Beef, lean, cooked, grilled 3 oz 30.6 Chicken, light meat, cooked 3 oz 25.8 Brown Rice, long grain, cooked 1 cup 19.1 Sunflower seed (dried) ¼ cup 18.5

Modification to Chrissi’s Diet & Nutritional Supplements Based on DNA Data Before DNA testing After DNA Testing Diet Ham Limit to only 1 serving per month Cheese Eliminate White Bread Eliminate and substitute with gluten free bread; limit to 3 to 4 times per week Eggs Egg white only Chicken White meat, without the skin Salad Colorful and fresh, 4‐6 servings per day Veggies Colorful and fresh, 4‐6 servings per day Yogurt Eliminate Beef Lean beef only Other proteins Beans, Protein Powders, Tofu, Nuts and Seeds Fruits Colorful and fresh, low sugar fruits Nutritional NONE EPA/DHA, Ubiquinol, Multiple Vitamin‐Mineral (50% Supplements Vitamin A Ester and 50% Carotenoids, alpha Tocopherols, Vitamin C) © Genoma International All rights reserved. 59

Case History: Glutathione (GSH), Selenium and GPx1 and DIO2 Genotype

Analyte Tissue Result Reference Range GSH Whole Blood 458 > or equal to 669 (micromole/L) Selenium RBC 0.22 0.25 to 0.76 mcg/g

Gene High Moderate Nutritional Intervention/ Impact Impact Lifestyle Choices/ Action Steps GPx1 X Increase intake of foods rich in Selenium Supplement with Selenomethionine DIO2 X Thyroid Panel

20 Modification to Chrissi’s Diet & Nutritional Supplements Based on DNA Data Before DNA testing After DNA Testing Diet Ham Limit to only 1 serving per month Cheese Eliminate White Bread Eliminate and substitute with gluten free bread; limit to 3 to 4 times per week Eggs Egg white only Chicken White meat, without the skin Salad Colorful and fresh, 4‐6 servings per day Veggies Colorful and fresh, 4‐6 servings per day Yogurt Eliminate Beef Lean beef only Other proteins Beans, Protein Powders, Tofu, Nuts and Seeds Fruits Colorful and fresh, low sugar fruits Nutritional NONE EPA/DHA, Ubiquinol, Multiple Vitamin‐Mineral (50% Supplements Vitamin A Ester and 50% Carotenoids, alpha Tocopherols, Vitamin C, selenomethionine) © Genoma International All rights reserved. 61

Take Aways: Monogenic Gene SNPs and Nutrient Requirements and Homeostasis

• Single Gene Variants Can Influence Nutrient Requirements. • Genomic Testing Can Personalize Single Nutrient Requirements.

Take Aways: Monogenic Gene SNPs and Nutrient Requirements and Homeostasis

21 Take Aways: Monogenic Gene SNPs and Nutrient Requirements and Homeostasis

• Single Gene Variants Can Influence Nutrient Requirements. • Genomic Testing Can Personalize Single Nutrient Requirements. • Single Gene Variants Can Lead to Nutritional Paradoxes. • Genomic Testing Can Resolve Nutritional Paradoxes. • Genomic Testing Can Provide DNA‐Directed Interventions to Address the Dysregulation in Cellular, Biochemical and/or Metabolic Mechanism(s) Behind a Person’s Signs and Symptoms.

Genomic Testing Benefits Both Clinician and Patient

22 Module 2 Lesson 7

Time Stamp Title / Slide Subject 0:00 Genomic Medicine Training Program 0:13 Module 2 Session 7 0:58 Polygenic Model: Nutrient Requirements Based on DNA 2:55 Antioxidant Versus Reactive Oxygen Species Fulcrum Graphic 5:50 Dietary and Endogenous Antioxidants 7:47 Free Radical Theory of Disease and Aging 9:43 Free Radical Theory of Disease and Aging‐Use of Antioxidants 11:40 Lipid Bilayer and Antioxidant Protection 15:34 Antioxidant Cascade Electron Transport System 17:30 Case Histories 18:29 Gene SNPs in Antioxidant Cascade‐Chrissi 37:56:00 Antioxidant, Gene Impact and Antioxidant Cascade 42:48:00 Estrogen Metabolism Graphic 47:40:00 Estrogen Metabolism Genotypes‐Chrissi 49:37:00 Polygenic Patterns and Priorities‐Chrissi 55:30:00 Antioxidant Cascade‐ Case History # 2 Ted 57:24:00 Polygenic Patterns and Priorities‐Ted 1:05:11 Antioxidant Cascade‐Case History #3 Details Margaret 1:09:05 Polygenic Patterns and Priorities‐Margaret 1:20:45 Antioxidant Cascade Case History #4 Details Ali 1:23:41 Polygenic Patterns and Priorities‐Ali 1:28:32 Role of Gene Function Polygenic Matrix 1:34:32 Take Aways, Gene SNPs, Nutrient Requirment, Homeostasis 1:37:18 End Genomic Medicine Training Program

Module 2: Personalizing Nutrient Requirements and Resolving Nutritional Paradoxes Using Genomic Medicine

Joe Veltmann PhD FAAIM DCCN Roberta L. Kline MD FACOG

Nutrient Utilization: Personalizing Nutrient Requirement Based on DNA Polygenic

Carotenoids Vitamin C Vitamin E Polyphenols Selenium

SOD Catalase GPx NFE2L2 HMOX1 Ubiquinol

Graphic adapted from: Da Costa L, et al., (2012). Progress in Molecular Biology and Translational Science Vol 108: 180‐199

Free Radical Theory of Disease and Aging

• Formally proposed by Denham Harman in 1956. • Postulates: aging and degenerative diseases associated with cumulative oxidative damage to cells by free radicals produced during aerobic respiration. • Free radicals are atoms and molecules with single unpaired electrons. Graphic source: http://www.biotek.com/resources/articles/reactive‐oxygen‐species.html © Genoma International All rights reserved. 6

2 Free Radical Theory of Disease and Aging

• Free radicals are unstable and highly reactive. • Steal electrons from nearby molecules to gain stability. • Disrupt cellular and metabolic processes. • Reactive oxygen species (ROS) not only cause molecular damage to cells but also act as modulators of physiological processes (ie. adaptation to physical exercise). • Over expression of antioxidant enzymes increases life‐span. • Antioxidant supplementation can lower incidence of many age‐ associated disease.

Function of Antioxidants (Cellular Membrane)

Water and fat soluble antioxidants

CoQ10 = Ubiquinone CoQH2 = Ubiquinol

Antioxidant Cascade‐Electron Transport System

CoQ10 recycles β‐Carotene, Polyphenols Lipoic Acid Polyphenols Vit E recycles Vit C recycles Vit C recycles GSH

Disarms Free Radicals Recycles Vit E Recycles Vit C

3 Case Histories

Multiple Gene SNPs in Antioxidant Cascade: Gene‐Function Matrix (Chrissi) Gene High Moderate Metabolic Consequence in Impact Impact Antioxidant Cascade BCMO1 X Impairs Ability to Recycle Vitamin E. α-TTP X Free Radicals Not Neutralized. SLC23A1 X Recycling Vitamin E Reduced. GPx X Vitamin C Recycling Reduced. NQO1 X Vitamin E Recycling Compromised. Total 2 3

Gene Variant Summary Anti‐oxidant Cascade‐Chrissi

Antioxidant Gene Impact CoQ10 NQO1 High Vitamin A/ Carotenoid BCMO1 Moderate Vitamin E Alpha TTP High Vitamin C SLC23A Moderate Se/ Glutathione Px GPx Moderate

Estrogen Metabolism Genotypes: Chrissi

Detox Gene High Moderate Impact Impact

Phase I CYP1B1 X

Phase II COMT X

GSTM1 Absent

CAT X

GPx X

SOD2 X

Polygenic Patterns & Priorities‐Chrissi • Patterns: – Three genes (NQO1, alpha TTP, GSTM1) have high impact on nutrient utilization and antioxidant cascade. – SLC23A1, BCMO1 and GPx have moderate impact on nutrient utilization and antioxidant cascade. • Priorities: – Supplement with ubiquinol: supports recycling of vitamin E in the lipid bilayer; first antioxidant in the electron chain. – Supplement with alpha tocopherol: supports the recycling of ascorbate and neutralizing free radicals in the both the lipid bilayer and electron transport chain. – Higher Vitamin C requirement (gene SNPs on SLC23A1 and GSTM1) since Estrogen Metabolism compromised. – Higher Vitamin A/ Carotenoids (BCMO1)and Selenium (GPx) requirement needed. – Comprehensive Thyroid Panel Needed.

5 Antioxidant Cascade: Case History # 2 (Ted)

• Ted: Caucasian, Male • Age: 81 Occupation: Retired Antioxidant Gene Impact CoQ10 NQO1 Low/ No • Heath Issues: Vitamin A/ Carotenoid BCMO1 Moderate – Prostate Cancer (Prostatectomy) Vitamin E Alpha TTP Low/ No – Osteopenia Vitamin C SLC23A1 Low/ No – Rheumatoid arthritis Glutathione Px GPx Moderate – Hypercholesterolemia • Medications: None

Polygenic Patterns & Priorities‐Ted • Patterns: – One gene (GSTM1) has high impact on nutrient utilization (Vitamin C) associated with estrogen metabolism. – Three genes (BCMO1, GPx, FADS1) have moderate impact on nutrient utilization, antioxidant cascade and pro‐inflammatory response. • Priorities: – Supplement with extra vitamin C to support estrogen metabolism. – Supplement with extra vitamin A and carotenoids to support electron transport chain and neutralizing free radicals (lipid bilayer). – Supplement with DIM to upregulate CYP1A1, re‐route estrogen metabolites. – Supplement with extra selenium to support GPx pathway. – Supplement with EPA/DHA to support anti‐ inflammatory process, quench free radicals, down regulate IL‐1B.

Anti‐oxidant Cascade: Case History #3 (Margaret)

• Margaret: Caucasian, Female • Age: 65 Occupation: Retired Antioxidant Gene Impact • Health issues: CoQ10 NQO1 Low/ No – Hypercholesterolemia Vitamin A/ Carotenoid BCMO1 Low/ No – Elevated TGs Vitamin E Alpha TTP High – Menopausal Symptoms Vitamin C SLC23A1 Low/ No – Osteopenia Glutathione Px GPx Moderate – Weight Gain • Medications: – Bio‐identical hormones

6 Polygenic Patterns & Priorities‐Margaret

• Patterns – One gene (alpha TTP) has high impact on nutrient utilization and antioxidant cascade. – One gene (CYP1B1*4) has high impact on estrogen metabolism. – Several genes encoding for protein associated with estrogen metabolism have moderate impact on this biochemical pathway. • Priorities – Supplement with alpha tocopherol to support recycling of ascorbate and neutralizing free radicals in both lipid bilayer and electron transport chain. – Supplement with DIM to support re‐routing of estrone to 2‐OH estrone rather than 4‐OH estrone. – Higher selenium requirement to support GPx function. – Supplement with SamE to support COMT function. – Comprehensive Thyroid Panel.

Anti‐oxidant Cascade: Case History #4 (Ali)

• Ali: Caucasian, female • Age: 24 Antioxidant Gene Impact • Occupation: Retail Sales CoQ10 NQO1 Moderate • Health Issues: None Vitamin A/ Carotenoid BCMO1 Moderate • Medications: None Vitamin E Alpha TTP Moderate • Dietary Preference: Vegan Vitamin C SLC23A1 Low/ No Glutathione Px GPx Moderate

Polygenic Patterns & Priorities‐Ali

• Patterns – Three genes (GSTM1, SULT1A1 and FADS2) have high impact on nutrient utilization. – Multiple genes in estrogen metabolism, antioxidant cascade and free radical quenching in the lipid bilayer have moderate impact. • Priorities – Supplement with extra vitamin C, alpha tocopherol, CoQ10, vitamin A, selenium, SamE to insure functionality of estrogen metabolism, antioxidant cascade and support neutralizing free radicals in the cell membrane. – Supplement with EPA/DHA (algae) to support omega 3 fatty acid pathway and anti‐inflammatory response.

7 Gene‐Function Polygenic Matrix

• Quantifies and Qualifies the Impact of Each Gene SNP on the Biological System. • Prioritizes Nutrigenomic Interventions for Healthcare Professional. • Identifies Biomarkers to Evaluate Nutrigenomic Interventions. • Helps Healthcare Professional Appreciate the Interconnectedness of One Biological System with Another: – Endogenous Free Radical Quenching, Pro‐inflammatory Response, Estrogen Metabolism, Thyroid Function. – Role of Nutrient Co‐factors and Dietary Bioactives in Functionality of Biological Systems.

Take Aways: Polygenic Gene SNPs and Nutrient Requirements and Homeostasis

• Polygenic Gene SNPs Can Influence Nutrient Requirements Across Multiple Biological Systems. • Genomic Testing Can Identify Polygenic SNPs. • A Gene‐Function Polygenic Matrix Utilizing Gene SNPs Crossed with its Impact on a Biochemical Processes or Metabolic Pathways Can Personalize Nutrient Requirements More Quickly and Effectively than a Monogenic Model.

8 Module 2 Lesson 8

Time Stamp Title / Slide Subject 0:00 Genomic Medicine Training Program 0:15 Module 2 Session 8 0:21 Vitamin D and Calcium Requirements in Era of Genomic Medicine 1:33 Learning Objectives 2:27 Learning Objectives #2 3:33 Factors Affecting Calcium Absorption 4:38 Role of Calcium in the Body 6:24 Classical Metabolic Functions Linked to Active Vitamin D Molecule 7:21 Metabolic Functions Linked to Active Vitamin D in Biological Systems 8:26 Non‐classical Metabolic Functions of Vitamin D‐Immunity 8:52 Non‐classical Metabolic Functions of Vitamin D‐Cell Growth Regulator 9:41 Calcium Homeostasis and Vitamin D‐ A Review 10:57 Hormones in Calcium and Phosphorus Metabolism 12:56 Graphic Calcium Homeostasis 15:30 Calcium Sensing Receptors in Calcium Homeostasis 17:45 Graphic CaSR Gene and Calcium Homeostasis 20:00 Impact of CaSR Gene SNPs 21:34 Genes Linked to Calcium Homeostasis and Vitamin D Metabolism 25:20:00 Graphic Vitamin D Absorption and Utilization‐ A Review 27:57:00 Graphic Genes Associated with Vitamin D Metabolism 28:37:00 Gene Function Matrix‐Vitamin D Metabolism 31:40:00 Case Report‐Chrissi 31:48:00 Case Report Details Chrissi 32:10:00 Case Report Health Issues Chrissi 33:35:00 Genomic Test Results for Chrissi Bone Formation 36:25:00 Gene, Protein, Biochem Pathway, Metabolic Consequence, Action Step for High/Mod Impact Genes 49:05:00 Dietary Sources of Vitamin D 51:05:00 Calcium Absorption (Requirement) with VDR Gene SNP‐Post Menopausal 53:50:00 Calcium Absorption (Requirement) with VDR Gene SNP‐Pre‐menopausal 54:48:00 Calcium Absorption (Requirement) with VDR Gene SNP‐Short Term Energy Restriction Obese Women 56:17:00 Nutritional Paradox Calcium and Vitamin D 57:03:00 Nutritional Paradox Calcium and Vitamin D‐ VDR Bsml MM Genotype‐Post Menopausal Women 58:02:00 Nutritional Paradox Calcium and Vitamin D‐ VDR Bsml MM Genotype‐Pre‐Menopausal Women 59:00:00 Graphic Origin of Osteoblasts and Osteoclasts 1:01:19 Case Report‐Chrissi 1:01:29 High/Moderate Impact Gene SNPs Associated with Bone Degradation‐Other Biological Systems 1:03:23 High/Moderate Impact Gene SNPs Associated with Bone Formation‐Vitamin D Metabolism 1:04:06 Gene Function Matrix‐Bone Health (Formation and Degradation) Chrissi 1:04:43 Cytokines on Osteoclastogenesis 1:05:06 Gene Function Matrix‐Bone Health (Formation and Degradation) Chrissi 1:05:17 Cytokines on Osteoclastogenesis 1:05:36 Gene, Protein, Biochem Pathway, Metabolic Consequence, Action Step for High/Mod Impact Genes‐Other Biological Systems 1:08:52 Gene, Protein, Biochem Pathway, Metabolic Consequence, Action Step for High/Mod Impact Genes‐Vitamin D Metabolism 1:09:21 Graphic Vitamin D Metabolism 1:10:22 Dietary Changes Chrissi 1:11:32 Lifestyle/Nutritional Supplement Changes 1:11:54 Take Aways for Calcium and Vitamin D Requirements in Era of Genomic Medicine 1:16:47 End Genomic Medicine Training Program

Module 2: Personalizing Nutrient Requirements and Resolving Nutritional Paradoxes Using Genomic Medicine

Joe Veltmann PhD FAAIM DCCN Roberta L. Kline MD FACOG

Vitamin D and Calcium Requirements in the Era of Genomic Medicine

1 Learning Objectives

• Role of Calcium and Vitamin D in the Body • Genes linked to Calcium Homeostasis and Vitamin D Metabolism • Gene Variants Influence Calcium Homeostasis and Vitamin D Metabolism (Bone Formation) • Gene Variants Influence Inflammation, Oxidative Stress and Thyroid Function (Bone Degradation)

Learning Objectives

• Case History (Chrissi): – DNA‐Directed Nutrigenomic Action Steps • Nutritional Paradoxes • Take‐Aways

Factors Associated with Calcium Absorption

• Vitamin D: • eggs, butter, fatty fish, liver, fortified foods‐milk, orange juice, cereal • Vitamin C • Vitamins E and K • Magnesium • Boron • Exercise

2 Calcium: Roles in the Body

• Vital element needed to prevent osteopenia/osteoporosis. • Calcium universal intracellular messenger (second messenger): – gene transcription – muscle contraction – cell proliferation – neurotransmitter release

Classical Metabolic Functions Associated with Active Vitamin D Molecule • Biological Systems: – Intestine: Increases Ca and P absorption. – GI Tract: Increases Smooth Muscle Proliferation (Less Inflammation). – Bone: Increases Mineralization & Osteoclastic Differentiation. – PTH: Decreases Hormone Synthesis and Release. – Kidney: Decreases Renin Expression and LV Hypertension. – Pancreas: Increases Insulin Secretion.

3 Non‐Classical Metabolic Functions Associated with Active Vitamin D Molecule • Biological Systems: – Immunity: • Promotes Innate Immunity/Inhibits Adaptive Immunity. –Increases anti‐microbial activity.

Non‐Classical Metabolic Functions Associated with Active Vitamin D Molecule • Biological Systems: – Immunity: • Promotes Innate Immunity/Inhibits Adaptive Immunity. – Increases anti‐microbial activity. – Cell Growth Regulator (Tumor Microenvironment): • Promotes Apoptosis. • Inhibits inflammation (Colon and GIT). • Induces Immune Cell Differentiation.

Calcium Homeostasis & Vitamin D Metabolism: A Review • Metabolic Players – Dietary Calcium Calcitonin CaSR (Ca Sensing Receptors) – Dietary Vitamin D Active Vitamin D Parathyroid Hormone – Ionized Calcium (Ca+2) • Organs – Parathyroid Gland Thyroid Gland Bone – Liver Intestine Kidney

4 Hormones in Calcium and Phosphate Metabolism

Bone: Calcium and Phosphate Metabolism ‐Skeleton (storage), Gut (uptake), Kidney (secretion) Parathyroid Hormone (PTH): Increase serum Ca • Stimulates bone resorption (osteoclasts) • Stimulates renal absorption of Ca • Stimulates synthesis of active Vitamin D 1,25 OH2 Vitamin D: Increase Ca and P • Stimulates Ca and P uptake from food in gut Calcitonin: Decrease serum Ca • Counteracts all actions of PTH • Inhibits bone resorption (osteoclasts) and renal reabsorption Fibroblast Growth Factor (FGF23): Decrease P • Reduces renal absorption of P • Inhibits synthesis of 1,25 OH2 Vitamin D Image source: https://www.slideshare.net/obanbrahma/fat‐ soluble‐vitamins‐30768043

Calcium Sensing Receptors (CaSR) and Calcium Homeostasis • Parathyroid gland: modulates PTH secretion • Kidneys: modifies calcium reabsorption • Bone: differentiation of osteoblasts and osteoclasts • Skin: promotes keratinocyte differentiation • Stomach: modulates acid secretion • Colon – Absorption, secretion, motility, immunity • Small Intestine: – Gut permeability and inflammation • Lungs: pulmonary function

5 Gene/Protein Linked to Calcium Homeostasis

• Ca Sensing Receptors(CaSR): – senses small changes in circulating calcium concentration and couples this information to intracellular signaling pathways – G protein‐coupled receptor • Regulates PTH output in response to subtle fluctuations in ionic calcium in real time.

Image source: https://basicmedicalkey.com/hormonal‐regulation‐of‐calcium‐and‐phosphate‐metabolism/; adapted from Porterfield SP, White BA: Endocrine Physiology, 3rd ed. Philadelphia, Mosby, 2007. © Genoma International All rights reserved. 16

Impact of CaSR Gene Polymorphisms

• Impact of Gene Polymorphisms: – Hyperparathyroidism – Hypercalcemia – Pulmonary Hypertension – Modifies Colonic Processes – Increases Gut Inflammation – Modifies Nutrient Absorption – Increases Kidney Stone Risk Image source: Pua F, Chenb N, et al. (2016). Food Science and Human Wellness 5: 8‐16.

Genes Linked to Calcium Homeostasis and Vitamin D Metabolism • Vitamin D Metabolism: 1,25 dihydroxy vitamin D – Reductase Enzyme: Cholesterol Molecule to Cholecalciferol (DHCR7) – Hydroxylation Enzymes: • 25‐OH lase in Liver (CYP2R1) • 1 alpha‐OH lase in Kidney (CYP27B1) – Transport Protein: Vitamin D Binding Protein (DBP or GC) – Nuclear Receptors: VDR (Fokl, Taql and Bsml)

6 Vitamin D Absorption and Utilization

Absorption of dietary Vitamin D similar to absorption process of dietary β-carotene & vitamin E.

Cholesterol

GC DHCR7

CYP2R1

CYP27B1

VDR Taql, VDR Fokl, VDR Bsml

Gene Protein Encoded by Biochemical Metabolic Gene Pathway Consequence of Gene SNP DHCR7 7‐dehydrocholesterol Conversion of Cholesterol Reduced enzyme reductase to Cholecalciferol (Skin) activity. CYP2R1 Vitamin D 25‐hydrolase Cholecalciferol to 25‐OH Reduced enzyme Vit D (Liver) activity. GC Vitamin D Binding Protein Carrier Protein Functionality of (DBP) (Liver) transport protein (Kidney) impaired. CYP27B1 Vitamin D 1‐hydroxylase 25‐OH Vit D to 1, 25‐diOH Reduced enzyme Vit D (Kidney) activity. VDR Vitamin D Receptors Inside nucleus Functionality of Bsml RXR (Vitamin A) receptor site Fokl Vit D Response compromised. Taql Element

7 Case History: Chrissi

Case History Details: Chrissi • Female, 45 years of age, resident of the Mid‐West • Occupation: Student/Stay‐at‐home Mom (2 children) • Height: 5’ 9; Weight: 190; Calculated BMI: 28; Polish Heritage • Diet: Ham, Cheese, Bread, Chicken, Salad, Veggies, Yogurt, Beef • Exercise: Biking, Running (1 hr/day); Runs ½ Marathons

Case History Details: Chrissi • Health Issues: – Rheumatoid Arthritis Yo‐yo Dieting—Weight Gain – Ruptured Achilles Tendon Poor Recovery from Exercise – Osteopenia Bi‐polar Disorder – Extreme Fatigue Stomach Issues (Inflammation – Insomnia • Meds: Lexapro (10 mg); Vyvance (20 mg) • Nutritional Supplements: None

8 Genomic Test Results: Chrissi (Bone Formation)

Gene Biological Function High Moderate Low/No Impact Impact Impact CYP2R1 25‐OH lase X VDR Taql Nuclear Receptor X COL1A1 “Rebar” X VDR‐Bsml Nuclear Receptor X VDR‐Fokl Nuclear Receptor X DHCR7 Cholesterol to 7‐ X dehydrocholesterol GC Vitamin D Transport X

Gene Protein Encoded by Biochemical Metabolic Nutritional Gene Pathway Consequence Action Step of Gene SNP COL1A1 Collagen‐1‐alpha‐1 Structural Integrity of Low BMD/Increase Hydroxyapatite bone Fracture Risk Supplementation CYP2R1 Vitamin D 25‐hydrolase Cholecalciferol to 25‐OH Reduced enzyme Vitamin D Vit D (Liver) activity Requirement Increased VDR Vitamin D Receptor Inside nucleus Functionality of Vitamin D Taql RXR (Vitamin A) receptor site Requirement Vit D Response impaired Increased Element VDR Vitamin D Receptor Inside nucleus Functionality of Vitamin D Bsml RXR (Vitamin A) linked to receptor site Requirement Vit D Response impaired Increased Element VDR Vitamin D Receptor Inside nucleus Functionality of Vitamin D Fokl RXR (Vitamin A) linked to receptor site Requirement Vit D Response impaired Increased Element © Genoma International All rights reserved. 26

Dietary Sources of Vitamin D

Cod liver oil Sardines Plant‐Based: Salmon Raw mushrooms‐ Shitake, Maitake, Swordfish Chanterelle Trout Egg yolk

9 Calcium Absorption and VDR Gene SNPs: Proof of Concept • VDR Bsml MM genotype (High Impact) associated with lower Ca absorption compared to Mm and mm genotypes (Low or No Imact) during very low calcium intake in 60 postmenopausal women given Ca 1500 mg/day for two weeks and <300 mg/day for two weeks. • Dawson‐Hughes B et al., (1995). J Clin Endo Met 80: 3657‐3661.

Calcium Absorption and VDR Gene SNPs: Proof of Concept • VDR Bsml MM genotype (High Impact) associated with lower Ca absorption compared to Mm and mm genotypes (Low or No Impact)during very low calcium intake in 60 postmenopausal women given Ca 1500 mg/day for two weeks and <300 mg/day for two weeks. • Dawson‐Hughes B et al., (1995). J Clin Endo Met 80: 3657‐3661. • VDR Bsml mm genotype (Low or No Impact) linked with increased Ca absorption in premenopausal women consuming Ca 1000 mg/day. • Wishart et al., (1997). Am J Clin Nutr 65: 798‐802. • Reduced calcium absorption during short term energy restriction among 168 obese or overweight women with VDR Bsml MM genotype (High Impact) compared to combined Mm/mm genotypes (Low or No Impact). • Chang B et al., (2015). Bone 81: 138‐144.

10 Nutritional Paradoxes: Calcium and Vitamin D

• Using 25‐OH vitamin D serum levels to determine vitamin D requirement may not be adequate if gene SNPs on VDR. – May need to also assess 1,25‐dihydroxyvitamin D level.

Nutritional Paradoxes: Calcium and Vitamin D

• Using 25‐OH vitamin D serum levels to determine vitamin D requirement may not be adequate if gene SNPs on VDR. – May need to also assess 1,25‐dihydroxyvitamin D level. • VDR Bsml MM genotype can decrease Ca absorption in post menopausal women: – When dietary calcium levels are low. – When caloric content of diet is low (weight loss programs).

Nutritional Paradoxes: Calcium and Vitamin D

• Using 25‐OH vitamin D serum levels to determine vitamin D requirement may not be adequate if gene SNPs on VDR. – May need to also assess 1,25‐dihydroxyvitamin D level. • VDR Bsml polymorphisms can decrease Ca absorption in post menopausal women: – When dietary calcium levels are low. – When caloric content of diet is low (weight loss programs). • VDR Bsml MM genotype gene variants can decrease Ca absorption among pre‐menopausal women given adequate dietary Ca (1000 mg) but energy restriction (weight loss program, intermittent feeding).

11 Image source: Weilbaecher KM, Guise TA, et al. (2011). Nat Rev Cancer 11:411‐425.

Case History: Chrissi

Genomic Test Results: Chrissi (Bone Degradation) Gene Biological Function High Moderate Low/No Impact Impact Impact DIO2 Thyroid Function X GSTM1/GSTT1 Oxidative Stress Absent TNF‐alpha Inflammation X CRP Inflammation X IL‐1B Inflammation X IL‐6 Inflammation X IL‐6R Receptor Site for IL‐6 X IL‐17A Inflammation X APOE Vitamin D Transport X MTHFR Transmethylation Cycle X

12 Genomic Test Results: Chrissi (Bone Formation)

Gene‐Function Matrix‐Bone Health (Chrissi)

Effect of Cytokines on Osteoclastogenesis

Graphic source: http://www.biochemia‐medica.com/2013/23/43

13 Gene‐Function Matrix‐Bone Health (Chrissi)

Effect of Cytokines on Osteoclastogenesis

Graphic source: http://www.biochemia‐medica.com/2013/23/43

Gene Protein Encoded Biochemical Metabolic Consequence Nutritional by Gene Pathway of Gene SNP Action Step

DIO2 5’ Deiodinase Conversion of T4 to T3 in Low BMD Guggul peripheral tissues (Indian Myrrh) GSTM1/ Glutathione Transferase Oxidative stress Enzymes are absent to Boost antioxidants and GSTT1 (Absent) quench free radicals dietary bioactives TNF‐α Tumor Necrosis Factor‐ Pro‐inflammatory cytokine Low BMD; risk for bone Dietary bioactives alpha fracture down regulate NF‐ƙB CRP C‐reactive Protein Systemic Inflammation Bone Loss Reduce pro‐ induced by IL‐6 inflammatory IL‐1B Interleukin 1B Pro‐inflammatory cytokine Low BMD EPA/DHA IL‐6/ Interleukin ‐6/ Pro‐inflammatory cytokine/ Increased bone Beta‐sitosterol/stress IL‐6R Interleukin 6 Receptor Receptor sensitivity to IL‐6 degradation in response management increased to all types of stress IL‐17A Interleukin 17A Pro‐inflammatory cytokine Increased bone Food sensitivity testing reabsorption

14 Gene Protein Encoded by Biochemical Metabolic Consequence of Nutritional Gene Pathway Gene SNP Action Step

COL1A1 Collagen‐1‐alpha‐1 Structural Integrity of bone Low BMD/Increase Fracture Hydroxyapatite Risk Supplementation

CYP2R1 Vitamin D 25‐hydrolase Cholecalciferol to 25‐OH Reduced enzyme activity Vitamin D Vit D (Liver) Requirement Increased VDR Vitamin D Receptor Inside nucleus Functionality of receptor site Vitamin D Taql RXR (Vitamin A) impaired Requirement Vit D Response Increased Element VDR Vitamin D Receptor Inside nucleus Functionality of receptor site Vitamin D Bsml RXR (Vitamin A) linked to impaired Requirement Vit D Response Increased Element VDR Vitamin D Receptor Inside nucleus Functionality of receptor site Vitamin D Fokl RXR (Vitamin A) linked to impaired Requirement Vit D Response Increased Element

Image source: http://www.mdpi.com/2072‐6694/3/1/213/htm

Dietary Changes‐Chrissi

• Diet: Ham, Cheese, Bread, • Colorful low‐sugar fruits rich in Chicken, Salad, Veggies, Yogurt, Vitamin C. Beef. • Vegetables rich in Carotenoids. • Vitamin E rich foods. • Brazil nuts. • Vitamin D rich foods. • Calcium rich products. • Limit intake of Meat, Eggs and Dairy Products to minimize Arachidonic Acid Production.

15 Lifestyle Changes, Nutritional Supplements‐Chrissi

• Stress management: • Ubiquinol (100 mg) – Meditation • EPA/DHA (1‐2 g/day) – Yoga • Multi‐Vitamin (Beta carotene, – Journaling Vitamin A, alpha Tocopherols, – Communicate using feelings words to Vitamin C; Selenomethionine) promote healthier relationships with husband and sons. • Vitamin D emulsion (1,000 IU/day) • Calcium/ Magnesium (1,000 mg/500 mg) • Beta‐sitosterol (100 mg) • Guggul (250 mg, 2 x/day) • DimPro+ (175 mg, 2x/day)

Take‐Aways: Calcium and Vitamin D Metabolism in the Era of Genomic Medicine

• Multiple factors influence calcium and vitamin D metabolism including gene polymorphisms. • Gene Polymorphisms Impact Functional Biomarkers and Can Create Nutritional Paradoxes: • 25‐OH Vitamin D levels versus 1,25 Dihydroxy Vitamin D. • Gene‐Function Matrix Identifies Patterns Disrupting Calcium and Vitamin D Homeostasis. • Prioritizes Nutrigenomic Interventions.

16 Module 2 Lesson 9

Time Stamp Title / Slide Subject 0:00 Genomic Medicine Training Program 0:29 Module 2 Session 9 0:37 B‐vitamin Requirement in Era of Genomic Medicine 2:17 B‐vitamin Requirement Polygenic Model 3:21 Learning Objectives 4:37 B‐vitamins in Biological Systems 6:10 Digestion and Absorption of B‐vitamins 8:17 Digestion and Absorption of Vitamin B12 9:35 Digestion and Absorption of B‐12 Stomach to Ileum 11:02 Digestion and Absorption of B‐12 Ilieum to Liver 12:31 Digestion and Absorption of B‐12 Stomach to Ileum to Liver 13:24 Digestion and Absorption of B‐12 Active and Inactive 14:02 Vitamin B12 Deficiency 17:24 Absorption and Digestion B‐12 Graphic 22:20 Transmethyation/Transsulfuration Pathway‐B‐vitamin Dependent 22:35 Methylation Biochemistry Graphic 22:57 Transmethyation/Transsulfuration (TM/TS) Pathway‐B‐vitamin Dependent 23:05 Methylation Biochemistry Graphic 25:23:00 Simple Version of TM/TS Graphic 26:40:00 Complex Version of TM/TS Graphic 32:07:00 Genes, Enzymes, Proteins, Linked to B‐vitamins in TM/TS 34:27:00 Complex Version of TM/TS Graphic 34:39:00 Genes, Enzymes, Proteins, Linked to B‐vitamins in TM/TS 34:47:00 Complex Version of TM/TS Graphic 35:01:00 Nutritionally Important Methyl Donors 35:37:00 Complex Version of TM/TS Graphic 36:45:00 Graphic TM/TS Co‐factors 37:56:00 Detailed TM/TS Pathway 43:00:00 Chrissi Case Report 44:02:00 Case History Details (Female, Occupation, Diet, Exercise) 44:58:00 Health Issues Associated with Chrissi, Meds, Supplements 45:42:00 High Impact Genomic Test Results‐Chrissi 48:41:00 Moderate Impact Genomic Test Results‐Chrissi 48:51:00 High Impact Genomic Test Results‐Chrissi 48:56:00 Moderate Impact Genomic Test Results‐Chrissi 49:24:00 Chrissi Genomic Test Results on Detailed TM/TS Graphic 52:10:00 High Impact Genomic Test Results‐Chrissi Matrix 52:43:00 Moderate Impact Genomic Test Results‐Chrissi Matrix 56:45:00 NBPF3 and MMAB and Vitamin B Requirement 58:41:00 Adenosyl Vitamin B12 and MMAB gene 59:32:00 Graphic Adenosyl Vitamin B12 and Krebs Cycle 59:52:00 Adenosyl Vitamin B12 and MMAB gene 1:00:12 High/Moderate Impact Nutritional Action Steps Chrissi 1:00:59 Vitamin B‐12 Digestion and Adenosyl Vitamin B‐12 1:02:49 Adenosyl Vitamin B12 and MMAB gene 1:04:00 High/Moderate Impact Nutritional Action Steps Chrissi 1:06:03 Gene Function Matrix Chrissi 1:06:32 Chrissi Genomic Test Results on Detailed TM/TS Graphic 1:07:58 Complex Version of TM/TS Graphic with Gene SNPs Chrissi 1:09:21 TM/TS Pathway Gene SNPs and Nutritional Requirements 1:10:29 TM/TS Pathway and GSH Graphic 1:12:01 TM/TS Pathway Free Radical Quenching 1:16:15 Detox Matrix and Graphic of Estrogen Metabolism Pathway 1:17:19 TM/TS Pathway and Free Radical Quenching GPx, GSTM1 1:17:55 Biochem Pathway, High/Mod SNPx, Nutritional Requirment Chrissi 1:18:17 Case History # 2 Ted Details 1:19:12 High/Moderate SNPs and Nutritional Action Steps Ted 1:20:13 Gene Function Matrix Ted 1:20:32 Detailed TM/TS Pathway with Gene SNPs‐ Ted 1:21:35 Complex TM/TS Pathway with Gene SNPs‐Ted 1:21:49 Biochemical Pathway, High/Mod SNPs, Nutritional Requirement Ted 1:21:57 TM/TS Pathway and Free Radical Quenching Ted 1:22:43 Detox Matrix and Graphic of Estrogen Metabolism Pathway Ted 1:24:30 Biochemical Pathway, High/Mod SNPs, Nutritional Requirement, Detox Ted 1:25:21 Margaret Case Report‐Details 1:25:27 Gene Function Matrix Margaret 1:26:16 Detailed TM/TS Pathway Margaret Results 1:26:41 Complex TM/TS Pathway with Gene SNPs‐Margaret 1:26:52 TM/TS Pathway, High/Moderate SNPs, Nutritional Requirement 1:27:07 TM/TS Pathway with Free Radical Quenching 1:27:22 Detox Matrix and Estrogen Metablism Graphic Margaret 1:28:07 TM/TS Pathway, Free Radical Quenching with GPx, GSTP1 Margaret 1:28:33 TM/TS, High/Mod Gene SNPs, Nutritional Requirements Margaret 1:28:45 Ali Case History Nutrient, Gene SNP and Action Step 1:29:18 TM/TS Pathway, High/Moderate SNPs, Nutritional Action Step Ali 1:29:47 Gene Function Matrix‐Ali 1:30:19 Detailed TM/TS Pathway with Gene SNPs‐ Ali 1:30:35 Complex TM/TS Pathway Ali Results, MMAB, FUT2, TCN2, NBPF3, SLC19A1 1:30:42 Biochemical Pathway, High/Mod Gene SNPs, Nutritional Requirements‐Ali 1:30:49 TM/TS Pathway with Free Radical Quenching‐Ali 1:31:09 Detox Matrix with Estrogen Metabolism Graphic Ali 1:31:54 TM/TS Pathway Free Radical Quenching Ali GPx, GSTM1 1:32:03 TM/TS Pathway with Detox Nutritional Requirements‐Ali 1:32:25 Take‐Aways Associated with TM/TS Pathway 1:33:04 Take‐Aways Associated with TM/TS Pathway Plus Free Radical Quenching 1:35:54 End Genomic Medicine Training Program

Module 2: Personalizing Nutrient Requirements and Resolving Nutritional Paradoxes Using Genomic Medicine

Joe Veltmann PhD FAAIM DCCN Roberta L. Kline MD FACOG

Module 2: B‐Complex Vitamin Requirements in the Era of Genomic Medicine

1 B‐Vitamin Requirements: Polygenic Model • Multiple Gene SNPs in a Metabolic Pathway Affect Biological Systems (Polygenic Model): • Antioxidant Cascade: Quench Free Radicals √ • Vitamin D Metabolism: Calcium & Vit D requirement √ • Transmethylation/Transsulfuration: – Nutrient Requirements for B2, B3, B6, B9, B12, Choline – Nutrient Requirements for Downstream Substrates Associated with Other Biological Systems (Detox, AA Requirements)

Learning Objectives

• Case History (Chrissi) – Genomic Test Results – Nutrigenomic Interventions • Additional Case Histories (Ted, Margaret, Ali). – Genomic Test Results – Gene Function Matrix – Biochemical Pathway Visuals – Nutrigenomic Interventions • Take‐aways

B‐Complex Vitamin Metabolism in Biological Systems

B‐complex vitamins (B2, B3, B6, B9, B12, Choline): – Cell metabolism: glycolysis – DNA synthesis: purines and pyrimidines – DNA repair of single or double strand breaks – Epigenetics (methyl groups) – Transmethylation/ Transsulfuration Pathway – Oxidized GSH‐Reduced GSH (redox homeostasis) – Mitochondrial function

2 Digestion and Absorption of B‐vitamins

• Water soluble Vitamins (B1, B2, B3, B6, and B9) bound to proteins in foods. • Stomach: low pH and pepsin degrade food proteins, releasing B‐Vitamins. – Small intestine: B‐vitamins absorbed by passive diffusion; picked up by capillaries in the mucosa and transported by the portal vein to liver and then whole body distribution. • Exception is Vitamin B12 that needs a carrier (intrinsic factor).

Digestion and Absorption of Vitamin B‐12: From the Stomach to the Ileum

• Vitamin B‐12 found in food products of animal origin. • Stomach: Pepsin and acid pH hydrolyze proteins, releasing Vit B‐12; free vitamin B‐12 binds to haptocorrin‐ protein (secreted by salivary glands and parietal cells). • Duodenum: Pancreatic proteases degrade haptocorrin‐ vitamin B‐12 complex, releasing vitamin B‐12, which then binds to Intrinsic Factor (secreted by the parietal cells of the stomach).

3 Digestion and Absorption of Vitamin B‐12: From the Stomach to the Ileum to Liver

• Vitamin B‐12 found in food products of animal origin. • Stomach: Pepsin and acid pH hydrolyze proteins, releasing Vit B‐ 12; free vitamin B‐12 binds to haptocorrin (secreted by salivary glands and parietal cells). • Duodenum: Pancreatic proteases degrade haptocorrin‐vitamin B‐12 complex, releasing vitamin B‐12, which then binds to Intrinsic Factor (secreted by the parietal cells of the stomach). • Distal ileum: Special receptors bind with B‐12‐Intrinsic Factor Complex—releasing B‐12 into the blood circulation bound to another protein—transcobalamin 2 or holotranscobalamin– and gets transported to the liver via the portal system. • TCN2 gene encodes for transcobalamin 2.

Digestion and Absorption of Vitamin B‐12: From the Stomach to the Ileum to Liver Vitamin B‐12

Intrinsic factor complex

Active B12

Digestion and Absorption of Vitamin B‐12: Active Versus Inactive Forms of B12

• 70 to 80% of Vitamin B‐12 bound in the blood to haptocorrin. • 20 to 30 % bound to transcobalamin 2 or holotranscobalamin. • Transcobalamin 2: only active fraction of B‐12 in the blood; delivered to all cells in the body via blood circulation.

4 Vitamin B12 (Cobalamin) Deficiency

• Inadequate intake, inadequate bioavailability or malabsorption. • Disruption of B12 transport in the blood, impaired cellular uptake, gene SNPs associated with metabolism (intracellular deficiency). • Signs of clinical B12 deficiency: hematological and neurological manifestations (relatively uncommon). • Subclinical B12 deficiency: 2.5% to 26 % of general population across all ages and genders, especially among vegans. • Biomarkers for B12 status: – circulating total B12 and transcobalamin‐bound B12. – abnormally increased levels of homocysteine and methylmalonic acid.

Transmethylation/ Tramssulfuration Pathway: Dependent on B‐vitamins • DNA Synthesis: purines and pyrimidines • Generate methyl groups for methylation reactions: – DNA/RNA (Epigenetics) – O‐methyltransferases (Catecholamines, Estrogen Detoxification) – Proteins (Arginine and Lysine Residues) – Methylcobalamin – Ubiquinone synthesis • Detoxification (Sulfation, Glutathione Transferase) • Amino Acid Synthesis (Homocysteine, Cysteine, Glutathione)

5 Image source: https://neuroendoimmune.wordpress.com/2014/01/14/genetic‐control‐of‐methylation‐can‐affect‐your‐health/

Image source: https://neuroendoimmune.wordpress.com/2014/01/14/genetic‐control‐of‐methylation‐can‐affect‐your‐health/

6 Transmethylation Transsulfuration Pathway Pathway

TCN2 MTR FUT2 B12 Choline B6

Choline MTRR Betaine Choline MTHFD MTHFD

B2 B6

Graphic adapted from: http://openi.nlm.nih.gov/detailedresult.php?img=3129576_1471‐2350‐12‐75‐1&req=4

Genes, Enzymes, Proteins Linked to B‐Vitamins in Transmethylation/ Transulfuration Pathway • MTHFR: Methylenetetrahydrofolate reductase (B2, B9) • MTRR: Methionine synthase reductase (B12) • MTR or MS: Methionine synthase (B12) • TCN2: Transcobalamin 2 (B12) • FUT2: Fucosyltransferase 2 (B12) • CβS: Cystathionine Beta Synthase (B6) • MTHFD: Methylenetetrahydrofolate dehydrogenase (Choline) • BHMT: Betaine hydroxymethyltransferase (TMG, Choline)

7 B6

TCN2 MTR FUT2 B12 Choline B6

Choline MTRR Betaine Choline MTHFD MTHFD

B2 B6

Graphic adapted from: http://openi.nlm.nih.gov/detailedresult.php?img=3129576_1471‐2350‐12‐75‐1&req=4

Nutritionally Important Methyl Donors

8 B6

TCN2 MTR FUT2 B12 Choline B6

Choline MTRR Betaine Choline MTHFD MTHFD

B2 B6

Graphic adapted from: http://openi.nlm.nih.gov/detailedresult.php?img=3129576_1471‐2350‐12‐75‐1&req=4

Vitamin B6 Vitamin B12 Cofactor for methyl Cofactor for Vitamin B3 Vitamin B9 (folate) transferase methionine synthase Methylation Methyl donor DNA synthesis

Vitamin B2 Helps recycle folate Choline Precursor to FAD Methyl donor / GlycineDNA repair Vitamin C Cofactor for Transmethylation/ methylation Serine/Glycine Transulfuration Methyl donor Zinc Cofactor for methylation Glutathione Selenium Methylation Cofactor for methionine SAMe DNA repair Magnesium Copper Cofactor for Cofactor for methylation methylation

FUT2

Choline

9 Case History Details: Chrissi

Case History Details: Chrissi • Health Issues: – Rheumatoid Arthritis Yo‐yo Dieting—Weight Gain – Ruptured Achilles Tendon Poor Recovery from Exercise – Osteopenia Bipolar Disorder – Extreme Fatigue Stomach Issues (Inflammation – Insomnia • Meds: Lexapro (10 mg); Vyvance (20 mg) • Nutritional Supplements: None

10 Transmethylation/Transsulfuration: Genomic Test Results (High Impact): Chrissi Gene Protein Biochemical Metabolic Encoded Pathway Consequence of Gene by Gene SNP MTR Methionine Synthase Converts homocysteine (Hcy) to Enzyme activity methionine (MET). decreased; less recycling of MET. MTRR Methionine Works with MTR to recycle Hcy Enzyme activity Synthetase molecule to MRT facilitated by decreased, compromising Reductase methylcobalamin, which donates a transsulfuration pathway. methyl group. SLC19A1 Solute carrier family Folate transport protein; important Decreased capacity to 19, member 1 regulator of intracellular folate transport folate; low concentrations. folate concentrations (intracellular).

Transmethylation/Transsulfuration: Genomic Test Results (Moderate Impact) for Chrissi Gene Protein Encoded Biochemical Metabolic Consequence of Gene SNP by Gene Pathway FUT2 Fucosyltransferase Absorption of B‐12, fucosylation of Increased enzyme activity; more intrinsic factor. bioavailable vitamin B‐12. TCN2 Transcobalamin 2 Binds dietary cobalamin to Decreased protein function, resulting in less transcobalamin, facilitating transport transport of vitamin B‐12 into cells. into cells. CβS Cystathionine Beta Catalyzes conversion of homocysteine Increased enzyme activity; more cysteine, Synthase plus serine to cystathionine. glutathione and taurine molecules produced. MTHFR Methylene‐ Transmethylation pathway; enzyme Reduced enzyme activity, leading to tetrahydrofolate critical in the conversion of MTHF into impaired transmethylation/ transulfuration (MTHF) reductase 5‐methyl folate. pathway; elevated Hcy. MTHFD1 Methylene‐ Catalyzes reversible reactions between Reduced enzyme activity, impairing the tetrahydrofolate 5,10‐ methylenetetrahydrofolate and transmethylation cycle and dehydrogenase 10‐ formyltetrahydrofolate. recycling of Hcy to Met in transsulfuration pathway. © Genoma International All rights reserved. 32

Transmethylation/Transsulfuration: Genomic Test Results (High Impact): Chrissi Gene Protein Biochemical Metabolic Encoded Pathway Consequence of Gene by Gene SNP MTR Methionine Synthase Converts homocysteine (Hcy) to Enzyme activity methionine (MET). decreased; less recycling of MET. MTRR Methionine Works with MTR to recycle Hcy Enzyme activity Synthetase molecule to MRT facilitated by decreased, compromising Reductase methylcobalamin, which donates a transsulfuration pathway. methyl group. SLC19A1 Solute carrier family Folate transport protein; important Decreased capacity to 19, member 1 regulator of intracellular folate transport folate; low concentrations. folate concentrations (intracellular).

11 Transmethylation/Transsulfuration: Genomic Test Results (Moderate Impact) for Chrissi Gene Protein Encoded Biochemical Metabolic Consequence of Gene SNP by Gene Pathway FUT2 Fucosyltransferase Absorption of B‐12, fucosylation of Increased enzyme activity; more intrinsic factor. bioavailable vitamin B‐12. TCN2 Transcobalamin 2 Binds dietary cobalamin to Decreased protein function, resulting in less transcobalamin, facilitating transport transport of vitamin B‐12 into cells. into cells. CβS Cystathionine Beta Catalyzes conversion of homocysteine Increased enzyme activity; more cysteine, Synthase plus serine to cystathionine. glutathione and taurine molecules produced. MTHFR Methylene‐ Transmethylation pathway; enzyme Reduced enzyme activity, leading to tetrahydrofolate critical in the conversion of MTHF into impaired transmethylation/ transulfuration (MTHF) reductase 5‐methyl folate. pathway; elevated Hcy. MTHFD1 Methylene‐ Catalyzes reversible reactions between Reduced enzyme activity, impairing the tetrahydrofolate 5,10‐ methylenetetrahydrofolate and transmethylation cycle and dehydrogenase 10‐ formyltetrahydrofolate. recycling of Hcy to Met in transsulfuration pathway. © Genoma International All rights reserved. 34

Chrissi

FUT2

NBPF3 = B6 absorption

MMAB = cobalamin

12 Transmethylation/Transsulfuration: Genomic Test Results (Moderate Impact) for Chrissi Gene Protein Encoded Biochemical Metabolic Consequence of Gene SNP by Gene Pathway FUT2 Fucosyltransferase Absorption of B‐12, fucosylation of Increased enzyme activity; more intrinsic factor. bioavailable vitamin B‐12. TCN2 Transcobalamin 2 Binds dietary cobalamin to Decreased protein function, resulting in less transcobalamin, facilitating transport transport of vitamin B‐12 into cells. into cells. CβS Cystathionine Beta Catalyzes conversion of homocysteine Increased enzyme activity; more cysteine, Synthase plus serine to cystathionine. glutathione and taurine molecules produced. MTHFR Methylene‐ Transmethylation pathway; enzyme Reduced enzyme activity, leading to tetrahydrofolate critical in the conversion of MTHF into impaired transmethylation/ transulfuration (MTHF) reductase 5‐methyl folate. pathway; elevated Hcy. MTHFD1 Methylene‐ Catalyzes reversible reactions between Reduced enzyme activity, impairing the tetrahydrofolate 5,10‐ methylenetetrahydrofolate and transmethylation cycle and dehydrogenase 10‐ formyltetrahydrofolate. recycling of Hcy to Met in transsulfuration pathway. © Genoma International All rights reserved. 37

Transmethylation/Transsulfuration: Nutrient Requirements for Chrissi Gene Protein Biochemical Metabolic (High Encoded Pathway Consequence of Gene Impact) by Gene SNP NBPF3 Neuroblastoma Vitamin B‐6 absorption. Reduced vitamin B‐6 breakpoint family, utilization for member 3 transmethylation/ transsulfuration pathway. MMAB Cobalamin Conversion of cobalamin into Decreased enzyme adenosyltransferase adenosylcobalamin (Vit B‐12 activity and conversion coenzyme for methylmalonyl‐ of cobalamin into CoA mutase). adensylcobalamin reduced. © Genoma International All rights reserved. 38

Conversion of Cobalamin into Adenosylcobalamin (AdoCbl) by MMAB Gene

Image Source: Hardy, CO et al. Potential for Misdiagnosis Due to Lack of Metabolic Derangement in Combined Methylmalonic Aciduria/Hyperhomocysteinemia (cblC) in the Neonate. July 2003Journal of Perinatology 23(5):384‐6.

13 Citric Acid or Krebs Cycle

Image source: Dahlhoff C, Desmarchelier C, et al. (2013). PLoS. Oct.

Conversion of Cobalamin into Adenosylcobalamin (AdoCbl) by MMAB Gene

Transmethylation/Transsulfuration, Gene SNPs, Nutritional Requirements (B‐Vitamins): Chrissi Gene High Impact Moderate Impact Nutritional Action Step

MTR X Methyl B12 MTRR X Methyl Folate SLC19A1 X Methyl Folate FUT2 X Methyl B12 TCN2 X Methyl B12 MTHFR X Methyl Folate CβSXCysteine MTHFD1 X Methyl Folate, Choline MMAB X Adenosyl B12 NBPF3 X Vitamin B‐6

Conversion of Cobalamin into Adenosylcobalamin (AdoCbl) by MMAB Gene

Transmethylation/Transsulfuration, Gene SNPs, Nutritional Requirements (B‐Vitamins): Chrissi Gene High Impact Moderate Impact Nutritional Action Step

15 Gene‐Function Matrix (B‐vitamins): Chrissi

Chrissi

FUT2

NBPF3 = B6 absorption

MMAB = cobalamin

Chrissi SLC19A FUT2 NBPF3 TCN2 MMAB

TCN2 MTR FUT2 B12 Choline B6

Choline MTRR Betaine Choline MTHFD MTHFD

CBS B2 B6

Graphic adapted from: http://openi.nlm.nih.gov/detailedresult.php?img=3129576_1471‐2350‐12‐75‐1&req=4

16 Transmethylation/Transsulfuration, Gene SNPs, Nutritional Requirements: Chrissi Biochemical High Moderate Nutritional Pathway Impact Impact Requirement Transmethylation MTR, MTRR FUT2, TCN2 Methyl B12 SLC19A1 MTHFR Methyl Folate, MTHFD1 Choline

Transsulfuration CβSCysteine NBPF3 Vitamin B6

Transmethylation & Transsulfuration Pathways: Glutathione and Free Radical Quenching

Transmethylation & Transsulfuration Pathways: Free Radical Quenching: Chrissi

Folate Pathway

17 Estrogen Metabolism Genotypes: Chrissi

Detox Gene High Moderate Impact Impact

Phase I CYP1B1 X

Phase II COMT X

GSTM1 Absent

CAT X

GPx X

SOD2 X

Transmethylation & Transsulfuration Pathways: Free Radical Quenching: Chrissi

Folate Pathway

GPx

GSTM1

Transmethylation/Transsulfuration, Gene SNPs, Nutritional Requirements: Chrissi Biochemical High Moderate Nutritional Pathway Impact Impact Requirement Transmethylation MTR, MTTR FUT2, TCN2 Methyl B12 SLC19A1 MTHFR, Methyl Folate, MTHFD1 Choline

Transsulfuration CβSCysteine NBPF3 Vitamin B6 Glutathione GSTM1 (Absent) GPx, CAT, SOD2 Glycine, Glutamate Glutathione

Other MMAB Adenosyl B12

18 Transmethylation/ Transsulfuration Pathway: Case History # 2 (Ted)

Nutrient Gene Impact • Ted: Caucasian, Male Vitamin B6 NBPF3 High • Age: 81 Occupation: Retired Vitamin B12 FUT2 Moderate • Heath Issues: Vitamin B12 TCN2 Moderate – Prostate Cancer (Prostatectomy) Betaine BHMT Moderate Cysteine C Beta S Moderate – Osteopenia Vitamin B12 MTR Moderate – Rheumatoid arthritis Vitamin B12 MTRR Moderate – Hypercholesterolemia Methyl Folate MTHFR‐2 Moderate • Medications: None Methyl Folate SLC19A1 Moderate Adenosyl B12 MMAB Low/ No Methyl Folate MTHFR‐1 Low/ No Choline MTHFD1 Low/ No

Transmethylation/Transsulfuration, Gene SNPs, Nutritional Requirements (B‐Vitamins): Ted Gene High Impact Moderate Impact Nutritional Action Step

NBPF3 X Vitamin B6 MTRR X Methyl Folate SLC19A1 X Methyl Folate FUT2 X Methyl B12 TCN2 X Methyl B12 MTHFR‐2 X Methyl Folate CβSXCysteine MTR X Methyl B12 BHMT X TMG, Choline

Gene‐Function Matrix (B‐vitamins): Ted

19 Ted

FUT2

NBPF3 = B6 absorption

MMAB= cobalamin

Ted NBPF3 FUT2 SLC19A1 TCN2

TCN2 MTR FUT2 B12 Choline B6

Choline MTRR Betaine Choline MTHFD MTHFD

B2 B6

Graphic adapted from: http://openi.nlm.nih.gov/detailedresult.php?img=3129576_1471‐2350‐12‐75‐1&req=4

Transmethylation/Transsulfuration, Gene SNPs, Nutritional Requirements: Ted Biochemical High Moderate Nutritional Pathway Impact Impact Requirement Transmethylation FUT2, MTR, Methyl B12 MTRR, TCN2 MTHFR‐2, Methyl Folate SLC19A1

Transsulfuration NBPF3 Vitamin B6 BHMT TMG, Choline CβSCysteine

20 Transmethylation & Transsulfuration Pathways: Glutathione and Free Radical Quenching (Ted)

Estrogen Metabolism Genotypes: Ted

Detox Gene High Moderate Impact Impact

Phase I CYP1B1 X

Phase II COMT X

GSTM1 Absent

GSTP1 X

GPx X

Transmethylation & Transsulfuration Pathways: Free Radical Quenching: Ted

Folate Pathway

SULT1A1

GPx

GSTM1 GSTP1

21 Transmethylation/Transsulfuration, Gene SNPs, Nutritional Requirements: Ted Biochemical High Moderate Nutritional Pathway Impact Impact Requirement Transmethylation FUT2, TCN2, Methyl B12 MTR, MTRR MTHFR‐2, Methyl Folate SLC19A1

Transsulfuration BHMT, CβS Betaine, Choline, Cysteine NBPF3 Vitamin B6 Glutathione GSTM1 (Absent) GSTP1, GPx Glycine, Glutamate, Glutathione Detox SULT1A1 Taurine, Sulfate, MSM

Transmethylation/ Transsulfuration Pathway: Case History #3 (Margaret) • Margaret: Caucasian, Female Nutrient Gene Impact • Age: 65 Occupation: Retired Vitamin B6 NBPF3 Moderate Vitamin B12 FUT2 Low/ No • Health issues: Vitamin B12 TCN2 High – Hypercholesterolemia Betaine BHMT High – Elevated TGs Cysteine C Beta S Moderate – Menopausal Symptoms Vitamin B12 MTR Low/ No – Osteopenia Vitamin B12 MTRR High – Weight Gain Methyl Folate MTHFR‐2 Low/ No • Medications: Methyl Folate SLC19A1 High Adenosyl B12 MMAB Low/ No – Bio‐identical hormones Methyl Folate MTHFR‐1 Moderate Choline MTHFD1 Moderate

Gene‐Function Matrix (B‐vitamins): Margaret

22 Margaret

FUT2

NBPF3 = B6 absorption

MMAB= cobalamin

Margaret

SLC19A, TCN2 NBPF3

TCN2 MTR FUT2 B12 Choline B6

Choline MTRR Betaine Choline MTHFD MTHFD

B2 B6

Graphic adapted from: http://openi.nlm.nih.gov/detailedresult.php?img=3129576_1471‐2350‐12‐75‐1&req=4

Transmethylation/Transsulfuration, Gene SNPs, Nutritional Requirements: Margaret

Biochemical High Moderate Nutritional Pathway Impact Impact Requirement Transmethylation TCN2, MTRR Methyl B12 MTRR, SLC19A1 MTHFR‐1 Methyl Folate MTHFD1 Choline

Transsulfuration BHMT CβS TMG, Choline, Cysteine NBPF3 Vitamin B6

23 Transmethylation & Transsulfuration Pathways: Glutathione and Free Radical Quenching

Hormones

Estrogen Metabolism Genotypes: Margaret

Detox Gene High Moderate Impact Impact

Phase I CYP1B1 X X

Phase II COMT X

GSTP1 X

GPX X

SOD2 X

SULT1A1 X

Transmethylation & Transsulfuration Pathways: Free Radical Quenching: Margaret

Hormones Folate Pathway

SULT1A1

GPx

GSTP1

24 Transmethylation/Transsulfuration, Gene SNPs, Nutritional Requirements: Margaret

Biochemical High Moderate Nutritional Pathway Impact Impact Requirement Transmethylation TCN2, MTRR Methyl B12 SLC19A1 MTHFR‐1 Methyl Folate MTHFD1 Choline

Transsulfuration BHMT CβS Betaine, Choline, Cysteine NBPF3 Vitamin B6 Glutathione GSTP1, GPx Glycine, Glutamate, Glutathione Detox SULT1A1 Taurine, Sulfate, MSM

Anti‐oxidant Cascade: Case History #4 (Ali)

Nutrient Gene Impact • Ali: Caucasian, female Vitamin B6 NBPF3 Moderate • Age: 24 Occupation: Retail Sales Vitamin B12 TCN2 Moderate • Health Issues: None Betaine BHMT Moderate • Medications: None Cysteine C Beta S Low/ No Vitamin B12 MTR Moderate Vitamin B12 MTRR Moderate Methyl Folate MTHFR‐2 Moderate Methyl Folate SLC19A1 Low/ No Adenosyl B12 MMAB High Methyl Folate MTHFR‐1 Moderate Choline MTHFD1 Low/ No Vitamin B12 FUT2 Moderate

Transmethylation/Transsulfuration, Gene SNPs, Nutritional Requirements (B‐Vitamins): Ali Gene High Impact Moderate Impact Nutritional Action Step

MMAB X Methyl B12 BHMT X Betaine TCN2 X Methyl B12 MTHFR‐1 X Methyl Folate MTHFR‐2 X Methyl Folate MTRR X Methyl Folate MTR X Methyl Folate NBPF3 X Vitamin B‐6 FUT2 X Methyl B12

25 Gene‐Function Matrix (B‐vitamins): Ali

TCN2, FUT2

Ali

FUT2

NBPF3 = B6 absorption

MMAB= cobalamin

Ali MMAB FUT2, TCN2, NBPF3, SLC19A1

TCN2 MTR FUT2 B12 Choline B6

Choline MTRR Betaine Choline MTHFD MTHFD

B2 B6

Graphic adapted from: http://openi.nlm.nih.gov/detailedresult.php?img=3129576_1471‐2350‐12‐75‐1&req=4

26 Transmethylation/Transsulfuration, Gene SNPs, Nutritional Requirements: Ali

Biochemical High Moderate Nutritional Pathway Impact Impact Requirement Transmethylation FUT2, TCN2, Methyl B12 MTR, MTRR MTHFR‐1, Methyl Folate MTHFR‐2

Transsulfuration BHMT Betaine, Choline NBPF3 Vitamin B6

Transmethylation & Transsulfuration Pathways: Glutathione and Free Radical Quenching: Ali

Estrogen Metabolism Genotypes: Chrissi

Detox Gene High Moderate Impact Impact

Phase II COMT X

GSTM1 Absent

GSTP1 X

GPx X

SOD2 X

27 Transmethylation & Transsulfuration Pathways: Glutathione and Free Radical Quenching: Ali

SULT1A1

GPx

GSTM1

Transmethylation/Transsulfuration, Gene SNPs, Nutritional Requirements: Ali Biochemical High Moderate Nutritional Pathway Impact Impact Requirement Transmethylation FUT2, TCN2, MTR, MTRR Methyl B12 MTHFR‐1, MTHFR‐2 Methyl Folate

Transsulfuration BHMT Betaine, Choline NBPF3 Vitamin B6 Glutathione GSTM1 (Absent) GSTP1, GPx Glycine, Glutamate, Glutathione Detox SULT1A1 Taurine, Sulfate, MSM

Other MMAB Adenosyl B12

Take‐Aways: B Vitamins, Transmethylation and Transsulfuration Pathways

• B complex vitamins play a critical role in the transmethylation/ transsulfuration pathway. • The transmethylation/ transsulfuration pathway is interrelated with many other biochemical pathways and biological systems, including DNA synthesis, detoxification, generation of methyl groups, and amino acid synthesis.

28 Take‐Aways: B Vitamins, Transmethylation and Transsulfuration Pathways

• Many genes are involved in these pathways, and SNPs can have significant effects on absorption, transport, and utilization of B‐ complex vitamins. • Using genomic test results, a healthcare professional can tailor and personalize nutritional requirements to optimize a person’s health (gene‐function matrix, visuals).

29 Module 2: Additional Resources

Wisneski, L., and Anderson, L. The Scientific Basis of Integrative Medicine. CRC Press, 2005.

Bidlack, W. and Rodriguez, R. Nutritional Genomics: The Impact of Dietary Regulation of Gene Function on Human Disease. CRC Press 2012.

Kaupt, J. and Rodriguez, R. editors. Nutritional Genomics: Discovering the Path to Personalized Nutrition. John Wiley & Sons Inc. 2006.

Young Joon, S., Cadenas, E., Dong, Z., and Packer, L. editors. Dietary Modulation of Cell Signaling Pathways. CRC Press, 2009.

Zempleni, J, and Dakshinamurti, K. editors. Nutrients and Cell Signaling. CRC Press, 2005.

Duke, James. Handbook of Biologically Active Phytochemicals and Their Activities. CRC Press, 1992.

Meskin, M., Bidlack, W., and Randolph, R. K., editors. Phytochemicals: Nutrient Gene Interactions. CRC Press, 2006.

Meskin, M., Bidlack, W., Davies, A., Lewis, D., Randolph, R.K. editors. Phytochemicals: Mechanisms of Action. CRC Press, 2004.

JAMA | Original Investigation Effect of Low-Fat vs Low-Carbohydrate Diet on 12-Month Weight Loss in Overweight Adults and the Association With Genotype Pattern or Insulin Secretion The DIETFITS Randomized Clinical Trial

Christopher D. Gardner, PhD; John F. Trepanowski, PhD; Liana C. Del Gobbo, PhD; Michelle E. Hauser, MD; Joseph Rigdon, PhD; John P. A. Ioannidis, MD, DSc; Manisha Desai, PhD; Abby C. King, PhD

Supplemental content

IMPORTANCE Dietary modification remains key to successful weight loss. Yet, no one dietary CME Quiz at strategy is consistently superior to others for the general population. Previous research jamanetwork.com/learning suggests genotype or insulin-glucose dynamics may modify the effects of diets. and CME Questions page 715

OBJECTIVE To determine the effect of a healthy low-fat (HLF) diet vs a healthy low-carbohydrate (HLC) diet on weight change and if genotype pattern or insulin secretion are related to the dietary effects on weight loss.

DESIGN, SETTING, AND PARTICIPANTS The Diet Intervention Examining The Factors Interacting with Treatment Success (DIETFITS) randomized clinical trial included 609 adults aged 18 to 50 years without diabetes with a body mass index between 28 and 40. The trial enrollment was from January 29, 2013, through April 14, 2015; the date of final follow-up was May 16, 2016. Participants were randomized to the 12-month HLF or HLC diet. The study also tested whether 3 single-nucleotide polymorphism multilocus genotype responsiveness patterns or insulin secretion (INS-30; blood concentration of insulin 30 minutes after a glucose challenge) were associated with weight loss.

INTERVENTIONS Health educators delivered the behavior modification intervention to HLF (n = 305) and HLC (n = 304) participants via 22 diet-specific small group sessions administered over 12 months. The sessions focused on ways to achieve the lowest fat or carbohydrate intake that could be maintained long-term and emphasized diet quality.

MAIN OUTCOMES AND MEASURES Primary outcome was 12-month weight change and determination of whether there were significant interactions among diet type and genotype pattern, diet and insulin secretion, and diet and weight loss. Author Affiliations: Stanford Prevention Research Center, Department of Medicine, Stanford RESULTS Among 609 participants randomized (mean age, 40 [SD, 7] years; 57% women; University Medical School, Stanford, mean body mass index, 33 [SD, 3]; 244 [40%] had a low-fat genotype; 180 [30%] had a California (Gardner, Trepanowski, low-carbohydrate genotype; mean baseline INS-30, 93 μIU/mL), 481 (79%) completed the Del Gobbo, Hauser, Ioannidis, King); trial. In the HLF vs HLC diets, respectively, the mean 12-month macronutrient distributions Quantitative Sciences Unit, Stanford University School of Medicine, were 48% vs 30% for carbohydrates, 29% vs 45% for fat, and 21% vs 23% for protein. Stanford, California (Rigdon, Desai); Weight change at 12 months was −5.3 kg for the HLF diet vs −6.0 kg for the HLC diet Department of Health Research and (mean between-group difference, 0.7 kg [95% CI, −0.2 to 1.6 kg]). There was no significant Policy, Stanford University School of diet-genotype pattern interaction (P = .20) or diet-insulin secretion (INS-30) interaction Medicine, Stanford, California (Ioannidis, Desai, King); Department (P = .47) with 12-month weight loss. There were 18 adverse events or serious adverse events of Statistics, Stanford University that were evenly distributed across the 2 diet groups. School of Humanities and Sciences, Stanford, California (Ioannidis, Desai); CONCLUSIONS AND RELEVANCE In this 12-month weight loss diet study, there was no Department of Biomedical Data Science, Stanford University School significant difference in weight change between a healthy low-fat diet vs a healthy of Medicine, Stanford, California low-carbohydrate diet, and neither genotype pattern nor baseline insulin secretion was (Ioannidis, Desai). associated with the dietary effects on weight loss. In the context of these 2 common weight Corresponding Author: Christopher loss diet approaches, neither of the 2 hypothesized predisposing factors was helpful in D. Gardner, PhD, Stanford Prevention identifying which diet was better for whom. Research Center, Department of Medicine, Stanford University Medical School, 1265 Welch Rd, TRIAL REGISTRATION clinicaltrials.gov Identifier: NCT01826591 Stanford, CA 94305 JAMA. 2018;319(7):667-679. doi:10.1001/jama.2018.0245 ([email protected]).

Downloaded From: by Roberta Kline on 02/21/2018 Research Original Investigation Low-Fat vs Low-Carbohydrate Diet on Weight Loss in Overweight Adults

besity is a 21st-century major public health challenge.1,2 Among many strategies studied for weight loss, Key Points a common contrast has been low-fat diets vs low- O Question What is the effect of a healthy low-fat (HLF) diet 3-5 carbohydrate diets. Most diet trials have reported modest vs a healthy low-carbohydrate (HLC) diet on weight change (ie, <5%) mean weight loss after 12 months and negligible mean at 12 months and are these effects related to genotype pattern weight loss differences between diet groups.6 In contrast, in- or insulin secretion? dividual weight losses have varied widely within diet groups Findings In this randomized clinical trial among 609 overweight in these studies, ranging from approximately 25 kg lost to ap- adults, weight change over 12 months was not significantly proximately 5 kg gained.3-5 different for participants in the HLF diet group (−5.3 kg) The substantial variability of weight loss response sug- vs the HLC diet group (−6.0 kg), and there was no significant gests some strategies may work better for some individuals diet-genotype interaction or diet-insulin interaction with 12-month than others, and that no one diet should be recommended weight loss. universally.7 Yet, interindividual differences in response to diet Meaning There was no significant difference in 12-month weight are poorly understood. Some studies have reported that geno- loss between the HLF and HLC diets, and neither genotype pattern type variation could predispose individuals to differential nor baseline insulin secretion was associated with the dietary weight loss that varies by diet type.8,9 effects on weight loss. In a preliminary retrospective study, a 3-fold difference was observed in 12-month weight loss for initially overweight women who were determined to have been appropriately matched (mean weight loss of 6 kg) vs mismatched low-carbohydrate diet for 12 months. Participant enrollment (mean weight loss of 2 kg) to a low-fat or low-carbohydrate began on January 29, 2013, and continued through April 14, diet based on multilocus genotype patterns with single- 2015. The date of final follow-up was May 16, 2016. Inter- nucleotide polymorphisms (SNPs) from 3 genes (PPARG, ventions consisted primarily of class-based instruction. Five ADRB2, and FABP2) relevant to fat and carbohydrate metabo- waves of recruitment (cohorts) had staggered start dates lism (a putative low-fat–responsive genotype and a low- between March 2013 and March 2015. The primary outcome carbohydrate–responsive genotype). The participants with was 12-month weight change. the low-fat–responsive genotype were observed to lose more The first primary hypothesis was that there is a signifi- weight when assigned to a low-fat diet than those assigned to cant diet × genotype pattern interaction for weight loss. a low-carbohydrate diet, and vice versa for those with the The second primary hypothesis was that there is a signifi- low-carbohydrate–responsive genotype.9,10 cant diet × insulin secretion interaction for weight loss. Sec- Similarly, several studies11-14 have reported that baseline ondary outcomes included anthropometric measures, insulin dynamics may explain differential weight loss suc- plasma lipid levels, insulin and glucose levels, and blood cess obtained via a low-fat diet vs a low-carbohydrate diet. pressure levels. The protocol update and statistical analysis For example, individuals with greater insulin resistance may plan are included in Supplement 1 and the full study proto- have better success with low-carbohydrate diets due to the de- col was published previously10 (the protocol included creased demand on insulin to clear a lower amount of dietary details regarding blood sampling, storage, and specific labo- carbohydrate delivered to the circulation. However, these stud- ratory assays). ies were limited by relatively small sample sizes or post hoc analyses of the results. Participants The primary objective of the Diet Intervention Examin- We aimed to recruit 600 adults from the Stanford and ing The Factors Interacting with Treatment Success San Francisco Bay areas of California using media advertise- (DIETFITS) study was to test whether (1) a set of 3 SNP geno- ments and email lists from previous recruitment for nutri- type patterns or (2) baseline differences in insulin secretion tion studies conducted by our laboratory group. We consid- (the blood insulin concentration at 30 minutes after a glu- ered men and premenopausal women aged 18 to 50 years cose challenge; INS-30),12,13 or both, predisposed individu- with a body mass index (calculated as weight in kilograms als to differential success in 12-month weight change while divided by height in meters squared) of 28 to 40. on a low-fat diet vs a low-carbohydrate diet. The major criteria for exclusion were having uncon- trolled hypertension or metabolic disease; diabetes; cancer; heart, renal, or liver disease; and being pregnant or lactating. Methods Individuals were excluded if taking hypoglycemic, lipid- lowering, antihypertensive, psychiatric, or other medica- The Stanford University human subjects committee ap- tions known to affect body weight or energy expenditure. Any proved the study. All study participants provided written in- medication type not noted was allowed if the individual had formed consent. been stable while taking such medication for at least 3 months prior to baseline data collection. Study Design Randomization to a healthy low-fat diet or a healthy This single-site, parallel-group, weight loss diet trial ran- low-carbohydrate diet was performed using an allocation domized individuals to a healthy low-fat diet or a healthy sequence determined by computerized random-number

668 JAMA February 20, 2018 Volume 319, Number 7 (Reprinted) jama.com

Downloaded From: by Roberta Kline on 02/21/2018 Low-Fat vs Low-Carbohydrate Diet on Weight Loss in Overweight Adults Original Investigation Research

Figure 1. Flow of Participants Through the Diet Intervention Examining The Factors Interacting with Treatment Success Trial

1057 Individuals screened for eligibility

254 Excluded 69 Did not meet eligibility criteria 60 Body mass index >40 or <28a 9 Blood glucose level >125 mg/dLb 137 No longer interested 29 Discontinued communication 19 Other

803 Attended study orientation and informed of study details

171 Excluded 142 Not interested 29 Other

632 Randomized

314 Randomized to receive a healthy 318 Randomized to receive a healthy low-fat diet low-carbohydrate diet

9 Withdrew prior to receiving diet 14 Withdrew prior to receiving diet assignment assignment 5 Scheduling conflict 9 Scheduling conflict 4 Other reasons 5 Other reasons

305 Informed of diet assignment 304 Informed of diet assignment

24 Lost to follow-up 29 Lost to follow-up 40 Discontinued intervention 37 Discontinued intervention 21 Personal reasons 13 Personal reasons 8 Scheduling conflict 12 Scheduling conflict 6 Health issues unrelated to study 11 Health issues unrelated to study 5 Unhappy with diet 1 Unhappy with diet

241 Completed study 238 Completed study a Body mass index is calculated as 305 Included in primary analysis 304 Included in primary analysis weight in kilograms divided by 9 Excluded (withdrew prior to 14 Excluded (withdrew prior to height in meters squared. receiving diet assignment) receiving diet assignment) b To convert glucose to mmol/L, multiply by 0.0555.

generation (Blockrand in R version 3.4.0; R Project for Sta- Weight Loss Intervention tistical Computing) in block sizes of 8 (with 4 individuals The protocol included a 1-month run-in period during which going to each diet) by a statistician not involved in interven- participants were instructed to maintain their habitual diet, tion delivery or data collection. Participants did not learn of physical activity level, and body weight. The intervention their diet group assignment until they completed all base- involved 22 instructional sessions held over 12 months in line measures and attended their first intervention class diet-specific groups of approximately 17 participants per (Figure 1). class. Sessions were held weekly for 8 weeks, then every 2 The original study design was a 2 × 2 factorial design weeks for 2 months, then every 3 weeks until the sixth (diet × genotype-pattern matching). However, near the month, and monthly thereafter. Classes were led by 5 regis- onset of the study, the initial funding was more than doubled, tered dietitian health educators who each taught 1 healthy allowing for a 50% increase in sample size, the addition low-fat class and 1 healthy low-carbohydrate class per of a second primary hypothesis for the assessment of a cohort. Dietitians were blinded to all laboratory measures diet × insulin secretion interaction, and an expanded set of and genotype. measurements. To test for both primary hypotheses, the study The dietary interventions were described previously.10 was changed to a simple parallel group design with testing Briefly, the main goals were to achieve maximal differentia- for 2 interactions (described in further detail in eAppendix 1 tion in intake of fats and carbohydrates between the 2 in Supplement 2). diet groups while otherwise maintaining equal treatment

jama.com (Reprinted) JAMA February 20, 2018 Volume 319, Number 7 669

Downloaded From: by Roberta Kline on 02/21/2018 Research Original Investigation Low-Fat vs Low-Carbohydrate Diet on Weight Loss in Overweight Adults

intensity and an emphasis on high-quality foods and bever- type; patterns 1-5), more sensitive to carbohydrates (low- ages. Thus, participants were instructed to reduce intake of carbohydrate genotype; patterns 6-14), or sensitive to nei- total fat or digestible carbohydrates to 20 g/d during the first ther genotype (pattern 15). Additional details are available in 8 weeks. Higher priorities for reduction were given to specific eAppendix 3 in Supplement 2. foods and food groups that derived their energy content pri- Before randomization and at months 6 and 12, each par- marily from fats or carbohydrates. For example, the reduc- ticipant completed an oral glucose tolerance test of 75 g. This tion of edible oils, fatty meats, whole-fat dairy, and nuts was included measurement of insulin concentration 30 minutes prioritized for the healthy low-fat group, whereas the reduc- after glucose consumption (ie, INS-30, which is a proxy mea- tion of cereals, grains, rice, starchy vegetables, and legumes sure of insulin secretion).10,21,22 When this study was first was prioritized for the healthy low-carbohydrate group. designed, insulin sensitivity was to be measured and used as Then individuals slowly added fats or carbohydrates a predictor of differential weight loss success. After the study back to their diets in increments of 5 to 15 g/d per week until was initiated, reports were published12,13,23,24 indicating they reached the lowest level of intake they believed could be INS-30 was a successful predictor of weight loss in the con- maintained indefinitely. No explicit instructions for energy text of low-carbohydrate diets or similar diets. There also was (kilocalories) restriction were given. Both diet groups were evidence25 that early-phase insulin secretion differed mark- instructed to (1) maximize vegetable intake; (2) minimize intake edly between diets that were similar to those tested in the of added sugars, refined flours, and trans fats; and (3) focus DIETFITS study. Prior to examining any data, we modified on whole foods that were minimally processed, nutrient dense, the primary hypothesis of our study and tested baseline and prepared at home whenever possible. Other components INS-30 rather than a measure of insulin sensitivity as the of the emphasis on high-quality food for both diet groups are putative effect modifier. No other glucose or insulin variables described elsewhere.10 were tested for effect modification. Participants were encouraged to follow current physical A set of related secondary outcomes was assessed. Con- activity recommendations.15 Health educators emphasized centrations of plasma lipids, glucose, and insulin were mea- emotional awareness and behavior modification to support sured in fasting samples, waist circumference was assessed by dietary adherence and weight loss. Behavioral modification measuring tape, blood pressure was measured via auto- strategies included empirically supported principles of self- mated device, and all of these were assessed using standard regulatory behavior change (eg, goal setting, self-efficacy assessment techniques.10 building, supportive environments, and relapse prevention) Body composition was assessed by dual-energy x-ray based on social cognitive theory and the transtheoretical absorptiometry and both respiratory exchange ratio model.10,16-18 (bounded by 0.7 [using solely fat for fuel] and 1.0 [using solely glucose for fuel]) and resting energy expenditure Outcome Measurements were assessed by metabolic cart (ie, measures respiratory All data were collected at baseline and at months 3, 6, and 12 exchange of oxygen and carbon dioxide while a participant is for all cohorts unless noted otherwise. Staff who measured out- supine and resting) at baseline and at months 6 and 12 in comes were blinded to diet assignment, genotype pattern, cohorts 2 through 5. Adequate funding became available for INS-30, and diet assignment. Dietary intake at each time point dual-energy x-ray absorptiometry, respiratory exchange was assessed using 3 unannounced 24-hour multiple-pass re- ratio, and resting energy expenditure only after cohort 1 call interviews (2 on weekdays and 1 on a weekend day).19 was enrolled. The metabolic syndrome was determined using Total energy expenditure was assessed using the Adult Treatment Panel III guidelines from the National Stanford Seven-Day Physical Activity Recall questionnaire.20 Cholesterol Education Program.26 Both the dietary intake and physical activity recall were self- reported measures. Weight was measured by digital scale Statistical Analysis at the Stanford Clinical Translational Research Unit and Based on the original study design, assuming 100 partici- 12-month weight change was the primary outcome. pants in each of the 4 relevant groups (genotype and dietary Genotype pattern and insulin secretion were assessed for assignment), and normally distributed values of weight interaction testing. The Affymetrix UK Biobank Axiom micro- change at 12 months, there was 80% power to detect clini- array was used for analysis of 820 967 SNPs and insertions or cally meaningful differences in treatment effect by genotype deletions. The array included 2 of the SNPs from the original (eg, whether dietary assignment had an effect on weight study design: PPARG (rs1801282) and ADRB2 (rs1042714). change at 12 months except for those assigned to the low- FABP2 (rs1799883) was imputed with an imputation quality carbohydrate diet who have the low-carbohydrate genotype score (r2 = 0.99). Additional details appear in eAppendix 2 in because such individuals lose 3.2 kg on average). This calcu- Supplement 2. The 3 SNP multilocus genotype patterns have lation was based on simulations, and assumed a 2-sided Wald been explored previously.9 test conducted at the .05 level of significance. Of 27 possible 3-locus genotypes that could arise from Under similar assumptions regarding the statistical the combination of the 3 SNPs, 15 were observed with 1% or testing and type I error, and assuming a sample size of only greater genotype frequency in previously studied samples 400 participants (200 in each treatment group), there was of adults. The multilocus genotypes were grouped into greater than 80% power to detect differences in the treat- those predicted to be more sensitive to fat (low-fat geno- ment effect with insulin secretion, including for example,

670 JAMA February 20, 2018 Volume 319, Number 7 (Reprinted) jama.com

Downloaded From: by Roberta Kline on 02/21/2018 Low-Fat vs Low-Carbohydrate Diet on Weight Loss in Overweight Adults Original Investigation Research

if for every 1-unit increase in insulin secretion, weight loss time point, and baseline INS-30. The INS-30 variable was at 12 months increases by 0.8 lb (0.36 kg). These power analyzed as a continuous variable, but is presented as tertiles calculations were performed a priori for the originally plan- for ease of presentation in parallel to the presentation of ned sample size of 400. As described in greater detail in genotype pattern data. The cutoffs for the tertiles were de- eAppendix 1 in Supplement 2, after initially being funded by termined using the baseline insulin concentrations of all the National Institutes of Health and the National Institute 609 participants. of Diabetes and Digestive and Kidney Diseases in 2012, A Satterthwaite approximation for denominator degrees additional funding was received to augment the trial, which of freedom was used in all Wald tests.29 All tests were 2-sided involved, among other modifications, increasing the sample and conducted at the .05 level of significance. Formal hypoth- size from 400 to 600, and adding INS-30 as a second vari- esis testing was performed only for the 2 primary hypoth- able for interaction testing. With the larger sample size, the eses. All other P values that were generated were purely de- study had even greater statistical power, estimated at 90% scriptive in nature and correspond to secondary and based on post hoc calculations. exploratory analyses. Statistical analyses were performed using The main hypotheses addressed 12-month weight change R version 3.4.0 (R Project for Statistical Computing). Specifi- by diet, diet and genotype, and diet and baseline INS-30. All cally, the lme430 package was used for mixed-effects models hypotheses were addressed using generalized, linear mixed- and the lmerTest29 package was used for hypothesis tests in effects models.27,28 We applied modified intent-to-treat prin- the mixed-effects models. ciples. This means that all participants who were randomized and had baseline information were included in the analysis and analyzed according to original treatment assignment, Results regardless of adherence or loss to follow-up (Figure 1). For the hypothesis about the effect of diet group on 12-month Among 609 participants randomized (mean age, 40 [SD, 7] weight change, a linear mixed-effects model for weight years; 57% women; mean body mass index, 33 [SD, 3]; 244 that accounted for missing data under flexible assumptions [40%] had a low-fat genotype; 180 [30%] had a low- regarding missingness was used with fixed effects for diet, carbohydrate genotype; mean baseline INS-30, 93 μIU/mL), time (baseline, 3, 6, and 12 months), and their interaction, 481 (79%) completed the trial. The flow of the participants along with a random effect for participant. For the hypoth- through the trial appears in Figure 1. Baseline characteristics eses involving diet and genotype (or diet and baseline by diet group appear in Table 1. Among participants in the INS-30), an additional fixed effect was added for genotype (or healthy low-fat diet group, 130 (42.6%) had the low-fat geno- baseline INS-30), along with all 2- and 3-way interactions type and 83 (27.2%) had the low-carbohydrate genotype, (model appears in eAppendix 4 in Supplement 2). whereas in the healthy low-carbohydrate group, 114 (37.5%) The validity of such an analysis relies on the assumption had the low-fat genotype and 97 (31.9%) had the low- that the missing outcome data measured at follow-up are un- carbohydrate genotype. related to unobserved values of weight conditional on ob- Of 22 assigned intervention instruction sessions for the full served variables such as treatment assignment and baseline study sample, the mean number of sessions attended was 14.4 and intermittent values of weight. The hypothesis about diet (SD, 5.3) for the healthy low-fat diet group and 14.6 (SD, 5.1) was tested using a Wald test for the 2-way interaction be- for the healthy low-carbohydrate diet group, which includes tween the 12-month time point and diet. The hypothesis about dropouts. Retention at 12 months, which was defined as par- genotype (or baseline INS-30) was tested using a Wald test for ticipants who provided any data at 12 months, was 79% for both the 3-way interaction between the 12-month time point, diet, groups. Participant ratings for health educator enthusiasm and and genotype (or baseline INS-30). Genotype was defined as knowledge of material was high and similar between diet matched for those participants with a 3-SNP combination sug- groups. The mean ratings were 4.6 to 5.0 on a scale of 1 to 5, gesting success on a low-carbohydrate diet who were random- with 5 as the highest rating. ized to the low-carbohydrate diet, or for those participants with Total energy intake was not different between diet a 3-SNP combination suggesting success on a low-fat diet who groups at baseline or at any subsequent time point (P ≥.10for were randomized to the low-fat diet. Genotype was other- all; Table 2). Despite not being instructed to follow a specific wise defined as mismatched and is described in eAppendices energy (kilocalorie) intake restriction, the mean reported 2and3inSupplement 2. energy intake reduction relative to baseline was approxi- There were 185 individuals who were not classified as hav- mately 500 to 600 kcal/d for both groups at each time point ing either the low-fat genotype pattern or a low-carbohydrate after randomization. genotype pattern (146 individuals with other 3-SNP patterns At baseline, there were no significant between-group and 39 with missing or compromised genotyping data) who differences for any nutrients examined. In contrast, there were excluded from the genotype analysis for the first were significant between-group differences after random- hypothesis as originally planned.10 An additional diet- ization at every time point (all P ≤ .001) for percentage of genotype analysis was performed, restricting the study popu- energy; intakes of carbohydrates, fat, protein, saturated fat, lation to whites only and focusing on only 1 ancestry group fiber, and added sugars; and glycemic index and glycemic as originally planned.10 The second hypothesis was tested load (Table 2). In the healthy low-fat diet vs the healthy using a Wald test for the interaction among diet, 12-month low-carbohydrate diet, respectively, the mean 12-month

jama.com (Reprinted) JAMA February 20, 2018 Volume 319, Number 7 671

Downloaded From: by Roberta Kline on 02/21/2018 Research Original Investigation Low-Fat vs Low-Carbohydrate Diet on Weight Loss in Overweight Adults

Table 1. Baseline Demographics and Anthropometric and Metabolic Variables

Healthy Low-Fat Diet Healthy Low-Carbohydrate Diet (n = 305) (n = 304) Sex, No. (%) Women 167 (54.8) 179 (58.9) Men 138 (45.2) 125 (41.1) Age, mean (SD), y 39.3 (6.8) 40.2 (6.7) Highest level of education achieved, No. (%)a SI conversion factors: To convert glucose to mmol/L, multiply by

672 JAMA February 20, 2018 Volume 319, Number 7 (Reprinted) jama.com

Downloaded From: by Roberta Kline on 02/21/2018 Low-Fat vs Low-Carbohydrate Diet on Weight Loss in Overweight Adults Original Investigation Research

Table 2. Dietary Intake by Time Point

Healthy Low-Fat Diet Healthy Low-Carbohydrate Diet No. of No. of Participants Mean (SD) Participants Mean (SD) Mean Between-Group Difference (95% CI)a Total Energy Intake, kcal Baseline 304 2148.1 (39.4) 304 2222.8 (37.5) −76.3 (−166.1 to 13.4) 3 mo 274 1515.0 (27.7) 275 1580.8 (29.1) −56.9 (−150.2 to 36.4) 6 mo 240 1624.4 (37.3) 251 1621.3 (33.2) 0.2 (−96.9 to 97.3) 12 mo 225 1716.1 (34.5) 224 1697.1 (32.1) 2.9 (−97.2 to 103.0) Carbohydrates, g Baseline 304 241.8 (5.0) 304 246.5 (4.5) −4.9 (−16.6 to 6.9) 3 mo 274 205.2 (4.3) 275 96.6 (3.4) 109.0 (96.8 to 121.2) 6 mo 240 211.2 (5.3) 251 113.2 (4.1) 95.6 (83.0 to 108.3) 12 mo 225 212.9 (5.0) 224 132.4 (4.2) 74.2 (61.2 to 87.2) Carbohydrates, % kcal Baseline 304 44.5 (0.5) 304 44.0 (0.4) 0.5 (−1.1 to 2.1) 3 mo 274 52.6 (0.6) 275 23.1 (0.7) 29.4 (27.8 to 31.0) 6 mo 240 50.8 (0.7) 251 26.5 (0.7) 24.1 (22.4 to 25.8) 12 mo 225 48.4 (0.7) 224 29.8 (0.7) 17.8 (16.0 to 19.5) Fat, g Baseline 304 87.0 (2.0) 304 92.6 (1.9) −5.6 (−10.4 to −0.8) 3 mo 274 42.0 (1.2) 275 88.8 (2.0) −46.2 (−51.2 to −41.2) 6 mo 240 50.3 (1.8) 251 86.6 (2.0) −36.0 (−41.2 to −30.8) 12 mo 225 57.3 (1.7) 224 86.2 (2.0) −28.4 (−33.8 to −23.0) Fat,%kcal Baseline 304 34.8 (0.4) 304 36.0 (0.3) −1.2 (−2.4 to 0.1) 3 mo 274 24.0 (0.5) 275 49.0 (0.5) −24.9 (−26.2 to −23.6) 6 mo 240 26.4 (0.6) 251 46.8 (0.6) −20.3 (−21.7 to −18.9) 12 mo 225 28.7 (0.5) 224 44.6 (0.6) −15.4 (−16.8 to −14.0) Protein, g Baseline 304 92.1 (1.7) 304 93.1 (1.6) −1.1 (−5.8 to 3.6) 3 mo 274 79.5 (1.6) 275 96.9 (2.0) −17.1 (−22.0 to −12.2) 6 mo 240 81.9 (1.9) 251 93.8 (1.9) −11.6 (−16.6 to −6.5) 12 mo 225 84.5 (1.8) 224 93.3 (2.0) −8.5 (−13.8 to −3.3) Protein, % kcal Baseline 304 17.9 (0.3) 304 17.3 (0.2) 0.6 (−0.4 to 1.5) 3 mo 274 21.5 (0.4) 275 25.9 (0.4) −4.4 (−5.4 to −3.5) 6 mo 240 20.8 (0.4) 251 24.3 (0.4) −3.5 (−4.5 to −2.5) 12 mo 225 20.6 (0.4) 224 22.9 (0.4) −2.1 (−3.2 to −1.1) Saturated Fat, g Baseline 304 28.9 (0.7) 304 30.8 (0.7) −1.9 (−3.7 to −0.1) 3 mo 274 12.7 (0.4) 275 29.1 (0.7) −16.2 (−18.1 to −14.3) 6 mo 240 15.3 (0.7) 251 27.9 (0.7) −12.4 (−14.4 to −10.5) 12 mo 225 18.2 (0.6) 224 28.2 (0.8) −9.8 (−11.8 to −7.8) Saturated Fat, % kcal Baseline 304 11.5 (0.2) 304 11.9 (0.2) −0.4 (−0.9 to 0.2) 3 mo 274 7.2 (0.2) 275 16.1 (0.3) −8.8 (−9.4 to −8.2) 6 mo 240 8.0 (0.2) 251 15.0 (0.3) −7.0 (−7.6 to −6.4) 12 mo 225 9.0 (0.2) 224 14.5 (0.3) −5.3 (−5.9 to −4.7) Fiber, g Baseline 304 22.0 (0.6) 304 21.6 (0.5) 0.4 (−1.2 to 1.9) 3 mo 274 24.2 (0.7) 275 16.7 (0.8) 7.5 (5.9 to 9.1) 6 mo 240 23.7 (0.7) 251 17.2 (0.5) 6.5 (4.8 to 8.2) 12 mo 225 23.0 (0.6) 224 18.6 (0.5) 4.1 (2.3 to 5.8)

(continued)

jama.com (Reprinted) JAMA February 20, 2018 Volume 319, Number 7 673

Downloaded From: by Roberta Kline on 02/21/2018 Research Original Investigation Low-Fat vs Low-Carbohydrate Diet on Weight Loss in Overweight Adults

Table 2. Dietary Intake by Time Point (continued)

Healthy Low-Fat Diet Healthy Low-Carbohydrate Diet No. of No. of Participants Mean (SD) Participants Mean (SD) Mean Between-Group Difference (95% CI)a Fiber, g/1000 kcal Baseline 304 10.8 (0.2) 304 10.3 (0.2) 0.4 (−0.5 to 1.4) 3 mo 274 16.5 (0.4) 275 11.3 (0.6) 5.3 (4.3 to 6.3) 6 mo 240 15.4 (0.4) 251 11.2 (0.3) 4.2 (3.1 to 5.2) 12 mo 225 14.2 (0.4) 224 11.6 (0.3) 2.5 (1.4 to 3.5) Added Sugars, g Baseline 304 49.3 (2.0) 304 52.2 (2.0) −3.0 (−7.3 to 1.3) 3 mo 274 28.5 (1.3) 275 16.2 (1.2) 12.4 (8.0 to 16.9) 6 mo 240 31.5 (1.7) 251 18.9 (1.3) 12.2 (7.5 to 16.9) 12 mo 225 33.1 (1.7) 224 22.8 (1.6) 9.6 (4.7 to 14.5) Sugar, g/1000 kcal Baseline 304 22.4 (0.7) 304 23.2 (0.7) −0.8 (−2.7 to 1.1) 3 mo 274 18.4 (0.7) 275 9.8 (0.6) 8.6 (6.6 to 10.6) 6 mo 240 18.9 (0.8) 251 11.2 (0.7) 7.6 (5.5 to 9.7) 12 mo 225 18.7 (0.9) 224 12.9 (0.8) 5.6 (3.5 to 7.8) Glycemic Indexb Baseline 304 57.8 (0.3) 304 58.2 (0.3) −0.4 (−1.4 to 0.6) 3 mo 274 56.0 (0.3) 275 50.1 (0.4) 5.9 (4.9 to 7.0) 6 mo 240 56.2 (0.4) 251 51.4 (0.5) 4.7 (3.6 to 5.8) 12 mo 225 56.1 (0.4) 224 52.7 (0.5) 3.1 (2.0 to 4.3) Glycemic Loadc Baseline 304 128.2 (2.9) 304 132.4 (2.7) −4.3 (−11.1 to 2.6) 3 mo 274 102.1 (2.3) 275 43.1 (2.0) 59.0 (51.9 to 66.2) 6 mo 240 107.0 (3.1) 251 52.4 (2.5) 53.0 (45.6 to 60.4) 12 mo 225 108.0 (2.8) 224 62.9 (2.6) 41.5 (33.8 to 49.1) a Healthy low-fat diet minus healthy low-carbohydrate diet from linear from that food to raise blood glucose relative to 50 g of glucose (scale of mixed-effects model. 0-100; a score of 100 refers to the same rate as glucose). b Indicates ranking of foods according to the potential of 50 g of carbohydrates c Indicates the actual amount of carbohydrates multiplied by the glycemic index.

macronutrient distributions were 48% vs 30% for carbohy- type to a low-fat genotype (Figure 2A). This indicates that drates, 29% vs 45% for fat, and 21% vs 23% for protein. there was no significant difference in weight change among participants matched vs mismatched to their diet assign- Primary Outcome ment based on their 3-SNP genotype pattern. In analyses The mean 12-month weight change was −5.3 kg (95% CI, restricted to participants of European descent only, no sig- −5.9 kg to −4.7 kg) for the healthy low-fat diet group and nificant interaction was observed by genotype pattern (the −6.0 kg (95% CI, −6.6 kg to −5.4 kg) for the healthy low- 3-way interaction for the main diet, genotype, and time carbohydrate diet group, which was not statistically different yielded a beta coefficient of 2.58 [95% CI, −0.18 to 5.34]; (Table 3). There was a similar range for weight change of P = .07). approximately 40 kg within each group (−30 kg to 10 kg; Similarly, the test for interaction among diet, baseline eFigure 1 in Supplement 2). insulin secretion (INS-30), and the 12-month time point was not statistically significant. The interpretation of the beta Interaction Testing coefficient for the 3-way interaction (beta coefficient, 0.08 The test for the interaction among diet, genotype pattern, [95% CI, −0.13 to 0.28], P = .47) is that 12-month weight and the 12-month time point was not statistically sig- change increases (estimated as 0.08 kg) when switching nificant. The interpretation of the beta coefficient for the from a healthy low-carbohydrate diet and x units of baseline 3-way interaction (beta coefficient, 1.38 [95% CI, −0.72 to INS-30 to a healthy low-fat diet and x + 10 units of baseline 3.49], P = .20) is that 12-month weight change increases INS-30 beyond the effects of changing from a healthy low- (estimated as 1.38 kg) when switching from a healthy low- carbohydrate diet to a healthy low-fat diet and increasing carbohydrate diet and a low-carbohydrate genotype to a baseline INS-30 by 10 μIU/mL (Figure 2B). Weight change healthy low-fat diet and low-fat genotype beyond the main trajectories for the diet-genotype pattern subgroups are pre- effects of switching from a healthy low-carbohydrate diet to sented in eFigure 2A and for diet and INS-30 tertile sub- a healthy low-fat diet and from a low-carbohydrate geno- groups in eFigure 2B in Supplement 2.

674 JAMA February 20, 2018 Volume 319, Number 7 (Reprinted) jama.com

Downloaded From: by Roberta Kline on 02/21/2018 Low-Fat vs Low-Carbohydrate Diet on Weight Loss in Overweight Adults Original Investigation Research

Table 3. 12-Month Change Estimates for Anthropometric Variables by Diet

12-mo Change Estimate (95% CI)a Healthy Low-Fat Diet Healthy Low-Carbohydrate Diet Between-Group Difference (n = 305) (n = 304) (95% CI)b Weight, kg −5.29 (−5.93 to −4.65) −5.99 (−6.63 to −5.35) 0.70 (−0.21 to 1.60) Body mass indexc −1.75 (−1.97 to −1.52) −2.07 (−2.30 to −1.85) 0.33 (0.01 to 0.64) Body fat %d −1.97 (−2.38 to −1.56) −2.15 (−2.54 to −1.75) 0.18 (−0.40 to 0.75) Waist circumference, cm −3.74 (−4.64 to −2.84) −4.41 (−5.31 to −3.51) 0.67 (−0.60 to 1.94) Lipid level, mmol/L High-density lipoprotein cholesterol 0.40 (−0.37 to 1.18) 2.64 (1.87 to 3.41) −2.24 (−3.33 to −1.15) Low-density lipoprotein cholesterol −2.12 (−4.70 to 0.47) 3.62 (1.04 to 6.19) −5.74 (−9.38 to −2.09) Triglycerides −9.95 (−17.46 to −2.44) −28.20 (−35.67 to −20.72) 18.25 (7.65 to 28.84) Blood pressure, mm Hg Systolic −3.18 (−4.33 to −2.03) −3.72 (−4.86 to −2.58) 0.54 (−1.07 to 2.16) Diastolic −1.94 (−2.65 to −1.22) −2.64 (−3.34 to −1.93) 0.70 (−0.31 to 1.71) Fasting glucose, mg/dL −3.67 (−4.90 to −2.44) −2.10 (−3.32 to −0.87) −1.58 (−3.31 to 0.16) Fasting insulin, μIU/mL −2.64 (−3.79 to −1.49) −2.33 (−3.48 to −1.19) −0.31 (−1.93 to 1.31) Insulin-30, μIU/mLe −15.38 (−21.13 to −9.62) −11.48 (−17.18 to −5.78) −3.90 (−12.00 to 4.20) Metabolic syndrome, No. (%)f Had metabolic syndrome at baseline 36 (11.8) 36 (11.8) but not at 12 mo Had metabolic syndrome at baseline and 12 mo 39 (12.8) 36 (11.8) Did not have metabolic syndrome at baseline 128 (42.0) 137 (45.1) or 12 mo Did not have metabolic syndrome at baseline 13 (4.3) 11 (3.6) but had metabolic syndrome at 12 mo Respiratory exchange ratiog −0.008 (−0.018 to 0.002) −0.027 (−0.037 to −0.018) 0.020 (0.006 to 0.033) Resting energy expenditure, kcalg −66.45 (−96.65 to −36.26) −76.93 (−106.68 to −47.19) 10.48 (−31.91 to 52.87) Energy expenditure, kcal/kg/d 0.55 (0.20 to 0.90) 0.49 (0.13 to 0.84) 0.06 (−0.44 to 0.56) SI conversion factors: To convert glucose to mmol/L, multiply by 0.0555; and 123 in the healthy low-carbohydrate diet group. This was due to high-density and low-density lipoprotein cholesterol to mg/dL, divide by a combination of dropout and not having any data for cohort 1. 0.0259; insulin to pmol/L, multiply by 6.945; triglycerides to mg/dL, e Indicates the blood concentration of insulin at the 30-minute time point of divide by 0.0113. an oral glucose tolerance test. a Data were missing for 91 participants in the healthy low-fat diet group and f Defined by Adult Treatment Panel III guidelines from the National Cholesterol 86 in the healthy low-carbohydrate diet group (almost exclusively due Education Program.26 to dropout). g There were missing data for 125 participants in the healthy low-fat diet group b Healthy low-fat diet minus healthy low-carbohydrate diet. and 121 in the healthy low-carbohydrate diet group due to a combination of c Calculated as weight in kilograms divided by height in meters squared. dropout and not having any data for cohort 1. d There were missing data for 138 participants in the healthy low-fat diet group

Secondary Outcomes Respiratory exchange ratio was not significantly differ- There were improvements in the secondary outcomes ent between the groups at baseline, but was lower for the for both diet groups. However, there were no significant healthy low-carbohydrate diet group than for the healthy between-group differences observed for body mass index, low-fat diet group at each time point after randomization body fat percentage, and waist circumference (Table 3). (P < .001; eTable 1 in Supplement 2). Resting energy expen- At 12 months relative to baseline, both diets improved lipid diture was not significantly different between groups at profiles and lowered blood pressure, insulin, and glucose lev- baseline or at 6 months or 12 months, but decreased sig- els, with the exception of low-density lipoprotein cholesterol nificantly from baseline in both diet groups. Total energy concentrations, which increased for participants in the healthy expenditure was not significantly different between groups low-carbohydrate group (Table 3). The 12-month changes in at baseline or any other time point. Relative to baseline, low-density lipoprotein cholesterol concentrations signifi- there was a small absolute mean increase in energy expen- cantly favored a healthy low-fat diet. High-density lipopro- diture for both diet groups that was not significantly differ- tein cholesterol concentrations increased significantly more ent than baseline. and concentrations of triglycerides decreased significantly more for the healthy low-carbohydrate diet group than for Adverse Events the healthy low-fat diet group. The decrease in the preva- During the trial, there were 7 serious adverse events, all lence of the metabolic syndrome was not significantly differ- requiring hospitalization; 2 of these could have been related ent between the diet groups. to the study (kidney stones and diverticulitis requiring

jama.com (Reprinted) JAMA February 20, 2018 Volume 319, Number 7 675

Downloaded From: by Roberta Kline on 02/21/2018 Research Original Investigation Low-Fat vs Low-Carbohydrate Diet on Weight Loss in Overweight Adults

Figure 2. Interaction Among Diet and Genotype and Diet and Insulin-30 Tertile at Baseline and 12-Month Weight Loss

A 12-mo Weight loss by diet and genotype No. of Participants Low-fat genotype Healthy low-fat diet 83 Healthy low-carbohydrate diet 70

Low-carbohydrate genotype Healthy low-fat diet 63 Healthy low-carbohydrate diet 81

Neither genotype Healthy low-fat diet 79 Healthy low-carbohydrate diet 60

–40 –30 –20 –10 0 10 20 12-mo Weight Change, kg

B 12-mo Weight loss by diet and insulin-30 tertile at baseline No. of Participants Lowest insulin-30 tertile Healthy low-fat diet 66 Healthy low-carbohydrate diet 85

Middle insulin-30 tertile Healthy low-fat diet 81 Healthy low-carbohydrate diet 71

Highest insulin-30 tertile Healthy low-fat diet 68 Healthy low-carbohydrate diet 64

–40 –30 –20 –10 0 10 20 12-mo Weight Change, kg

The black solid circle indicates the mean, the left and right borders of the box these 2 genotype patterns. By design, as described in the initial National mark the first and third quartiles, the black vertical line indicates the median, Institutes of Health grant application, those individuals with neither of the main the error bars indicate the 5th and 95th percentiles, and the hollow circles 2 genotype patterns were not included in the main analyses. There were 39 indicate the individuals whose values were outside the 5th or 95th percentiles. participants who had compromised or missing DNA. The No. of participants reflect data for the individuals who had weight data at B, Three-way interaction term among diet, insulin, and the 12-month time both baseline and 12 months. Statistical analyses include data from all point was not statistically significant (beta coefficient for 10-μIU/mL individuals randomized (described in the Statistical Analysis section). increase in insulin, 0.08 [95% CI, −0.13 to 0.28]; P = .47). Insulin-30 A, Three-way interaction term among diet, genotype, and the 12-month time is the blood concentration of insulin 30 minutes after consuming 75 g of glucose point was not statistically significant (beta coefficient, 1.38 [95% CI, −0.72 to as part of a standard oral glucose tolerance test. Insulin-30 was treated as a 3.49]; P = .20). As described in Stanton et al,10 of all the possible combinations continuous variable in the statistical model. Tertiles were used in this Figure for of variance in 3 single-nucleotide polymorphism multilocus genotype patterns, ease of presentation. The mean for the lowest tertile was 40.8 μIU/mL (range, some were considered consistent with the low-fat genotype pattern, 7.3-60.6 μIU/mL); middle, 80.1 μIU/mL (range, 60.7-103.1 μIU/mL); and highest, some with the low-carbohydrate genotype pattern, and some with neither of 159.6 μIU/mL (range, 103.4-562.5 μIU/mL).

surgery). There were 11 adverse events; 9 of these were there was no significant difference in weight loss at 12 related to the study or possibly related (eg, hypoglycemia months. In addition, there were no significant interactions following oral glucose tolerance test). Combined serious between diet and 3 SNP multilocus genotype patterns or adverse events and adverse events were evenly distributed diet and baseline insulin secretion on 12-month weight loss. across the 2 diet groups. These results were observed in the context of similar mean 12-month weight loss in both diet groups that was greater than 5% of baseline body weight, and a similar and substan- Discussion tial range of weight change, reflecting approximately 40 kg within each diet group (from losing approximately 30 kg to In this clinical trial of 609 generally healthy overweight or gaining approximately 10 kg). obese adults without diabetes who were randomly assigned Dietary intake of fats and carbohydrates was well differ- to a healthy low-fat vs a healthy low-carbohydrate diet, entiated between the 2 diet groups, as confirmed by diet

676 JAMA February 20, 2018 Volume 319, Number 7 (Reprinted) jama.com

Downloaded From: by Roberta Kline on 02/21/2018 Low-Fat vs Low-Carbohydrate Diet on Weight Loss in Overweight Adults Original Investigation Research

assessment, and corroborated by changes in blood lipid para- ticipants, Cornier et al11 observed a significant diet × fasting meters and respiratory exchange ratio, indicating strong treat- insulin interaction for weight loss. A post hoc analysis from ment fidelity. With the large sample size, good retention, the A TO Z Study revealed a significant diet × fasting insulin substantial weight loss and weight loss variability, and good interaction on 12-month weight loss among a subset of 81 adherence to and differentiation of diets, the study was overweight and obese women.14 well positioned to detect significant interactions by the pri- However, in a recent pilot study conducted in prepara- mary variables of interest if they existed. However, no such ef- tion for the DIETFITS study, a significant effect modification fects were observed. Differences in weight loss between the 2 was not detected for INS-30 status.35 In each case in which a groups were nonsignificant and not clinically meaningful. significant interaction was reported, investigators proposed a Among the secondary outcomes, the clinical variables mechanism involving insulin secretion status, insulin sensi- that were significantly different between the diet groups were tivity, or insulin resistance interacting with glycemic load to the blood lipid results, which were more favorable in the differentially affect weight loss response with low-fat diets high healthy low-fat diet group for changes in low-density lipopro- in carbohydrates vs high-fat diets low in carbohydrates.12,36 tein cholesterol and were more favorable in the healthy low- In these studies, the consistent direction of the finding was carbohydrate diet group for changes in high-density lipopro- that a lower carbohydrate diet was superior for those indi- tein cholesterol and triglycerides. The magnitude of the viduals with higher insulin secretion or higher insulin resis- between-group differences were 5% for low-density lipopro- tance; the putative mechanism involves a lower demand or bur- tein cholesterol, 5% for high-density lipoprotein cholesterol, den on insulin-mediated glucose disposal for those with and 15% for triglycerides. impaired insulin metabolism while maintaining a lower car- There is considerable scientific interest in identifying bohydrate and higher fat diet. Despite mechanistic plausibil- genetic variants that help explain interindividual differences ity, studies to date have involved relatively small sample sizes. in weight loss success in response to diet interventions,31,32 Effect modification claims observed in single random- particularly diets with varying macronutrient compositions. ized trials are often spurious and this result is even more fre- Multiple secondary analyses of low-fat and low-carbohydrate quent when small sample sizes and post hoc analyses are in- weight loss diet trials, including the Preventing Over- volved; validation of such claims is infrequent.37-39 The current weight Using Novel Dietary Strategies (POUNDS LOST) study with a larger sample, a low-carbohydrate diet that was and the Nutrient-Gene Interactions in Human Obesity also a low glycemic load diet, and using INS-30 could not rep- (NUGENOB) trials,8,32-34 have reported effect modification by licate findings from prior studies using smaller numbers of pa- SNPs on associations of dietary fat and carbohydrates with tients or those studies with a shorter duration. We consider the weight loss. differences between the current findings and the studies cited For example, Qi et al8 reported that individuals with the to potentially involve diet quality beyond simply differenti- IRS1 rs2943641 CC genotype were more successful with weight ating fat and carbohydrate intake. In this regard, refined grains loss than those without this genotype when assigned to are low in fat but considered of poor nutritional quality due a low-fat and high-carbohydrate diet vs a low-carbohydrate to low-nutrient density relative to energy content. In con- and high-fat diet. Grau et al32 reported that individuals with trast, vegetables are high in nutrient density, and relatively high the FTO rs9939609 TT genotype had greater decreases in the in proportional carbohydrate content, but low in calories. Both homeostatic assessment model of insulin resistance on low- diet groups in the current study were instructed to minimize fat vs low-carbohydrate diets; however, the diet-genotype in- or eliminate refined grains and added sugars and maximize interaction for weight loss was not statistically significant. Most take of vegetables. We conclude that when equal emphasis is prior studies examined single SNPs, with few replication at- given to high dietary quality for both low-fat and low- tempts. The intent in the current study was to replicate the post carbohydrate eating plans, it is not helpful to preferentially di- hoc findings from the A TO Z (Atkins, Traditional, Ornish, Zone) rect an individual with high insulin secretion status who is seek- Weight Loss Study.3 ing weight loss to follow a lower-carbohydrate eating plan The finding of no significant difference in weight loss in instead of a lower-fat eating plan. genotype-matched vs mismatched groups in the current study This study had several strengths. Study design strengths highlights the importance of conducting large, appropriately included the similarly intensive demands on both diet groups powered trials such as DIETFITS for validating early explor- in making changes to baseline diets, similar focus on dietary atory analyses. Analyses of all the genomic data obtained are quality, repeated major time points of data collection, and the under way to evaluate whether other genetic signatures may extensive range of types of data collected. Strengths in study demonstrate effect modification. conduct included meeting and exceeding the sample size tar- Several research groups previously reported observ- get of 600 participants, the nearly equal proportions of women ing a differential effect of low-fat vs low-carbohydrate diets and men enrolled, high and equivalent retention for both diet on weight loss by baseline insulin status. In both a 6-month groups, and comparability of change between groups in po- feeding study with 32 participants and an 18-month free- tentially important outcomes related to weight loss, such as living study with 56 participants, effect modification physical activity. In addition, the collective loss of approxi- between diet assignment (low-fat vs low-carbohydrate or low mately 3000 kg among study participants, and the wide indi- glycemic load) and INS-30 was reported.12,13 Using fasting vidual variability of weight loss, provided the opportunity to insulin cutoffs in a 4-month feeding study involving 20 par- meaningfully test for effect modification.

jama.com (Reprinted) JAMA February 20, 2018 Volume 319, Number 7 677

Downloaded From: by Roberta Kline on 02/21/2018 Research Original Investigation Low-Fat vs Low-Carbohydrate Diet on Weight Loss in Overweight Adults

Limitations participant burden were determined to be beyond the scope The study also has several limitations. First, generalizability of the study. In addition, self-reported diet assessment meth- of the findings may be limited by the conduct of the study in ods are all known to have limited accuracy; therefore, we chose a geographic area with individuals who have attained rela- to use the Nutrition Data System for Research, which is rec- tively high education levels, and have personal resources and ognized as a top method. high accessibility to high-quality food options. To address this, Fifth, even though insulin sensitivity was well assessed in the study was broadly advertised and successfully enrolled par- this study, assessment of genetic characteristics as effect modi- ticipants with relatively good ethnic and racial diversity, and fiers of diet response need better and increased study in the a range, albeit limited, of educational attainment. future because there has been much progress in understand- Second, in regard to the possible role of insulin-glucose dy- ing the genetic architecture of metabolic phenotypes such as namics as an effect modifier in low-fat vs low-carbohydrate obesity since the current trial was designed. Other explana- studies, there are many possible indices to consider other than tions for heterogeneity besides insulin dynamics and genetic INS-30,36 a proxy measure of insulin secretion selected for rea- characteristics also need to be assessed. sons described elsewhere.12,13 But others have reported find- Sixth, by not randomizing or conducting stratification ac- ing significant effect modification according to prestudy fast- cording to genotype or insulin secretion status, the level of ing insulin concentrations.11,14 causal inference to be drawn from the analyses of interac- Third, there were 3 missing secondary anthropometric tions was limited; however, this allowed us to test for 2 pri- and metabolic variables (percentage of body fat, resting mary interaction associations in the same study. energy expenditure, and respiratory exchange ratio) for the first 78 participants enrolled in the study due to inadequate initial funding. This funding situation subsequently changed Conclusions (described in eAppendix 1 in Supplement 2), which allowed the addition of these measurements for the remaining par- In this 12-month weight loss diet study, there was no signifi- ticipants enrolled. cant difference in weight change between a healthy low-fat diet Fourth, the Stanford 7-day Physical Activity Recall tool vs a healthy low-carbohydrate diet, and neither genotype pat- (which was used to determine total energy expenditure) pro- tern nor baseline insulin secretion was associated with the di- vides only a relatively crude assessment of total energy ex- etary effects on weight loss. In the context of these 2 com- penditure. Another method of measuring energy expendi- mon weight loss diet approaches, neither of the 2 hypothesized ture, such as the doubly labeled water method, would have predisposing factors was helpful in identifying which diet was provided greater accuracy; however, the overall cost and added better for whom.

ARTICLE INFORMATION Role of the Funder/Sponsor: The funders had no Benjamin Chrisinger, MUEP, PhD, and Michael Accepted for Publication: January 17, 2018. role in the design and conduct of the study; Stanton, PhD (aided various phases of the study). collection, management, analysis, and We also acknowledge the 609 study participants Author Contributions: Dr Gardner had full access interpretation of the data; preparation, review, or without whom this investigation would not to all of the data in the study and takes approval of the manuscript; and decision to submit have been possible. responsibility for the integrity of the data and the the manuscript for publication. accuracy of the data analysis. REFERENCES Concept and design: Gardner, Rigdon, Ioannidis, Disclaimer: The content is solely the responsibility Desai, King. of the authors and does not necessarily represent 1. Flegal KM, Kruszon-Moran D, Carroll MD, Fryar Acquisition, analysis, or interpretation of data: All the official views of the National Institutes of Health CD, Ogden CL. Trends in obesity among adults in authors. or the other funders. the United States, 2005 to 2014. JAMA. 2016;315 Drafting of the manuscript: Gardner, Trepanowski, Additional Contributions: This study would not (21):2284-2291. Del Gobbo, Hauser, Rigdon, Desai. have been possible without the work of the 2. Zylke JW, Bauchner H. The unrelenting challenge Critical revision of the manuscript for important following individuals who were affiliated with of obesity. JAMA. 2016;315(21):2277-2278. intellectual content: All authors. Stanford University at the time of the study and 3. Gardner CD, Kiazand A, Alhassan S, et al. Statistical analysis: Gardner, Del Gobbo, Hauser, who received compensation for their work: Jennifer Comparison of the Atkins, Zone, Ornish, and LEARN Rigdon, Ioannidis, Desai. Robinson, PhD, and Antonella Dewell, MS, RD diets for change in weight and related risk factors Obtained funding: Gardner. (served as study coordinators), Rise Cherin, MS, RD, among overweight premenopausal women: Administrative, technical, or material support: Susan Kirkpatrick, RD, CDE, Jae Berman, MS, RD, the A TO Z Weight Loss Study: a randomized trial. Gardner, Hauser, King. CSSD, Dalia Perelman, MS, RD, CDE, and Mandy JAMA. 2007;297(9):969-977. Supervision: Gardner, Desai, King. Murphy Carroll, MPH, RD (health educators), Sarah Farzinkhou, MPH, Valerie Alaimo, BS, Margaret 4. Sacks FM, Bray GA, Carey VJ, et al. Comparison Conflict of Interest Disclosures: The authors have of weight-loss diets with different compositions of completed and submitted the ICMJE Form for Shimer Lawton, MPH, and Diane Demis, BS (diet assessment team), Josephine Hau, MPH, RD, fat, protein, and carbohydrates. N Engl J Med. Disclosure of Potential Conflicts of Interest and 2009;360(9):859-873. none were reported. Erin Avery, MS, Alexandra Rossi, BS, Katherine Dotter, BS, RD, and Sarah Mummah, PhD 5. Shai I, Schwarzfuchs D, Henkin Y, et al; Dietary Funding/Support: This study was supported by (involved in recruitment, screening, blood Intervention Randomized Controlled Trial (DIRECT) grant 1R01DK091831 from the National Institute of sample management, innovation, and Group. Weight loss with a low-carbohydrate, Diabetes and Digestive and Kidney Diseases, grant other tasks), Ariadna Garcia, MS, FeiFei Qin, MPH, Mediterranean, or low-fat diet. N Engl J Med. 2008; T32HL007034 from the Nutrition Science and Vidhya Balasubramanian, MS (involved in 359(3):229-241. Initiative, grant 1K12GM088033 from the National statistical support), Alana Koehler, BA Heart, Lung, and Blood Institute, and the Stanford 6. Johnston BC, Kanters S, Bandayrel K, et al. (administrative support), and Lucia Aronica, PhD, Comparison of weight loss among named diet Clinical and Translational Science Award. Jennifer Hartle, DrPH, MHS, CIH, Lisa Offringa, PhD, Kenji Nagao, PhD, Marily Oppezzo, PhD, MS, RD,

678 JAMA February 20, 2018 Volume 319, Number 7 (Reprinted) jama.com

Downloaded From: by Roberta Kline on 02/21/2018 Low-Fat vs Low-Carbohydrate Diet on Weight Loss in Overweight Adults Original Investigation Research

programs in overweight and obese adults: 18. King AC, Frey-Hewitt B, Dreon DM, Wood PD. 30. Bates D, Mächler M, Bolker BM, Walker SC. a meta-analysis. JAMA. 2014;312(9):923-933. Diet vs exercise in weight maintenance: the effects Fitting linear mixed-effects models using lme4. 7. Field AE, Camargo CA Jr, Ogino S. The merits of of minimal intervention strategies on long-term J Stat Softw. 2015;67(1):1-48. subtyping obesity: one size does not fit all. JAMA. outcomes in men. Arch Intern Med. 1989;149(12): 31. Bray MS, Loos RJ, McCaffery JM, et al; 2013;310(20):2147-2148. 2741-2746. Conference Working Group. NIH working group 8. Qi Q, Bray GA, Smith SR, Hu FB, Sacks FM, Qi L. 19. Feskanich D, Sielaff BH, Chong K, Buzzard IM. report-using genomic information to guide weight Insulin receptor substrate 1 gene variation modifies Computerized collection and analysis of dietary management: from universal to precision insulin resistance response to weight-loss diets in a intake information. Comput Methods Programs treatment. Obesity (Silver Spring). 2016;24(1):14-22. 2-year randomized trial: the Preventing Overweight Biomed. 1989;30(1):47-57. 32. Grau K, Cauchi S, Holst C, et al. TCF7L2 Using Novel Dietary Strategies (POUNDS LOST) 20. Sallis JF, Haskell WL, Wood PD, et al. Physical rs7903146-macronutrient interaction in obese trial. Circulation. 2011;124(5):563-571. activity assessment methodology in the Five-City individuals’ responses to a 10-wk randomized 9. Dopler Nelson M, Prabakar P, Kondragunta V, Project. Am J Epidemiol. 1985;121(1):91-106. hypoenergetic diet. Am J Clin Nutr. 2010;91(2):472- Kornman KS, Gardner CD. Genetic phenotypes 21. Chiu KC, Martinez DS, Yoon C, Chuang LM. 479. predict weight loss success: the right diet does Relative contribution of insulin sensitivity and 33. Heianza Y, Ma W, Huang T, et al. Macronutrient matter. Paper presented at: joint conference of the beta-cell function to plasma glucose and insulin intake-associated FGF21 genotype modifies effects 50th Cardiovascular Disease Epidemiology and concentrations during the oral glucose tolerance of weight-loss diets on 2-year changes of central Prevention and Nutrition, Physical Activity, and test. Metabolism. 2002;51(1):115-120. adiposity and body composition: the POUNDS Lost Metabolism; March 2-3, 2010; San Francisco, CA. 22. Phillips DI, Clark PM, Hales CN, Osmond C. trial. Diabetes Care. 2016;39(11):1909-1914. 10. Stanton MV, Robinson JL, Kirkpatrick SM, et al. Understanding oral glucose tolerance: comparison 34. Qi Q, Bray GA, Hu FB, Sacks FM, Qi L. DIETFITS study (Diet Intervention Examining The of glucose or insulin measurements during the oral Weight-loss diets modify glucose-dependent Factors Interacting With Treatment Success): study glucose tolerance test with specific measurements insulinotropic polypeptide receptor rs2287019 design and methods. Contemp Clin Trials. 2017;53: of insulin resistance and insulin secretion. Diabet Med. genotype effects on changes in body weight, 151-161. 1994;11(3):286-292. fasting glucose, and insulin resistance: the 11. Cornier MA, Donahoo WT, Pereira R, et al. 23. Hron BM, Ebbeling CB, Feldman HA, Preventing Overweight Using Novel Dietary Insulin sensitivity determines the effectiveness of Ludwig DS. Relationship of insulin dynamics to Strategies trial. Am J Clin Nutr. 2012;95(2):506-513. dietary macronutrient composition on weight loss body composition and resting energy expenditure 35. Gardner CD, Offringa LC, Hartle JC, in obese women. Obes Res. 2005;13(4):703-709. following weight loss. Obesity (Silver Spring). 2015; Kapphahn K, Cherin R. Weight loss on low-fat 12. Ebbeling CB, Leidig MM, Feldman HA, Lovesky 23(11):2216-2222. vs low-carbohydrate diets by insulin resistance MM, Ludwig DS. Effects of a low-glycemic load vs 24. Chaput JP, Tremblay A, Rimm EB, Bouchard C, status among overweight adults and adults with low-fat diet in obese young adults: a randomized Ludwig DS. A novel interaction between dietary obesity: a randomized pilot trial. Obesity (Silver trial. JAMA. 2007;297(19):2092-2102. composition and insulin secretion: effects on Spring). 2016;24(1):79-86. 13. Pittas AG, Das SK, Hajduk CL, et al. weight gain in the Quebec Family Study. Am J Clin 36. Pittas AG, Roberts SB. Dietary composition A low-glycemic load diet facilitates greater weight Nutr. 2008;87(2):303-309. and weight loss: can we individualize dietary loss in overweight adults with high insulin secretion 25. Ludwig DS, Majzoub JA, Al-Zahrani A, Dallal GE, prescriptions according to insulin sensitivity but not in overweight adults with low insulin Blanco I, Roberts SB. High glycemic index foods, or secretion status? Nutr Rev. 2006;64(10 pt 1): secretion in the CALERIE Trial. Diabetes Care. 2005; overeating, and obesity. Pediatrics. 1999;103(3):E26. 435-448. 28(12):2939-2941. 26. Goni L, Cuervo M, Milagro FI, Martínez JA. 37. Sun X, Ioannidis JP, Agoritsas T, Alba AC, 14. McClain AD, Otten JJ, Hekler EB, Gardner CD. Future perspectives of personalized weight loss Guyatt G. How to use a subgroup analysis: Users’ Adherence to a low-fat vs low-carbohydrate diet interventions based on nutrigenetic, epigenetic, Guide to the Medical Literature. JAMA. 2014;311(4): differs by insulin resistance status. Diabetes Obes and metagenomic data. J Nutr. 2016;146(4):905S- 405-411. Metab. 2013;15(1):87-90. 912S. 38. Wallach JD, Sullivan PG, Trepanowski JF, 15. Physical Activity Guidelines Advisory 27. Diggle P. Analysis of Longitudinal Data. Oxford, Steyerberg EW, Ioannidis JPA. Sex based subgroup Committee, US Office of Disease Prevention England: Oxford University Press; 2002. differences in randomized controlled trials: and Health Promotion. 2008 physical activity empirical evidence from Cochrane meta-analyses. 28. Batterham MJ, Tapsell LC, Charlton KE. BMJ. 2016;355:i5826. guidelines for Americans. https://health.gov Analyzing weight loss intervention studies with /paguidelines/guidelines/. Accessed missing data: which methods should be used? 39. Brookes ST, Whitely E, Egger M, Smith GD, November 21, 2017. Nutrition. 2013;29(7-8):1024-1029. Mulheran PA, Peters TJ. Subgroup analyses in randomized trials: risks of subgroup-specific 16. Bandura A. Self-Efficacy: The Exercise of Control. 29. Kuznetsova A, Brockhoff PB, Christensen RHB. New York, NY: WH Freeman and Co; 1997. analyses; power and sample size for the interaction lmerTest: tests in linear mixed effects models. R test. J Clin Epidemiol. 2004;57(3):229-236. 17. Foreyt JP, Goodrick GK. Impact on behavior package version 2.0-33. https://cran.r-project.org/. therapy on weight loss. Am J Health Promot. 1994;8 Accessed January 10, 2018. (6):466-468.

jama.com (Reprinted) JAMA February 20, 2018 Volume 319, Number 7 679

Downloaded From: by Roberta Kline on 02/21/2018 Sugar and Fiber Content in Fruits

Amount Types of Sugar Dietary fiber

g/100 g grams/ 100 grams grams/100 grams

Fruit Total Glucose Fructose Sucrose Total Soluble Insoluble

Banana 19.6 26.5 30.1 43.4 1.6 0.3 1.3 Apple 11.0 10.9 54.5 34.5 2.2 0.6 1.6 Pineapple 10.6 21.7 13.2 65.0 1.2 0.3 0.9 Grapes 9.3 51.6 46.2 15.4 0.7 0.2 0.5 Pears 8.7 28.7 57.4 13.8 2.6 0.4 2.2 Orange 8.5 29.4 21.2 49.4 2.4 1.5 0.9 Peach 7.5 13.3 13.3 73.3 2.6 0.8 1.8 Mango 7.4 ------100 2.3 NA NA Blackberry 7.0 45.7 41.4 12.8 2.3 0.1 2.2 Cantaloupe 6.5 16.9 20.0 63.0 0.8 0.3 0.5 Apricot 6.1 29.5 13.1 57.4 1.9 NA NA Grapefruit 5.2 36.5 23.1 40.4 0.6 0.3 0.3 Strawberry 4.5 37.8 40.0 22.2 2.6 1.2 1.4 Raspberry 4.3 20.9 30.2 48.8 4.5 NA NA Honeydew 4.1 ------100 0.9 NA NA Tangerine 3.8 ------100 2.0 NA NA Watermelon 3.2 ------100 0.4 0.3 0.1

12 oz Cola 11.0 ------100 0 0 0

Adapted from Foods and Nutrition Encyclopedia, Ensminger et al., (Vol 1: 341-342, 1983), by Joe Veltmann Jr. PhD, FAAIM DCCN, GENESIS Center for Integrative Medicine.

Clinical Thyroidology / Original Paper

Eur Thyroid J 2017;6:143–151 Received: December 5, 2016 DOI: 10.1159/000469709 Accepted after revision: March 7, 2017 Published online: April 24, 2017

Hypothyroid Patients Encoding Combined MCT10 and DIO2 Gene Polymorphisms May Prefer L-T3 + L-T4 Combination Treatment – Data Using a Blind, Randomized, Clinical Study

a c, d b a Allan Carlé Jens Faber Rudi Steffensen Peter Laurberg c, d Birte Nygaard a b Department of Endocrinology, Aalborg University Hospital, Aalborg, Denmark; Department of Clinical c Immunology, Aalborg University Hospital, Aalborg, Denmark; Department of Endocrinology, Herlev Hospital, d University of Copenhagen, Copenhagen, Denmark; Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark

Keywords (rs17606253), and asked in which of the 2 treatment periods Hypothyroidism · Thyroid failure · Single nucleotide patients felt better (i.e., which treatment was preferred). Re- polymorphisms · L-T3 treatment · Randomized clinical trial sults: 27 out of 45 patients (60%) preferred the combination therapy. Two polymorphisms (rs225014 (DIO2, Thr92Ala) and rs17606253 (MCT10)) were combined yielding 3 groups: Abstract none vs. 1 of 2 vs. both SNPs present, and 42 vs. 63 vs. 100% Objectives: In previous studies, around half of all hypothy- of our patients in the 3 groups preferred the combined treat- roid patients preferred levo-thyroxine (L-T4) + levo-triiodo- ment (Jongheere-Terpstra trend test, p = 0.009). Conclusion: thyronine (L-T3) combination therapy, 25% preferred T4, The present study indicates that the combination of poly- and 25% had no preference. The reason for this is yet to be morphisms in DIO2 (rs225014) and MCT10 (rs17606253) en- explored. Methods: A total of 45 overtly autoimmune, hypo- hances hypothyroid patients’ preference for L-T4 + L-T3 re- thyroid patients – now euthyroid on ≥6 months’ L-T4 thera- placement therapy. In the future, combination therapy may py – participated in a prospective, double-blind, cross-over be restricted or may be even recommended to individuals study. The patients were randomized into 2 groups of either harbouring certain polymorphisms. 3 continuous months’ L-T4 therapy followed by 3 months’ © 2017 European Thyroid Association combination therapy or vice versa. In all periods, 50 μg L-T4 Published by S. Karger AG, Basel was blindly replaced by either (identical) 50 μg L-T4 or by 20 μg T3. L-T4 was hereafter adjusted to obtain normal serum Introduction TSH values. We investigated 3 single nucleotide polymorphisms (SNPs) on the type II iodothyronine deiodinase Hypothyroidism is very common and affects up to (DIO2) gene (rs225014 (Thr92Ala), rs225015, and rs12885300 10% of the female population [1]. Many patients do not (ORFa-Gly3Asp)) and 1 SNP on the cellular membrane trans- feel well on standard levo-thyroxine (L-T4) treatment [2– port-facilitating monocarboxylate transporter (MCT10) gene 5], and many patients may benefit from experimental

© 2017 European Thyroid Association Allan Carlé, MD, PhD Published by S. Karger AG, Basel Department of Endocrinology Aalborg University Hospital E-Mail [email protected] DK–9000 Aalborg (Denmark) www.karger.com/etj E-Mail carle @ dadlnet.dk treatment with combined levo-triiodothyronine (L-T3) + tients’ preference for one therapy form over the other L-T4 preparations [6]. In the 2012 European Thyroid As- may be explained by local intracerebral factors, again per- sociation guidelines for the use of L-T4 and L-T3 in the haps influenced by different gene expressions of DIO2 treatment of hypothyroidism, Wiersinga et al. [6] re- and/or the thyroid hormone transporters. Several gene viewed 5 cross-over studies [7–11] and showed that 48% polymorphisms may contribute to the understanding of of all hypothyroid patients preferred the L-T3 + L-T4 the different experiences reported by hypothyroid pa- combination therapy, 27% preferred L-T4 monotherapy, tients. and 25% had no preference. No study has ever explained We investigated 3 polymorphisms in the DIO2 genes: why so many patients prefer L-T3. In the meta-analysis rs225014 known as Thr92Ala, rs22501, and rs12885300 by Grozinsky-Glasberg et al. [12], no difference was ob- originally entitled D2 ORFa-Gly3Asp, which may affect served in symptom scores or various serum lipid profiles the intracerebral T4 to T3 conversion. We also investi- in those treated with L-T3 + L-T4 versus those treated gated the role of the MCT10 rs17606253 polymorphism, with L-T4 alone. Almost 30 years ago, Carr et al. [13] re- which may affect the transport of thyroid hormones into ported better well-being in patients treated with a supra- the brain. MCT10 is a T-type aromatic amino acid trans- optimal dose of 50 μg L-T4. However, they had all serum porter known both as solute carrier family 16 aromatic TSH values of <0.2 mU/L, and were thus slightly over- amino acid transporter member 10 (SLC16A10) and also treated. known as T-type amino acid transporter (TAT1). MCT10 Several other studies, in which patients were treated is a member of a family of plasma membrane amino acid with a high dose of thyroid hormone with subsequently transporters that mediate the Na(+)-independent trans- low serum TSH values, support these findings [9, 14]. On port of aromatic amino acids across the plasma mem- the other hand, no differences in hypothyroid symptoms, brane. However, MCT10 has also been found to transport quality of life, or cognitive function were seen in a double- thyroid hormones across membranes [18]. The MCT10 blind, randomized, clinical trial with cross-over design on polymorphism has not been investigated as extensively as 52, L-T4-treated, hypothyroid patients receiving 3 L-T4 the MCT8 polymorphism. However, the MCT8 and doses in random order for 8 weeks: usual dose, 25 μg less, MCT10 transporters have similar structures [19], wide or 25 μg more resulting in serum TSH values of 2.8 ± 0.4, tissue distributions [20], and facilitate both the uptake 1.1 ± 0.2, and 0.3 ± 0.1 mU/L, respectively (p < 0.001). and efflux of T3 and T4 [21]. MCT10 has been shown to Walsh et al. [7] also studied hypothyroid patients with transport T3 better than the MCT8 transporter [22]. To biochemically restored euthyroidism due to L-T4 substi- our knowledge, no other study has ever elaborated wheth- tution and found that the group of patients dissatisfied er polymorphisms in the MCT10 gene may explain why with their treatment had the same serum TSH, T3, and T4 some patients have better treatment response when L-T3 concentrations as the patients having no complaints. A is combined with the classic L-T4 treatment. different co-morbidity status could perhaps explain some of the difference [7]. Weight loss may also be part of the explanation [2, 14–16], but although statistically signifi- Subjects and Methods cant, the weight loss in L-T3 + L-T4-treated patients (vs. standard L-T4 monotherapy) was shown to be only very The present study is based on patients previously recruited to a Danish, randomized, double-blind, cross-over study [8]. For the modest [12]. present study, the original patient group was re-invited to a blood Differential local expression and activity of iodothyro- test for genetic testing. nine deoidinases and thyroid hormone transmembrane transporters may be an explanation. Panicker et al. [17] Sample Population showed that hypothyroid patients harbouring the CC The precise criteria for selecting 59 patients in the original study have been previously outlined [8]. In short, patients diag- genotype of the type II iodothyronine deiodinases (DIO2) nosed with overt autoimmune hypothyroidism (TSH values >20 rs225014 polymorphism, seen in 16% in a population- mU/L, serum T4 <60 nmol/L, TPOAb >60 mU/L, no other evident based study from the UK, had worse baseline general nosological type of thyroid failure than spontaneous hypothyroid- health scores and did also benefit more on L-T3 + L-T4 ism [1]), between 18 and 76 years of age, who had been treated with combination therapy compared to patients harbouring L-T4 substitution dose, and with a 6-month stable serum TSH value (range of 0.1–5.0 mU/L, determined on morning blood samples the wild type (WT) alleles. However, the serum T3 and T4 before taking thyroid hormone substitution) were included in the levels did not differ between WT and CC polymorphism original study. Exclusion criteria were a previous L-T3 treatment patients in that study. This indicates that perhaps pa- or a current or planned pregnancy.

144 Eur Thyroid J 2017;6:143–151 Carlé/Faber/Steffensen/Laurberg/Nygaard DOI: 10.1159/000469709 As previously described in details [8], patients were block ran- for trend testing. The Hardy-Weinberg equation was tested under domized into 1 of 2 treatment arms. One group of patients had 50 the assumption of an additive model, with a null hypothesis of no μg of their usual L-T4 dose replaced by 20 μg L-T3 for a 12-week linkage and no association [24]. A p value <0.05 was considered period and were then shifted into their usual L-T4 dose for an- statistically significant. Minor allele frequencies (MAF) were cal- other 12 weeks. The other group initially continued on their usual culated according to standard procedures. dose – blind by a “50 to 50 μg L-T4 shift” – for 12 weeks, after that, 50 μg of their usual L-T4 dose was replaced by 20 μg L-T3 for an- Ethical Approval other 12 weeks. As previously stated [8], within the 12-week peri- The project was accepted by the Danish Medicines Agency (No. od, the L-T4 doses were further adjusted to obtain stable serum 2612-1939), the Danish National Committee on Biochemical Re- TSH values between 0.1 and 5.0 mU/L. search Ethics (No. KA02022ms and H-1-22011-145), and the Dan- ish Data Protection Agency (No. 2002-41-2236). Questionnaires Participants filled out a questionnaire obtaining information on their general well-being (Short Form Survey, SF-36). They were asked for the treatment period in which they felt better. The results Results were blinded until the time of statistical analyses. Then, it was revealed whether they preferred the combined L-T3 + L-T4 treatment, L-T4 monotherapy alone, or had no preference. Of the 59 patients, who originally attended the double- blind, randomized, clinical study [8], 45 participants ac- Single Nucleotide Polymorphism Genotyping cepted the invitation for a second visit and blood test for All patients from the original study were re-invited to a blood test for genetic testing. Of those 59, 45 patients gave blood for sin- genotyping. Of these, 27 patients (60%) preferred the com- gle nucleotide polymorphism (SNP) genotyping. Genomic DNA bined L-T3 + L-T4 treatment, 7 patients preferred the L-T4 was extracted from EDTA-treated peripheral blood using the monotherapy, whereas 11 patients had no preference. The QiaSymphony SP System (Qiagen, Hilden, Germany). The prede- ® baseline characteristics and serum concentrations of TSH, signed TaqMan SNP Genotyping Assays C_15819951_10, T3, and T4 during the study are depicted in Table 1. A C_568127_10, C_31755153_30, and C_33173970_10 (Applied Biosystems, Foster City, CA, USA) were used to investigate the comparison between those preferring the combined L-T3 4 SNPs – of which 3 are located in the DIO2 gene (rs225014 + L-T4 treatment with the rest of the patients revealed no (Thr92Ala), rs225015, and rs12885300 (ORFa-Gly3Asp)), and 1 is difference in any of the parameters investigated. In par- located in the MCT10 gene (rs17606253). Allelic discrimination ticular, no difference was observed in serum TSH, T3, or plots clearly distinguish between WT, heterozygous polymor- T4 levels between those 2 groups – a finding consistent at phism alleles, and homozygous polymorphism alleles [23]. the time of diagnosis, at the time of randomization which Serum TSH, T3 and T4 Analyses was performed when patients had stable thyroid function Serum levels of TSH and peripheral thyronines (T3 and T4) during standard L-T4 treatment, and during both the L-T4 were all measured on blood samples drawn in the morning. The ® monotherapy and the combined L-T3 + L-T4 treatment. DPC Immulite 2500 (Siemens, DPC, Bad Nauheim, Germany) However, both groups of patients had significantly higher was used for measuring serum TSH levels (normal range 0.4–4.0 mU/L, inter- and intra-assay coefficients of variation (CVs) of 5%), T3 (2.40 vs. 1.60 nmol/L, p < 0.001, Wilcoxon test) and total T3 (normal range 1.0–2.6 nmol/L, CVs 4%), and total T4 lower T4 (75 vs. 126 nmol/L, p < 0.001) but equal serum (normal range 60–140 nmol/L, CVs 5%). The total T3 and T4 con- TSH levels (0.80 vs. 1.06 mU/L, p = 0.38) during the period centrations will be referred to as T3 and T4 in this paper, respec- in which they were treated with L-T3 + L-T4 combination tively. Conversion factors from SI (nmol/l) to American (ng/dl) therapy versus L-T4 monotherapy. units are ×0.0778 for total T4 and ×0.0643 for total T3. TPOAb (normal range <60 kU/L, CVs 4%) was measured by TPOAb- The distribution of the alleles of the 3 DIO2 SNPs DYNO test (previously BRAHMS, now Thermo Fisher). (rs225014, rs225015, and rs12885300) and the MCT10 gene (rs17606253) is shown in Table 2. No deviation from Statistical Analysis the Hardy-Weinberg equation was observed. The poly- We used IBM Statistical Package for Social Sciences version morphisms investigated were rather common, judged by 15.0 (SPSS, Chicago, IL, USA) for statistical analyses. Means and standard error of means (SEM) or when appropriate medians and their MAF varying between 17 and 47%. interquartile range (25 to 75% range) were calculated. The Fisher The association between the 4 gene polymorphisms exact t test, independent sample t test (Gaussian-distributed val- and the patients’ preference for L-T3 + L-T4 combination ues), Mann-Whitney U test (no Gaussian distribution), and Wil- versus L-T4 monotherapy is depicted in Figure 1. Calcu- coxon test (no Gaussian distribution, dependent test between 2 lations were performed under the assumption of an addi- groups) were used for intergroup comparisons. Odds ratios (OR) with 95% confidence intervals was used to describe the association tive model comparing WT (no polymorphism alleles) between the various SNP distributions and L-T3 treatment prefer- with non-WT patients (1 or 2 polymorphism alleles pres- ences. Linear by linear test and Jongheere-Terpstra test were used ent). Patients preferring the combination therapy had a

Hypothyroidism, MCT10 and DIO2 Gene Eur Thyroid J 2017;6:143–151 145 Polymorphisms, and L-T3 + L-T4 DOI: 10.1159/000469709 Table 1. Patient characteristics according to treatment preference (L-T3 + L-T4 vs. L-T4) of hypothyroid patients (n = 45)

Characteristics All L-T3 preference No L-T3 preference p value (n = 27) (n = 18)

At diagnosis Women/MenA 42/3 25/2 17/1 0.65D Age, yearsA 46.7±1.62 45.6±2.21 48.4±2.35 0.39E Serum TSH, mU/LB, C 58 (35–100) 69 (39–135) 50 (30–82) 0.22F Serum TPOAb, kU/LA, C 1,142 (233–3,000) 1,271 (206–3,000) 916 (225–2,110) 0.46F Before randomization BMIA 25.9±0.54 26.4±0.80 25.1±0.57 0.20E Height, cmA 170±1.01 171±1.46 170±1.30 0.55E Weight, kgA 75.1±1.68 76.9±2.47 72.3±1.87 0.35E Serum TSH, mU/LA 1.12 (0.63–2.22) 1.65 (0.55–2.85) 0.95 (0.77–1.34) 0.15F Serum T3, nmol/L 1.65 (1.30–2.00) 1.75 (1.28–2.00) 1.55 (1.28–1.93) 0.50F Serum T4, nmol/L 125 (101–149) 127 (103–153) 124 (89–152) 0.73F T3/T4 ratio, ×10–3 12.9 (11.3–15.3) 13.6 (11.6–15.5) 12.2 (10.4–14.9) 0.24F On L-T3 + L-T4 therapy Weight, kgA 74.8±1.73 76.7±2.56 72.0±1.88 0.36E Serum TSH, mU/L 0.80 (0.31–1.75) 0.86 (0.37–2.14) 0.65 (0.24–1.70) 0.39F Serum T3, nmol/L 2.40 (1.78–3.20) 2.80 (1.75–3.25) 2.30 (1.75–3.05) 0.84F Serum T4, nmol/L 74.5 (54.3–101) 71 (56–105) 83 (51–104) 0.85F T3/T4 ratio, ×10–3 32.2 (23.9–44.6) 28.6 (23.0–47.7) 33.8 (26.4–40.8) 0.65F On L-T4 monotherapy Weight, kgA 75.7±1.83 77.5±2.70 73.0±2.05 0.41E Serum TSH, mU/L 1.06 (0.66–2.00) 1.22 (0.68–2.10) 0.98 (0.58–1.95) 0.30F Serum T3, nmol/L 1.60 (1.30–1.98) 1.75 (1.45–2.30) 1.55 (1.30–1.90) 0.94F Serum T4, nmol/L 126 (96–149) 120 (97–149) 130 (92–150) 0.28F T3/T4 ratio, ×10–3 13.3 (11.7–15.1) 13.6 (12.5–15.2) 12.3 (10.6–15.1) 0.13F

Values are mean ± SEM or median (interquartile range) unless indicated otherwise. A Information at randomization. B Information when hypothyroidism was diagnosed. C Missing information on TSH at diagnosis (n = 11) and TPOAb (n = 1), T3 at inclusion/after L-T3 combination therapy (after L-T4 monotherapy (n = 1/3/1), T4 at inclusion (n = 2/3/1), weight at inclusion (n = 14). Tests for comparisons: D Fisher exact t test; E independent sample t test (Gaussian distributed values), F Mann-Whitney U test (no Gaussian distribution). Conversion factors from SI (nmol/L) to American (ng/dL) units are ×0.0778 for total T4 and ×0.0643 for total T3.

Table 2. Distribution of iodothyronine deiodinase (DIO2) and 6.40 (1.06–49.7) higher odds for harbouring at least 1 monocarboxylate transporter (MCT10) single nucleotide poly- polymorphism in the rs17606253 (MCT10) gene (p = morphisms (SNPs) in hypothyroid patients (n = 45) 0.018). Similarly, patients who felt better during L-T3 +

SNPs, n Hetero- Homo- Hardy- L-T4 combination therapy had a 2.80 (0.66–12.3) border- (wild type) zygous, n zygous, n Weinberg line higher odds for harbouring at least 1 polymorphism (wild type; (double equation in the rs225014 (DIO2) gene (p = 0.11). Collinearity was 1 SNP) SNP) observed between the 2 DIO2 polymorphisms rs225014 DIO2 and rs12885300 (p < 0.001). rs225014 26 (TT) 18 (CT) 1 (CC) 0.29 Accordingly to standard procedures [25, 26], we com- rs225015 23 (GG) 18 (AG) 4 (AA) 0.86 bined the 2 gene polymorphisms with the highest OR sug- rs12885300 14 (CC) 20 (CT) 11 (TT) 0.47 gesting an association with L-T3 treatment preference, rs225011 41 (TT) 6 (GT) 0 (GG) namely the rs17606253 (MCT10) and the rs225014 MCT10 rs17606253 31 (TT) 13 (CT) 1 (CC) 0.79 (DIO2) polymorphism. Patients were classified into 3 groups: those with no polymorphism in neither of the 2 genes (WT in both genes, “double” WT, n = 19), those

146 Eur Thyroid J 2017;6:143–151 Carlé/Faber/Steffensen/Laurberg/Nygaard DOI: 10.1159/000469709 rs17606253 (MCT10) p = 0.011A/p = 0.009B

100 rs225014 (DIO2)

rs225015 (DIO2) 80

rs12885300 (DIO2)

60 0.1 0.5 1 5 10 50 Odds ratio for T3 treatment preference

40 Fig. 1. Odds ratios (OR) for harbouring various single nucleotide polymorphisms (SNPs) in case of favouring the combined L-T3 + L-T4 over standard L-T4 treatment in hypothyroid patients (n = 45). OR for the 4 SNPs investigated were: 6.40 (1.06–49.7), p = % T3 + T4 treatment, favouring Patients 20 0.018 for rs17606253 (MCT10); 2.80 (0.66–12.3), p = 0.11 for rs225014 (DIO2); 1.96 (0.50–7.94), p = 0.27 for rs225015 (DIO2); and 0.49 (0.10–2.24), p = 0.29 for rs12885300 (DIO2). Test for collinearity: rs12885300 vs. rs225014, p < 0.001; all other combina- 0 tions, p > 0.05. None One of two Both Number of SNPs (rs225014 and rs17606253 combined) with a polymorphism in only 1 of the 2 genes (heterozy- Fig. 2. Patients’ preference for the combined L-T3 + L-T4 treat- gous as well as homozygous, n = 19), and those harbour- ment over no preference was depicted according to their MCT10 ing polymorphisms in both the rs17606253 (MCT10) and and DIO2 gene polymorphisms. Three groups of patients were the rs225014 (DIO2) gene (n = 7). In the group of “dou- compared: those with wild-type nucleotides on both rs225014 ble” WT patients with no polymorphism present, 42.1% (DIO2) and rs17606253 (MCT10) genes, those with 1 of the 2 pos- preferred the combination L-T3 + L-T4 therapy (Fig. 2). sible polymorphisms, and those with both polymorphisms present (number of patients in the 3 groups were 19, 19, and 7). In the 3 The corresponding prevalences were 63.2 and 100% in groups 8, 12, and 7 patients (42.1, 63.2, 100%) preferred the com- the groups with polymorphisms present in 1 and 2 genes. bined L-T3 + L-T4 treatment. Linear by linear test (A) and Jong- Trend tests indicated a statistically significant dose-re- heere-Terpstra test (B) were used for testing the dose-response as- sponse pattern (p values between 0.009 and 0.011). sociation.

Weight Changes The 45 patients included in the present study had an average weight (mean ± SEM) of 75.1 kg (± 1.68) at the Discussion time of randomization i.e. after half a year with normal thyroid status (Table 1). Only modest and statistically In this randomized, clinical, cross-over trial encom- nonsignificant weight differences were observed between passing hypothyroid patients treated with L-T4 mono- the various groups: a 0.78 (± 0.23) kg higher weight when therapy or with L-T3 + L-T4 combination therapy, we treated with L-T4 versus L-T3 + L-T4 combination, a 0.95 demonstrated that subjects harbouring the combination (± 0.28) kg higher weight on L-T4 only (vs. on L-T3 + of 2 polymorphisms (rs225014 encoding the DIO2 en- L-T4) in those preferring L-T3. The last finding was not zyme and rs12885300 encoding the MCT10 transporter) statistically different (p = 0.28) from the modest weight were more prone to benefit from the experimental L-T3 differences observed in patients not preferring L-T3 who + L-T4 combination therapy. This is the first study show- weighed 0.51 (± 0.38) kg more on L-T4 monotherapy ing that the MCT10 polymorphism is associated with a compared to combination therapy. preference for combined L-T3 + L-T4 treatment.

Hypothyroidism, MCT10 and DIO2 Gene Eur Thyroid J 2017;6:143–151 147 Polymorphisms, and L-T3 + L-T4 DOI: 10.1159/000469709 Treatment Preferences – Well-Being phisms in the MCT8 gene. However, Friesema et al. [18] In the 2005 study by Appelhof et al. [27] encompassing demonstrated that MCT10 is at least as active as MCT8 in hypothyroid patients on a stable L-T4 dose for more than terms of in- and efflux of thyroid hormones with a prefer- 6 months, no differences in well-being, neurocognitive ence for L-T3 over L-T4. MCT10 is expressed in many effect, or treatment preference were found between sub- organs – also in the brain. It is also expressed in the small jects harbouring different rs225014 or rs12885300 poly- intestine [18], but our study does not indicate that pa- morphisms (both DIO2 SNPs also investigated in the tients with MCT10 polymorphisms may have different present study). However, the authors apparently did not intestinal absorption with a subsequent lower hormone try to combine any of the genes under investigation in level in blood. their statistical work-up. DIO2 is a cytosolic protein. It catalyzes the conversion In the 2009 study by Panicker et al. [17], 16 type I, II, of thyroxine (3,5,3 ,5 -tetraiodothyronine, T4) to the bio- and III deiodinase gene polymorphisms were investigat- logically more active thyroid hormone (3,5,3 -triiodothy- ed. Patients with SNPs in the 3 DIO2 genes investigated ronine, T3) by outer′ ring′ 5 -deiodination. The DIO2 gene (rs225011, rs225014, and rs225015) felt better when treat- is widely expressed in the thyroid, placenta, pituitary,′ and ed with the combination therapy compared to the stan- brain [42]. Thus, it is responsible′ for the “local” produc- dard L-T4 treatment [17]. However, all findings were tion of T3 in brain tissue. Theoretically, the T3 concentra- borderline significant with p values ranging from 0.02 to tion in brain cells may depend more on the MCT10 and 0.06, and the authors concluded that the results should be DIO2 activity than on the T3 or T4 concentrations in verified. This is the first study to replicate their findings. blood. A recently published paper has reviewed various stud- Thyroid Hormone Concentrations ies, and it was concluded that an overall discrepancy be- The results on serum T3 and T4 concentrations in med- tween the symptom burden and thyroid hormone levels ically substituted hypothyroid patients harbouring differ- in overt hypothyroidism was present [43]. This was con- ent blood-brain transmembrane and intracerebral iodo- firmed in a study comprising patients newly diagnosed thyronine deiodinase polymorphisms are highly inconsis- with overt autoimmune hypothyroidism [44]. It is impor- tent. Almost all studies have shown identical serum TSH tant to remember that the concentration of thyroid hor- and thyroid hormone concentrations in hypothyroid mones in serum may not necessarily indicate the hor- patients harbouring different DIO2 polymorphism in mone status of all individual tissues. rs225014 [17, 25–37] and rs12885300 [27, 28, 30]. How- ever, a few exceptions have also been published [25, 38, 39]. Implication The role of various MCT10 polymorphisms is still in Whereas the expression of the DIO1, the PDE8B, and its research infancy, and more studies have been advo- to some extent also the thyroid hormone receptor may cated for [19]. A few studies have demonstrated different have a pivotal role in determining the serum T3 to T4 bal- serum thyroid hormone concentrations in subjects har- ance [40, 45], the results from the present study and oth- bouring different rs17706253 (MCT10) polymorphisms ers suggest that DIO2 and MCT activity may play a role [19, 40], but the findings are not consistent [19, 41]. in patients’ response to thyroid hormone treatment. We have reported that the presence of combined rs17606253 Body Weight (MCT10) and rs225014 (DIO2) gene polymorphisms As already mentioned, the small changes in body may predict better outcome if hypothyroid patients are weight did not explain why some patients felt better on treated with the combined L-T3 + L-T4 regime. the experimental L-T3 + L-T4 combination therapy. This Perhaps it will be possible in the future to individually is in good line with the findings from the Grozinsky-Glas- tailor a specific treatment based on a patient’s genetic berg meta-analysis [12] which only showed very modest composition. Patients harbouring certain polymor- differences in body weight. phisms may need additional L-T3 treatment to obtain better outcome. In the future, more polymorphisms will Pathophysiological Mechanisms Involved be studied. The study by Taylor et al. [46] indicated that This is the first study to evaluate the role of MCT10 uncommon polymorphisms (MAF <1%) may have a very polymorphisms on the clinical L-T3 + L-T4 treatment re- large impact on the associations studied. sponse in hypothyroid patients. The MCT10 polymor- The relative importance of the thyroid hormone trans- phism is not yet as extensively investigated as polymor- porter in species may differ substantially [47] and so may

148 Eur Thyroid J 2017;6:143–151 Carlé/Faber/Steffensen/Laurberg/Nygaard DOI: 10.1159/000469709 their relative expression in various tissues and at different have complaints despite biochemically restored euthy- points of time during human life [48]. It has been claimed roidism. Some of these patients may seek help from new that the OATP1C1 gene is relatively less expressed in hu- treatment modalities. Thus, the 45 patients included in mans compared to other species and restricted to very the present study may be just a small and highly selected specific areas of the human brain. The present study may fraction of patients diagnosed with hypothyroidism. indicate that MCT10 expression plays a large role in adult human life. Conclusion

Limitations We found that the combination of 2 polymorphisms in MCT10 (rs17606253) and DIO2 (rs225014) enhances A limitation of the present study and others searching hypothyroid patients’ preference for L-T3 + L-T4 re- for associations between SNPs and subject characteristics placement therapy. Hopefully, these findings will be re- is the possibility of introducing false positive conclusions tested and perhaps also confirmed in future studies. If so, (type I error) [49]. We found p values around 1% for the it may be possible in the future to genetically test and se- association between the 2 combined polymorphisms and lect those hypothyroid patients who may benefit from the treatment preference. However, this is just below the combined L-T3 + L-T4 treatment. By such an approach, Bonferroni adjusted p value of 1.3% (p = 1 – 4√0.95 [50]). treatment of hypothyroidism may become individually In observational studies, an association does not nec- tailored to fit all individuals harbouring different genetic essarily imply causation. However, several of the Brad- compositions. ford-Hill criteria are met [51]. Firstly, the dose-response criterion is met (Fig. 2). Secondly, the biological mechanism for the finding is plausible as impaired MCT10-fa- Disclosure Statement cilitated thyroid hormone influx may influence brain me- The authors have no conflicts of interest to disclose. The fund- tabolism and may cause symptoms. Thirdly, an OR of 6 ing received had no impact on study design and methods, data col- (for the MCT10 polymorphism) indicates a rather strong lection and analysis, or manuscript preparation. association, and interestingly positive findings were achieved with only 45 patients included in the study. Last, the fact that 2 previous studies [17, 27] have reported sim- Author Contributions ilar findings studying 3 of the potential 4 associations between DIO2 polymorphisms and preference of the L-T3 A.C., B.N., and P.L. had full access to all of the data in the study and take responsibility for the integrity of the data and the accu- combination therapy also supports our findings. racy of the data analysis. Study concept and design: all authors. The present study may hamper from selection bias. Of Data acquisition, analysis, or interpretation: all authors. Manu- all hypothyroid patients treated in a population, some still script preparation: all authors.

References

1 Carlé A, Laurberg P, Pedersen IB, Knudsen N, 4 Wekking EM, Appelhof BC, Fliers E, Schene 7 Walsh JP, Shiels L, Lim EM, Bhagat CI, Ward Perrild H, Ovesen L, Rasmussen LB, Jor- AH, Huyser J, Tijssen JG, Wiersinga WM: LC, Stuckey BG, Dhaliwal SS, Chew GT, Bha- gensen T: Epidemiology of subtypes of hypo- Cognitive functioning and well-being in eu- gat MC, Cussons AJ: Combined thyroxine/ thyroidism in Denmark. Eur J Endocrinol thyroid patients on thyroxine replacement liothyronine treatment does not improve

2006;154: 21–28. therapy for primary hypothyroidism. Eur J well-being, quality of life, or cognitive func-

2 Saravanan P, Chau WF, Roberts N, Vedhara Endocrinol 2005;153: 747–753. tion compared to thyroxine alone: a random- K, Greenwood R, Dayan CM: Psychological 5 Walsh JP: Dissatisfaction with thyroxine ther- ized controlled trial in patients with primary well-being in patients on “adequate” doses of apy – could the patients be right? Curr Opin hypothyroidism. J Clin Endocrinol Metab

L-thyroxine: results of a large, controlled Pharmacol 2002;2: 717–722. 2003;88: 4543–4550. community-based questionnaire study. Clin 6 Wiersinga WM, Duntas L, Fadeyev V, 8 Nygaard B, Jensen EW, Kvetny J, Jarlov A, Fa-

Endocrinol (Oxf) 2002;57: 577–585. Nygaard B, Vanderpump MP: 2012 ETA ber J: Effect of combination therapy with thy- 3 Ladenson PW: Psychological wellbeing in pa- Guidelines: the use of L-T4 + L-T3 in the roxine (T4) and 3,5,3 -triiodothyronine versus

tients. Clin Endocrinol (Oxf) 2002;57: 575– treatment of hypothyroidism. Eur Thyroid J T4 monotherapy in patients with hypothyroid-

576. 2012;1: 55–71. ism, a double-blind,′ randomised cross-over

study. Eur J Endocrinol 2009;161: 895–902.

Hypothyroidism, MCT10 and DIO2 Gene Eur Thyroid J 2017;6:143–151 149 Polymorphisms, and L-T3 + L-T4 DOI: 10.1159/000469709 9 Bunevicius R, Kazanavicius G, Zalinkevicius 19 van der Deure WM, Peeters RP, Visser TJ: Ge- 30 Panicker V, Cluett C, Shields B, Murray A, R, Prange AJ Jr: Effects of thyroxine as com- netic variation in thyroid hormone transport- Parnell KS, Perry JR, Weedon MN, Singleton pared with thyroxine plus triiodothyronine in ers. Best Pract Res Clin Endocrinol Metab A, Hernandez D, Evans J, Durant C, Ferrucci

patients with hypothyroidism. N Engl J Med 2007;21: 339–350. L, Melzer D, Saravanan P, Visser TJ, Ceresini

1999;340: 424–429. 20 Dayan CM, Panicker V: Novel insights into G, Hattersley AT, Vaidya B, Dayan CM, Fray- 10 Escobar-Morreale HF, Botella-Carretero JI, thyroid hormones from the study of common ling TM: A common variation in deiodinase 1

Gomez-Bueno M, Galan JM, Barrios V, San- genetic variation. Nat Rev Endocrinol 2009;5: gene DIO1 is associated with the relative lev- cho J: Thyroid hormone replacement therapy 211–218. els of free thyroxine and triiodothyronine. J

in primary hypothyroidism: a randomized 21 Nishimura M, Naito S: Tissue-specific mRNA Clin Endocrinol Metab 2008;93: 3075–3081. trial comparing L-thyroxine plus liothyro- expression profiles of human solute carrier 31 Mentuccia D, Thomas MJ, Coppotelli G, Re- nine with L-thyroxine alone. Ann Intern Med transporter superfamilies. Drug Metab Phar- inhart LJ, Mitchell BD, Shuldiner AR, Celi FS:

2005;142: 412–424. macokinet 2008;23: 22–44. The Thr92Ala deiodinase type 2 (DIO2) vari- 11 Bunevicius R, Jakuboniene N, Jurkevicius R, 22 Visser TJ: Cellular uptake of thyroid hor- ant is not associated with type 2 diabetes or Cernicat J, Lasas L, Prange AJ Jr: Thyroxine vs mones; in: De Groot LJ, Chrousos G, Dungan indices of insulin resistance in the old order of

thyroxine plus triiodothyronine in treatment K, Feingold KR, Grossman A, Hershman JM, Amish. Thyroid 2005;15: 1223–1227. of hypothyroidism after thyroidectomy for Koch C, Korbonits M, McLachlan R, New M, 32 Torlontano M, Durante C, Torrente I, Cro-

Graves’ disease. Endocrine 2002;18: 129–133. Purnell J, Rebar R, Singer F, Vinik A (eds): cetti U, Augello G, Ronga G, Montesano T, 12 Grozinsky-Glasberg S, Fraser A, Nahshoni E, Endotext [Internet]. South Dartmouth, Travascio L, Verrienti A, Bruno R, Santini S, Weizman A, Leibovici L: Thyroxine-triiodo- MDText.com, Inc., 2000– (last update: June D’Arcangelo P, Dallapiccola B, Filetti S, thyronine combination therapy versus thy- 15, 2016). Trischitta V: Type 2 deiodinase polymor- roxine monotherapy for clinical hypothyroid- 23 Steffensen R, Baech J, Nielsen KR: Allelic dis- phism (threonine 92 alanine) predicts L-thy- ism: meta-analysis of randomized controlled crimination by TaqMan-PCR for genotyping roxine dose to achieve target thyrotropin lev-

trials. J Clin Endocrinol Metab 2006;91: 2592– of human neutrophil antigens. Methods Mol els in thyroidectomized patients. J Clin Endo-

2599. Biol 2015;1310: 205–212. crinol Metab 2008;93: 910–913. 13 Carr D, McLeod DT, Parry G, Thornes HM: 24 Rodriguez S, Gaunt TR, Day IN: Hardy- 33 Butler PW, Smith SM, Linderman JD, Brych- Fine adjustment of thyroxine replacement Weinberg equilibrium testing of biological as- ta RJ, Alberobello AT, Dubaz OM, Luzon JA, dosage: comparison of the thyrotrophin re- certainment for Mendelian randomization Skarulis MC, Cochran CS, Wesley RA, Pucino

leasing hormone test using a sensitive thyro- studies. Am J Epidemiol 2009;169: 505–514. F, Celi FS: The Thr92Ala 5 type 2 deiodinase trophin assay with measurement of free thy- 25 Peeters RP, van den Beld AW, Attalki H, Toor gene polymorphism is associated with a de- roid hormones and clinical assessment. Clin H, de Rijke YB, Kuiper GG, Lamberts SW, layed triiodothyronine secretion′ in response

Endocrinol (Oxf) 1988;28: 325–333. Janssen JA, Uitterlinden AG, Visser TJ: A new to the thyrotropin-releasing hormone-stimu- 14 Appelhof BC, Fliers E, Wekking EM, Schene polymorphism in the type II deiodinase gene lation test: a pharmacogenomic study. Thy-

AH, Huyser J, Tijssen JG, Endert E, van Weert is associated with circulating thyroid hor- roid 2010;20: 1407–1412. HC, Wiersinga WM: Combined therapy with mone parameters. Am J Physiol Endocrinol 34 Fadeyev VV, Madiyarova M, Morgunova: levothyroxine and liothyronine in two ratios, Metab 2005;289:E75–E81. Evaluation the association of polymor- compared with levothyroxine monotherapy 26 Peeters RP, van Toor H, Klootwijk W, de phismes in DIO1 and DIO2 genes with men- in primary hypothyroidism: a double-blind, Rijke YB, Kuiper GG, Uitterlinden AG, Visser tal, biochemical parameters and quality of life randomized, controlled clinical trial. J Clin TJ: Polymorphisms in thyroid hormone path- in women with primary hypothyroidism (ab-

Endocrinol Metab 2005;90: 2666–2674. way genes are associated with plasma TSH stract OP77). 38th Annual Meeting of the 15 Celi FS, Zemskova M, Linderman JD, Smith and iodothyronine levels in healthy subjects. European Thyroid Association. Santiago de

S, Drinkard B, Sachdev V, Skarulis MC, Koz J Clin Endocrinol Metab 2003;88: 2880–2888. Compostela, September 2014. losky M, Csako G, Costello R, Pucino F: Met- 27 Appelhof BC, Peeters RP, Wiersinga WM, 35 Heemstra KA, Hoftijzer HC, van der Deure abolic effects of liothyronine therapy in hypo- Visser TJ, Wekking EM, Huyser J, Schene WM, Peeters RP, Fliers E, Appelhof BC, thyroidism: a randomized, double-blind, AH, Tijssen JG, Hoogendijk WJ, Fliers E: Wiersinga WM, Corssmit EP, Visser TJ, Smit crossover trial of liothyronine versus levothy- Polymorphisms in type 2 deiodinase are not JW: Thr92Ala polymorphism in the type 2 de-

roxine. J Clin Endocrinol Metab 2011;96: associated with well-being, neurocognitive iodinase is not associated with T4 dose in 3466–3474. functioning, and preference for combined athyroid patients or patients with Hashimoto

16 Hoang TD, Olsen CH, Mai VQ, Clyde PW, thyroxine/3,5,3 -triiodothyronine therapy. J thyroiditis. Clin Endocrinol (Oxf) 2009;71:

Shakir MK: Desiccated thyroid extract com- Clin Endocrinol Metab 2005;90: 6296–6299. 279–283. pared with levothyroxine in the treatment of 28 de Jong FJ, Peeters′ RP, den Heijer T, van der 36 Roef G, Taes Y, Toye K, Goemaere S, Fiers T, hypothyroidism: a randomized, double- Deure WM, Hofman A, Uitterlinden AG, Verstraete A, Kaufman JM: Heredity and life- blind, crossover study. J Clin Endocrinol Visser TJ, Breteler MM: The association of style in the determination of between-subject

Metab 2013;98: 1982–1990. polymorphisms in the type 1 and 2 deiodinase variation in thyroid hormone levels in euthy-

17 Panicker V, Saravanan P, Vaidya B, Evans J, genes with circulating thyroid hormone pa- roid men. Eur J Endocrinol 2013;169: 835– Hattersley AT, Frayling TM, Dayan CM: rameters and atrophy of the medial temporal 844.

Common variation in the DIO2 gene predicts lobe. J Clin Endocrinol Metab 2007;92: 636– 37 Wouters HJ, van Loon HC, van der Klauw baseline psychological well-being and re- 640. MM, Elderson MF, Slagter SN, Muller KA, sponse to combination thyroxine plus triio- 29 Gumieniak O, Perlstein TS, Williams JS, Hop- Kema IP, Links TP, Vliet-Ostaptchouk JV, dothyronine therapy in hypothyroid patients. kins PN, Brown NJ, Raby BA, Williams GH: Wolffenbuttel BH: No effect of the Thr92Ala

J Clin Endocrinol Metab 2009;94: 1623–1629. Ala92 type 2 deiodinase allele increases risk polymorphism of deiodinase-2 on thyroid 18 Friesema EC, Jansen J, Jachtenberg JW, Viss- for the development of hypertension. Hyper- hormone parameters, health-related quality

er WE, Kester MH, Visser TJ: Effective cellu- tension 2007;49: 461–466. of life, and cognitive functioning in a large lar uptake and efflux of thyroid hormone by population-based cohort study. Thyroid

human monocarboxylate transporter 10. Mol 2017;27: 147–155.

Endocrinol 2008;22: 1357–1369.

150 Eur Thyroid J 2017;6:143–151 Carlé/Faber/Steffensen/Laurberg/Nygaard DOI: 10.1159/000469709 38 Yalakanti D, Dolia PB: Association of type II single nucleotide polymorphisms in thyroid association analysis of 68 thyroid hormone 5 monodeiodinase Thr92Ala single nucleo- hormone transporter genes (MCT8, MCT10 pathway genes with serum TSH and FT4 lev-

tide gene polymorphism and circulating thy- and OATP1C1) and circulating thyroid hor- els. Eur J Endocrinol 2011;164: 781–788.

roid′ hormones among type 2 diabetes melli- mones. Clin Chim Acta 2013;425: 227–232. 46 Taylor PN, Porcu E, Chew S, et al: Whole-ge-

tus patients. Indian J Clin Biochem 2016;31: 42 Bianco AC, Salvatore D, Gereben B, Berry MJ, nome sequence-based analysis of thyroid

152–161. Larsen PR: Biochemistry, cellular and molec- function. Nat Commun 2015;6: 5681. 39 Peltsverger MY, Butler PW, Alberobello AT, ular biology, and physiological roles of the io- 47 Wirth EK, Schweizer U, Kohrle J: Transport Smith S, Guevara Y, Dubaz OM, Luzon JA, dothyronine selenodeiodinases. Endocr Rev of thyroid hormone in brain. Front Endocri-

Linderman J, Celi FS: The –258A/G (SNP 2002;23: 38–89. nol (Lausanne) 2014;5: 98. rs12885300) polymorphism of the human 43 Carlé A, Pedersen IB, Knudsen N, Perrild H, 48 Friesema EC, Visser TJ, Borgers AJ, Kalsbeek type 2 deiodinase gene is associated with a Ovesen L, Andersen S, Laurberg P: Hypothy- A, Swaab DF, Fliers E, Alkemade A: Thyroid shift in the pattern of secretion of thyroid hor- roid symptoms fail to predict thyroid insuffi- hormone transporters and deiodinases in the mones following a TRH-induced acute rise in ciency in old people: a population-based case- developing human hypothalamus. Eur J En-

TSH. Eur J Endocrinol 2012;166: 839–845. control study. Am J Med 2016;129: 1082– docrinol 2012;167: 379–386. 40 Medici M, van der Deure WM, Verbiest M, 1092. 49 Wacholder S, Chanock S, Garcia-Closas M, El Vermeulen SH, Hansen PS, Kiemeney LA, 44 Carlé A, Pedersen IB, Knudsen N, Perrild H, Ghormli L, Rothman N: Assessing the prob- Hermus AR, Breteler MM, Hofman A, Hege- Ovesen L, Laurberg P: Hypothyroid symp- ability that a positive report is false: an ap- dus L, Kyvik KO, den Heijer M, Uitterlinden toms and the likelihood of overt thyroid fail- proach for molecular epidemiology studies.

AG, Visser TJ, Peeters RP: A large-scale asso- ure: a population-based case-control study. J Natl Cancer Inst 2004;96: 434–442.

ciation analysis of 68 thyroid hormone path- Eur J Endocrinol 2014;171: 593–602. 50 Rothman KJ, Greenland S: Modern Epidemi- way genes with serum TSH and FT4 levels. 45 Medici M, van der Deure WM, Verbiest M, ology, ed 2. Philadelphia, Lippincott-Raven

Eur J Endocrinol 2011;164: 781–788. Vermeulen SH, Hansen PS, Kiemeney LA, Publishers, 1988. 41 Roef GL, Rietzschel ER, De Meyer T, Bekaert Hermus AR, Breteler MM, Hofman A, He- 51 Hill AB: The environment and disease: asso-

S, De Buyzere ML, Van daele C, Toye K, gedüs L, Kyvik KO, den Heijer M, Uitterlin- ciation or causation? Proc R Soc Med 1965; 58: Kaufman JM, Taes YE: Associations between den AG, Visser TJ, Peeters RP: A large-scale 295–300.

Hypothyroidism, MCT10 and DIO2 Gene Eur Thyroid J 2017;6:143–151 151 Polymorphisms, and L-T3 + L-T4 DOI: 10.1159/000469709 ORIGINAL ARTICLE

Endocrine Care

Common Variation in the DIO2 Gene Predicts Baseline Psychological Well-Being and Response to Combination Thyroxine Plus Triiodothyronine Therapy in Hypothyroid Patients

Vijay Panicker, Ponnusamy Saravanan, Bijay Vaidya, Jonathan Evans, Andrew T. Hattersley, Timothy M. Frayling, and Colin M. Dayan

Henry Wellcome Laboratories for Integrative Neurosciences and Endocrinology (V.P., P.S., C.M.D.), University of Bristol BS1 3NY, Bristol, United Kingdom; Faculty of Medicine, Dentistry and Health Sciences (V.P.), University of Western Australia, Crawley, Western Australia 6009, Australia; Clinical Sciences Research Institute (P.S.), University of Warwick, Coventry CV4 7AL, United Kingdom; Department of Endocrinology (B.V.), Royal Devon and Exeter Hospital, Exeter EX2 5DW, United Kingdom; Academic Unit of Psychiatry (J.E.), University of Bristol, Bristol BS6 6JL, United Kingdom; and Genetics of Complex Traits (A.T.H., T.M.F.), Institute of Biomedical and Clinical Science, Peninsula Medical School, Exeter EX1 2LU, United Kingdom

Introduction: Animal studies suggest that up to 80% of intracellular T3 in the brain is derived from

circulating T4 by local deiodination. We hypothesized that in patients on T4 common variants in the deiodinase genes might influence baseline psychological well-being and any improvement on

combined T4/T3 without necessarily affecting serum thyroid hormone levels. Methods: We analyzed common variants in the three deiodinase genes vs. baseline psychological

morbidity and response to T4/T3 in 552 subjects on T4 from the Weston Area T4 T3 Study (WATTS). Primary outcome was improvement in psychological well-being assessed by the General Health Questionnaire 12 (GHQ-12).

Results: The rarer CC genotype of the rs225014 polymorphism in the deiodinase 2 gene (DIO2) was present in 16% of the study population and was associated with worse baseline GHQ scores in ϭ patients on T4 (CC vs. TT genotype: 14.1 vs. 12.8, P 0.03). In addition, this genotype showed greater

improvement on T4/T3 therapy compared with T4 only by 2.3 GHQ points at 3 months and 1.4 at 12 months (P ϭ 0.03 for repeated measures ANOVA). This polymorphism had no impact on circulating thyroid hormone levels.

Conclusions: Our results require replication but suggest that commonly inherited variation in the

DIO2 gene is associated both with impaired baseline psychological well-being on T4 and enhanced

response to combination T4/T3 therapy, but did not affect serum thyroid hormone levels. (J Clin Endocrinol Metab 94: 1623–1629, 2009)

p to 3% of the population in Western countries is on thyroid levels of thyroid hormone (T4 and T3) by replacement with T4 hormone replacement (1), the majority on T alone. How- alone (2, 3) or T alone (4). In humans, patients on T mono- U 4 3 4 ever, the adequacy of this to replace physiological requirements therapy have a significantly higher serum T4 to T3 ratio for a and reverse patients’ symptoms remains controversial due to similar TSH than people with normal thyroid function (5, 6). several observations. In thyroidectomized rats, Escobar-Morre- Some markers of thyroid hormone action, such as IGF-1 may not

ale et al. reported that it was not possible to normalize tissue normalize on T4 monotherapy (7). Finally, a significant number

ISSN Print 0021-972X ISSN Online 1945-7197 Abbreviations: C, Cytosine; fT3, free T3; fT4, free T4; GHQ-12, General Health Question- Printed in U.S.A. naire, 12-question version; HAD, Hospital Anxiety and Depression Scale questionnaire; Copyright © 2009 by The Endocrine Society SNP, single-nucleotide polymorphism; T, thymine; TSQ, thyroid symptom questionnaire; doi: 10.1210/jc.2008-1301 Received June 16, 2008. Accepted January 26, 2009. WATTS, Weston Area T4/T3 Study. First Published Online February 3, 2009

For editorial see page 1521

J Clin Endocrinol Metab, May 2009, 94(5):1623–1629 jcem.endojournals.org 1623

Downloaded from https://academic.oup.com/jcem/article-abstract/94/5/1623/2598196 by guest on 22 March 2018 1624 Panicker et al. DIO2 Is Associated with Well-Being on T4 J Clin Endocrinol Metab, May 2009, 94(5):1623–1629

of patients report persistent symptoms despite titration of T4 lect a set of SNPs that capture the majority of common variation across the replacement to adequate serum levels of thyroid hormone and three deiodinase genes (DIO1, DIO2, and DIO3) including 50 kb either normalization of TSH levels (8, 9). These observations led to the side of the genes. We used a minor allele frequency of at least 10%. The 21, seven, and seven SNPs in the DIO1,DIO2, and DIO3 genes required nine, proposition that combination therapy with both T4 and the T3 four, and six SNPs, respectively, to capture all common variants with might be more effective. However, although in the initial study an r2 Ͼ 0.8. These were: D1, rs11206237, rs11206244, rs2235544,

of combination therapy (T4 and T3), patients treated with the rs2268181, rs2294511, rs2294512, rs4926616, rs731828, and combination appeared to have improved well-being (10), 10 sub- rs7527713; D2, rs12885300, rs225011, rs225014, and rs225015; sequent larger studies failed to confirm this benefit (11–20) and and D3, rs1190716, rs17716499, rs7150269, rs8011440, rs945006, and rs1190715. We used only SNPs that were in Hardy Weinberg equi- in at least one case demonstrated a large and sustained placebo librium (P Ͼ 0.05) and were genotyped in at least 97.5% of the samples effect (18). Metaanalysis of these trials (21) showed no benefit in the final analyses. We examined the association between these SNPs from combination therapy and a carefully controlled study of and baseline (before randomisation) psychological well-being. Genotyping was performed by KBiosciences (http://www.kbioscience.co. overreplacement with T4 also failed to show benefit (22). Doubt remains though because an excess psychological morbidity uk) using their own novel fluorescence-based competitive allele-specific PCR (KASPar). Assays were designed for each of the 19 SNPs by among patients on T4 has been documented in three separate KBiosciences. Design of an assay for the SNP rs1190715 was not large community-based studies (9, 23, 24), and also anecdotally possible and two further SNPs (rs1190716 and rs12885300) failed there remain patients who feel better on combination therapy. quality control. The percentage of duplicate samples included was One possibility that might resolve these issues is the existence 20% and concordance between duplicate samples was 99% or of a subgroup of patients who require combination therapy due greater. Use of these 16 SNPs resulted in coverage of 100, 85, and 71% of the common (minor allele frequency Ͼ 10%), HapMap based, to an inherited abnormality. If this group represented less than variation in DIO1, DIO2, and DIO3, respectively, at r2 Ͼ 0.8. 20% of the population, such patients may be too infrequent for their presence to be detected in the intervention trials but could Psychological assessments still account for significant morbidity in patients on T4. The three At each visit, the patient’s psychological well-being was assessed by iodothyronine deiodinases represent possible candidate loci for the following self-report scales: the General Health Questionnaire, such genetic variation, because these are responsible for the in- 12-question version (GHQ-12) (26, 27), a disease-specific thyroid symptom questionnaire (TSQ) (9), and the Hospital Anxiety and Depression terconversion of T4 and T3 (25). The Weston Area T4/T3 Study is the largest study of thyroid hormone replacement yet con- Scale questionnaire (HAD) (28). In addition, patients completed a sat- ϭ isfaction question on a 5-point scale on the second and third visits. The ducted (n 697) (18). We have taken advantage of the greater GHQ-12 and TSQ were scored by both linear (Likert method) and cat- statistical power of this study to explore the role of common egorical methods (score Ն3 by GHQ method) (26, 27). HAD is divided polymorphisms in the three deiodinase genes in determining psy- into seven questions for anxiety (HAD-A) and seven questions for depression (HAD-D) all scored 0–3. HAD anxiety or depression caseness chological well-being and the response to partial T3 replacement. was defined as a total score from the seven questions 8 or greater, which has been shown to provide the best sensitivity and specificity for case finding (28). Subjects and Methods For response to T4/T3 or T4-only therapy, improvement in GHQ-12 score was used as the primary end point as this was the endpoint that the Sample population original study was powered to measure (18). All other measures were analyzed as secondary end points. Subjects were 552 people in the Weston Area T4/T3 Study (WATTS) who had DNA available for genotyping (total study participants ϭ 697). The study design has been previously described (18), but briefly, subjects Statistical analysis ␮ on a stable dose of T4 therapy 100 g or more per day were recruited from To ensure there were no significant differences between the two study 28 primary care practices in the Weston-superMare and Bristol areas of groups, baseline characteristic means were compared by ANOVA and 2 the United Kingdom and randomized to either combination T4/T3 ther- proportions by ␹ test. Initial analysis of the relationship between psy- ␮ ␮ apy (original dose minus 50 gofT4 and added 10 gT3) or original dose chological well-being and genotype at baseline was performed by linear of T4 alone. Biochemical, physical, and psychological assessments were regression, with total GHQ Likert score as the dependent variable and made at baseline and 3 and 12 months. The trial was double blinded and genotype as the independent variable, with each allele considered addi- results analyzed on an intention-to-treat basis. The study protocol was tive. Logistic regression was also used with GHQ, HAD-D, or HAD-A approved by the local research ethics committee. caseness as the dependent variable. For response to therapy repeated- measures ANOVA was used with the scales, which were normally dis- Biochemical methods tributed (total GHQ Likert score, total TSQ Likert score, and satisfaction question score). The total number of missing values was low (Ͻ5%). To Serum TSH and free T4 (fT4) were measured from a serum sample by ensure this did not create error missing values were imputed using the RIA (Diagnostic Product Corp., Los Angeles, CA). Free T3 (fT3) was measured by chemiluminescence assay (Elecsys system 1010; Roche Di- missing values analysis function on SPSS (Chicago, IL), using regression agnostics, Mannheim, Germany). The laboratory reference ranges were: methods to estimate values and adding a random regression residual. TSH, 0.3–4.0 mU/liter; fT4, 10–24 pmol/liter; and fT3, 2.8–7.1 pmol/ This did not significantly change the original estimates, which are dis- liter. Coefficients of variation were: TSH, 5.5–8.0%; fT4, 7.7–10.0%; played in the results section. For GHQ and TSQ, the scores at 3 and 12 and fT3, 11.7–12.6%. months were the repeated measures, the study treatment arm, and genotype were the between-subject effects and baseline (before randomization) score was adjusted for as a covariate. The repeated-measures Tag single-nucleotide polymorphism (SNP) selection analysis includes two-way interactions between study treatment arm and and genotyping genotype and genotype and baseline score. For rs225014, genotype was We used genotype data from the Caucasian European individuals in the analyzed as both additive (each allele increases the association) and with International Haplotype Mapping Project (http://www.hapmap.org) to se- the T-allele as dominant (TT and TC combined vs. CC) as a dominant

Downloaded from https://academic.oup.com/jcem/article-abstract/94/5/1623/2598196 by guest on 22 March 2018 J Clin Endocrinol Metab, May 2009, 94(5):1623–1629 jcem.endojournals.org 1625

A – GHQ *p = 0.03 Genotype and psychological well-being T T T C C C at baseline 15 15 15 The relationship between the 16 SNPs 14 14 14 covering the three deiodinase genes and 13 13 13 baseline psychological well-being is shown 12 12 12 in Table 2. Two SNPs in the DIO2 gene; 11 11 11 rs225014 and rs225015 showed an associ- 10 10 10 ation at the P Ͻ 0.05 level of significance. By 9 9 9 contrast, the other SNPs analyzed, including 1 2 3 1 2 3 1 2 3 all of those from DIO1 and DIO3 did not Visit Visit Visit show any association. Because the two D2 B – TSQ *p = 0.03 SNPs that had showed an association, T T TC C C rs225014 and rs225015, are in strong link- 18 18 18 age disequilibrium with an r2 of 0.88 in this 17 17 17 population, and rs225014 is also in linkage 16 16 16 disequilibrium with the third SNP studied in 15 15 15 this gene, rs225011 (r2 of 0.59), further 14 14 14 analysis is shown on rs225014 alone. In this 13 13 13 SNP the possible base combinations of thy- 12 12 12 mine (T) and cytosine (C) are TT, TC and 11 11 11 CC. For GHQ-12, each C allele of rs225014 1 2 3 1 2 3 1 2 3 Visit Visit Visit was associated with an average increase of C – Satisfaction *p = 0.02 0.71 GHQ points (worse well-being, P for the trend ϭ 0.02) with a difference between T T TC C C the CC and TT alleles of 1.3 points. 3.5 3.5 3.5 Table 3 shows the relationship between 3.4 3.4 3.4 rs225014 genotype and baseline psycholog- 3.3 3.3 3.3 ical well-being for other parameters measured in WATTS. The scores for GHQ-12 3.2 3.2 3.2 from Table 2 are included for comparison. 3.1 3.1 3.1 An association with HAD-D (depression) 3.0 3.0 3.0 caseness in the same direction as GHQ was 1 2 3 1 2 3 1 2 3 seen (P ϭ 0.01). Each C allele was associated Visit Visit Visit with a 49% increase in odds of being a FIG. 1. Response to therapy by genotype rs225014 as measured by GHQ (A), TSQ (B), and satisfaction HAD-D case (P ϭ 0.01) and as a result case-

score (C). Squares and continuous line,T4/T3 group; triangles and dashed line,T4-only group. P values ness was almost twice as great in subjects reflect the significance of an effect of the CC genotype on difference in scores by treatment arm using homozygous for the CC genotype as in sub- repeated-measures ANOVA. *, P Ͻ 0.05. There was a significant effect of the interaction between the CC genotype and treatment arm at follow-up (visits 2 and 3) on GHQ scores, TSQ scores, and satisfaction. jects with the TT genotype (24 vs. 13%). No There are no baseline (visit 1) scores for satisfaction with therapy (C) because this was not assessed at significant differences were seen in the other baseline. For GHQ and TSQ scores, higher scores indicate worse well-being, whereas for satisfaction higher psychological scores, although all the scores scores indicate more satisfied. appeared to increase in the same direction across the genotypes (Table 3), with the TT effect was suggested by the graphs (Fig. 1) and has been proposed pre- genotype having the lowest score and TC intermediate and CC viously (29). For satisfaction score, no baseline score was adjusted for because there was no baseline assessment. No correction was made for the worst score. We published previously that rs225014 did not multiple testing because, despite being the largest study to date, it is still have any detectable effect on baseline thyroid function in this underpowered to detect all but very large differential gene-treatment cohort, and hence, this effect appears to be independent of serum effects. Instead, we have chosen to report the P values and associations, thyroid hormone levels (30). which should be considered suggestive, and have qualified our findings by stating clearly that the results need replicating as a risk of type I statistical error exists. Analyses were performed on Stata version 9.0 Genotype and response to therapy (www.stata.com) and SPSS version 14.0 (www.spss.com). Results of repeated-measures ANOVA for response to treatment by genotype and treatment arm for rs225014 are shown in Fig. 1. P values indicate an effect of an interaction between treat- Results ment arm (T4/T3 vs. T4) and the CC genotype on mean scores at both follow-up visits. Note the higher baseline scores for GHQ Descriptive statistics of the two groups are displayed in Table 1. in the CC genotype as in Table 2. As described in the initial report The treatment groups were not significantly different at baseline (18), both treatments resulted in an improvement from baseline in any of the parameters studied. consistent with a strong placebo effect. However, when analyzed

Downloaded from https://academic.oup.com/jcem/article-abstract/94/5/1623/2598196 by guest on 22 March 2018 1626 Panicker et al. DIO2 Is Associated with Well-Being on T4 J Clin Endocrinol Metab, May 2009, 94(5):1623–1629

TABLE 1. Baseline characteristics of sample population

a T4/T3 treatment T4 only P n 270 282 Sex (percent female) 82.2 84.4 0.49 Age (yr), mean 56.7 Ϯ 11.1 57.6 Ϯ 10.5 0.35 BMI (kg/m2) 29.0 Ϯ 5.8 29.5 Ϯ 6.3 0.37 Serum TSH, median (interquartile range) (mU/liter) 1.00 (0.30, 2.15) 0.86 (0.27, 2.03) 0.92 fT4 (mean Ϯ SD) (pmol/liter) 21.2 Ϯ 3.6 21.0 Ϯ 3.7 0.58 fT3 (mean Ϯ SD) (pmol/liter) 3.85 Ϯ 0.69 3.84 Ϯ 0.75 0.81 Positive TPOAb 53.2% 57.1% 0.35 GHQ case 42.0% 42.9% 0.94 HAD-D case 19.6% 16.3% 0.31 HAD-A case 45.2% 42.9% 0.59 TSQ case 60.0% 65.3% 0.20

Characteristics compared between two study arms by ANOVA or ␹2 test as appropriate. TPOAb, Anti-thyroid peroxidase antibody. a ANOVA or ␹2 test as appropriate.

by dominant effects (P values represent interaction between provement in mean HAD-D or HAD-A scores in either of the treatment arm and presence of CC genotype), there was a sig- study groups. HAD-D and HAD-A cases improved on treatment nificant interaction between treatment arm and genotype on im- in both treatment arms, with no difference in improvement due provement in GHQ (P ϭ 0.03), TSQ (P ϭ 0.03), and satisfaction to genotype (P for trend all Ͼ0.20). scores (P ϭ 0.02) (Fig. 1). This suggests an improved response to The rs225014 genotype frequencies were not significantly combination therapy in this genotype, the same genotype that different between the two study groups (frequency of TT, TC,

had the poorest psychological well-being at baseline (on T4 and CC genotypes: 40.6, 45.5, and 13.9%, respectively, in T4/T3 ϭ only). In addition, when analyzed as an additive model, trends group and 41.1, 41.1, and 17.9% in T4 only group, P 0.38). toward an interaction between treatment arm and genotype on improvement in GHQ (P ϭ 0.07), TSQ (P ϭ 0.06), and satis- Genotype and thyroid function in response to treatment faction scores (P ϭ 0.06) were also seen. In the CC genotype, the Although no significant differences were seen in baseline thyroid mean fall in GHQ score was 2.33 (95% confidence interval function when analyzed by genotype across the whole cohort (30),

0.38–4.38) points greater with T4/T3 than with T4 only at 3 it remained possible that serum levels might alter differently by Ϫ months and 1.44 ( 0.25 to 3.12) points greater at 12 months. genotype in response to T4/T3 treatment. Table 4 shows thyroid No difference was seen in response to the two treatments for the function by genotype for each of the treatment groups at the three subjects with the other genotypes. time points. No significant difference in thyroid function by geno-

By contrast with the above effects, rs225014 genotype did not type was seen after T4/T3 treatment. There was a statistically sig- predict decrease in numbers of HAD-D or HAD-A cases, or im- nificant difference (at the P Ͻ 0.05 level) in serum TSH levels by

TABLE 2. Relationship between genotype and GHQ-12 scores at baseline in all studied SNPs

Common homozygous Heterozygous Minor homozygous n Mean (95% CI) n Mean (95% CI) n Mean (95% CI) P DIO1 rs11206237 399 13.4 (12.9, 13.9) 130 13.5 (12.6, 14.4) 16 11.7 (9.5, 13.9) 0.56 rs11206244 239 13.2 (12.5, 13.8) 238 13.6 (12.9, 14.3) 69 13.2 (12.1, 14.2) 0.70 rs2235544 143 13.6 (12.8, 14.4) 288 13.4 (12.8, 14.0) 111 13.1 (12.2, 14.1) 0.50 rs2268181 387 13.4 (12.9, 14.0) 140 13.3 (12.5, 14.2) 19 11.9 (10.0, 13.9) 0.37 rs2294511 240 13.2 (12.6, 13.8) 248 13.6 (12.9, 14.3) 56 13.2 (11.9, 14.5) 0.71 rs2294512 252 13.7 (13.1, 14.4) 245 13.0 (12.3, 13.7) 50 13.2 (11.8, 14.6) 0.18 rs4926616 245 13.2 (12.5, 13.8) 236 13.7 (13.0, 14.4) 56 13.1 (11.7, 14.5) 0.63 rs731828 183 13.4 (12.7, 14.1) 278 13.2 (12.6, 13.8) 84 13.8 (12.7, 14.9) 0.70 rs7527713 362 13.5 (12.9, 14.0) 157 13.1 (12.3, 13.9) 26 13.0 (11.1, 15.0) 0.46 DIO2 rs225011 172 12.6 (11.9, 13.3) 264 13.7 (13.1, 14.3) 107 13.7 (12.5, 14.9) 0.06 rs225014 223 12.8 (12.2, 13.4) 236 13.6 (13.0, 14.3) 87 14.1 (12.8, 15.5) 0.02 rs225015 237 12.9 (12.3, 13.5) 236 13.6 (12.9, 14.3) 71 14.3 (12.8, 15.8) 0.03 DIO3 rs17716499 202 12.9 (12.3, 13.6) 248 13.6 (12.9, 14.3) 95 13.7 (12.7, 14.6) 0.19 rs7150269 121 13.6 (12.6, 14.5) 254 13.4 (12.8, 14.1) 169 13.1 (12.4, 13.8) 0.44 rs8011440 215 13.1 (12.5, 13.8) 246 13.5 (12.8, 14.2) 85 13.4 (12.3, 14.5) 0.51 rs945006 438 13.4 (12.9, 13.8) 92 13.4 (12.3, 14.4) 10 13.4 (9.7, 17.1) 0.98

Bold values indicate P Ͻ 0.05. CI, Confidence interval.

Downloaded from https://academic.oup.com/jcem/article-abstract/94/5/1623/2598196 by guest on 22 March 2018 J Clin Endocrinol Metab, May 2009, 94(5):1623–1629 jcem.endojournals.org 1627

TABLE 3. Genotype of rs225014 and all psychological parameters at baseline

Common homozygous (TT) Heterozygous (TC) Minor homozygous (CC) Mean (95% CI) Mean (95% CI) Mean (95% CI) or percent of or percent of or percent of n cases n cases n cases P GHQ Likert score 223 12.8 (12.2, 13.4) 236 13.6 (13.0, 14.3) 87 14.1 (12.8, 15.5) 0.02 GHQ case 217 39.2% 235 43.8% 86 46.5% 0.20 HAD-D case 223 13.0% 236 20.8% 87 24.1% 0.01 HAD-A case 223 40.4% 236 46.6% 87 48.3% 0.14 TSQ Likert score 216 16.5 (15.9, 17.1) 231 16.9 (16.3, 17.4) 83 17.2 (16.2, 18.1) 0.23 TSQ case 216 62.0% 231 62.3% 83 66.3% 0.56

CI, Confidence interval.

genotype in the T4-only group at baseline; however, this is likely to rare genotype (CC with rs225014) was associated with poorer

represent a chance finding because it is not present in the T4/T3 well-being. There was also a significant association at baseline group at baseline and does not persist at the 3- and 12-month visits, with HAD-D caseness, and four other measures showed stepwise

despite the T4 group remaining on the same treatment. changes in the same direction but did not reach statistical significance (Table 3). After intervention there was evidence of a signifi-

cant effect of the CC genotype on the response to T4/T3 compared Discussion with T4 only as measured by GHQ, TSQ, and satisfaction score (Fig. 1). By contrast, there was a consistent failure to observe any Our results suggest that common variation in the DIO2 gene as effect of intervention by genotype on HAD scores, despite the as- tagged by the SNP rs225014 may predict both poorer psychological sociations with HAD-D seen at baseline (Table 3). This may relate

well-being on T4 monotherapy and improved response to combi- to a lack of sensitivity to change in the HAD scoring system, which,

nation T4/T3 therapy in patients on thyroid hormone replacement. unlike the other scales, asks subjects to score particular symptoms This result needs replicating because we have tested 16 SNPs across without reference to a time period or how they would normally feel.

three genes. Although our study is the largest available T4/T3 trial, Wide variation exists in deiodinase expression between tis- it is still underpowered to reliably detect gene-treatment interac- sues resulting in important variation in the relative contribution

tions. However, if replicated, this is likely to reflect an effect on local of the serum concentrations of T4 and T3 to thyroid hormone

deiodination of T4 by the D2 deiodinase in the brain because vari- action. In rodent studies it is estimated that serum T3 contributes

ation in rs225014 has no effect on circulating thyroid hormone 87% of intracellular T3 in the kidney but only 50% in the pitu- levels (30). The CC genotype of rs225014 is present in a relatively itary and just 20% in the cerebral cortex, the remainder coming

small proportion of the population on T4, approximately 16% in from local deiodination of serum T4 by D2 (25). Our observation our study population, and hence, previous studies are likely to have that common genetic variation in the DIO2 gene but not the been underpowered to see this effect. DIO1 or DIO3 genes could be relevant to psychological well-

At baseline on T4, there is an association with genotype of all being is interesting because the D1 enzyme is not expressed in the three DIO2 SNPs studied and psychological well-being, as mea- brain (25) and D3 is a deactivating enzyme. Therefore, D2 is the Ͻ sured by GHQ score, at a P 0.1 level; however, no association only enzyme able to convert T4 to T3 in the brain and is likely to with any of the DIO1 or DIO3 SNPs (Table 2). In each case the play a key role in determining the ability of the brain to respond

TABLE 4. Thyroid function by genotype rs225014 and intervention at three time points

T4/T3 treatment T4 only Genotype TT TC CC P a TT TC CC P a Baseline fT4 21.3 Ϯ 3.7 21.1 Ϯ 3.6 20.9 Ϯ 3.9 0.54 21.4 Ϯ 3.9 20.9 Ϯ 3.6 20.5 Ϯ 3.8 0.14 3-month fT4 13.7 Ϯ 3.5 13.9 Ϯ 3.4 14.2 Ϯ 4.3 0.43 19.8 Ϯ 3.5 19.7 Ϯ 3.4 19.1 Ϯ 3.4 0.32 12-month fT4 14.6 Ϯ 3.5 14.4 Ϯ 3.4 14.8 Ϯ 3.7 0.99 20.7 Ϯ 3.5 20.2 Ϯ 3.4 19.5 Ϯ 2.9 0.06 Baseline fT3 3.87 Ϯ 0.67 3.78 Ϯ 0.73 4.10 Ϯ 0.59 0.38 3.86 Ϯ 0.86 3.87 Ϯ 0.64 3.74 Ϯ 0.70 0.34 3-month fT3 3.84 Ϯ 0.84 3.81 Ϯ 0.82 4.10 Ϯ 0.72 0.19 3.83 Ϯ 0.62 3.84 Ϯ 0.70 3.89 Ϯ 0.65 0.63 12-month fT3 3.74 Ϯ 0.58 3.70 Ϯ 0.90 3.88 Ϯ 0.77 0.56 3.64 Ϯ 0.68 3.63 Ϯ 0.65 3.52 Ϯ 0.56 0.31 Baseline TSH 1.00 1.08 0.87 0.39 0.69 0.92 1.10 0.04 (0.30, 2.00) (0.34, 2.18) (0.10, 2.67) (0.24, 1.58) (0.26, 2.33) (0.47, 2.14) 3-month TSH 2.83 2.14 1.56 0.07 0.71 0.64 0.69 0.68 (0.88, 4.97) (0.88, 4.16) (0.55, 3.81) (0.26, 1.36) (0.16, 1.69) (0.26, 2.38) 12-month TSH 2.36 2.35 1.52 0.48 0.55 0.71 0.84 0.36 (0.88, 5.37) (0.87, 4.40) (0.71, 4.71) (0.20, 1.60) (0.15, 1.94) (0.24, 2.30)

Values are mean Ϯ SD for fT4 and fT3 and median (interquartile range) for TSH. a Adjusted for age and sex.

Downloaded from https://academic.oup.com/jcem/article-abstract/94/5/1623/2598196 by guest on 22 March 2018 1628 Panicker et al. DIO2 Is Associated with Well-Being on T4 J Clin Endocrinol Metab, May 2009, 94(5):1623–1629

ϭ to circulating T4 levels. Indeed, common variation in D2 activity of these studies (n 141) did examine DIO2 polymorphisms

may represent the best available marker of intracellular T3 levels and showed a nonsignificant trend toward the CC genotype of in the brain. The lack of effect of the DIO2 polymorphisms on rs225014, giving poorer scores in baseline assessments of general serum thyroid hormone levels (30–32) means that without per- well-being such as the Rand and Symptom Checklist, a 90-item forming genetic testing, it is impossible to select out the group self-report scale, which would be consistent with our findings, likely to respond to combination therapy for subgroup analysis (39). In the WATTS study itself (total n ϭ 697), we would esti- in intervention trials. One previous study has shown an associ- mate that approximately 50 subjects in the intervention group ation between a DIO2 polymorphism, rs12885300, and serum (of 344) would have had the genotype, which would respond to thyroid hormone levels in young blood donors but not an elderly combination therapy, but given the large overall placebo effect, population (33) or in a subsequent study (31). This SNP was the differential change in these subjects was not detected in the excluded from analysis in our sample due to failed quality con- initial analysis of the whole cohort. trol; however, it is unlikely to have influenced the results because The observation that genetic variation in DIO2 may influence we have shown no difference in serum thyroid hormone levels by an individual’s ability to respond to T3 (and T4) is important for genotype during the trial (Table 4), supporting the view that several reasons. First, it may help to explain the excess of patients differences in serum hormone levels are not responsible for the who do not feel back to normal on T replacement therapy alone differences in well-being seen. 4 (9, 23, 24). In our initial cross-sectional study on well-being on D2 activity is regulated according to local thyroid hormone T replacement, subjects on T had an average score 0.7 GHQ levels in the brain and other tissues and hence can protect against 4 4 points higher compared with age- and sex-matched controls (9); hypothyroidism at a local level by increasing its activity (34, 35). hence, our finding of an extra 0.71 GHQ score points at baseline This regulation is mostly brought about by substrate (T )-in- 4 for each C allele suggests that subjects with one or two copies of duced ubiquitination (36). The DIO2 SNP of interest in the cur- this allele might explain a significant part of this difference. Sub- rent study is located in exon 3 of the D2 gene resulting in a jects on T replacement are likely to be particularly susceptible Thr92Ala substitution in the instability loop in D2, which is 4 closely linked to ubiquitination and a key determinant of turn- to impaired D2 function because they have a lower circulating T3 over rate (37). In vitro studies have not shown any difference in to T4 ratio compared with subjects with an intact thyroid axis (5, the enzymatic function of D2 compared with the wild-type when 6). Second, this effect provides evidence that thyroid hormone this polymorphism is transfected into cells (29, 32); however, an activity, presumably in the brain, plays a role in psychological in vivo study did show decreased D2 velocity in skeletal muscle well-being and mood, something that has previously been shown and thyroid biopsy samples in type 2 diabetes with the rare ge- only in animal studies (40). Finally, there are other candidate notype (29). In addition, a recent report has shown an associa- genes in addition to DIO2, notably thyroid hormone transport- tion between the CC genotype and osteoarthritis, which given ers that may also influence psychological well-being, and our the action of D2 at the growth plate is again consistent with findings suggest that common genetic variation in these other decreased function (38). Potentially the Thr92Ala substitution loci should also be explored. may effect ubiquitination impairing its ability to increase its activity with low T levels, reducing the ability to maintain 4 Conclusion homeostasis and increasing dependence on serum T as a 3 Genetic polymorphisms in the DIO2 gene may affect psy- source of T in the brain. While in the rat model brain, T 3 3 chological well-being in patients on T replacement and predict shows remarkable stability to different concentrations of in- 4 those who will have improved well-being in response to combi- fused T4 (2), it is likely that a significant amount of this homeostasis is derived from the ability of D2 within the brain to nation therapy with T3. Replication of this result, including pro- up-regulate in conditions of low local thyroid hormone spective studies with genotype-selected populations, are required concentrations. before changes in treatment approach can be recommended in Previous studies on the benefits of combination therapy on routine practice. psychological well-being in patients on thyroid hormone replacement, including the WATTS study itself and a metaanalysis of the major studies (21), have not shown any significant ad- Acknowledgments vantage over T4 monotherapy. The key issue here is likely to be one of statistical power. Our results suggest that the relevant Address all correspondence and requests for reprints to: Dr. C. M. DIO2 alleles that might confer responsiveness to combination Dayan, Henry Wellcome Laboratories for Integrative Neurosciences and therapy are only present in 16% of the population on T4. Other Endocrinology, University of Bristol, Dorothy Hodgkin Building, Whitson than WATTS, the studies of combination therapy have all had Street, Bristol BS1 3NY, United Kingdom. E-mail: colin.dayan@ study populations of less than 141 subjects and more commonly bristol.ac.uk. The Weston Area T /T Study was funded by grants from the South around 40 subjects. Hence, only between three and 10 subjects 4 3 West National Health Service Research and Development and Gold- in each study arm would have the potentially responsive geno- shield Pharmaceuticals PLC. V.P. is supported by an Athelstan & Amy type, giving little power to detect effects on baseline scores and Saw Medical Research Fellowship through the Faculty of Medicine, Den- even less power to detect differences in response to therapy in tistry and Health Sciences at the University of Western Australia. which subjects are divided into intervention groups. The largest Disclosure Summary: The authors have nothing to disclose.

Downloaded from https://academic.oup.com/jcem/article-abstract/94/5/1623/2598196 by guest on 22 March 2018 J Clin Endocrinol Metab, May 2009, 94(5):1623–1629 jcem.endojournals.org 1629

References Thyroxine-triiodothyronine combination therapy versus thyroxine monotherapy for clinical hypothyroidism: meta-analysis of randomized controlled 1. Flynn RW, MacDonald TM, Morris AD, Jung RT, Leese GP 2004 The thyroid trials. J Clin Endocrinol Metab 91:2592–2599 epidemiology, audit, and research study: thyroid dysfunction in the general 22. Walsh JP, Ward LC, Burke V, Bhagat CI, Shiels L, Henley D, Gillett MJ, population. J Clin Endocrinol Metab 89:3879–3884 Gilbert R, Tanner M, Stuckey BG 2006 Small changes in thyroxine dosage do 2. Escobar-Morreale HF, Obregon MJ, Escobar del Rey F, Morreale de Escobar not produce measurable changes in hypothyroid symptoms, well-being, or G 1995 Replacement therapy for hypothyroidism with thyroxine alone does quality of life: results of a double-blind, randomized clinical trial. J Clin En- not ensure euthyroidism in all tissues, as studied in thyroidectomized rats. docrinol Metab 91:2624–2630 J Clin Invest 96:2828–2838 23. Engum A, Bjoro T, Mykletun A, Dahl AA 2002 An association between de- 3. Escobar-Morreale HF, del Rey FE, Obregon MJ, de Escobar GM 1996 Only pression, anxiety and thyroid function—a clinical fact or an artefact? Acta the combined treatment with thyroxine and triiodothyronine ensures euthy- Psychiatr Scand 106:27–34 roidism in all tissues of the thyroidectomized rat. Endocrinology 137:2490– 24. Wekking EM, Appelhof BC, Fliers E, Schene AH, Huyser J, Tijssen JG, Wiersinga 2502 WM2005Cognitivefunctioningandwell-beingineuthyroidpatientsonthyroxine 4. Escobar-Morreale HF, Obregon MJ, Escobar del Rey F, Morreale de Escobar replacement therapy for primary hypothyroidism. Eur J Endocrinol 153:747–753 G 1999 Tissue-specific patterns of changes in 3,5,3Ј-triiodo-L-thyronine con- 25. Bianco AC, Salvatore D, Gereben B, Berry MJ, Larsen PR 2002 Biochemistry, centrations in thyroidectomized rats infused with increasing doses of the hor- cellular and molecular biology, and physiological roles of the iodothyronine mone. Which are the regulatory mechanisms? Biochimie (Paris) 81:453–462 selenodeiodinases. Endocr Rev 23:38–89 5. Woeber KA 2002 Levothyroxine therapy and serum free thyroxine and free 26. Goldberg DP, Gater R, Sartorius N, Ustun TB, Piccinelli M, Gureje O, Rutter triiodothyronine concentrations. J Endocrinol Invest 25:106–109 C 1997 The validity of two versions of the GHQ in the WHO study of mental 6. Jonklaas J, Davidson B, Bhagat S, Soldin SJ 2008 Triiodothyronine levels in illness in general health care. Psychol Med 27:191–197 athyreotic individuals during levothyroxine therapy. JAMA 299:769–777 27. Pan PC, Goldberg DP 1990 A comparison of the validity of GHQ-12 and CHQ-12 7. Alevizaki M, Mantzou E, Cimponeriu AT, Alevizaki CC, Koutras DA 2005 in Chinese primary care patients in Manchester. Psychol Med 20:931–940 TSH may not be a good marker for adequate thyroid hormone replacement 28. Bjelland I, Dahl AA, Haug TT, Neckelmann D 2002 The validity of the Hos- therapy. Wien Klin Wochenschr 117:636–640 pital Anxiety and Depression Scale. An updated literature review. J Psychosom 8. Walsh JP 2002 Dissatisfaction with thyroxine therapy–could the patients be Res 52:69–77 right? Curr Opin Pharmacol 2:717–722 29. Canani LH, Capp C, Dora JM, Meyer EL, Wagner MS, Harney JW, Larsen PR, 9. Saravanan P, Chau WF, Roberts N, Vedhara K, Greenwood R, Dayan CM Gross JL, Bianco AC, Maia AL 2005 The type 2 deiodinase A/G (Thr92Ala) 2002 Psychological well-being in patients on ‘adequate’ doses of L-thyroxine: polymorphism is associated with decreased enzyme velocity and increased results of a large, controlled community-based questionnaire study. Clin En- insulin resistance in patients with type 2 diabetes mellitus. J Clin Endocrinol docrinol (Oxf) 57:577–585 Metab 90:3472–3478 10. Bunevicius R, Kazanavicius G, Zalinkevicius R, Prange Jr AJ 1999 Effects of 30. Panicker V, Cluett C, Shields B, Murray A, Parnell KS, Perry JR, Weedon MN, thyroxine as compared with thyroxine plus triiodothyronine in patients with Singleton A, Hernandez D, Evans J, Durant C, Ferrucci L, Melzer D, Saravanan hypothyroidism. N Engl J Med 340:424–429 P, Visser TJ, Ceresini G, Hattersley AT, Vaidya B, Dayan CM, Frayling TM 11. Bunevicius R, Jakubonien N, Jurkevicius R, Cernicat J, Lasas L, Prange Jr AJ 2008 A common variation in deiodinase 1 gene DIO1 is associated with the 2002 Thyroxine vs thyroxine plus triiodothyronine in treatment of hypothy- relative levels of free thyroxine and triiodothyronine. J Clin Endocrinol Metab roidism after thyroidectomy for Graves’ disease. Endocrine 18:129–133 93:3075–3081 12. Appelhof BC, Fliers E, Wekking EM, Schene AH, Huyser J, Tijssen JG, Endert 31. de Jong FJ, Peeters RP, den Heijer T, van der Deure WM, Hofman A, Uitterlinden E, van Weert HC, Wiersinga WM 2005 Combined therapy with levothyroxine AG, Visser TJ, Breteler MM 2007 The association of polymorphisms in the type and liothyronine in two ratios, compared with levothyroxine monotherapy in 1 and 2 deiodinase genes with circulating thyroid hormone parameters and atro- primary hypothyroidism: a double-blind, randomized, controlled clinical trial. phy of the medial temporal lobe. J Clin Endocrinol Metab 92:636–640 J Clin Endocrinol Metab 90:2666–2674 32. Peeters RP, van Toor H, Klootwijk W, de Rijke YB, Kuiper GG, Uitterlinden 13. Clyde PW, Harari AE, Getka EJ, Shakir KM 2003 Combined levothyroxine AG, Visser TJ 2003 Polymorphisms in thyroid hormone pathway genes are plus liothyronine compared with levothyroxine alone in primary hypothy- associated with plasma TSH and iodothyronine levels in healthy subjects. roidism: a randomized controlled trial. JAMA 290:2952–2958 J Clin Endocrinol Metab 88:2880–2888 14. Escobar-Morreale HF, Botella-Carretero JI, Gomez-Bueno M, Galan JM, 33. Peeters RP, van den Beld AW, Attalki H, Toor H, de Rijke YB, Kuiper GG, Barrios V, Sancho J 2005 Thyroid hormone replacement therapy in primary Lamberts SW, Janssen JA, Uitterlinden AG, Visser TJ 2005 A new polymor- hypothyroidism: a randomized trial comparing L-thyroxine plus liothyronine phism in the type II deiodinase gene is associated with circulating thyroid with L-thyroxine alone. Ann Intern Med 142:412–424 hormone parameters. Am J Physiol Endocrinol Metab 289:E75–E81 15. Sawka AM, Gerstein HC, Marriott MJ, MacQueen GM, Joffe RT 2003 Does 34. Leonard JL, Kaplan MM, Visser TJ, Silva JE, Larsen PR 1981 Cerebral cortex Ј a combination regimen of thyroxine (T4) and 3,5,3 -triiodothyronine improve responds rapidly to thyroid hormones. Science 214:571–573 Ј depressive symptoms better than T4 alone in patients with hypothyroidism? 35. Silva JE, Larsen PR 1982 Comparison of iodothyronine 5 -deiodinase and Results of a double-blind, randomized, controlled trial. J Clin Endocrinol other thyroid-hormone-dependent enzyme activities in the cerebral cortex of Metab 88:4551–4555 hypothyroid neonatal rat. Evidence for adaptation to hypothyroidism. J Clin 16. Siegmund W, Spieker K, Weike AI, Giessmann T, Modess C, Dabers T, Kirsch Invest 70:1110–1123 G, Sanger E, Engel G, Hamm AO, Nauck M, Meng W 2004 Replacement 36. Gereben B, Goncalves C, Harney JW, Larsen PR, Bianco AC 2000 Selective pro- therapy with levothyroxine plus triiodothyronine (bioavailable molar ratio 14 teolysis of human type 2 deiodinase: a novel ubiquitin-proteasomal mediated : 1) is not superior to thyroxine alone to improve well-being and cognitive mechanism for regulation of hormone activation. Mol Endocrinol 14:1697–1708 performance in hypothyroidism. Clin Endocrinol (Oxf) 60:750–757 37. Dentice M, Bandyopadhyay A, Gereben B, Callebaut I, Christoffolete MA, 17. Walsh JP, Shiels L, Lim EM, Bhagat CI, Ward LC, Stuckey BG, Dhaliwal SS, Kim BW, Nissim S, Mornon JP, Zavacki AM, Zeold A, Capelo LP, Curcio- Chew GT, Bhagat MC, Cussons AJ 2003 Combined thyroxine/liothyronine Morelli C, Ribeiro R, Harney JW, Tabin CJ, Bianco AC 2005 The Hedgehog- treatment does not improve well-being, quality of life, or cognitive function inducible ubiquitin ligase subunit WSB-1 modulates thyroid hormone activation compared to thyroxine alone: a randomized controlled trial in patients with and PTHrP secretion in the developing growth plate. Nat Cell Biol 7:698–705 primary hypothyroidism. J Clin Endocrinol Metab 88:4543–4550 38. Meulenbelt I, Min JL, Bos S, Riyazi N, Houwing-Duistermaat JJ, van der Wijk 18. Saravanan P, Simmons DJ, Greenwood R, Peters TJ, Dayan CM 2005 Partial HJ, Kroon HM, Nakajima M, Ikegawa S, Uitterlinden AG, van Meurs JB, van

substitution of thyroxine (T4) with tri-iodothyronine in patients on T4 re- der Deure WM, Visser TJ, Seymour AB, Lakenberg N, van der Breggen R, placement therapy: results of a large community-based randomized controlled Kremer D, van Duijn CM, Kloppenburg M, Loughlin J, Slagboom PE 2008 trial. J Clin Endocrinol Metab 90:805–812 Identification of DIO2 as new susceptibility locus for symptomatic Osteoar- 19. Rodriguez T, Lavis VR, Meininger JC, Kapadia AS, Stafford LF 2005 Substi- thritis. Hum Mol Genet 17:1867–1875 tution of liothyronine at a 1:5 ratio for a portion of levothyroxine: effect on 39. Appelhof BC, Peeters RP, Wiersinga WM, Visser TJ, Wekking EM, Huyser J, fatigue, symptoms of depression, and working memory versus treatment with Schene AH, Tijssen JG, Hoogendijk WJ, Fliers E 2005 Polymorphisms in type levothyroxine alone. Endocr Pract 11:223–233 2 deiodinase are not associated with well-being, neurocognitive functioning, Ј 20. Levitt JA, Silverberg J, T4 plus T3 for hypothyroidism: a double-blind com- and preference for combined thyroxine/3,5,3 -triiodothyronine therapy. J Clin parison with usual T4. Proc 74th Annual Meeting of the American Thyroid Endocrinol Metab 90:6296–6299 Association, Los Angeles, CA, 2002, p 4 (Abstract 4) 40. Bauer M, Heinz A, Whybrow PC 2002 Thyroid hormones, serotonin and mood: 21. Grozinsky-Glasberg S, Fraser A, Nahshoni E, Weizman A, Leibovici L 2006 of synergy and significance in the adult brain. Mol Psychiatry 7:140–156

Downloaded from https://academic.oup.com/jcem/article-abstract/94/5/1623/2598196 by guest on 22 March 2018 Module 2 References

Overview

NIH Office of Dietary Supplements: https://ods.od.nih.gov/Health_Information/Dietary_Reference_Intakes.aspx Da Costa L, et al., (2012). Genetic Determinants of Dietary Antioxidant Status. Progress in Molecular biology and Translational Science Vol 108: 180-199.

Muto M, et al. (2002). Association between aldehyde dehydrogenase gene polymorphisms and the phenomenon of field cancerization in patients with head and neck cancer. Carcinogenesis 23(10):1759- 65.

Stover, P, 2004. Nutritional Genomics. American Physiological Society; Physiological Genomics. Vol 16, no. 2, 161-165.

Berna, G., et al., (2014). Nutrigenetics and Nutrigenomics Insights into Diabetes Etiopathogenesis. Nutrients 6(11), 5338-5369.

Park, S., et al., 2012. Resveratrol ameliorates aging-related metabolic phenotypes by inhibiting cAMP phosphodiesterases. Cell. 148(3):421-33.

CoQ10

Dos Santos, G., et al., 2009. Coenzyme Q10 and its effects in the treatment of neurodegenerative diseases. Braz. J. Pharm. Sci. Vol. 45 no. 4.

Deichmann, R., et al., 2010. Coenzyme Q10 and Statin-Induced Mitochondrial Dysfunction. Ochsner J., 10(1):16-21.

Fischer, A., et al., 2011. Association between genetic variants in the Coenzyme Q10 metabolism and Coenzyme Q10 status in humans. BMC Research Notes. 4:245.

Atia, A., et al., 2014. A Review of NAD(P)H:Quinone Oxidoreductase 1 (NQO1); A Multifunctional Antioxidant Enzyme. Journal of Applied Pharmaceutical Science. Vol 4 (12), 118-122.

Zahid, M., et al., 2009. NAD(P)H:quinone oxidoreductase 1 Arg139Trp and Pro187Ser polymorphisms imbalance estrogen metabolism towards DNA adduct formation in human mammary epithelial cells. J Steroid Biochem Mol Biol. 117(1-3);56-66.

© GENO MA INTERNATIONAL I ALL RIGHTS RESERVED I genomainternational.com Omega FA Azrad, M., et al., 2013. Current Evidence Linking Polyunsaturated Fatty Acids with Cancer Risk and Progression. Front Oncol. 3:224.

Gillingham, L., et al., 2013. Dietary oils and FADS1-FADS2 genetic variants modulate [13C]a-linolenic acid metabolism and plasma fatty acid composition. Am J Clin Nutr 97:195-207.

Li, S, .e t al., 2013. FADS Gene Polymorphisms Confer the Risk of Coronary Artery Disease in a Chinese Han Population through the Altered Desaturase Activities: Based on High-Resolution Melting Analysis. PLOS ONE Vol 8, Issue 1, e55869.

Roke, K., et al., 2017. Evaluating Changes in Omega-3 Fatty Acid Intake after Receiving Personal FADS1 Genetic Information: A Randomized Nutrigenetic Intervention. Nutrients. 9, 240.

Al-Hilal, M., et al., 2013. Genetic variation at the FADS1-FADS2 gene locus influences delta-5 desaturase activity and LC-PUFA proportions after fish oil supplement. Journal of Lipid Research, Vol 54.

DaCosta, L., et al., 2012. Genetic Determinants of Dietary Antioxidant Status. Progress in Molecular Biology and Translational Science. Vol. 108.

Wisneski, L., and Anderson, L. The Scientific Basis of Integrative Medicine. CRC Press, 2005.

Bidlack, W. and Rodriguez, R. Nutritional Genomics: The Impact of Dietary Regulation of Gene Function on Human Disease. CRC Press 2012.

Kaupt, J. and Rodriguez, R. editors. Nutritional Genomics: Discovering the Path to Personalized Nutrition. John Wiley & Sons, Inc 2006.

Carotenoids

Bonet,M., et al., 2015. Carotenoids and their conversion products in the control of adipocyte function, adiposity and obesity. Arch Biochem Biophys 572:112-25.

Leung, W., et al., 2009. Two common single nucleotide polymorphisms in the gene encoding Beta- carotene 15,15’-monoxygenase alter Beta-carotene metabolism in female volunteers. FASEB Journal. Vol 23.

DaCosta, L., et al., 2012. Genetic Determinants of Dietary Antioxidant Status. Progress in Molecular Biology and Translational Science. Vol. 108.

© GENO MA INTERNATIONAL I ALL RIGHTS RESERVED I genomainternational.com Ferrucci,L., et al., 2009. Common Variation in the b-Carotene 15,150-Monooxygenase 1 Gene Affects Circulating Levels of Carotenoids: A Genome-wide Association Study. Amer J Human Genetics. 84:123- 133.

Vitamin E

DaCosta, L., et al., 2012. Genetic Determinants of Dietary Antioxidant Status. Progress in Molecular Biology and Translational Science. Vol. 108.

Bauer, Scott, 2013. Antioxidant and vitamin E transport genes and risk of high-grade prostate cancer and prostate cancer recurrence. The Prostate. 73(16).

X, W., et al., 2017. Dietary Antioxidants: Potential Anticancer Agents. Nutr Cancer. 69(4):521-533.

Rizvi, S., et al., 2014. The role of vitamin e in human health and some diseases. Sultan Qaboos Univ Med J. 14(2):e157-65.

Brigelious-Flohe, R., et al., Vitamin E: function and metabolism. The FASEB Journal. Vol. 23, no. 10 1145- 55.

Schmolz, L., et al., 2016. Complexity of vitamin E metabolism. World J Biol Chem. 7(1): 14-43.

Ulatowski, L., et al., 2012. Expression of the Alpha Tocopherol Transfer Protein gene is regulated by Oxidative Stress and Common Single Nucleotide Polymorphisms. Free Radic Biol Med. 53(12).

Wright, M., et al., 2009. Association of Variants in Two Vitamin E Transport Genes with Circulating Vitamin E Concentrations and Prostate Cancer Risk. American Assoc for Cancer Research.

Vitamin C

DaCosta, L., et al., 2012. Genetic Determinants of Dietary Antioxidant Status. Progress in Molecular Biology and Translational Science. Vol. 108.

Figeuroa-Mendez, R., Rivas-Arancibia, S. 2015. Vitamin C in Health and Disease: Its Role in the Metabolism of Cells and Redox State in the Brain. Front. Physiol.

Schwartz, Betty. 2014. New criteria for supplementation of selected micronutrients in the era of nutrigenetics and nutrigenomics. Int J Food Sci Nutr.

Burzle, M., Hediger, MA. 2012. Functional and physiological role of vitamin C transporters. Curr Top Membr. 70:357-75.

© GENO MA INTERNATIONAL I ALL RIGHTS RESERVED I genomainternational.com © GENO MA INTERNATIONAL I ALL RIGHTS RESERVED I genomainternational.com Burzle, M., et al. 2013. The sodium-dependent ascorbic acid transporter family SLC23. Mol Aspects Med. 34(2-3):436-54.

Timpson, NJ., et al., 2010. Genetic variation at the SLC23A1 locus is associated with circulating concentrations of L-ascorbic acid (vitamin C): evidence from 5 independent studies with >15,000 participants. Am J Clin Nutr Vol. 92 No. 2:375-82.

Selenium Burk, RF., et al., 2003. Selenoprotein metabolism and function: evidence for more than one function for selenoprotein P. J Nutr, 133(5 Suppl 1):1517S-20S.

Meplan, C., et al., 2013. Association between Polymorphisms in Glutathione Peroxidase and Selenoprotein P Genes, Glutathione Peroxidase Activity, HRT Use and Breast Cancer Risk. PLOS ONE. Vol. 8, Issue 9, e73316.

Jablonska, E., et al., 2015. DNA damage and oxidative stress response to selenium yeast in the non-smoking individuals: a short-term supplementation trial with respect to GPX1 and SEPP1 polymorphism. Eur J Nutr.

Bellinger, F., et al., 2010. Regulation and function of selenoproteins in human disease. Biochem J; 422(1);11-22.

Cominetti, C., et al., 2010. Associations between glutathione peroxidase-1 Pro198Leu polymorphism, selenium status, and DNA damage levels in obese women after consumption of Brazil nuts. Nutrition 27: 891-896.

Steinbrecher, A., et al., 2010. Effects of Selenium Status and Polymorphisms in Selenoprotein Genes on Prostate Cancer Risk in a Prospective Study of European Men. America Association for Cancer Research.

Toppo, S., et al., 2008. Evolutionary and structural insights into the multifaceted glutathione peroxidase (Gpx) superfamily. Antioxid Redox Signal. 10(9):1501-14.

Antioxidants DaCosta, L., et al., 2012. Genetic Determinants of Dietary Antioxidant Status. Progress in Molecular Biology and Translational Science. Vol. 108.

Wisneski, L., and Anderson, L. The Scientific Basis of Integrative Medicine. CRC Press, 2005.

© GENO MA INTERNATIONAL I ALL RIGHTS RESERVED I genomainternational.com Bidlack, W. and Rodriguez, R. Nutritional Genomics: The Impact of Dietary Regulation of Gene Function on Human Disease. CRC Press 2012.

Kaupt, J. and Rodriguez, R. editors. Nutritional Genomics: Discovering the Path to Personalized Nutrition. John Wiley & Sons, Inc 2006.

Vitamin D and Calcium Young-Joon, S., Cadenas, E., Dong, Z., and Packer, L. editors. Dietary Modulation of Cell Signaling Pathways. CRC Press, 2009.

Zempleni, J, and Dakshinamurti, K. editors. Nutrients and Cell Signaling. CRC Press, 2005.

Porterfield SP, White BA: Endocrine Physiology, 3rd ed. Philadelphia, Mosby, 2007

Carvalho, A., et al, 2013. The Role of Vitamin D Level and Related Single Nucleotide Polymorphisms in Crohn’s Disease. Nutrients. 5;3898-3909.

Laing BB and Ferguson LR, 2015. Genetic Variations in Vitamin D Metabolism Genes and the Microbiome, in the Presence of Adverse Environmental Changes, Increase Immune Dysregulation. Austin J Nutr Metab. Vol. 2, Issue 4.

Moran-Auth, Y., et al., 2013. Vitamin D status and gene transcription in immune cells. J Steroid Biochem Mol Biol. 136:83-5.

Peacock, M., 2010. Calcium Metabolism in Health and Disease. CJASN Vol. 5 No. Supplement 1:523-530

Cole, D.E., et al., 20. Association between Total Serum Calcium and the A986S Polymorphism of the Calcium-Sensing Receptor Gene. Mol Gen Metab Volume 72, Issue 2: 168-174.

Tfelt-Hansen, J. and Brown, E., 2008. The Calcium-Sensing Receptor in Normal Physiology and Pathophysiology: A Review. Crit Rev Clin Lab Sci. Vol 42, Issue 1.

Alfadda, T., et al., 2014. Calcium-sensing receptor 20 years later. Am J Physiol Cell Physiol 307: C221- C231.

Deeb, K., et al., 2007. Vitamin D signalling pathways in cancer: potential for anticancer therapeutics. Nature Reviews Cancer 7, 684-700.

Jensen, MB 2014. Vitamin D and male reproduction. Nature Reviews Endocrinology 10, 175-186.

Dawson-Hughes B, et al., 1995. Calcium absorption on high and low calcium intakes in relation to vitamin D receptor genotype. J Clin Endocrinol Metab 80:3657–3661.

© GENO MA INTERNATIONAL I ALL RIGHTS RESERVED I genomainternational.com Wishart, JM et al., 1997. Relations between calcium intake, calcitriol, polymorphisms of the vitamin D receptor gene, and calcium absorption in premenopausal women. Am J Clin Nutr. Vol. 65, No. 3:798- 802.

Chang, B., et al. 2015. Influence of vitamin D an d estrogen receptor gene polymorphisms on calcium absorption: BsmI predicts a greater decrease during energy restriction. Bone. Vol. 81:138-144.

B complex vitamins Berdanier, Carolyn. Advanced Nutrition: Micronutrients. C RC Press, 1998

Pantuso, T. 2014. MTHFR Clinical Considerations: A Review. CA MH edia. Integrative Medicine Alert.

Aneiros-Guerrero A., et al., Genetic polymorphisms in folate pathway enzymes, DRD4 and GSTM1 are related to temporomandibular disorder. BMCd M Gee net.

Loor, J, et al.,. 20 R 15 umen-protected methyl donorsd a nt e hge nome: beyond nutrigenomics. PAXXI AS Congress.

Lim,, U. et al., 2007. Gene-nutrient interactions among determinants of folate and one-carbon metabolism on the risk of non -Hodgkin lymphoma: NCI-SEER Case-Control Study. Blood. 109:3050-3059.

Obeid, Rima 2013.e T hM etabolic Burden of Methyl Donor Deficiency with Focus on the Betaine Homocysteine Methyltransferase Pathway. Nutrients. 5:3481-3495.

Tanaka,, T. et al., 2009. Genome-wide Association Study of Vitamin B6, Vitamin B12, Folate, and Homocysteine Blood Concentrations. Am J Hum Genet 84:477-482.

Wemimont, S., et al, 2011. Folate network genetic variation, plasma homocysteine, and global genomic methylation content: a genetic association study.C B M edical Genetics. 12:150.