리차드 존슨 '과당' 대사 흡수, 대사의 문제점 2026 네이처!!

[문형철]

주요 저자: Richard J. Johnson, Miguel A. Lanaspa, Dean R. Tolan, Marcus D. Goncalves, Samir Softic, Kimber L. Stanhope, Laura G. Sánchez-Lozada, Mark A. Herman, Joshua D. Rabinowitz

Richard J. Johnson, MD 상세 소개
기본 프로필

• 현재 직함: Emeritus Professor of Medicine (명예 교수), University of Colorado Anschutz Medical Campus
• 전문 분야: 신장내과(Nephrology), 내과학, 감염병
• 주요 연구 주제: 과당(fructose)과 요산(uric acid)이 비만, 대사증후군, 당뇨, 고혈압, 신장 질환에 미치는 역할
  
  
  
  

학력 및 경력

• 1975년: University of Wisconsin (인류학 전공, 우등 졸업)
• 1979년: University of Minnesota 의과대학 졸업
• University of Washington에서 내과 전공의 + 신장내과·감염병 전임의 수련
• 주요 직책:
  • Baylor College of Medicine 신장내과 과장 (2000–2003)
  • University of Florida 신장내과 과장 (2003–2008)
  • University of Colorado 신장내과·고혈압 과장 (2008–2017)
• 2024년 7월부터 임상 진료에서 은퇴하고 명예 교수로 연구·저술·강연 활동 지속

연구 업적

• 논문: 800편 이상 발표 (Google Scholar h-index 약 156)
• NIH 연구비 지원을 장기간 받음
• 주요 연구 방향:
  • 과당이 ‘지방 스위치(fat switch)’를 켜서 지방 축적과 인슐린 저항성을 유도하는 기전
  • 요산이 과당 대사의 핵심 매개체로 작용한다는 가설 제시 및 검증
  • 내인성 과당 생성 경로(polyol pathway)의 역할
  • 진화적 관점의 ‘생존 스위치(survival switch)’ 가설

저서

• The Sugar Fix (2008) – 저과당 식단의 개념을 대중화
• The Fat Switch (2012)
• Nature Wants Us to Be Fat (2022) – 과당과 진화적 지방 축적 메커니즘을 종합

핵심 메시지

과당(fructose)은

단순한 ‘빈 칼로리’가 아니다.

‘풍요의 신호(metabolic signal of plenty)’로서 작용하며,

현대의 과잉 영양 환경에서는 대사 증후군의 주요 구동 요인이 된다.

1. 포도당 vs 과당의 근본적 차이

구분                            포도당 (Glucose)                                                              과당 (Fructose)

주요 작용	인슐린 분비 자극 → 에너지 사용 또는 저장	인슐린과 거의 무관하게 지방 합성 촉진
대사 조절	엄격한 피드백 조절 (PFK 등)	조절이 거의 없는 경로 (KHK → F1P)
결과	에너지 공급 중심	지방 축적 + ATP 고갈 유도

2. 과당 대사의 특징적 기전

• Ketohexokinase (KHK)에 의해 빠르게 과당-1-인산(F1P)으로 전환
• F1P가 축적되면 영양 감지 신호로 작용 → 간에서 해당과정과 지방생성을 촉진
• 세포 내 ATP 고갈 → AMP 증가 → 요산 생성 증가
• 미토콘드리아 산화 스트레스 유발 → 지방 합성 경로 활성화
• ChREBP 전사인자 활성화 → 지방합성 유전자 발현 증가

3. 내인성 과당 생성 (Endogenous Fructose Pathway)

외부에서 과당을 먹지 않아도 체내에서 만들어질 수 있다.

• 포도당 → 알도스 환원효소(aldose reductase) → 소르비톨 → 과당
• 고혈당, 고염식, 저산소 상태에서 활성화됨
• 이 경로가 활성화되면 외부 과당 섭취와 유사한 대사 장애를 유발할 수 있음

4. 현대 사회에서의 위험

진화적으로 과당은 계절적으로 풍부한 시기(과일·꿀)에 지방을 저장하도록 하는 ‘생존 신호’였다. 그러나 현재는:

• 설탕(sucrose)과 고과당 옥수수 시럽(HFCS)의 상시 섭취
• 지속적인 과당 신호 노출

→ 대사증후군(비만, 인슐린 저항성, 지방간, 고중성지방혈증, 고혈압)을 촉진 → 최근 연구들은 암과 치매와도 연관성을 시사

5. 임상적·공중보건적 함의

• 과당은 칼로리 그 이상으로 대사 조절 분자로 인식해야 함
• 단순 칼로리 제한만으로는 과당의 신호 효과를 막기 어려움
• KHK 억제제가 치료제로 개발 중 (일부 임상시험 진행)

한 줄 요약

과당은 진화적으로 ‘풍요의 신호’였으나, 현대의 지속적인 과잉 섭취 환경에서는 지방 축적과 대사 질환을 적극적으로 유도하는 위험한 대사 신호로 작용한다.

이 논문은 Richard Johnson 그룹의 과당 연구 성과를 종합한 권위 있는 리뷰로, 단순 당 섭취를 넘어 신호 분자로서의 과당을 강조한 점이 특징입니다.

Figure 1. Hepatocyte Glucose Metabolism (간세포 포도당 대사)

이 도식은 식사 후 상태(fed state)에서 인슐린이 간세포의 포도당 대사를 어떻게 조절하는지 보여줍니다.

주요 경로

1. 포도당 유입 및 초기 대사
   • Glucose → GCK(글루코키나아제) → G6P
   • G6P는 세 갈래로 나뉨:
     • Glycogen 합성 (GS 효소, 인슐린이 촉진)
     • PPP (오탄당 인산 경로)
     • Glycolysis (해당과정)
2. 해당과정
   • G6P → F6P → PFK-1 → F1,6P₂ → 여러 단계 → Pyruvate
   • Pyruvate는 젖산(Lactate)과 상호 전환 가능
3. 미토콘드리아 내 대사
   • Pyruvate → PDH → Acetyl CoA
   • Acetyl CoA는 TCA 회로로 들어가거나 지방산 합성에 이용
   • 지방산(Fatty Acids) → CPT1a → β-oxidation → Acetyl CoA
   • TCA + OxPhos를 통해 ATP 생성
4. 지방산 합성 (Fatty-Acid Synthesis)
   • 인슐린이 SREBP-1c를 증가시켜 FAS와 ACC를 촉진
   • Citrate → ACLY → Acetyl CoA → Malonyl CoA → 지방산 합성

조절 신호

• 초록색 점선 화살표 (인슐린 자극) → 글리코겐 합성 촉진 → 지방산 합성 촉진 (에너지 저장 방향)
• 빨간색 점선 화살표 (억제) → ATP가 많을 때 포도당 이용을 억제 → PFK-1, PDH 등 주요 효소를 억제하여 과도한 대사 방지

핵심 메시지

식사 후 인슐린은 간에서 포도당을 글리코겐과 지방으로 저장하도록 유도하고, ATP가 충분해지면 포도당 사용을 억제하여 균형을 맞춥니다.

Figure 2. Fructose Metabolism (과당 대사)

이 도식은

과당(Fructose)이 간에서 어떻게 대사되어 최종 산물로 이어지는지를 보여줍니다.

주요 대사 경로

1. 과당 인산화
   • Fructose → Ketohexokinase-C → Fructose 1-P
2. 분해
   • Fructose 1-P → Aldolase-B에 의해 두 개로 쪼개짐
     • Dihydroxyacetone 3-P (DHAP)
     • Glyceraldehyde
3. 이후 경로

중간산물주요 경로최종 결과

Dihydroxyacetone 3-P	Glycerol-3-P dehydrogenase → Glycerol-3-P	DNL (새로운 지방 합성, de novo lipogenesis)
Glyceraldehyde	Triose Kinase → Glyceraldehyde 3-P	Glycolysis (해당과정)
Glyceraldehyde	Aldehyde dehydrogenase → Glycerate	다른 경로

• Dihydroxyacetone 3-P와 Glyceraldehyde는 Triosephosphate Isomerase로 서로 전환 가능
• Glycerol도 Glycerol kinase를 통해 Glycerol 3-P로 들어가 DNL에 기여

핵심 포인트

• 과당은 인슐린 조절을 거의 받지 않고 빠르게 대사됩니다.
• 특히 Fructose 1-P → Aldolase-B 단계는 조절이 약해, 과다 섭취 시 → Glycerol-3-P와 Acetyl-CoA가 많이 생성되어 → 지방 합성(DNL)이 촉진됩니다.
• 이 경로가 Excess fructose 섭취 시 간 지방 축적과 대사 이상을 일으키는 중요한 이유로 여겨집니다.

Figure 3. Hepatocyte Fructose Metabolism (간세포 과당 대사)

이 도식은

과당이 간세포에서 대사될 때 포도당 대사와 어떻게 다르게 조절되는지를 보여줍니다.

색깔 상자와 X 표시가 핵심 차이점을 강조합니다.

주요 특징

1. 과당 유입과 초기 대사

• Dietary Fructose → GLUT5/2 → Liver Fructose
• Fructolysis: Fructose → F1P (Fructose-1-Phosphate)
• F1P가 여러 단계를 거쳐 Trioses (Dihydroxyacetone-P, Glyceraldehyde)로 분해

2. 조절되지 않는 대사 (Unregulated Triose Metabolism)

• 과당에서 생긴 Trioses는 → Glycerol3P → DNL (새로운 지방 합성)을 강하게 자극
• ChREBP-1c 증가 → FAS & ACC 증가 → 지방산 합성 촉진

3. 포도당 대사 경로에 미치는 영향 (빨간색 억제)

• PFK-1이 억제됨 (X 표시)
• CPT1a가 억제되어 지방산 산화(β-oxidation)가 차단
• TCA 회로와 OxPhos도 영향을 받음

4. 녹색 촉진 신호

• 일부 경로(글리코겐 합성, PPP 등)는 촉진되지만, 전체적으로는 지방 합성 쪽으로 기울어짐

핵심 메시지 (포도당 대사와의 차이)

구분                                  포도당 대사                                                                    과당 대사

주요 조절	인슐린과 ATP에 의해 엄격히 조절	상대적으로 조절이 약함 (unregulated)
지방 합성	조절됨	강하게 촉진 (DNL Stimulated)
지방산 산화	가능	차단됨 (Fatty Acid Oxidation Blocked)
결과	에너지 저장과 사용의 균형	지방 축적 쪽으로 치우침

과당은 인슐린 조절을 거의 받지 않고 빠르게 Triose로 전환되어, 간에서 지방 합성을 촉진하고 지방 분해를 억제하는 방향으로 작용합니다. 이것이 과당 과다 섭취 시 지방간과 대사 이상을 유발하는 주요 기전으로 설명됩니다.

Figure 4. Fructose Induced ATP Depletion and Uric Acid Generation (과당에 의한 ATP 고갈과 요산 생성)

이 도식은

과당이 간에서 ATP를 빠르게 소모하고,

그 결과 요산(uric acid)이 생성되는 과정을 보여줍니다.

핵심 과정

1. 과당의 빠른 인산화
   • Fructose → KHK-C (Ketohexokinase-C) → F1P
   • 이 과정에서 ATP가 대량으로 소모되어 ADP → AMP로 전환됨
2. AMP의 분해와 요산 생성 (Purine Catabolism Pathway)
   • AMP → AMPD2 → IMP → INO → HX → XAN → Uric acid
   • 인간에게는 Uricase가 없기 때문에 (X 표시) 요산이 알란토인으로 더 분해되지 못하고 축적됨
3. 피드백 증폭
   • 생성된 요산이
     • Aldose reductase (AR) 활성화
     • KHK-C 활성화
     • SREBP-1c, ChREBP 활성화 → 과당 대사와 지방 합성을 더욱 촉진
   • IMP가 Aldolase B를 억제하여 F1P 축적을 더 강화
4. Polyol Pathway와의 연결
   • Glucose ↔ Sorbitol ↔ Fructose 경로도 연결되어 있음

핵심 메시지

과당은 KHK-C에 의해 조절 없이 빠르게 인산화되면서 간세포의 ATP를 일시적으로 고갈시킵니다.

이 ATP 고갈이 AMP 분해를 촉발하고, 결국 요산 생성을 증가시킵니다.

요산은

다시 과당 대사를 증폭시키는 악순환을 만들고,

인간에서는 uricase가 없어 요산 반응이 다른 포유류보다 더 크게 나타납니다.

이 기전은

과당 과다 섭취가 고요산혈증, 지방간, 인슐린 저항성 등과 연결되는 중요한 설명 중 하나입니다.

Figure 5. Sites of Metabolism for Fructose (과당이 대사되는 부위)

이 그림은

식이 과당과 체내 생성 과당이 어디서 어떻게 대사되는지를 비교해서 보여줍니다.

A) Dietary fructose (식이 과당)

1. 소장 상피세포 (Enterocyte)
   • GLUT5로 흡수
   • 30–60%가 여기서 1차 대사됨 (First pass effect)
   • 나머지는 GLUT2를 통해 문맥 순환으로 이동
2. 간세포 (Hepatocyte)
   • 문맥을 통해 들어온 과당의 30–60%가 간에서 대사
   • KHK-C에 의해 빠르게 F1P로 전환 → ATP 소모
   • Trioses → Pyruvate, Glycerol-3P → 중성지방·지방산 합성 (DNL 촉진)
3. 전신 순환 (Systemic circulation)
   • 약 10%만 전신으로 나감

B) Endogenous fructose (체내 생성 과당)

• 포도당이 Polyol pathway를 통해 과당으로 전환됨 (Glucose → Sorbitol → Fructose)

상시 발현 (Constitutive)

• 생식 기관 (전립선, 정낭, 태반 등)

유도성 발현 (Inducible triggers)

• 저산소, 허혈, 열 스트레스, 외상
• 고혈당, 고요산혈증, 고장성
• GCK 활성화 등

핵심 메시지

• 식이 과당은 소장과 간에서 대부분 처리되며, 간에서 지방 합성을 강하게 유도합니다.
• 체내에서도 스트레스 상황이나 특정 조직에서 포도당으로부터 과당이 만들어질 수 있습니다.
• 이 두 경로가 합쳐져 과당의 대사 부담을 증가시킬 수 있습니다.

핵심 내용

이 리뷰 논문은

케토헥소키나아제(Ketohexokinase, KHK)가

과당(fructose)으로 인한 질병 발생의 중심 매개체이며,

이를 억제하는 것이 유망한 치료 전략임을 강조합니다.

주요 포인트

1. KHK의 역할
   • 과당 대사(fructolysis)의 첫 번째 효소.
   • 과당을 fructose-1-phosphate로 인산화시키는 핵심 단계.
2. 두 가지 아이소폼(isoform)의 기능 차이
   • KHK-C (고활성형): 간, 신장, 소장에 주로 발현. → 대사 질환의 주요 원인 (대사이상 관련 지방간질환(MASLD), 당뇨성 신장질환, 비만 등)
   • KHK-A (저활성형): 전신에 널리 발현. → 종양의 대사 재프로그래밍, 증식, 전이에 관여 (대사적 역할 + 비전형적 단백질 키나아제 활성)
3. 조절 요인
   • 식이 구성, 요산/ChREBP 전사 축, 저산소유도인자(HIF), 중쇄지방산 등에 의해 동적으로 조절됨.
   • 대사 조절 외에 심장 항상성, 뼈 발달, 신경인지 기능에도 영향을 미침.
4. 치료적 개발 현황
   • 1세대 억제제 PF-06835919: MASLD, 제2형 당뇨, 유전성 과당불내증(HFI)을 대상으로 Phase 2 임상시험 완료.
   • 2세대 억제제 LY3522348: 건강한 지원자 대상 Phase 1 평가 완료.
   • 아직 해결해야 할 과제: 아이소폼 선택성(selectivity)과 조직 특이적 전달(tissue-specific delivery).

결론

KHK는 과당 관련 질환의 핵심 매개체로, 그 억제가 대사 질환 치료에 유망한 접근법임을 제시합니다. 특히 KHK-C를 선택적으로 억제하는 전략이 중요하며, 향후 아이소폼 선택성과 조직 특이적 약물 전달 기술 개발이 필요합니다.

https://pmc.ncbi.nlm.nih.gov/articles/PMC12220530/

핵심 내용

과당이

비알코올성 지방간질환(NAFLD)의 주요 원인으로 주목받는 가운데,

외부에서 과당을 섭취하지 않아도 체내에서 만들어질 수 있다는 점을 강조한 리뷰 논문입니다.

주요 포인트

• 과당은 포도당으로부터 polyol pathway를 통해 내인성으로 생성될 수 있음 (포도당 → 알도스 환원효소(AR) → 소르비톨 → 과당)
• 알도스 환원효소(AR)가 이 경로의 핵심 관문 효소
• 고혈당 상태에서 AR이 활성화되면 내인성 과당이 증가하고, 이어서 과당-요산 경로가 작동하여 간에서 지방 합성을 촉진
• 결과적으로 NAFLD 및 기타 만성 대사 질환을 유발할 수 있음

결론

내인성 과당 생성이

기존에 생각했던 것보다 훨씬 더 중요한 역할을 할 수 있으며,

알도스 환원효소의 조절이 지방간 발생의 핵심 메커니즘이라는 새로운 시각을 제시합니다.

논문 그림 설명

(Aldose reductase, fructose and fat production in the liver)

이 그림들은

간에서 polyol pathway

→ 내인성 과당 생성

→ 요산 증가

→ 지방 합성이 일어나는 전체 과정을 단계별로 보여줍니다.

핵심 경로 요약 (Figure 1 & 2)

1. Polyol pathway 시작
   • 포도당이 알도스 환원효소(AR)에 의해 소르비톨로 전환
   • 소르비톨이 소르비톨 탈수소효소(SD)에 의해 과당으로 전환
2. 과당 대사 (Fructolysis)
   • 과당이 케토헥소키나제(FK/KHK)에 의해 빠르게 Fru-1-P로 인산화
   • 이 과정에서 ATP가 급격히 소모됨
3. 요산 생성
   • ATP 고갈 → AMP 증가 → AMP deaminase(AMPD) 활성화
   • 결과적으로 요산(UA)이 증가
4. 지방 합성 촉진
   • 요산이 NOX4를 활성화 → 미토콘드리아 aconitase 억제
   • 시트르산이 세포질로 이동 → de novo lipogenesis(지방 신생합성) 증가
   • 동시에 PFK1이 억제되어 해당과정이 막히고, Warburg 효과처럼 젖산 생성이 증가

주요 그림별 의미

• Figure 4: AR 발현을 증가시키는 요인들 → 고혈당, 저산소, 알코올, 우마미, 설탕, 소금 등이 AR을 유도
• Figure 5 & 6: 스위치 개념
  • 저혈당 상태: AR과 GCK가 비활성 → 정상 해당과정
  • 고혈당 상태: AR과 GCK가 활성화 → polyol pathway가 열려 과당-요산 경로로 전환

임상적 의미

외부에서 과당을 섭취하지 않아도, 고혈당·고염·알코올·저산소 같은 상황에서 AR이 활성화되면 체내에서 과당이 만들어지고, 이것이 지방간과 요산 상승을 유발할 수 있습니다.

이 그림들은 “포도당만으로도 과당 효과가 나타날 수 있다”는 논문의 핵심 주장을 시각적으로 잘 정리하고 있습니다.

Nat Metab

. Author manuscript; available in PMC: 2026 May 16.

Published in final edited form as: Nat Metab. 2026 Apr 17;8(5):1019–1033. doi: 10.1038/s42255-026-01506-y

Fructose: Metabolic signal and modern hazard

Richard J Johnson 1, Miguel A Lanaspa 1, Dean R Tolan 2, Marcus D Goncalves 3, Samir Softic 4, Kimber L Stanhope 5, Laura G Sánchez-Lozada 6, Mark A Herman 7, Joshua D Rabinowitz 8,9,10

• Author information
• Article notes
• Copyright and License information

PMCID: PMC13178449  NIHMSID: NIHMS2168142  PMID: 41998215

The publisher's version of this article is available at Nat Metab

Abstract

There is much interest in the role of sweeteners like table sugar (sucrose) and high-fructose corn syrup in obesity and metabolic disease. Both sweeteners consist of glucose and fructose, two six-carbon isomeric sugars, that account for 15% of the Western diet. While glucose ingestion may promote obesity through its effects to stimulate insulin secretion, fructose has unique metabolic effects that promote triglyceride synthesis and fat accumulation. These effects arise from fructose’s well known role as a signal of metabolic plenty. Under modern conditions of overnutrition, chronic excess fructose drives features of metabolic syndrome. Emerging evidence further links fructose to cancer and dementia. Here we review the biochemical, molecular, and physiological distinctions between fructose and glucose, as well as the endogenous fructose pathway that makes fructose from glucose. Through this review, we highlight the role of fructose not only as a caloric source, but also as a regulator of metabolic health and disease.

---

Intake of added sugars, such as table sugar (sucrose) and high fructose-corn syrup (HFCS), have increased markedly in the last few centuries in parallel with the rise in obesity and diabetes, with the strongest association being with sugar-sweetened beverages (SSBs)1. The World Health Organization recommends a total daily energy intake from ‘free’ sugars of <10%2. In the United States and many high-income countries, SSB intake is dropping but overall free sugar consumption still exceeds 10%. In low- and middle-income countries, SSB and overall free sugar consumption continue to rise.3,4

One hypothesis for how added sugars might cause obesity is that the sweet and highly palatable characteristics of sugar encourage excessive caloric intake that leads to positive energy balance 5. Another hypothesis is that the glucose component in sugar results in chronic stimulation of insulin that favors fat storage 5. While both hypotheses carry merit, this review focuses on the metabolic effects of fructose, the other sugar present in sucrose and HFCS. Fructose has also been implicated in driving metabolic syndrome6, and while a large exposure comes from added sugars, fructose can also be generated from glucose within the body. Despite the evidence for dietary fructose, and the potential importance of endogenous fructose production in metabolic disease, the role of fructose is often less emphasized than the energy-balance or glucose-insulin models in reviews discussing the pathogenesis of obesity and metabolic disease 5.

The physiological effects of fructose metabolism are distinct from glucose (Table 1). Fructose catabolism bypasses the regulated steps of glycolysis, with high fructose doses triggering ATP depletion, lactate and glycerate production, and fat synthesis7. The metabolic effects of fructose appear to serve as a physiological signal to trigger storage of carbohydrate as fat, to enable survival during future times of limited food availability8. While these effects were evolutionarily beneficial, the excessive intake of fructose today makes it a hazardous nutrient that contributes to obesity and metabolic disease.

Table 1.

Differences in Fructose and Glucose Actions

Glucose Effects Substantially Greater than FructoseFructose Effects Substantially Greater than Glucose

Insulin Secretion

Stimulate satiety

Increased Carbohydrate Oxidation37

Impaired Fatty Acid Oxidation25

Increased De novo Lipogenesis25,32

Increased Circulating Triglycerides (VLDL-triglycerides) 25,28

Increased Lactate production25

Acute ATP depletion121

Rapid Uric Acid Generation187

Induce Hyperinsulinemia and Insulin Resistance28,32

Stimulate Gluconeogenesis (Increase G6P phosphatase) 28

Increase Intestinal Surface (Villi) Area to Enhance Fat Absorption47

Induce Leptin Resistance128

GCK Activation28

Vasopressin Release70

FGF21 Release130

Increased Systemic Blood Pressure188

Increase Glomerular Hydrostatic Pressure189

Leaky Gut and Endotoxemia43

Reduce Satiety

Open in a new tab

DIFFERENCES IN GLUCOSE AND FRUCTOSE METABOLISMGlucose Metabolism.

Starches are polysaccharides composed of long chains of glucose molecules, serving as the primary storage form of carbohydrate in plants (like grains, potatoes, and legumes) and the major source of dietary calories worldwide. Starch is broken down into glucose in the intestinal lumen. Some of the glucose is metabolized as it passes through the intestinal wall and liver with the rest entering the systemic circulation where it is the most abundant circulating carbon source (5 – 10 mM). The rise in blood glucose stimulates a release in insulin that is further enhanced by the gut stimulation of incretins such as glucagon-like peptide-1 (GLP-1).

Glucose is primarily catabolized to generate ATP through glycolysis and oxidative phosphorylation (Figure 1). After entering cells via glucose transporters, glucose is phosphorylated to glucose 6-phosphate (G6P), converted to fructose 6-phosphate (F6P), and then to fructose 1,6-bisphosphate (F1,6P2) by phosphofructokinase-1 (PFK-1). This is cleaved into trioses that yield 2 ATP and 2 lactate (or pyruvate) per glucose. Subsequent oxidation of pyruvate through the TCA cycle and oxidative phosphorylation produces an additional 20–30 ATP per glucose molecule.

Figure 1. Hepatocyte Glucose Metabolism.

Open in a new tab

The schematic highlights insulin-stimulated (green dashed arrows) hepatic metabolic flux in the fed state that promotes energy storage. This is balanced by inhibition of glucose usage when ATP flux is high (red dashed arrows). The available pathways are denoted with light-blue boxes, the important enzymes in italics, and metabolic intermediates in bold. Abbreviations in alphabetical order are as follows: ACC, acetyl-CoA carboxylase; ACLY, ATP-citrate lyase; CoA, Coenzyme-A; CPT1a, carnitine palmitoyltransferase-1a; FAS, fatty acid synthase; F6P, fructose 6-phosphate; F1,6P2, fructose 1,6-bisphosphate; FADH2, flavin adenine dinucleotide (reduced); G6P, glucose 6-phosphate; GCK, glucokinase (hexokinase IV); GS, glycogen synthase; NADH, nicotinamide adenine dinucleotide (reduced); OxPhos, oxidative phosphorylation pathway; PDH, pyruvate dehydrogenase complex; PFK-1, phosphofructokinase-1; PPP, pentose phosphate pathway; SREBP-1c, sterol regulatory element binding protein-1c; TCA, tricarboxylic acid cycle.

Glucose metabolism is tightly regulated via actions of insulin, which promotes both glucose catabolism and glucose storage as glycogen and lipids. Glycolysis is further regulated by energy balance, with ATP inhibiting and AMP and ADP promoting the key regulated step of phosphofructokinase-1. Glycolysis is feedback inhibited by key downstream carbonaceous products including lactate and citrate (Figure 1). The balance between positive and negative regulation of glucose catabolism helps maintain energy and metabolite homeostasis. Yet glucose can still have pernicious metabolic effects beyond its calories. Specifically, insulin drives fat and glycogen storage, providing an argument for the glucose-insulin model of obesity. So how is fructose different?

Fructose Metabolism.

In contrast with glucose, fructose metabolism is not tightly regulated by insulin or feedback signals. While in principle, fructose can be metabolized by hexokinase to generate fructose 6-phosphate (F6P) with further catabolism by the same glycolytic pathway as glucose, in practice fructose mainly follows an alternative pathway (Figure 2). This alternative pathway begins with fructokinase (also known as ketohexokinase, KHK), which has higher affinity for fructose than do standard hexokinases. Crucially, KHK phosphorylates fructose on carbon one, making fructose 1-phosphate (F1P), a distinctive metabolite with signaling effects. KHK has two major isoforms. KHK-C has high affinity (Km = 0.5 mM) for fructose and is expressed in small intestine, liver, and kidney, as well as brain microglia and pancreatic islet cells9,10. KHK-A is lower affinity (Km = 10 mM) and is ubiquitously expressed but at low levels. Thus, most fructose is metabolized by KHK-C 9.

Figure 2. Fructose Metabolism.

Open in a new tab

The chemical pathway from fructose to triose phosphates is diagramed with all known end products from fructose. The structures of metabolic intermediates are shown with names in bold. The structures of the hexoses are in the b-anomeric forms utilized by the enzymes, which are noted in italics. Glycerate enters glycolysis as 2-phosphoglycerate via glycerate kinase (not shown). The major pathways, glycolysis and de novo lipogenesis (DNL) to which these triose phosphates enter (thick arrows) are noted by gray boxes.

The resulting F1P, in addition to serving as a signal, can be broken down by the enzyme F1P aldolase (also known as aldolase B) without further phosphorylation into three carbon units, bypassing the heavily regulated glycolytic enzyme PFK-1 (Figure 2). One of these three carbon units is a standard lower glycolytic intermediate (dihydroxyacetone phosphate) and the other is free phosphorylated glyceraldehyde, which has multiple potential fates: phosphorylation by triose kinase to enter lower glycolysis, reduction to glycerol, or oxidation to glycerate (see Figure 3). Consequences of the distinctive metabolic pathway for fructose include production of the pro-lipogenic signaling metabolite F1P; flow of fructose carbon into F1P, and inhibition of PFK-1 in the classic glycolytic pathway, and generation of lipogenic substrates without the positive and negative regulatory signals that normally tune glycolytic flux to energy status; and thus promotion of de novo lipogenesis (DNL). 11

Figure 3. Hepatocyte Fructose Metabolism.

Open in a new tab

The major differences with glucose metabolism are highlighted in the colored boxes, which are consequences of regulatory effects that activate (green dashed arrows) or inhibit (red dashed arrows) key enzymes or transcription factors as indicated. This regulation leads to inhibition of these pathways as indicated with red X’s. The stimulation of glucokinase (GCK) facilitates enhances glucose uptake and might further enhance fructose generation via the polyol pathway. The impacted pathways are denoted with gray boxes, the important enzymes in italics, and metabolic intermediates in bold. Other abbreviations in alphabetical order are as follows: ACC, acetyl-CoA carboxylase; ACLY, ATP-citrate lyase; AR, aldose reductase; ChREBP, carbohydrate-response element binding protein; CoA, Coenzyme-A; CPT1a, carnitine palmitoyltransferase-1a; DNL, de novo lipogenesis; FAS, fatty acid synthase; F1P, fructose 1-phosphate; F6P, fructose 6-phosphate; F1,6P2, fructose 1,6-bisphosphate; FADH2, flavin adenine dinucleotide (reduced); FBPase-1, fructose 1,6-bisphosphatase-1; GCK, glucokinase (hexokinase IV); G6Pase, glucose 6-phosphase; G6P, glucose 6-phosphate; GLUT, solute carrier family or glucose transporter; NADH, nicotinamide adenine dinucleotide (reduced); NADPH, nicotinamide adenine dinucleotide phosphate (reduced); OxPhos, oxidative phosphorylation pathway; PDH, pyruvate dehydrogenase complex; PFK-1, phosphofructokinase-1; PPP, pentose phosphate pathway; SREBP-1c, sterol regulatory element binding protein-1c; TCA, tricarboxylic acid cycle; Xu5P, xyulose 5-phosphate.

Fructose-associated signaling and non-caloric effects1]. Rapid ATP Consumption, Nucleotide Degradation and Uric Acid Generation.

The phosphorylation of fructose to F1P by KHK-C is rapid and results in a transient drop in ATP concentration in the liver (Figure 4)12,13. While seemingly paradoxical – an energy substrate depleting ATP – this reflects initial ATP-consuming steps outpacing downstream ATP-generation. This occurs because of the rapidity of the reaction and because there is no regulatory feedback to slow the reaction to preserve ATP levels.

Figure 4. Fructose Induced ATP Depletion and Uric Acid Generation.

Open in a new tab

The schematic highlights the relationship among the fructolytic, polyol, and purine degradation pathways (gray boxes). The relative fluxes are indicated by thickness of arrows with the results on metabolites indicated with up or down arrows. Activation (green dotted arrows) and inhibition (red lines) of key enzymes or transcription factors are indicated. The rapid consumption of ATP by ketohexokinase-C (KHK-C) and resulting sequestration of phosphate in fructose 1-phosphate (F1P) results in transient ATP depletion and activation of AMP deaminase-2 (AMPD2) to eventually generate uric acid. Uric acid stimulates aldose reductase (AR), KHK-C, and the transcription factors sterol regulatory element binding protein-1c (SREBP-1c) and carbohydrate-response element binding protein (ChREBP). Inosine monophosphate (IMP) inhibits aldolase B perhaps enhancing the amplification of F1P sequestration. Because humans lack uricase, which converts uric acid to allantoin, the uric acid response is greater in humans than most mammals. Other abbreviations in alphabetic order: ADP, adenosine diphosphate; AMP, adenosine monophosphate; GTP, guanidylate triphosphate; Pi, inorganic phosphate; SDH, sorbitol dehydrogenase.

Fructose-induced ATP depletion has been demonstrated in the livers of subjects taking fructose orally (75 g) as measured by magnetic resonance spectroscopy 14, as well as intravenously using lower doses15,16. ATP typically recovers within one hour. This effect is most severe in pericentral hepatocytes and is not observed in mammals similarly administered glucose.17

The fall in ATP levels is accompanied by a rise in adenosine monophosphate (AMP) and fall in guanosine triphosphate (GTP) and inorganic phosphate (Pi) levels (see Figure 4). The low GTP and Pi levels activate liver and intestinal AMP deaminase-2 (AMPD2) to convert the AMP to inosine monophosphate (IMP) 13,18. The IMP is further degraded, generating uric acid. Both intracellular and serum uric acid levels rise rapidly—within 15 to 60 minutes—following fructose ingestion, coinciding with a drop in intracellular ATP levels19. While the acute postprandial rise in serum uric acid lasts only a few hours, both fasting and postprandial serum uric acid levels rise when fructose is administered long-term20. This is mediated by stimulation of urate synthesis21.

2]. Unregulated Fructolysis.

F1P is cleaved by aldolase B to generate trioses that enter into glycolysis or gluconeogenesis but bypass PFK-1, which is the site where glycolysis is normally regulated 22. This allows for the rapid and unregulated production of glycolytic end products (see Figure 3). Isotope studies document as much as 25% of fructose is converted to lactate; other major fructose fates are resynthesis into glucose/glycogen, glycerate generation, and oxidation in the liver23.

3]. F1P as a Signaling Molecule that Enhances Glucose Metabolism.

F1P accumulates when KHK makes F1P faster than F1P is cleaved by aldolase B. Such accumulation is augmented by an inhibition of aldolase B by the IMP24. In the setting of fructose consumption, the accumulated F1P functions as a nutrient-sensing signal. One of the signaling effects of F1P is activation of liver glucose catabolism via glucokinase, the low-affinity liver-specific hexokinase. F1P dissociates glucokinase from its regulatory protein (GCKR) in the hepatocyte nucleus. Glucokinase then translocates to the cytosol where it stimulates liver glucose uptake and production of G6P that can feed glycogen synthesis or the pentose phosphate pathway; however, glycolysis is attenuated due to PFK-1 inhibition (Figure 2).25 The result is a buildup of these hexoses that activates the polyol pathway.

Thus, in addition to being made from dietary fructose, F1P may also arise from an alternative pathway of glucose catabolism, in which glucose is first converted to fructose before being broken down. Hepatic flux from glucose to F1P has been demonstrated by isotope tracing.26 The first enzyme of this pathway, aldose reductase (AR), has a similar affinity for glucose as GCK and hence this pathway may be substantial in the liver (Figure 2). Thus, in addition to directly being catabolized into pyruvate/lactate, fructose can also promote liver glucose catabolism.

4]. ChREBP Orchestrates Fructose-mediated Events.

Fructose rapidly activates the carbohydrate-sensing transcription factor (ChREBP) and induces expression‎ of the ChREBP-β isoform in the intestine and liver6,27,28. In the liver, ChREBP may be activated by metabolites produced from fructose like glycerol 3-phosphate29 or xylulose 5-phosphate30. It may also be triggered by fructose’s induction of uric acid31 and of glucose-derived metabolites whose levels rise in response to glucokinase activation by F1P28. ChREBP drives many fructose-dependent pathways, including glycolysis, fructolysis, gluconeogenesis, and de novo lipogenesis (DNL) (see Figures 3)28. DNL is stimulated by fructose to a greater extent than glucose32-35 and this is mediated in part by ChREBP28 and the sterol regulatory element-binding protein (SREBP1c)32. Activation of both transcriptional pathways is greater with fructose than glucose32.

Cellular and Physiological role of Fructose signaling1]. Shift from Fatty Acid to Carbohydrate Oxidation.

Fructose ingestion causes a remarkable shift from fatty-acid to carbohydrate oxidation as measured by an increase in the respiratory exchange ratio (RER) that is greater than that observed with glucose36-38. This occurs despite glucose more strongly inducing circulating insulin, a major effector of increased RER. The decrease in fatty-acid oxidation following fructose ingestion is distinct from glucose and is greatest in the postprandial period36,37, and may be mediated, in part, by fructose’s ability to increase hepatocellular malonyl CoA, which drives fatty acid synthesis and inhibits fat oxidation (see Figure 3)25,38. However, inhibition of hepatic fatty acid oxidation is likely insufficient to account for this acute, systemic shift to carbohydrate oxidation and reduced fatty acid oxidation in skeletal muscle is also likely required36.

2]. De Novo Lipogenesis (DNL).

Several mechanisms have been identified for how fructose metabolism stimulates DNL in the liver. First, fructose metabolism is associated with citrate accumulation, perhaps by inhibition of mitochondrial aconitase. Citrate is a substrate for ATP citrate lyase (ACLY) and subsequent DNL (Figure 3)39. Excess fructose intake can also lead to passage of fructose to the colon, where it can be metabolized to acetate, that subsequently can reach the portal vein and liver, to be converted to acetyl CoA and fat40. Fructose metabolism in the intestine also causes local ATP depletion, F1P accumulation, and ChREBP activation27,41, which may disrupt tight junctions and cause leaky gut42. This leakiness can produce endotoxemia that stimulates cytokine release by hepatic macrophages and Kupffer cells that can contribute to upregulation of enzymes involved in DNL43. Suppression of endotoxemia with antibiotics in fructose-fed mice reduces hepatic fat accumulation, as can blocking KHK metabolism specifically in the liver44. This suggests that the fatty liver induced by fructose is dependent on multiple mechanisms including substrate availability, enzymatic activity, and inflammation.

3]. Increased Circulating Triglycerides.

Fructose consumption elevates circulating triglycerides, which typically peak 4 to 6 hours after ingestion. The magnitude of triglyceride increase depends on the dose and duration of fructose exposure and is associated with elevated postprandial serum apolipoprotein B34,45 and in the long-term with increased visceral obesity34. These effects are not observed to the same extent with equivalent glucose consumption34,45.

Whether or not fructose carbons are incorporated into the circulating triglycerides is a matter of debate. Fructose increases the synthesis of triglycerides as noted by studies using labeled fatty acids (palmitate) or fructose37. One isotope tracer study found that nearly 40 percent of the glycerol fraction of triacylglycerol contained fructose carbons (probably from the glycerol-3-phosphate generated during fructose catabolism, see Figure 2), while only a small amount of fructose carbon (<1%) were present in the fatty acids of the VLDL-triacylglycerol37. However, fructose carbons are incorporated into lysophosphatidyl-cholines, and these lipid species have not been the focus of prior isotopic tracing studies46. Thus, fructose contributes to DNL both through substrate provision and lipogenic signaling.

Other mechanisms for the increase in circulating triglycerides have been identified. First, fructose-containing SSBs increases intestinal absorption of fats in mice in part by lengthening villi47. Metabolism of fructose also results in impaired clearance of VLDL-triglycerides48, possibly by altering lipolysis in adipose, or by decreased removal of VLDL-triglycerides by the liver37 that is linked with a fructose-dependent rise of apolipoprotein CIII 49.

4]. Hepatic and Systemic Insulin Resistance.

While dietary fructose does not elicit an insulin response due to lack of the fructose transporter, GLUT5, on the islet cells, over time there is development of both hepatic and peripheral insulin resistance and associated hyperinsulinemia32,34. In contrast glucose-enriched diets induce hepatocellular insulin resistance to a lesser degree32. Hepatic insulin resistance is linked with fructose-induced steatosis and accumulation of specific lipid species like ceramides50, a reduction in insulin signaling due to a reduction in insulin receptor and insulin-receptor substrate-2 (IRS2) expression‎, increased protein-tyrosine phosphatase 1B (PTP1b),32 and inhibition of the PI3K/Akt signaling pathway32. In addition, fructose-induced glucose production is enhanced by ChREBP-mediated upregulation of G6P phosphatase (G6Pase), the final step in gluconeogenesis28. Hepatic insulin resistance in humans has been reported two to three weeks after ingestion of fructose-based beverages, but not glucose-based beverages51,52.

Systemic insulin resistance has been documented with fructose feeding in rats, with impaired insulin signaling in skeletal muscle and adipocytes53. In skeletal muscle there is impaired translocation of the glucose transporter, GLUT4, to the plasma membrane, and some reports of impaired insulin mediated Akt phosphorylation, possibly related to skeletal muscle accumulation of advanced glycation end products (AGEs)54,55. Similarly impaired Akt phosphorylation also develops in adipose tissue of fructose-fed rats53. There is evidence that the peripheral insulin resistance may develop secondary to events downstream of KHK-mediated fructose metabolism in the liver.56,57 Similarly, the administration of fructose to humans alters fat oxidation and mitochondrial function in skeletal muscle58, and, similar to rats, hepatic insulin resistance occurs first followed by systemic insulin resistance51,59.

Of interest, cerebral insulin resistance develops in response to dietary fructose manifesting as reduction in the signaling of the insulin receptor and IRS160. The cerebral effects appear to be mediated by local fructose metabolism by microglia expressing KHK10.

5]. Biologic Effects from Nucleotide Degradation.

Activation of the nucleotide-degradation pathway by fructose may lead to biological consequences beyond altered nucleotide levels. These may include the stimulation of intracellular and mitochondrial oxidative stress, which can occur via uric acid-mediated translocation of NADPH oxidase to the mitochondria39. Uric acid and associated oxidative stress may also impact the activity of other key metabolic enzymes and regulatory factors, including inhibiting aconitase in the TCA cycle50,61 and enoyl CoA hydratase, an enzyme in β-oxidation31. A rise in intracellular uric acid can also activate ChREBP31 and SREBP-1c62, as well as stimulate expression‎ and activation of KHK31 and AR,63 which can lead to further fructose generation (Figure 4). Uric acid also inhibits AMP-activated protein kinase (AMPK)64, and stimulates proinflammatory and vasoconstrictive factors 65.

6]. Effects of Fructose on Mitochondrial Function.

Studies in fructose-fed animals document that fructose-induced ATP consumption is commonly associated with a compensatory stimulation of the TCA cycle and enhanced mitochondrial respiration and coupling26,50,66. Mitochondrial oxidative stress due to translocation of NADPH oxidase to the mitochondria also occurs10,39 associated with downregulation of nuclear factor erythroid 2-related factor-2 (NRF2) and antioxidant pathways50. Possibly as a consequence, long-term consumption in rats of high-fructose corn syrup has been reported to adversely affect liver mitochondrial function, with decreased mitochondrial numbers, disrupted cristae, reduced ATP production and decreased energy efficiency 61,67,68. While studies in humans remain limited, there is some evidence that fructose impairs mitochondrial function in muscle58. Accordingly, reducing fructose intake could potentially enhance mitochondrial health 69.

7]. Vasopressin Regulation.

Much like glucose stimulates a hormonal response in the form of insulin, fructose stimulates production and release of the hormone vasopressin, also known as antidiuretic hormone70. Vasopressin is synthesized in the hypothalamus and released by the pituitary to increase water retention and blood pressure. The rise in plasma vasopressin levels in humans in response to fructose does not occur with equimolar glucose70. Studies in mice show that the release is dependent on fructose metabolism by KHK71. SSBs also increase vasopressin (as reflected by an increase in plasma copeptin, a stable cleavage product of the vasopressin pro-protein)72, and circulating copeptin levels both predict and are elevated in subjects with metabolic syndrome and diabetes73.

The arginine vasopressin receptor 1a (V1a) receptor is responsible for the increased systemic blood pressure triggered by vasopressin. In contrast, the V1b receptor mediates vasopressin-induced production of adrenocorticotropic hormone (ACTH). ACTH stimulates the release of cortisol and glucagon, powerful glucose-elevating hormones. Mice lacking vasopressin V1b receptor (but not V1a receptor) have lower cortisol, glucagon, and liver KHK and are protected from fructose-induced metabolic syndrome71. This metabolic improvement may not be specific to fructose, as mice lacking V1b show reduced ACTH and corticosterone responses in response to other stressors such as forced swimming. The role of vasopressin suggests that better hydration, which lowers vasopressin, could potentially mitigate metabolic syndrome, with proof-of-concept shown in mice71. Thus, vasopressin is a fructose-induced hormone with strong connections to blood pressure and metabolic syndrome. Both the ancient reasons for evolution of the fructose-vasopressin circuit and the clinical potential of its targeting to treat modern metabolic problems merit further investigation.

DIETARY FRUCTOSE ABSORPTION AND METABOLISM

Table sugar (sucrose) is a disaccharide containing equimolar fructose and glucose, whereas high fructose corn syrup (HFCS) is most commonly a 55:45% mixture of fructose and glucose. The major source of dietary fructose comes from SSBs and ultra-processed foods with abundant sucrose or HFCS, but fructose is also high in fruits, fruit juices, and honey. Small amounts of fructose are also present in certain vegetables (such as sweet potatoes and carrots).

Whereas HFCS sugars are ingested as monosaccharides, the sucrose disaccharide is cleaved into glucose and fructose in the small intestine by sucrase-isomaltase. Despite this extra enzymatic step, there appears to be minimal difference in the physiological effects of sucrose and HFCS74. Glucose is absorbed by enterocytes in the small intestine by the sodium dependent transporter SGLT1, while fructose is taken up by GLUT5, and both exit into the portal venous system through the GLUT2 transporter.

Once in the enterocyte, fructose is metabolized by KHK-C, generating primarily glucose and organic acids (Figure 5).75 With low intake, most fructose is metabolized in the intestine, but higher intakes can saturate intestinal metabolism and lead to greater fructose delivery to the liver. Thus, intestinal metabolism functions to shield the liver from small quantities of ingested fructose.75 The fructose clearance rate of the human intestine is uncertain. Fructose metabolism in the gut also enlarges intestinal villi, increases absorption of high-fat foods, and may induce leaky gut by disrupting tight junctions.

Figure 5. Sites of Metabolism for Fructose.

Open in a new tab

A) Dietary Fructose: Fructose metabolism begins with absorption by enterocytes of the small intestine, gets exported to the portal circulation, and picked up by the liver and systemic circulation. B) Endogenous Fructose: Fructose can also be produced endogenously from glucose through the polyol pathway (blue box). Fructose production is regulated by Aldose reductase. Those tissues with constitutive expression‎ are listed in black in box on the left. Other tissues require induced expression‎ by various stresses listed in brown in middle box. Created in BioRender. Sánchez-Lozada, L.G. (2025)

Fructose absorption is affected by many factors. For example, it is enhanced by the presence of glucose via mechanisms still not fully understood76. Expression‎ of GLUT5, the essential luminal fructose transporter, is low in infancy, but expression‎ increases in response to dietary fructose and sugar77. This increase is dependent on ChREBP27. Higher GLUT5 expression‎ is associated with a higher risk for obesity in children78.

Fructose-based drinks are also more effective than fructose-based foods to induce KHK-dependent metabolism and downstream events in the liver79. Not only is this due to the amount ingested, but the speed of ingestion72, likely because rapid intake leads to saturation of intestinal fructose clearance and higher fructose concentrations in the liver. The presence of fiber can also slow absorption, and, along with ascorbate, potassium, and flavanols, slow or inhibit fructose metabolism. This may help explain why ingestion of fruit, the main source of natural fructose, does not typically result in adverse metabolic effects8,80. Ingestion of inulin (a fructan polymer consisting of a linear chain of fructose molecules) can stimulate the growth of bacteria that will degrade fructose in the gut and protect mice from fructose-induced fatty liver81. However, other fructans (such as levans that form net-like structures of fructose) can be degraded by bacteria such as Streptococcus that releases fructose, and may induce metabolic syndrome, at least in horses42.

Of the dietary fructose that reaches the liver, the majority is metabolized on first pass. Many of the detrimental metabolic effects of fructose are mediated by its metabolism in the liver, and liver-specific KHK knockout mice are protected from fructose-induced fatty liver, insulin resistance, and obesity57. Only 10 to 20 percent of the ingested fructose enters the systemic circulation6,75,82, where it is metabolized by other tissues (with contributions from the kidney, muscle, heart, adipose tissue, lung, brain, and circulating monocytes9,10,83-85) or excreted in the urine. Overall, the first-pass metabolism of fructose aligns with its role as an ancient signal to store carbohydrate as fat, with intestinal clearance ensuring that the liver ‘fat switch’ is turned on only when fructose is sufficiently abundant.

ENDOGENOUS FRUCTOSE PRODUCTION AND METABOLISM

Fructose can be converted to glucose via gluconeogenesis where fructolytic products are diverted to F1,6P2, F6P, G6P and then to glucose (see Figure 3). Conversely, glucose can be transformed to fructose via the polyol pathway: glucose is first converted to sorbitol by aldose reductase (AR) followed by conversion to fructose by sorbitol dehydrogenase (SDH) (see Figure 4).

Constitutive production of fructose has been documented in human reproductive organs including in the placenta in early pregnancy where fructose is present in amniotic fluid86 and in the prostate gland and seminal vesicles where fructose concentrates in semen87. Sperm use hexokinase and not KHK to metabolize fructose87, and both mice and humans lacking KHK are fertile.

AR, the committal enzyme, is not normally expressed in non-reproductive tissues, except for the renal medulla where it generates sorbitol as an intracellular osmolyte that protects kidney epithelial cells from osmotic stress. Importantly however, AR can be induced in other tissues by stimuli including hyperglycemia, hyperosmolarity, hypoxia, ischemia, heat, trauma, hyperuricemia, and glucokinase activation (Figure 5 and Table 2) 88-97.

Table 2.

Factors that Activate Endogenous Fructose Production

Type of StressExamples

Hyperosmolarity	Water Deprivation102
	Salt intake88
	Alcohol intake149
Hyperglycemia	Hyperglycemia and Diabetes100
	High glycemic diet98
Low Oxygen	Hypoxia (Sleep apnea)116
Ischemia	Myocardial Infarction, Acute Kidney Injury 92,101
Fructose Metabolism by KHK (such as GCK activation)	Western diet-induced steatosis148
Heat Stress	Heat Stress-induced kidney injury 95
Physical or Emotional Stress	Following major surgery96
Pregnancy	Brain production97
Uric acid	Hyperuricemia63

Open in a new tab

Historically, the role of endogenous fructose was considered minimal. However, a study in mice found that chronic glucose ingestion led to obesity and metabolic syndrome accompanied by induction of AR with accumulation of fructose and sorbitol in the liver98. Importantly, mice lacking either AR or KHK were protected from developing fatty liver or insulin resistance98.

Endogenous fructose production is increased with high salt88 or alcohol intake99 and this endogenously produced fructose may contribute to metabolic syndrome and fatty liver in these contexts. Experimental studies have now identified increased endogenous fructose production in many contexts, including diabetes, ischemic or heat-stressed kidneys, ischemic heart, and the brains of animals with diabetes, head trauma, dehydration, or intake of high salt or high sugar diets84,92,100-102. Systemic hypoxia can also increase serum fructose levels in the naked mole rat, and this facilitates survival in hypoxic burrows, nominally by promoting glycolysis103.

Evidence for endogenous fructose production in humans is also emerging. Francey et al documented a tripling of the endogenous fructose production rate following administration of an oral glucose-fructose beverage in healthy people82. ‘Clamping’ blood glucose at hyperglycemic levels in healthy individuals induced fructose production in human brain104. Interestingly, fructose and sorbitol levels are elevated in brain tissue or cerebral spinal fluid in multiple sclerosis, cranial hypertension, Alzheimer’s disease, bipolar disease, pregnancy, and individuals with brain tumors 105-111. Various cancers have also been shown to have AR expression‎ associated with fructose production112. Diabetes, obstructive sleep apnea, and chronic kidney disease are all associated with increased fasting circulating fructose levels in human populations suggestive of increased endogenous fructose production in these conditions. 113-116

These studies suggest that the potential exposure to fructose is high in western societies, not only from added sugars in the diet, but due to the high dietary content of carbohydrate that can provide substrate for fructose production, a process amplified by intake of salty foods and alcohol.

BIOLOGICAL CONSEQUENCES OF FRUCTOSE METABOLISM

An important question is how fructose metabolism translates into its biological effects. Indeed, it is the unique aspects related to F1P signaling, ATP depletion, uric acid generation, and stimulation of glycolytic and lipogenic pathways that drive the biological effects of fructose metabolism.

Taste, Hunger and Food Intake

Sugar (sucrose) and HFCS are sweet and palatable, and the sweet signals activate orexigenic pathways, including an amygdala to hypothalamic circuit that stimulates consumption of sweet foods117. However, the increase in food intake cannot fully explain the effects of fructose on systemic metabolism. For example, mice lacking sweet taste receptors still develop a dopamine response to sucrose whereas this does not occur with artificial sugars118. Indeed, mice lacking taste still seek sugar (or fructose), and, although they drink less than normal mice, they still develop metabolic syndrome119.

KHK knockout mice no longer prefer fructose, although they continue to prefer glucose. Mice lacking KHK only in the intestine show reduced preference for fructose, but demonstrate enhanced delivery of fructose to the liver that leads to increased propensity for metabolic syndrome 57,75. It is possible that mice lacking intestinal KHK may have some impaired absorption of fructose that could cause subclinical gastrointestinal side-effects that lead to less intake57. Mice with liver-specific KHK knockout continue to show marked preference for and excessive intake of fructose, but are protected from weight gain or metabolic syndrome57. These studies show that while the taste and palatability of fructose encourages intake, it is the metabolism of fructose in the liver that drives metabolic disease.

Central orexigenic and other pathways are also likely important for how fructose induces hunger and food intake120. For example, glucose injected into the cerebral ventricles of mice causes a rise in brain ATP levels, induces satiety and curtails food intake121. In contrast, injection of fructose lowers ATP, stimulates hunger, and increases food intake121. Fructose ingestion also causes in humans a reduction in cerebral cortical blood flow to brain areas involved in self-control while stimulating occipital visual centers to identify appealing foods122,123, consistent with the stimulation of foraging-like activity124,125. In contrast, glucose acutely increases cortical blood flow and satiety. The liver ATP depletion that occurs with fructose also stimulates hunger126.

Nevertheless, mice provided fructose in drinking water initially compensate for the fructose caloric intake by lowering their intake of chow to maintain overall caloric balance127. This can last for several weeks, but then food intake gradually increases with an increase in weight. Experimentally this is associated with the development of central leptin resistance.128,129

While experimental studies favor fructose metabolism in driving hunger and weight gain, fructose (and sucrose) stimulates hepatic fibroblast growth factor 21 (FGF21) that acts to reduce sugar intake and sweet preference in human. Ingestion of glucose is less potent than fructose in stimulating FGF21130. Studies in laboratory mice suggest the stimulation of FGF21 is dependent on fructose metabolism by hepatic KHK57. These studies suggest fructose metabolism may also have a negative feedback system to limit excessive sugar intake.

Metabolic Syndrome and Obesity

The administration of fructose to mice and rats is commonly used to induce obesity, metabolic syndrome, diabetes, and fatty liver. However, large doses are often required when incorporated into food (50 to 60%) and the most effective method is to provide fructose in the drinking water as HFCS at concentrations similar to SSBs131. Studies in mice have found that the metabolic syndrome can be prevented if KHK-C is knocked out or inhibited 32,127,132. While this could be attributed to that fact that KHK KO mice drink less fructose than wild type controls, the protection persists when fructose consumption is matched127.

In humans the metabolic effects of fructose are also best observed with administration of fructose in drinks rather than in food, and especially in individuals who are older, overweight or have borderline insulin resistance133. One study in healthy, physically active and lean young adults found minimal effects when given packets of high dose (150 g/d) crystalline fructose daily for 8 weeks134, while another study in overweight men found that high doses of fructose in the drinking water (200g/d) resulted in increases in blood pressure, fasting insulin and serum triglycerides within 2 weeks 135.

In a prospective clinical trial, fructose-sweetened or glucose-sweetened beverages were administered as 25 percent of the energy requirement to overweight or obese individuals for 10 weeks. Fructose supplementation promoted features of metabolic syndrome including decreased fat oxidation and insulin sensitivity and increased visceral obesity, DNL, postprandial plasma triglycerides and uric acid, and fasting apolipoprotein B, LDL cholesterol, plasminogen activator inhibitor-1 (PAI-1), and monocyte chemotactic protein-2 (MCP-1) compared to the group receiving glucose20,36,136.

While fructose induces worse metabolic effects than glucose in humans, most SSBs include both fructose and glucose. Moreover, and as mentioned earlier, experimental studies suggest that over time glucose-sweetened beverages may increase endogenous fructose production.57 This can potentially confound studies that attempt to distinguish effects of these two sugars.

Another confounding issue is the common cooccurrence of excess calorie intake and fructose consumption. If caloric intake is controlled, fructose consumption has a minimal effect on body weight in both experimental77,137 and human138 studies. Any observed modest weight increases are likely attributable to higher expression‎ of transporters (GLUT5) for fructose 78,139 or greater intestinal villous area that allows more efficient absorption of fat47.

As evidenced by pair-feeding studies, however, sugar, HFCS, glucose, or fructose can produce metabolic disease even if weight does not increase77,98,140,141. For example, one study found that rats can develop diabetes, fatty liver and hypertension on a high sugar diet despite caloric restriction77. Consistent with this observation, men consuming a weight-maintaining diet that included fructose-sweetened beverage for 9 days (versus a weight-maintaining diet with the fructose beverage) showed decreased fat oxidation and increased DNL, liver fat, postprandial triglyceride and hepatic insulin resistance59. In another study, isocaloric restriction of fructose in obese adolescents improved features of metabolic syndrome142. Thus, even in the absence of excessive caloric intake, sugar can drive metabolic disease.

Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) and Alcoholic Liver Disease

SSBs are strongly linked with the development and severity of MASLD, likely due to the effect of fructose to stimulate DNL and to inhibit fatty-acid oxidation 143. Subjects with MASLD have been found to have elevations of KHK mRNA in the liver biopsies144. Administration of SSBs has also been found to increase liver fat 74,145 while reducing fructose intake can decrease liver fat 146,147. Experimental studies have also found that knocking out KHK can substantially reduce fatty liver in mice given a high fat, high sugar diet as well as a diet supplemented with HFCS57,148. Fatty liver from high glycemic or high salt diets are also associated with endogenous fructose production and are prevented in mice lacking KHK88,98. These studies provide strong support for fructose as a key driver of MASLD.

Alcohol intake can also lead to fatty liver and cirrhosis with histological and biochemical similarities to MASLD. Indeed, the combination of fructose with alcohol can markedly accelerate development of liver disease99. More recently it has been shown that alcohol-induced liver disease also involves endogenous fructose, likely resulting from induced AR due to alcohol’s osmotic effects.99,149,150. AR is also induced in the liver of patients with alcoholic liver disease. Mice lacking AR or KHK are largely protected from alcohol-induced liver disease. Of interest, mice lacking KHK or given KHK inhibitors also drink less ethanol99.

Hypertension, Kidney Disease, and Other Conditions

A variety of other conditions have been linked with fructose metabolism. Primary hypertension, for example, is strongly linked with intake of SSBs. There is also an interesting interaction, as fructose enhances sodium absorption in the gut151 and sodium reabsorption in the kidney tubules151,152, and high salt intake raises osmolality and induces endogenous fructose production88. Perhaps most strikingly, mice lacking KHK are protected from salt-induced blood pressure elevation and cardiac hypertrophy88,153.

Systemic inflammation can potentially also be induced by fructose, by effects on circulating monocytes83,85 and the release of chemokines into the circulation136. Local inflammation can also be induced, such as the release of inflammatory mediators by kidney epithelial cells154, or by the activation of local macrophage-like cells including Kupffer cells 155 and microglia10.

Acute and chronic kidney disease may also be dependent on endogenous fructose production, including in experimental models of acute ischemic tubular injury, contrast nephropathy, heat stress induced chronic kidney disease, and diabetic nephropathy92,100,102. In mouse models of these conditions, endogenous fructose production was shown to occur in the kidney, and mice lacking the ability to metabolize the fructose (i.e., KHK KO mice) were protected from kidney damage.

Ingestion of SSBs also increases the risk of gout. Humans and great apes are much more susceptible to gout from fructose as they lack the hepatic enzyme, uricase, that helps regulate uric acid by degrading it to allantoin. Indeed, mice or rats that have uricase knocked out or inhibited not only have higher uric acid levels but show enhanced risk for hypertension, kidney disease and metabolic syndrome and also a relatively greater metabolic response to fructose (reviewed in 156).

One of the more interesting findings involved the process of aging. Mice and rats are known to develop aging-associated kidney disease, similar to humans. However, a study of aging KHK knockout mice on a high carbohydrate but sugar-free chow found that they were protected from age-associated kidney disease and hypertension157. Since the dietary carbohydrates were corn, soybean and wheat, all with minimal fructose content, it is likely that the aging-associated changes in the wild-type mice resulted from endogenous fructose.

Cancer

Fructose metabolism has emerged as tumor promoter, supporting both tumor-intrinsic and systemic processes that promote disease progression. In multiple cancers, including breast, gastric, lung, hepatic, pancreatic, and prostate, fructose metabolism has been implicated in cancer growth, proliferation, and metastasis.158-162 In several instances, fructose is produced endogenously by tumor cells via the polyol pathway, which facilitates cancer cell metabolism, migration, and resistance to apoptosis112. In cases where tumors have access to fructose-rich environments, such as the intestine and liver, fructose can be directly obtained and utilized from dietary sources163,164. That said, hepatocellular carcinoma loses fructolytic capacity compared to healthy liver and due to the expression‎ of KHK-A instead of KHK-C, may contribute to expression‎ of the oncogenic antioxidant transcription factor Nrf2165,166. Dietary fructose can also promote tumor growth in organs without direct access to dietary nutrients via fructose-derived metabolites, like lactate and lipids, which are produced by the liver following first-pass metabolism46. Within cancer cells, fructose fuels glycolysis and DNL to support macromolecule synthesis and proliferation46,163,167. Moreover, the major product of fructolysis, F1P, functions as a survival signal by increasing HIF1α transactivation, thereby promoting adaptation to hypoxic conditions163. Furthermore, fructose metabolism may shape the tumor microenvironment by influencing the activity of immunosuppressive cell populations168,169. In sum, through multiple mechanisms that merit further investigation, fructose likely promotes cancer.

Brain Disorders

The human brain makes endogenous fructose (such as from hyperglycemia)104 and also responds to dietary fructose. Ingestion of fructose stimulates foraging-like behavior with activation of occipital centers involved in food cues while suppressing blood flow to areas associated with self-control (cortex) and recent memory (hippocampus), while glucose has opposing effects122,123,170. Chronic inhibition of cortical and hippocampal function from excessive ingestion or production of fructose has been postulated to increase the risk of behavioral disorders and dementia124,125. High fructose and sorbitol levels in the brain occur in both bipolar disease106,171 and dementia111. Microglia contain KHK-C and are activated by fructose 10, and when this occurs in pregnant mice, the progeny display anxiety disorders172.

Fructose metabolism in the brain may have a role in Alzheimer’s disease125. Rats ingesting fructose chronically develop cognitive dysfunction, with cerebral insulin resistance, mitochondrial dysfunction, neuroinflammation, and eventually amyloid plaques and tau protein aggregates60,173-175. Diabetic mice also have increased endogenous fructose production with cognitive dysfunction that is improved by targeted knockout of KHK-C in the hippocampus10. Intake of sugar and high glycemic foods are also risk factors for Alzheimer’s disease, and this disease is associated with 5 to 6-fold higher levels of fructose and sorbitol in the brain in autopsy studies. 111

Genetic Disorders

Two rare genetic disorders of fructose metabolism are essential fructosuria (KHK deficiency) and hereditary fructose intolerance (HFI, aldolase B deficiency). The former is asymptomatic, while the latter results in a pathological reaction to fructose, characterized by hypoglycemia and lactic acidosis acutely and MASLD chronically (Box 1)176-178. Both experimental and pilot clinical studies suggest that inhibition of KHK could be a promising therapy for subjects with HFI 177,179. There are also ongoing experimental studies to genetically edit some of the common mutations that cause HFI in the hope of a permanent cure (M Lanaspa and D Tolan, unpublished data).

Box 1. Genetic Disorders of Fructose Metabolism.

Essential FructosuriaHereditary Fructose Intolerance

  • Fructokinase (KHK) Deficiency; Autosomal recessive (< 1 per 130,000).

  • Normal lifespan, Asymptomatic.

  • Prefer salty over sweet foods.

  • May be protective against obesity or type 2 diabetes (no reports of obesity of type 2 diabetes in affected individuals)

  • Excrete 10 percent of ingested fructose load in urine. Most ingested fructose is metabolized in adipose and other tissues via other hexokinases.

  • Aldolase B deficiency; Autosomal recessive (1:22,000).

  • Manifests after weaning with exposure to fructose.

  • Presents with acute hypoglycemia upon fructose intake, due to impaired hepatic glucogenesis.

  • Biochemically, F1P accumulates pathologically, ATP is depleted.

  • Causes chronic liver disease and kidney disease (Fanconi syndrome).

  • Rescued by inhibition or knocking out KHK.

  • Management consists of dietary fructose and sorbitol avoidance.

  • Associated with fructose, sorbitol and ethanol aversion.

  • Restriction of high glycemic carbohydrate to reduce endogenous fructose generation may also be helpful.

Open in a new tab

CONCLUSIONS AND FUTURE DIRECTIONSOpen questions

The role of endogenously produced fructose in metabolic disease remains controversial. While the risk of SSBs on metabolic health is robust, there is less evidence that fructose from other sources carries risk. Interpretation of epidemiological studies is confounded by fruits, which are rich in fructose but carry countering substances. There are also only limited studies in humans to determine the contribution of endogenous fructose.

The temporal relationship between sugar intake and metabolic disease trends remains debated. If fructose and sugar are important drivers of metabolic syndrome, then why has decreased intake of added sugars during the last two decades failed, prior to the common use of GLP-1 agonists, to slow the obesity and diabetes epidemic? Historically, there has been a 10–20 year lag between sugar exposure and metabolic outcomes180, and changes in incidence often precede changes in preval‎ence. Indeed, while the preval‎ence of diabetes has not decreased, the incidence of diabetes began to fall soon after sugar reduction181, while obesity preval‎ence plateaued around 2020 and began decreasing by 2023182. Importantly, while recent changes in obesity preval‎ence are impacted by widespread GLP-1 therapy, the decrease in diabetes incidence predates this.

The role of uric acid as a mediator of fructose’s metabolic effects is contentious. Experimental studies link fructose metabolism to increased uric acid and metabolic dysfunction, but Mendelian randomization and large clinical trials do not support uric acid as a causal mediator of cardiovascular or renal disease. These discrepancies may stem from genetic studies measuring serum rather than tissue uric acid, and from trial designs that inadequately capture tissue-specific effects156.

Finally, whether targeting fructose metabolism through KHK inhibition offers therapeutic benefit remains uncertain. Two pharmaceutical companies abandoned KHK inhibitor programs after phase 2 studies yielded modest outcomes183,184. Pfizer’s PF-06835919 initial studies produced only mild reductions in liver fat (~20%) without significant improvements in insulin, uric acid, inflammation, or weight. These limited benefits may reflect a short duration of treatment, suboptimal dosing, off-target effects, the need to use it in a more select population (such as those with high dietary intake of fructose), and also the possibility that KHK inhibition is more effective for disease prevention than treatment185. As mentioned earlier, there is some clinical evidence that KHK inhibition may benefit subjects with HFI 179. Given KHK’s broad metabolic role, its inhibition may still hold future promise for improving metabolic health.

Conclusion

In conclusion, fructose has multiple distinctive metabolic effects beyond standard carbohydrate metabolism. It produces signals that promote hepatic glycolysis and lipogenesis. This pro-lipogenic response presumably reflects an ancient evolutionary adaptive program intended to prepare for periods of food scarcity (Figure 6). Fructose’s induction of vasopressin may similarly reflect preparation for water scarcity, with fat storage also supplying metabolic water186. In modern society, however, this ancient signaling role can backfire. In the context of consistently abundant food, fructose intake is a hazard, promoting insulin resistance, hypertriglyceridemia, fatty liver, and elevated blood pressure8.

Figure 6. The Fructose Survival Hypothesis.

Open in a new tab

Fructose likely provided an evolutionary advantage by activating a “survival pathway” that increased hunger, stimulated foraging, and promoted efficient fat storage, traits that helped early hominids endure periods of famine. In modern environments with constant food availability, this once-beneficial mechanism can contribute to metabolic disease. This concept is described extensively in the literature8,125 and has been proposed as the “fructose survival hypothesis Created in BioRender. Lanaspa, M. (2025) https://BioRender.com/7oh5q9r.

Acknowledgements

We thank Soyoung Bae, PhD, for her helpful comments and help with the figures.

Supported by NIH CA258697 and DK132427 (M.D.G.) and UO1 AA027997 (R.J.J)

Footnotes

Competing Interests. RJJ, MAL, DRT and JDR have equity with Colorado Research Partners LLC, which is developing KHK inhibitors. RJJ also consults for Amgen, Dynamicure LLC, Soba Pharmaceuticals, and Kibow LLC and is on the Scientific Board of Santa Barbara Nutrients and RxSugar, Inc. MDG holds equity in Faeth Therapeutics and Skye Biosciences; reports consulting or advisory roles with Almac Discovery, Genentech Inc, Faeth Therapeutics, Scorpion Therapeutics, and Skye Biosciences; patents, royalties, and other intellectual property with Weill Cornell Medicine and Faeth Therapeutics. JDR is a member of the Rutgers Cancer Institute (RCI) and the University of Pennsylvania Diabetes Research Center (U Penn DRC); director of the U Penn DRC-Princeton inter-institutional metabolomics core and RCI metabolomics core; advisor and stockholder in Bantam Pharmaceuticals, Rafael Pharmaceuticals, and Empress Therapeutics; a founder, director, and stockholder of Farber Partners and Raze Therapeutics; a founder, advisor, and stockholder in Marea Therapeutics and Fargo Biotechnologies; and inventor of patents held by Princeton University. SS and MAH have no disclosures.

References

• 1.Malik VS & Hu FB The role of sugar-sweetened beverages in the global epidemics of obesity and chronic diseases. Nat Rev Endocrinol (2022). 10.1038/s41574-021-00627-6 [DOI] [Google Scholar]
• 2.Organization, W. H. 1–59 (World Health Organization, Geneva, Switzerland, 2015). [Google Scholar]
• 3.Yang Q. et al. Added Sugar Intake and Cardiovascular Diseases Mortality Among US Adults. JAMA Intern Med (2014). [Google Scholar]
• 4.Russell C, Baker P, Grimes C, Lindberg R & Lawrence MA Global trends in added sugars and non-nutritive sweetener use in the packaged food supply: drivers and implications for public health. Public Health Nutr 26, 952–964 (2023). 10.1017/S1368980022001598 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 5.Magkos F. et al. On the pathogenesis of obesity: causal models and missing pieces of the puzzle. Nat Metab (2024). 10.1038/s42255-024-01106-8 [DOI] [Google Scholar]
• 6.Herman MA & Birnbaum MJ Molecular aspects of fructose metabolism and metabolic disease. Cell Metab 33, 2329–2354 (2021). 10.1016/j.cmet.2021.09.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 7.Woods HF & Krebs HA Lactate production in the perfused rat liver. Biochem J 125, 129–139 (1971). 10.1042/bj1250129 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 8.Johnson RJ et al. The Fructose Survival Hypothesis for Obesity. Philos Trans R Soc Lond B Biol Sci 378 (2023). 10.1098/rstb.2022.0230 [DOI] [Google Scholar]
• 9.Diggle CP et al. Ketohexokinase: expression‎ and localization of the principal fructose-metabolizing enzyme. J Histochem Cytochem 57, 763–774 (2009). jhc.2009.953190 [pii] 10.1369/jhc.2009.953190 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 10.Li Y. et al. Ketohexokinase-dependent metabolism of cerebral endogenous fructose in microglia drives diabetes-associated cognitive dysfunction. Exp Mol Med (2023). 10.1038/s12276-023-01112-y [DOI] [Google Scholar]
• 11.Helsley RN et al. Tissue-Specific Fructose Metabolism in Obesity and Diabetes. Curr Diab Rep 20, 64 (2020). 10.1007/s11892-020-01342-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 12.Maenpaa PH, Raivio KO & Kekomaki MP Liver adenine nucleotides: fructose-induced depletion and its effect on protein synthesis. Science 161, 1253–1254 (1968). [DOI] [PubMed] [Google Scholar]
• 13.van den Berghe G, Bronfman M, Vanneste R & Hers HG The mechanism of adenosine triphosphate depletion in the liver after a load of fructose. A kinetic study of liver adenylate deaminase. Biochem J 162, 601–609 (1977). [DOI] [PMC free article] [PubMed] [Google Scholar]
• 14.Bawden SJ et al. Investigating the effects of an oral fructose challenge on hepatic ATP reserves in healthy volunteers: A (31)P MRS study. Clin Nutr 35, 645–649 (2016). 10.1016/j.clnu.2015.04.001 [DOI] [PubMed] [Google Scholar]
• 15.Abdelmalek MF et al. Higher dietary fructose is associated with impaired hepatic adenosine triphosphate homeostasis in obese individuals with type 2 diabetes. Hepatology 56, 952–960 (2012). 10.1002/hep.25741 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 16.Bode JC, Zelder O, Rumpelt HJ & Wittkamp U Depletion of liver adenosine phosphates and metabolic effects of intravenous infusion of fructose or sorbitol in man and in the rat. Eur J Clin Invest 3, 436–441 (1973). [DOI] [PubMed] [Google Scholar]
• 17.Samarah LZ et al. Spatial metabolic gradients in the liver and small intestine. Nature (2025). 10.1038/s41586-025-09616-5 [DOI] [Google Scholar]
• 18.Van den Berghe G. Fructose: metabolism and short-term effects on carbohydrate and purine metabolic pathways. Prog Biochem Pharmacol 21, 1–32 (1986). [Google Scholar]
• 19.Petrie JL et al. The rate of production of uric acid by hepatocytes is a sensitive index of compromised cell ATP homeostasis. Am J Physiol Endocrinol Metab 305, E1255–1265 (2013). 10.1152/ajpendo.00214.2013 [DOI] [PubMed] [Google Scholar]
• 20.Cox CL et al. Consumption of fructose- but not glucose-sweetened beverages for 10 weeks increases circulating concentrations of uric acid, retinol binding protein-4, and gamma-glutamyl transferase activity in overweight/obese humans. Nutr Metab (Lond) 9, 68 (2012). 10.1186/1743-7075-9-68 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 21.Zhang P. et al. Dietary intake of fructose increases purine de novo synthesis: A crucial mechanism for hyperuricemia. Front Nutr 9, 1045805 (2022). 10.3389/fnut.2022.1045805 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 22.Evans PR, Farrants GW & Hudson PJ Phosphofructokinase: structure and control. Philos Trans R Soc Lond B Biol Sci 293, 53–62 (1981). 10.1098/rstb.1981.0059 [DOI] [PubMed] [Google Scholar]
• 23.Sun SZ & Empie MW Fructose metabolism in humans - what isotopic tracer studies tell us. Nutr Metab (Lond) 9, 89 (2012). 10.1186/1743-7075-9-89 [DOI] [PMC free article] [PubMed] [Google Scholar]
• 24.Woods HF, Eggleston LV & Krebs H