Food · Misleading · 被妖魔化的分子
Gluten-Free
乳糜泻 (约 1% 人群) 和小麦过敏者必须避开麸质, 这是医学事实 · 但无麸质饮食对无上述疾病者没有显示健康获益 · 无麸质包装食品往往纤维更少、糖和精制淀粉更多 · 吃无麸质感觉更好的真正原因可能是减少了 FODMAP 或超加工小麦零食
Last updated
Story path
- 1The claim · 'gluten-free is healthier'The claim · 'gluten-free is healthier'
- 2Who must avoid · state this clearly, firstWho must avoid · state this clearly, first
- 3Evidence for the general population · no health benefit shownEvidence for the general population · no health benefit shown
- 4Behind 'feeling better' · FODMAPs and ultra-processed wheat snacksBehind 'feeling better' · FODMAPs and ultra-processed wheat snacks
- 5Who should avoid · a precise listWho should avoid · a precise list
- 6Practical conclusions · whole grains are the directionPractical conclusions · whole grains are the direction
Chapter 1
The claim · 'gluten-free is healthier'
The claim · 'gluten-free is healthier'
Gluten-free has evolved from a serious medical dietary therapy to a marketing label across entire supermarket aisles — covering cookies, bread, soy sauce, snacks, and even beer.
The current popular claim roughly holds that gluten has some degree of 'inflammatory' effect on everyone, and avoiding it helps with fat loss, digestion, energy, and skin, as a modern healthy lifestyle.
This conflates four very different situations:
1. Celiac disease — strictly gluten-free is a medical requirement
2. Wheat allergy — avoidance is required, an immune reaction
3. Non-celiac gluten/wheat sensitivity (NCGS) — contested; some cases may be FODMAP reactions
4. The general population without any of the above — 'gluten-free is healthier' has no scientific basis here
This chapter first separates 'must avoid' from 'no need to avoid', then unpacks what's really behind 'feeling better'.
The current popular claim roughly holds that gluten has some degree of 'inflammatory' effect on everyone, and avoiding it helps with fat loss, digestion, energy, and skin, as a modern healthy lifestyle.
This conflates four very different situations:
1. Celiac disease — strictly gluten-free is a medical requirement
2. Wheat allergy — avoidance is required, an immune reaction
3. Non-celiac gluten/wheat sensitivity (NCGS) — contested; some cases may be FODMAP reactions
4. The general population without any of the above — 'gluten-free is healthier' has no scientific basis here
This chapter first separates 'must avoid' from 'no need to avoid', then unpacks what's really behind 'feeling better'.
Chapter 2
Who must avoid · state this clearly, first
Who must avoid · state this clearly, first
This scene needs to be clear and stated first: the following people must strictly avoid gluten or wheat — this is medical fact, not a lifestyle choice.
Celiac Disease — approximately 1% of the population
Celiac disease is an autoimmune condition. Gliadin (a component of gluten) is misidentified by the immune system, triggering a T-cell-mediated attack on the small-intestinal villi. Continued exposure → villous atrophy → malabsorption (iron, calcium, folate, and B12 all affected) → untreated in the long term, it raises risk of osteoporosis, infertility, neuropathy, and intestinal lymphoma. Diagnosis requires serological antibody testing (tTG-IgA) plus small-bowel biopsy. Treatment: lifelong strict gluten-free diet is currently the only effective management (Lebwohl 2018, NEJM).
Wheat Allergy
An IgE-mediated immune reaction that can produce hives, breathing difficulty, or anaphylaxis within minutes to hours of ingestion. Strict avoidance of wheat-containing foods, diagnosed and managed by an allergy specialist.
Non-Celiac Gluten/Wheat Sensitivity (NCGS)
Genuinely contested. Some people reporting digestive discomfort may be reacting to FODMAPs in wheat (especially fructans), not gluten itself (Skodje 2018, Gastroenterology). Diagnosis requires ruling out celiac disease and wheat allergy first, followed by a systematic elimination and reintroduction protocol under medical guidance.
If you have any of the above symptoms, see a doctor for formal diagnosis rather than starting a gluten-free diet on your own — this matters because self-starting a gluten-free diet interferes with the accuracy of celiac diagnosis.
Celiac Disease — approximately 1% of the population
Celiac disease is an autoimmune condition. Gliadin (a component of gluten) is misidentified by the immune system, triggering a T-cell-mediated attack on the small-intestinal villi. Continued exposure → villous atrophy → malabsorption (iron, calcium, folate, and B12 all affected) → untreated in the long term, it raises risk of osteoporosis, infertility, neuropathy, and intestinal lymphoma. Diagnosis requires serological antibody testing (tTG-IgA) plus small-bowel biopsy. Treatment: lifelong strict gluten-free diet is currently the only effective management (Lebwohl 2018, NEJM).
Wheat Allergy
An IgE-mediated immune reaction that can produce hives, breathing difficulty, or anaphylaxis within minutes to hours of ingestion. Strict avoidance of wheat-containing foods, diagnosed and managed by an allergy specialist.
Non-Celiac Gluten/Wheat Sensitivity (NCGS)
Genuinely contested. Some people reporting digestive discomfort may be reacting to FODMAPs in wheat (especially fructans), not gluten itself (Skodje 2018, Gastroenterology). Diagnosis requires ruling out celiac disease and wheat allergy first, followed by a systematic elimination and reintroduction protocol under medical guidance.
If you have any of the above symptoms, see a doctor for formal diagnosis rather than starting a gluten-free diet on your own — this matters because self-starting a gluten-free diet interferes with the accuracy of celiac diagnosis.
临床 · 乳糜泻不治会怎样
乳糜泻的麻烦不只在肠道。绒毛被磨平之后, 吸收面积和挂在绒毛上的消化酶一起减少, 短板就从吃了什么变成吸得进什么:骨头: 钙吸收不良, 骨密度长期跟不上, 骨质疏松风险升高血: 铁和叶酸吸收不良, 贫血反复出现, 口服补铁也常常补不上去神经: B12 长期不足, 出现手脚发麻、平衡变差这类神经病变生育: 长期营养不良与不孕相关肠道本身: 长年不断的免疫刺激, 与肠道淋巴瘤风险升高有关
诊断要靠抽血加活检来确诊。顺序很关键: 抽血查的是免疫系统对麸质起反应时留下的抗体, 活检看的是绒毛有没有被磨平。你一旦提前自行停掉麸质, 免疫反应会平息、绒毛会开始修复, 两项检查都可能转成阴性——于是你既拿不到诊断, 又要终身怀疑自己到底该不该吃。这就是先确诊、再改饮食这个顺序背后的机制原因。
临床 · 小麦过敏与 NCGS
小麦过敏 (Wheat Allergy)IgE 介导的免疫反应, 可在摄入后数分钟至数小时出现荨麻疹、呼吸困难甚至过敏性休克。严格回避含小麦食物, 由过敏科医生诊断和管理。
它和乳糜泻的区别, 在于哪一套免疫在动手。 过敏走的是快通道: 一类叫 IgE 的抗体事先就挂在肥大细胞表面, 小麦蛋白一碰上去, 细胞当场把组胺等物质倒出来, 血管渗漏、气道收窄——所以是分钟级的、全身性的。乳糜泻走的是慢通道: T 细胞被召集到小肠壁, 损伤一点点累积在绒毛上, 按周和月计。同样叫对小麦起反应, 一个可能当场危及生命, 一个可能很久都没被发现。
非乳糜泻麸质/小麦敏感 (NCGS)
真实存在争议。部分报告消化不适的人可能是对小麦中的 FODMAP (尤其是果聚糖 fructans) 反应, 而非麸质本身 (Skodje 2018, Gastroenterology)。诊断需排除乳糜泻和小麦过敏后, 在医生指导下进行系统的食物排除和重新引入。
Chapter 3
Evidence for the general population · no health benefit shown
Evidence for the general population · no health benefit shown
For people without celiac disease, wheat allergy, or diagnosed NCGS, does a gluten-free diet provide health benefits? The evidence is clear: no.
Observational evidence: Lebwohl et al. in the BMJ (2017) analyzed three large US cohorts (over 110,000 people followed for over 26 years): a gluten-free diet was not associated with lower coronary-heart-disease risk; in fact, long-term low-gluten intake correlated with lower whole-grain intake, and lower whole-grain intake associates with higher cardiovascular risk. This is associational, but it points in the opposite direction from 'gluten-free is healthier'.
Absence of intervention evidence: There are no rigorously designed RCTs demonstrating any reproducibly measurable health benefit of a gluten-free diet in people without the above conditions.
The nutritional problem with gluten-free packaged foods: This point is often overlooked. Gluten-free bread, cookies, and pasta must compensate for the texture provided by wheat gluten, typically by adding more sugar, fat, and refined starch (tapioca, rice flour, cornstarch). The result: less fiber, higher glycemic index, lower nutrient density than conventional equivalents — and often more expensive. 'Buying gluten-free products' does not equal 'eating healthier'.
Evidence grade: health-benefit claims for a gluten-free diet in the general population without celiac disease are marketing claims with no supporting evidence (by contrast, the need for gluten-free in celiac disease is a Grade A medical fact).
Observational evidence: Lebwohl et al. in the BMJ (2017) analyzed three large US cohorts (over 110,000 people followed for over 26 years): a gluten-free diet was not associated with lower coronary-heart-disease risk; in fact, long-term low-gluten intake correlated with lower whole-grain intake, and lower whole-grain intake associates with higher cardiovascular risk. This is associational, but it points in the opposite direction from 'gluten-free is healthier'.
Absence of intervention evidence: There are no rigorously designed RCTs demonstrating any reproducibly measurable health benefit of a gluten-free diet in people without the above conditions.
The nutritional problem with gluten-free packaged foods: This point is often overlooked. Gluten-free bread, cookies, and pasta must compensate for the texture provided by wheat gluten, typically by adding more sugar, fat, and refined starch (tapioca, rice flour, cornstarch). The result: less fiber, higher glycemic index, lower nutrient density than conventional equivalents — and often more expensive. 'Buying gluten-free products' does not equal 'eating healthier'.
Evidence grade: health-benefit claims for a gluten-free diet in the general population without celiac disease are marketing claims with no supporting evidence (by contrast, the need for gluten-free in celiac disease is a Grade A medical fact).
误区 · 无麸质包装食品为什么更差
无麸质包装食品的营养问题: 这个点常被忽视。无麸质面包、饼干、面条为了弥补小麦面筋结构带来的口感, 通常需要添加更多糖、脂肪、精制淀粉 (木薯粉、大米粉、玉米淀粉)。其结果是: 比普通版本纤维更少、血糖指数更高、营养密度更低 — 在某些产品中还更贵。买了无麸质产品≠吃得更健康。为什么非得这么补? 面筋是小麦粉遇水揉搓后形成的一张有弹性的蛋白网。发酵产生的气体被这张网兜住, 面包才发得起来、咬下去才有嚼劲。把这张网抽掉, 面团就散成一摊淀粉糊, 烤出来又硬又碎。厂家要把口感做回来, 只能找别的东西替代这张网的功能: 更多的油和水让它显得湿润, 更多的糖和精制淀粉让它松软。
这就是无麸质这个标签在信息上的不对称: 它只承诺拿走什么, 从不说要拿什么补回来。 判断一件包装食品好不好, 该看的仍然是配料表的前几项——精制淀粉和糖排在前面, 它就不会因为少了麸质而变得更健康。
证据级别: 对无乳糜泻一般人群, 无麸质饮食的健康获益主张属营销主张、无支持证据 (与之相对, 乳糜泻需要无麸质是 Grade A 的医学事实)。
Chapter 4
Behind 'feeling better' · FODMAPs and ultra-processed wheat snacks
Behind 'feeling better' · FODMAPs and ultra-processed wheat snacks
'I feel much better since going gluten-free' is a genuine report — the question is whether improvement was actually caused by gluten.
Possible alternative explanations:
1. FODMAP effect: Wheat contains fructans, the F in FODMAP (fermentable oligosaccharides). Fructans are a common irritable-bowel-syndrome (IBS) trigger. Skodje 2018 (Gastroenterology, double-blind crossover RCT): in self-reported gluten-sensitive individuals, fructans provoked more IBS symptoms than gluten — suggesting the reactive target may be a FODMAP, not gluten. Low-FODMAP and gluten-free diets heavily overlap for this group, so avoiding 'gluten-containing foods' may effectively be running a low-FODMAP protocol. Dive to carbs-fiber for FODMAP and gut-fermentation mechanisms.
2. Reduction of ultra-processed wheat snacks: People who actively 'go gluten-free' usually simultaneously reduce cookies, pastries, instant noodles, and ultra-processed wheat bread. Per Hall 2019 evidence, reducing these foods on its own improves appetite regulation and energy intake. The perceived improvement may come from 'less ultra-processing', not 'less gluten'.
3. Placebo effect and attentional bias: People actively changing their diet pay closer attention to their body; the expectation of improvement itself influences symptom perception.
This is not saying the experience is invalid — it's that 'felt better' cannot automatically lead to 'gluten was the problem'.
Possible alternative explanations:
1. FODMAP effect: Wheat contains fructans, the F in FODMAP (fermentable oligosaccharides). Fructans are a common irritable-bowel-syndrome (IBS) trigger. Skodje 2018 (Gastroenterology, double-blind crossover RCT): in self-reported gluten-sensitive individuals, fructans provoked more IBS symptoms than gluten — suggesting the reactive target may be a FODMAP, not gluten. Low-FODMAP and gluten-free diets heavily overlap for this group, so avoiding 'gluten-containing foods' may effectively be running a low-FODMAP protocol. Dive to carbs-fiber for FODMAP and gut-fermentation mechanisms.
2. Reduction of ultra-processed wheat snacks: People who actively 'go gluten-free' usually simultaneously reduce cookies, pastries, instant noodles, and ultra-processed wheat bread. Per Hall 2019 evidence, reducing these foods on its own improves appetite regulation and energy intake. The perceived improvement may come from 'less ultra-processing', not 'less gluten'.
3. Placebo effect and attentional bias: People actively changing their diet pay closer attention to their body; the expectation of improvement itself influences symptom perception.
This is not saying the experience is invalid — it's that 'felt better' cannot automatically lead to 'gluten was the problem'.
机制 · 麸质在正常肠道里的四步
把上一屏那句正面回答展开成一条完整的链。第一步 · 剪。 你吃下的任何蛋白, 到了胃和小肠都要先被蛋白酶剪成小段才能吸收。剪刀认的是氨基酸与氨基酸之间的接口, 大多数接口都好剪。
第二步 · 有一段剪不动。 麸质里的醇溶蛋白 (gliadin) 有个结构特点: 一种叫脯氨酸的氨基酸在它身上特别密集。脯氨酸的形状是一个环, 卡在链上会把那一段拧出一个硬弯, 人的蛋白酶伸不进去。于是这一段被剩下来, 以肽 (几个到几十个氨基酸连成的碎片) 的形式留在肠腔里。这一步对所有人都一样, 有没有乳糜泻都一样。
第三步 · 分岔就在这里。 碎片要变成一场攻击, 免疫系统得先认出它。这需要肠壁里的免疫细胞把碎片举起来, 展示给 T 细胞看; 而举不举得起来, 取决于一种由基因决定的展示分子。绝大多数人身上的版本抓不住这些碎片, 举不起来, T 细胞也就什么都没看见。乳糜泻患者身上是能抓住的那个版本——而且这只是必要条件, 不是充分条件: 带着这个版本的人, 远多于真正发病的人。
第四步 · 没被认出的碎片去哪了。 继续往下走: 一部分被肠道菌利用, 其余随粪便排出。没有 T 细胞被召集, 绒毛就没有被攻击, 也就没有绒毛萎缩、没有随之而来的吸收障碍。
所以那句营销话术究竟错在哪? 麸质对所有人都有炎症把第二步 (人人都有碎片剩下) 直接当成了结论, 中间那个真正决定命运的第三步被跳过了。麸质不是靠难消化致病的——难消化是所有人共同的起点, 致病要的是那把只有一部分人带着的钥匙。这一层说清了, 那句宣称才是被正面拆穿, 而不是被绕开。
证据 · 你戒掉的可能是果聚糖
1. FODMAP 效应: 小麦含有果聚糖 (fructans), 是 FODMAP 分类里的 F (fermentable oligosaccharides)。果聚糖对肠易激综合征 (IBS) 患者是常见触发物。Skodje 2018 (Gastroenterology, 双盲交叉 RCT): 在自报对麸质敏感者中, 果聚糖比麸质引发了更多 IBS 症状 — 说明敏感者反应的目标可能是 FODMAP, 而非麸质。低 FODMAP 饮食和无麸质饮食在这批人里有大量重叠, 因此戒掉含麸质食物可能实际上是在执行低 FODMAP 方案。dive 到 carbs-fiber 了解 FODMAP 和肠道发酵的机制。果聚糖为什么会让人胀? 它是果糖一颗颗串成的短链, 而人的小肠没有能剪断这种串的酶。于是它整条走到大肠, 被那里的细菌当饭吃。细菌发酵会产气, 气把肠壁撑开; 这些短链同时还有渗透作用, 把水往肠腔里拉。胀、咕噜、大便变稀 —— 这些恰恰是很多人戒掉小麦后最先消失的感觉, 而它们和麸质那个蛋白毫无关系。
注意这条链和上一页那条的卡点完全不同: 一个卡在蛋白剪不断, 一个卡在糖链剪不断, 而它们恰好装在同一粒小麦里。这就是为什么戒小麦有效这件事, 完全不能反推出问题在麸质。
证据 · 另外两个解释, 以及怎么自己分辨
2. 超加工小麦零食的减少: 主动无麸质化的人通常同时减少了饼干、点心、即食面条、超加工面包等超加工小麦产品。按 Hall 2019 的证据, 这类食物的减少本身就能改善食欲调节和能量摄入。感受到的改善可能来自减超加工, 而不是去麸质。3. 安慰剂效应和注意力偏差: 主动改变饮食的人会对自己的身体更留意; 预期会好本身影响对症状的感知。
这不是说感受无效 — 而是感受好了不能自动推出麸质是问题。
这三条各自留下的痕迹不一样, 可以拿来对照:
症状主要是胀气、肠鸣、排便改变, 而且吃洋葱、大蒜、豆类时也照样出现 → 更像果聚糖那条链, 因为这些食物不含麸质却同样富含果聚糖。改善的是精力、体重、皮肤这类全身感受, 而你同时也把零食柜清空了 → 更像超加工那条链。只在你知道自己吃了麸质时才不舒服, 不知情时不出现 → 这正是双盲设计要排除的东西。
⚠️ 但这三条只是帮你想清楚, 不是自我诊断的替代品。真要分清, 顺序只有一个: 先由医生排除乳糜泻 (必须在开始无麸质饮食之前), 再在医生或营养师指导下做系统的排除与重新引入。
Chapter 5
Who should avoid · a precise list
Who should avoid · a precise list
Bringing the earlier scenes together into a precise layered recommendation.
Must strictly avoid gluten (medical necessity):
Celiac disease patients (confirmed by serology + biopsy): lifelong strict gluten-free, including avoiding cross-contaminationWheat allergy patients (IgE-mediated, confirmed by an allergy specialist): strict avoidance of all wheat, monitor labels
Should be systematically evaluated under medical guidance (do not self-navigate):
Suspected NCGS: first have celiac disease ruled out by a gastroenterologist (must happen before starting a gluten-free diet), then systematically run low-FODMAP or gluten-free elimination with physician or dietitian follow-up
No need for gluten-free (without the above diagnoses):
Generally healthy adults: no evidence supports benefit of a gluten-free diet; gluten-free packaged foods are usually nutritionally inferiorPeople trying to lose weight: 'gluten-free' is not a fat-loss tool; reducing ultra-processed food isPeople with low energy or digestive complaints: more likely causes are sleep, insufficient fiber, ultra-processed diet — address those first
If you have persistent digestive symptoms, seek formal evaluation from a gastroenterologist or allergy specialist — do not substitute self-diagnosis. This site's information does not replace medical advice.
Must strictly avoid gluten (medical necessity):
Celiac disease patients (confirmed by serology + biopsy): lifelong strict gluten-free, including avoiding cross-contaminationWheat allergy patients (IgE-mediated, confirmed by an allergy specialist): strict avoidance of all wheat, monitor labels
Should be systematically evaluated under medical guidance (do not self-navigate):
Suspected NCGS: first have celiac disease ruled out by a gastroenterologist (must happen before starting a gluten-free diet), then systematically run low-FODMAP or gluten-free elimination with physician or dietitian follow-up
No need for gluten-free (without the above diagnoses):
Generally healthy adults: no evidence supports benefit of a gluten-free diet; gluten-free packaged foods are usually nutritionally inferiorPeople trying to lose weight: 'gluten-free' is not a fat-loss tool; reducing ultra-processed food isPeople with low energy or digestive complaints: more likely causes are sleep, insufficient fiber, ultra-processed diet — address those first
If you have persistent digestive symptoms, seek formal evaluation from a gastroenterologist or allergy specialist — do not substitute self-diagnosis. This site's information does not replace medical advice.
Chapter 6
Practical conclusions · whole grains are the direction
Practical conclusions · whole grains are the direction
Pulling this chapter into a take-away judgment framework:
'Gluten-free = healthier eating': does not hold. For people without celiac disease, no evidence supports it; gluten-free packaged foods are usually nutritionally inferior.
'Felt better, so gluten was the problem': a logical leap. More likely mechanisms are reduced FODMAPs or reduced ultra-processed wheat snacks.
'Gluten-free labeling means it's safer': not necessarily. Read the ingredient list — refined starch and sugar are not healthier than wheat flour.
What genuinely helps most people:
Replace refined wheat (white bread, white noodles, cookies) with whole grains (whole-wheat bread, oats, brown rice): this raises fiber, B-vitamin, and mineral intake, with extensive cohort evidence supporting cardiovascular and metabolic healthReduce ultra-processed wheat products (packaged cookies, puffed snacks, instant noodles) — this is the dietary adjustment worth makingGluten in whole grains is harmless for most people; gluten is just one protein fraction in whole grain
Related reading: fiber and gut health → dive to carbs-fiber · whole grains and bread nutrition overview → bread · ultra-processed food RCT evidence → ultra-processed-foods.
'Gluten-free = healthier eating': does not hold. For people without celiac disease, no evidence supports it; gluten-free packaged foods are usually nutritionally inferior.
'Felt better, so gluten was the problem': a logical leap. More likely mechanisms are reduced FODMAPs or reduced ultra-processed wheat snacks.
'Gluten-free labeling means it's safer': not necessarily. Read the ingredient list — refined starch and sugar are not healthier than wheat flour.
What genuinely helps most people:
Replace refined wheat (white bread, white noodles, cookies) with whole grains (whole-wheat bread, oats, brown rice): this raises fiber, B-vitamin, and mineral intake, with extensive cohort evidence supporting cardiovascular and metabolic healthReduce ultra-processed wheat products (packaged cookies, puffed snacks, instant noodles) — this is the dietary adjustment worth makingGluten in whole grains is harmless for most people; gluten is just one protein fraction in whole grain
Related reading: fiber and gut health → dive to carbs-fiber · whole grains and bread nutrition overview → bread · ultra-processed food RCT evidence → ultra-processed-foods.
References · 5
- Lebwohl, B., Sanders, D. S., & Green, P. H. R. (2018). Coeliac disease. The Lancet, 391(10115), 70-81. A strict gluten-free diet is the established treatment for celiac disease; diagnosis requires serology and biopsy. 10.1016/S0140-6736(17)31796-8
- Skodje, G. I., Sarna, V. K., Minelle, I. H., et al. (2018). Fructan, rather than gluten, induces symptoms in patients with self-reported non-celiac gluten sensitivity. Gastroenterology, 154(3), 529-539. A double-blind crossover trial implicating FODMAP fructans over gluten in many self-reported cases. 10.1053/j.gastro.2017.10.040
- Lebwohl, B., Cao, Y., Zong, G., et al. (2017). Long term gluten consumption in adults without celiac disease and risk of coronary heart disease: prospective cohort study. BMJ, 357, j1892. Restricting gluten in people without celiac disease was not associated with lower cardiovascular risk and may reduce whole-grain intake. 10.1136/bmj.j1892
- Halmos, E. P., Power, V. A., Shepherd, S. J., Gibson, P. R., & Muir, J. G. (2014). A diet low in FODMAPs reduces symptoms of irritable bowel syndrome. Gastroenterology, 146(1), 67-75. Randomized controlled cross-over trial; a low-FODMAP diet significantly reduced overall IBS symptoms versus a typical Australian diet. Onion and garlic are among the highest-fructan (FODMAP) foods. 10.1053/j.gastro.2013.09.046
- Hall, K. D., Ayuketah, A., Brychta, R., Cai, H., Cassimatis, T., Chen, K. Y., et al. (2019). Ultra-processed diets cause excess calorie intake and weight gain: an inpatient randomized controlled trial of ad libitum food intake. Cell Metabolism, 30(1), 67-77.e3. 10.1016/j.cmet.2019.05.008