Nutrition · 人与自我 / 营养 / 代谢 / 神经科学

同样的 300 卡路里,为何身体反应不同:一项随机对照试验把"营养配比"与"加工本身"分开

Two meals were matched within one per cent on every nutrient — yet a randomized crossover trial found they behaved differently in the body and the brain, hinting that how food is made may matter as much as what it contains
来源 Nature Metabolism(2026 年 10 月 5 日)作者 外刊精读专栏日期 2026-10-08难度 难词数 约 515 词 / 9 段
超市货架上摆满各类包装食品与超加工食品(配图)
配图:超市货架上琳琅满目的包装食品——本文研究比较的"超加工餐"即由这类工业化加工食品构成(图片来源:AI 生成配图,仅供教学使用)。

中文导读

长久以来,营养学的默认前提是:只要 calories、脂肪、碳水、蛋白质相当,两餐在体内就该"表现一致"。但 2026 年 10 月 5 日发表在《自然·代谢》(Nature Metabolism)上的一项随机对照交叉试验,把这个前提本身变成了一根被研究的变量。美国弗吉尼亚理工大学生物医学研究所 DiFeliceantonio 实验室(第一作者 Zach Hutelin)招募了 57 名 18—45 岁的健康成年人,让受试者在不同日子里分别吃下"同样的 300 卡"——一份超加工餐、一份最低限度加工餐,两者在碳水、脂肪、蛋白质、热量、水分、盐与重量上被校准到相差不超过约 1%,连营养标签都几乎一模一样。结果令人意外:代谢的每一项指标都出现了差异;超加工餐带来更强的胰岛素反应、让血糖略高且持续略久、并让身体在消耗更多能量时反而燃烧更少的碳水。更耐人寻味的是,在另一天的脑成像中,受试者看到两类营养相当的食物图片时,大脑奖赏回路(腹侧纹状体、伏隔核)的反应差异,竟与餐后碳水燃烧的差异"同向同步"。本文循"预设前提—实验设计—代谢发现—大脑证据—意义—争议与局限—结论"的链条,追问一个正在重塑营养科学的问题:决定健康的,也许不只是食物里有什么,还有它被怎样制造出来。

一、原文精读

鼠标悬停在带 蓝色虚线下划线 的生词上,即可查看词性和释义。

🔊 英音朗读 · British English
由英音语音合成(en-GB)朗读全文原文,共 9 段。播放时原文会随进度逐句高亮,建议先盲听抓大意,再对照下方原文跟读,留意重音、连读与停顿。
【1】Nutrition science tends to assume that two meals supplying the same calories, fat, carbohydrate and protein will behave much the same way inside the body. A tightly controlled new experiment suggests that this tidy equation leaves out a variable that may matter enormously: how heavily the food itself has been processed.
【2】Published on 5 October 2026 in Nature Metabolism, the study was led by Alexandra DiFeliceantonio's laboratory at Virginia Tech's Fralin Biomedical Research Institute, with doctoral researcher Zach Hutelin as first author. Rather than surveying diets in the wild, the team ran a crossover trial in which 57 healthy adults aged 18 to 45 each ate a minimally processed meal and an ultraprocessed one on separate days. Thirty-two completed both the metabolic sessions and a further round of brain imaging.
【3】The meals were engineered to be as alike as possible, with carbohydrate, fat, protein, calories, water, salt and weight all held equal to within about 1 per cent, so that their labels would look virtually identical. The ultraprocessed version combined a bite of peanut-butter-and-jelly sandwich, deli turkey, veggie chips, a cookie, cereal and instant mashed potatoes; the minimally processed plate offered banana, dried cranberries, cheese and egg.
【4】To capture metabolism, participants arrived after fasting overnight and lay inside a sealed, airtight chamber that measures how much energy the body expends. Blood was drawn immediately after eating and six more times over the next three hours, tracking insulin, blood sugar, energy expenditure and the rate at which the body burned carbohydrate.
【5】The results startled the researchers: every metabolic measure differed. The ultraprocessed meal triggered a far larger insulin response, left blood sugar slightly higher for slightly longer, and pushed the body to burn less carbohydrate while expending more energy overall.
【6】On a separate day the same volunteers viewed photographs of nutritionally matched processed and unprocessed foods inside an MRI scanner, reporting what they would be willing to pay for each. Differences in how the body had burned carbohydrate after a meal tracked differences in how reward regions of the brain, including the ventral striatum and nucleus accumbens, responded to pictures of food.
【7】Crucially, participants did not say they would pay more for either kind of food, implying that the effect is not simply a matter of conscious desire. The work hints instead that processing itself, quite apart from nutrient content, can nudge the body and the brain into different states.
【8】The finding speaks to a long-running argument, since ultraprocessed foods now supply more than half of the average American's daily calories, yet the mechanisms linking them to obesity and disease remain fiercely debated. The authors are candid about the limits of one set of matched 300-calorie meals tested in young, healthy adults, and note that the two meals differed in their protein sources and additives.
【9】For Carlos Monteiro of the University of São Paulo, whose NOVA classification underpins the field, the brain data are what make the study valuable, because the brain ultimately governs eating. The broader lesson is that how food is made may be as worth studying as what it contains.

原文出处:本文根据 Nature Metabolism 论文《Metabolic and neural responses to ultraprocessed foods: a randomized, controlled, crossover study》(Zach Hutelin et al.;Virginia Tech, Fralin Biomedical Research Institute;Nature Metabolism, published online 5 October 2026, DOI: 10.1038/s42255-026-01619-4)改写整理,供学习使用。

三、长难句分析(三步拆解法)

Rather than surveying diets in the wild, the team ran a crossover trial in which 57 healthy adults aged 18 to 45 each ate a minimally processed meal and an ultraprocessed one on separate days.
主干:the team ran a crossover trial(主语 + 谓语 + 宾语)
rather than 状语:Rather than surveying diets in the wild("与其去调查现实中的饮食,不如……",置于句首表取舍)
介词 + 定语从句:in which 57 healthy adults ... ate a minimally processed meal and an ultraprocessed one(修饰 a crossover trial;in which = where)
后置定语与替代:aged 18 to 45(修饰 adults);an ultraprocessed one 中 one 替代 meal
考点:rather than 表取舍;"介词 + which"定语从句;one 作替代词避免重复;each 作同位语强调"每一个人"
译文:团队没有去调查现实生活中的真实饮食,而是开展了一项交叉试验:57 名 18 至 45 岁的健康成年人在不同的日子分别吃了一顿最低限度加工餐和一顿超加工餐。
The meals were engineered to be as alike as possible, with carbohydrate, fat, protein, calories, water, salt and weight all held equal to within about 1 per cent, so that their labels would look virtually identical.
主干:The meals were engineered to be as alike as possible(被动语态 + as ... as possible)
with 复合结构:with carbohydrate, fat, protein, calories, water, salt and weight all held equal to within about 1 per cent("with + 宾语 + 过去分词",表伴随;held 为 hold 的过去分词,"使保持")
结果状语从句:so that their labels would look virtually identical("以至于标签看起来几乎一模一样")
考点:as ... as possible 结构;with 复合结构作状语;so that 引导结果状语从句;hold 作"使保持"的熟词僻义
译文:这两餐被设计得尽可能相似——碳水、脂肪、蛋白质、热量、水分、盐与重量都被控制在约 1% 的差异之内,以至于它们的营养标签看起来几乎毫无二致。
Differences in how the body had burned carbohydrate after a meal tracked differences in how reward regions of the brain, including the ventral striatum and nucleus accumbens, responded to pictures of food.
主干:Differences ... tracked differences(主语 + 谓语 + 宾语,track 此处作"与……对应/同向变化")
介词 + 宾语从句(一):in how the body had burned carbohydrate after a meal(修饰第一个 Differences)
介词 + 宾语从句(二):in how reward regions of the brain ... responded to pictures of food(修饰第二个 differences)
插入成分:including the ventral striatum and nucleus accumbens(对 reward regions 补充举例)
考点:how 引导的名词性从句作介词宾语;过去完成时 had burned 表"餐后先行发生";插入语后置举例
译文:餐后身体燃烧碳水化合物的差异,与大脑奖赏相关脑区(包括腹侧纹状体和伏隔核)对食物图片的反应差异,呈现出同向的变化。
The finding speaks to a long-running argument, since ultraprocessed foods now supply more than half of the average American's daily calories, yet the mechanisms linking them to obesity and disease remain fiercely debated.
主干:The finding speaks to a long-running argument(主语 + 谓语 + 宾语)
原因状语从句:since ultraprocessed foods now supply more than half of the average American's daily calories("既然、因为")
yet 并列转折:yet the mechanisms ... remain fiercely debated(与前半句构成对比转折)
现在分词作定语:linking them to obesity and disease(修饰 the mechanisms)
考点:since 引导原因状语从句;yet 作并列连词表转折;现在分词短语作后置定语;speak to 意为"涉及、印证"
译文:这一发现触及一场旷日持久的争论:既然超加工食品如今提供了普通美国人每日热量的一半以上,那么把它们与肥胖和疾病联系起来的机制,却仍处在激烈的争论之中。
For Carlos Monteiro of the University of São Paulo, whose NOVA classification underpins the field, the brain data are what make the study valuable, because the brain ultimately governs eating.
主干:the brain data are what make the study valuable(主语 + 系动词 + 表语从句)
介词短语作状语:For Carlos Monteiro of the University of São Paulo("在……看来",引出评价者)
非限制性定语从句:whose NOVA classification underpins the field(修饰 Monteiro)
表语从句:what make the study valuable(what 引导,且在从句中作主语)
原因状语从句:because the brain ultimately governs eating
考点:whose 引导定语从句;what 引导表语从句并在从句中充当主语;For sb 表"在某人看来"
译文:在圣保罗大学的卡洛斯·蒙泰罗(其 NOVA 食品分类体系为该领域奠定基础)看来,真正让这项研究有价值的是那些大脑数据,因为归根结底,是大脑在调控进食。

四、全文大意

营养学长期默认一条简单的假设:只要两餐提供相同的热量、脂肪、碳水化合物和蛋白质,它们在体内的作用就应当大体相同。然而一项控制极严的新实验提示,这个整洁的等式漏掉了一个或许关系重大的变量——食物本身被加工的程度。2026 年 10 月 5 日发表在《自然·代谢》上的这项研究,由弗吉尼亚理工大学生物医学研究所的 DiFeliceantonio 实验室领衔,博士研究员 Zach Hutelin 为第一作者。团队没有像通常那样去调查现实生活中的饮食,而是开展了一项严格控制的交叉试验:57 名 18 至 45 岁的健康成年人,在不同的日子里分别吃下一顿"最低限度加工餐"和一顿"超加工餐";其中 32 人完成了全部的代谢测试与另一轮脑部成像。两餐被设计得尽可能相似——碳水化合物、脂肪、蛋白质、热量、水分、盐与总重量都被控制在约 1% 的差异之内,营养标签几乎一模一样。超加工那顿由一小份花生酱果酱三明治、熟食火鸡、蔬菜脆片、一块饼干、麦片和速食土豆泥组成,而最低限度加工那顿则提供了香蕉片、蔓越莓干、奶酪和鸡蛋。为了测量代谢,受试者禁食一整夜后进入一个密闭不透气的代谢舱,以测量身体消耗了多少能量;进食后立即抽血,并在随后三小时内再抽六次,以追踪胰岛素、血糖、能量消耗以及身体燃烧碳水化合物的速率。结果令研究者大感意外:几乎每一项代谢指标都不相同。超加工餐引发了强得多的胰岛素反应,让血糖略高且持续略久,并促使身体在总体上消耗更多能量的同时,反而燃烧更少的碳水化合物。更耐人寻味的是大脑这一侧:在另一天,同一批志愿者躺在核磁共振扫描仪中观看营养配比相当的加工与未加工食物图片,并报告自己愿意为每种食物付多少钱;结果显示,餐后身体燃烧碳水化合物的差异,与大脑奖赏相关脑区(包括腹侧纹状体和伏隔核)对食物图片的反应差异,呈现出同向的对应关系。关键在于,受试者并未表示愿意为任何一类食物多付钱,这说明效应并不只是主观欲望的问题;研究反而提示,撇开营养成分不谈,加工过程本身就能把身体和大脑悄然推入不同的状态。这一发现触及一场旷日持久的争论:超加工食品如今提供了普通美国人每日热量的一半以上,但把它们与肥胖和疾病联系起来的机制,仍处在激烈争论之中。作者对研究的局限直言不讳:这只是一组被匹配到 300 卡路里的餐食,受试者是年轻健康人群,而且两餐在蛋白质来源和添加剂上也并不完全相同。在圣保罗大学的蒙泰罗看来(其 NOVA 分类体系为该领域奠定基础),真正让这项研究有价值的是那些大脑数据,因为归根结底是大脑在调控进食。更广义的启示或许是:食物"如何被制作",可能和它"含有什么"一样值得研究。

五、思考与仿写

思考题:结合第 5、6 段,作者为什么强调"代谢指标的差异"与"大脑奖赏回路的差异"之间的对应关系?这一对应关系若成立,会如何改变人们对"超加工食品为何让人吃得更多"的解释?
仿写题:仿照第 3 段 "The meals were engineered to be as alike as possible, with ... all held equal to within about 1 per cent, so that ..."(as ... as possible + with 复合结构 + so that 结果状语从句)的句式,围绕「公平的实验设计」「控制变量」或「标准化流程」写一个句子。

六、文章结构图解

文章段落结构(共 9 段) 1 · 预设:同等营养的两餐"应当"表现一致,或漏掉"加工"变量 2 · 研究:弗吉尼亚理工大学团队做严格控制的交叉试验 3 · 设计:两餐营养素与重量校准到 1% 以内,标签几乎相同 4 · 测量:代谢舱 + 三小时七次抽血,追踪代谢全程 5 · 发现:代谢指标全不同,胰岛素更强、燃碳更少 6 · 大脑:奖赏回路反应与代谢差异同向对应 7 · 关键:受试者并未愿多付钱,效应不限于主观欲望 8 · 争议与局限:机制仍激烈争论,样本与餐食有限 9 · 结论:食物"如何制作"与"含有什么"同样值得研究
全文共 9 段,按「预设前提 → 研究概况 → 实验设计 → 测量方法 → 核心发现 → 大脑证据 → 关键细节 → 争议与局限 → 总结升华」的递进逻辑层层展开。