大麻对健康的影响与潜在风险 | Dr. Matthew Hill

大麻对健康的影响及潜在风险

摘要

卡尔加里大学 Hotchkiss 脑科学研究所教授 Matthew Hill 博士与 Andrew Huberman 深入探讨大麻的生物学机制。对话涵盖 THC 如何与大脑内源性大麻素系统相互作用、精神活性效应的机制、食欲刺激、记忆，以及大麻使用与心理健康之间的关系。本期节目的起因是 Hill 博士公开批评了 Huberman Lab 此前一期关于大麻的个人播客中的某些说法。

---

核心要点

• THC 并非”大锤” — 它实际上是 CB1 受体的部分激动剂，亲和力高但效能相对较低，与花生四烯酸乙醇胺（anandamide）类似，并非机体自身内源性大麻素的超强版本
• 大麻”地毯式轰炸”整个内源性大麻素系统 — 与在特定突触和特定时间精准发挥作用的内源性大麻素不同，THC 同时淹没所有 CB1 受体，这正是产生中毒效应的原因
• 抑制花生四烯酸乙醇胺的分解并不会产生快感 — 通过抑制其代谢酶（FAAH 抑制剂）来提升花生四烯酸乙醇胺水平的药物不显示任何精神活性，这表明 THC 的效应来源于其对全脑受体无差别的激活方式
• “嘴馋”涉及对饱腹感信号的压制 — 大麻在动物完全饱食的状态下仍能重新激活食欲，其机制是维持食物的奖励价值，并阻断瘦素的抑食效应
• 多巴胺神经元是唯一的例外 — CB1 受体几乎遍布大脑所有区域，唯独多巴胺神经元上没有；大麻通过解除对抑制性神经元的抑制，间接影响多巴胺
• 食用型大麻与吸入型大麻有本质区别 — 两种给药途径产生截然不同的药代动力学特征、起效时间和主观体验
• 规律使用者能有效自我调节剂量 — 即便是现代高效能大麻（THC 含量 20–30%），有经验的使用者与使用低效能产品的人相比，血液中 THC 水平相近（约 100 ng/mL）
• 中毒状态下的短期记忆损害已有充分证据，但目前的证据尚不能清楚地支持规律使用者存在长期认知损害

---

详细笔记

什么是大麻？

• 大麻是一种植物，在各种文化中拥有悠久的药用、精神和娱乐使用历史
• 主要精神活性成分是德尔塔-9-四氢大麻酚（THC） — THC 浓度越高，中毒效应越强
• 大麻二酚（CBD） 在结构上与 THC 相似，但无致醉性；可能通过影响焦虑或情绪而具有一定的精神活性
• 植物中存在 70 余种次要大麻素（如大麻酚、大麻萜酚），其生物学特性大多尚不明确
• 萜烯类化合物（如柠檬烯、蒎烯、β-石竹烯、月桂烯）是挥发性芳香化合物，并非大麻所独有；部分已知具有生物活性
• “协同效应”（entourage effect）是指 THC、次要大麻素和萜烯的组合所产生的效应与单独使用 THC 不同——这一说法仍在积极研究中

内源性大麻素系统

• 大脑中对大麻最敏感的主要受体是 CB1（1 型大麻素受体） — 是整个大脑中表达最广泛的受体之一
• CB2 受体主要存在于免疫细胞（小胶质细胞）上，调节炎症；在神经元中表达极少
• 这两种受体的进化并非为了应对大麻，而是为了服务机体自身的内源性大麻素系统

逆行信号传导：

• 与大多数神经递质（从神经元 A 向神经元 B 正向传递）不同，内源性大麻素在突触后神经元（B）中合成，并逆向传递，以调节突触前神经元（A）的神经递质释放
• 主要功能：稳态维持 — 防止过度兴奋（类似癫痫发作的状态）或过度抑制

两种主要内源性大麻素：

• 花生四烯酸乙醇胺（Anandamide） — 名称源自梵语 ananda（极乐）；对 CB1 亲和力高、效能低；被认为具有张力性（稳态调节作用，如同恒温器）
• 2-花生四烯酰甘油（2-AG） — 亲和力较低，效能较高；被认为具有时相性（按需启动，尤其在神经活动旺盛或可塑性事件期间）

THC 与内源性大麻素的区别

• THC 是 CB1 的部分激动剂 — 亲和力和效能与花生四烯酸乙醇胺相近，并非显著更强
• 关键区别在于：内源性大麻素在特定时间作用于特定突触；而 THC 经血液循环分布全身，无差别地同时激活所有脑区网络中的 CB1 受体
• 这种全面激活——尤其是遍及皮层回路——扰乱了正常的信息处理，产生中毒状态
• FAAH 抑制剂的研究印证了这一点：大幅提升花生四烯酸乙醇胺水平不会产生精神活性，证明 THC 的致醉效应并非简单源于受体激活程度，而在于其空间和时间上的无差别性

大麻、多巴胺与欣快感

• 多巴胺神经元上明显缺乏 CB1 受体 — 是大脑中唯一一类缺乏该受体的主要神经元类群
• 大麻通过间接方式影响多巴胺：THC 激活**腹侧被盖区（VTA）**中围绕多巴胺神经元的抑制性神经元上的 CB1 受体，解除对多巴胺神经元的抑制，使其爆发式放电
• 这一机制与μ-阿片受体响应阿片类药物的方式类似
• 这种多巴胺释放是否直接导致大麻的欣快感，目前尚未得到清晰的证明

大麻与食欲（“嘴馋”效应）

• 大麻通过多种汇聚机制刺激食欲：
  1. 去抑制下丘脑中的 AgRP 神经元（下丘脑摄食回路富含 CB1 受体）
  2. 奖励回路激活 — 伏隔核中的花生四烯酸乙醇胺增强对美味食物的摄入
  3. 增强味觉加工 — CB1 激活选择性放大味觉皮层中的甜味反应（对咸味、苦味或酸味无此效应）
• THC 可能欺骗大脑使其进入禁食状态 — 内源性大麻素在禁食期间自然升高以促进觅食行为；THC 激活这些相同的回路，而不管实际能量状态如何
• 饱腹感压制：预先饱食（或经条件训练因恶心而回避食物）的动物，在给予大麻后会重新开始进食，甚至为获取食物而进行大量操作行为（如压杆）
• 瘦素信号阻断：THC 可压制瘦素信号——正常情况下，瘦素会抑制内源性大麻素活性以减少食物摄入；大麻素水平升高则会逆转这一效应

大麻与记忆

• 急性中毒期间短期记忆损害有据可查 — 尤其影响记忆提取和巩固；动物及人体研究均有充分支持
• 规律使用者的长期认知损害尚未得到明确证实 — 证据缺乏一致性和可靠性
• 状态依赖性学习是一个混淆因素：长期使用者曾多次在中毒状态下完成认知任务，测试时表现出的损害较少，这并非说明大麻增强了认知，而是适应或状态依赖性习惯化的结果
• 规律使用者通常比偶尔使用者表现出更少的急性记忆损害，可能源于耐受性或状态依赖性习惯化

给药途径：吸入与食用型大麻

• 吸入型大麻：起效迅速（主观感受约 2–5 分钟出现）；使用者可根据快速反馈有效调节剂量
• 食用型大麻：药代动力学有本质差异 — 起效缓慢，效应延迟，往往更为强烈且持续时间更长；剂量调节难度大得多
• THC 效能提升：现代商业大麻的 THC 含量通常为 20–30%，而 1970 年代约为 5% — 大致相当于从啤酒/葡萄酒升级到烈酒
• 尽管产品效能更高，有经验的使用者通过自我调节仍能达到相近的血液 THC 水平（约 100 ng/mL）— 浓缩物（dabs）除外，使用浓缩物时血液

---

English Original 英文原文

How Cannabis Impacts Health & the Potential Risks

Summary

Dr. Matthew Hill, a professor at the University of Calgary’s Hotchkiss Brain Institute, joins Andrew Huberman to discuss the biology of cannabis in depth. The conversation covers how THC interacts with the brain’s endocannabinoid system, the mechanisms behind psychoactive effects, appetite stimulation, memory, and the relationship between cannabis use and mental health. This episode was prompted by Dr. Hill’s public criticism of claims made in a previous Huberman Lab solo episode on cannabis.

---

Key Takeaways

• THC is not a sledgehammer — it is actually a partial agonist at CB1 receptors with high affinity but relatively low efficacy, similar to anandamide, not a supercharged version of the body’s own cannabinoids
• Cannabis “carpet bombs” the entire endocannabinoid system — unlike endogenous cannabinoids that act precisely at specific synapses and times, THC floods all CB1 receptors simultaneously, which is what produces intoxication
• Blocking anandamide breakdown does NOT produce a high — drugs that elevate anandamide by inhibiting its metabolic enzyme (FAAH inhibitors) show no psychoactivity, suggesting THC’s effects come from its indiscriminate, whole-brain receptor activation
• The “munchies” involve overriding satiety signals — cannabis reactivates appetite even in fully satiated animals by maintaining the reward value of food and blocking leptin’s anorectic effects
• Dopamine neurons are the one exception — CB1 receptors are found nearly everywhere in the brain except on dopamine neurons; cannabis influences dopamine indirectly by disinhibiting inhibitory neurons
• Edibles and inhaled cannabis are fundamentally different — the two routes produce very different pharmacokinetic profiles, onset times, and subjective experiences
• Regular users self-titrate effectively — even with higher-potency modern cannabis (20–30% THC), experienced users achieve similar blood THC levels (~100 ng/mL) as those using lower-potency products
• Short-term memory impairment during intoxication is well-established, but long-term cognitive deficits in regular users are not clearly supported by current evidence

---

Detailed Notes

What Is Cannabis?

• Cannabis is a plant with a long history of medicinal, spiritual, and recreational use across cultures
• The primary psychoactive compound is Delta-9-tetrahydrocannabinol (THC) — the higher the THC concentration, the more intense the intoxication
• Cannabidiol (CBD) is structurally similar to THC but is non-intoxicating; may have some psychoactive properties through effects on anxiety or mood
• 70+ minor cannabinoids exist in the plant (e.g., cannabinol, cannabigerol); their biology is largely unknown
• Terpenes (e.g., limonene, pinene, beta-caryophyllene, myrcene) are volatile aromatic compounds, not unique to cannabis; some have known biological activity
• The “entourage effect” refers to the idea that the combination of THC, minor cannabinoids, and terpenes produces different effects than isolated THC — this remains under active investigation

The Endocannabinoid System

• The brain’s primary cannabis-sensitive receptor is CB1 (cannabinoid type 1 receptor) — one of the most widely expressed receptors in the entire brain
• CB2 receptors are found mainly on immune cells (microglia) and regulate inflammation; minimal expression in neurons
• Neither receptor evolved for cannabis — they exist for the body’s own endocannabinoid system

Retrograde Signaling:

• Unlike most neurotransmitters (which travel forward, neuron A → neuron B), endocannabinoids are made in the postsynaptic neuron (B) and travel backward to regulate neurotransmitter release from the presynaptic neuron (A)
• Primary function: homeostasis — preventing runaway excitation (seizure-like states) or runaway inhibition

Two Primary Endocannabinoids:

• Anandamide — named from Sanskrit ananda (bliss); high affinity, low efficacy at CB1; thought to be tonic (steady-state regulator, like a thermostat)
• 2-Arachidonoylglycerol (2-AG) — lower affinity, higher efficacy; thought to be phasic (brought online on demand, especially during high neural activity or plasticity events)

How THC Differs from Endocannabinoids

• THC is a partial agonist at CB1 — similar affinity and efficacy to anandamide, not dramatically stronger
• The critical difference: endocannabinoids act at specific synapses at specific times; THC distributes throughout the bloodstream and activates CB1 receptors indiscriminately across all brain networks simultaneously
• This blanket activation — especially across cortical circuits — disrupts normal information processing and produces the intoxicated state
• FAAH inhibitors (drugs that block anandamide breakdown) confirm this: dramatically elevating anandamide produces no psychoactivity, demonstrating that THC’s intoxication is not simply about receptor activation level but about spatial and temporal indiscrimination

Cannabis, Dopamine, and Euphoria

• Dopamine neurons are notably absent of CB1 receptors — the only major neuron class in the brain that lacks them
• Cannabis influences dopamine indirectly: CB1 receptors on inhibitory neurons surrounding dopamine neurons in the ventral tegmental area (VTA) are activated by THC, removing inhibition and allowing dopamine neurons to burst-fire
• This mechanism parallels how mu-opioid receptors work with opioids
• Whether this dopamine release directly causes the euphoria of cannabis has not been cleanly demonstrated

Cannabis and Appetite (“The Munchies”)

• Cannabis stimulates appetite through multiple convergent mechanisms:
  1. Disinhibition of AgRP neurons in the hypothalamus (hypothalamic feeding circuits rich in CB1 receptors)
  2. Reward circuit engagement — anandamide in the nucleus accumbens enhances intake of palatable foods
  3. Enhanced gustatory processing — CB1 activation selectively amplifies sweet taste responses in the gustatory cortex (not salty, bitter, or sour)
• THC may trick the brain into a fasted state — endocannabinoids naturally rise during fasting to promote food seeking; THC activates those same circuits regardless of actual energy state
• Satiety override: Animals pre-satiated (or conditioned to avoid food via nausea) will re-initiate eating and even perform high levels of work (lever pressing) for food when given cannabis
• Leptin blockade: THC can override leptin signaling — normally, leptin suppresses endocannabinoid activity to reduce food intake; cannabinoid elevation reverses this

Cannabis and Memory

• Acute intoxication reliably impairs short-term memory — particularly recall and consolidation; well-supported by both animal and human research
• Long-term cognitive deficits in regular users are not clearly established — evidence is not consistently replicated or reliable
• State-dependent learning is a confound: chronic users who have repeatedly performed cognitive tasks while intoxicated may show less impairment when tested high, not because cannabis enhances cognition but because of adaptation
• Regular users tend to show less acute memory impairment than occasional users, likely due to tolerance or state-dependent habituation

Routes of Administration: Inhaled vs. Edibles

• Inhaled cannabis: Rapid onset (approximately 2–5 minutes to subjective effects); users can titrate dose effectively based on rapid feedback
• Edible cannabis: Fundamentally different pharmacokinetics — slower onset, delayed and often more intense and prolonged effects; much harder to titrate
• THC potency increase: Modern commercial cannabis typically runs 20–30% THC vs. ~5% in 1970s cannabis — roughly analogous to moving from beer/wine to spirits
• Despite higher-potency products, experienced users still achieve similar blood THC levels (~100 ng/mL) through self-titration — except with concentrates (dabs), where blood