Evidence level: B
三个靶蛋白里,STAT1 是唯一在哺乳动物有硬证据的那个。但正因为它太”活”,要在人血里读出它的月相成分,比读钟基因难得多——难点不在采样密度,在于你分不清读到的是零件还是传令兵。
正文
上一篇理出一份清单:环节虫月亮钟的靶蛋白里,proctolin 在脊椎动物没有直系同源,ependymin 家族善变到找不到清晰对应物,唯有 STAT1 不同——它在哺乳动物里有确凿的”时钟相关节律功能”文献支撑(PMID 29996098、PMID 32477351)。它因此成了最值得先做的一个。本文要讲的是:它也是最容易被做错的一个。
第一个坑:PBMC 不是一种细胞,是一个池子
外周血单个核细胞(PBMC)是单核细胞、T 细胞、B 细胞、NK 细胞的混合物。这几类细胞的 STAT1 基线表达量本来就不同,而它们在血液里的比例本身就在动——免疫细胞亚群的转运与功能有明确的昼夜依赖(PMID 31109762),这种转运由器官间通讯网络控制(PMID 33751050)。后果是:若月相真的改变了单核细胞与淋巴细胞的配比,整个 PBMC 池的平均 STAT1 信号就会跟着变,而每一种细胞内部什么都没变。你采到的是”配方变了”,报出来却是”STAT1 表达有月相成分”。不做亚群分选或单细胞测序,这个混淆解不开。
第二个坑:p-STAT1 是个比值,两个端点要分开报
磷酸化是昼夜控制的中心机制之一(PMID 27818261),STAT1 的功能状态通常靠 Tyr701 与 Ser727 位点的磷酸化比例来读。但”比例”这个词藏着陷阱:p-STAT1/STAT1 上升,可能来自磷酸化加强(调电压),也可能来自总蛋白减少(分母变小)。这是两种不同的调控逻辑,一个说”通路开大了”,一个说”蛋白被降解了”。报”STAT1 活性有月相节律”,会把两者糊成一团。必须做的是总蛋白谱、磷酸化谱、比值谱三条线分别报。全血流式测 p-STAT1 已有临床级方案(PMID 40265078),早年信号谱检测也在新生儿样本上跑通过(PMID 23809143)——技术上可行,但那些方案意在把采样时刻固定住以便比较病人,不是把时刻当自变量研究。
第三个坑:它被多少已知因素改写
STAT1 是一个随时被外部输入改写的效应器,不是安静待着等你读的钟。至少这些因素会动它:进食(餐后 PBMC 基因表达存在日间差异,PMID 33518404)、睡眠(睡眠不足会改写整个人血转录组的表达振幅,PMID 23440187)、性别与性激素(月经周期对免疫功能与炎症的影响有系统综述与 meta 分析支撑,PMID 37309068、PMID 36195867;免疫与月经周期的关系早被综述过,PMID 22155200;自身免疫病的性别差异机制仍在梳理,PMID 40806232)、个体间差异(单细胞层面对病毒的反应在人群间差异极大,PMID 37558883)。再叠上健康人免疫指标可观的个体内波动(PMID 28786508)与多重免疫测定的方法学变异(PMID 27810706),”多大的月相成分才算超过噪声”必须先有答案。
最核心的张力:一个谱峰,分不开两种身份
假设你真做完了连续三个周期、固定昼夜时刻、分层性别的采样,并用 JTK_CYCLE 或 RAIN(PMID 20876817、PMID 25326247)跑出了 STAT1 总量的 29.5 天成分。这仍然回答不了本文最关心的问题:STAT1 是”近月钟的零件”,还是”月亮信号的传令兵”?
两种身份在周期图上给出同一个峰,判据却完全不同。若是传令兵(下游执行器),它该跟着别的月级节律走——你该去看它与其他月级读数的相位关系。若是零件(内部计时器),它该具备自主性——相位在个体间自持、与昼夜相位解耦、在缺少外部月相标记时仍走。一句”STAT1 有月相节律”,把这两种可能一起说了,等于什么都没说。 这就是本文要立的判据:不是”有没有峰”,而是”峰的相位从哪来、被谁牵”。人血转录组已能推断昼夜相位(PMID 28218891),能否推到月相层尚无人试。
最小可行方案
不追求一步到位:①先在分选后的单核细胞上做(避开混合池陷阱);②总量/磷酸化/比值三条线分报;③至少纳入空腹、固定昼夜时刻,同步记录睡眠与月经周期相位作为协变量;④以个体为自身对照、用随机效应模型,让”月相成分”去和”个体内噪声”正面比大小;⑤女性组里 28 天与 29.5 天两个成分必须同时拟合(PMID 38598621 主张卵巢周期由内源月级计时系统驱动,PMID 42380750 从演化与生理两侧论证人类近月钟的存在,PMID 35567349 则在儿童褪黑素上给了一个真实的近月实证)。
一个参照系的问题
《素问·脉要精微论》给诊法定了条硬规矩:”诊法常以平旦,阴气未动,阳气未散,饮食未进,经脉未盛,络脉调匀,气血未乱,故乃可诊有过之脉。”道理很清楚:你为什么必须在一个固定时刻测?因为别的时刻读数已被”阴气动、阳气散、饮食进、气血乱”改写过——你读到的不是病,是刚刚发生的事。 现代的做法是空腹、固定采血时间。
但这段规矩管的只是一天里的那一格。本文要补的是:这个读数身上还压着另一把尺子,刻度是 29.5 天。
更要紧的是”平人”这个概念。《素问·平人气象论》讲”平人者,不病也”,用平人的脉当标准,一切偏离都相对它算。可如果连”平人”的 STAT1 基线自己就在走一个 29.5 天的圈呢? 那参照系就是动的——你测到的每一次”异常”,都可能只是尺子在晃。所有测量的前提是参照系不动。中医把平人当参照系,现代医学把健康对照的基线当参照系,而这两个参照系,都没被检查过自己是不是平的。
证据与局限
已验证: ①STAT1 是哺乳动物昼夜钟-免疫交叉的直接执行者(PMID 29996098、PMID 32477351);②磷酸化是昼夜控制的中心机制(PMID 27818261);③免疫细胞亚群转运与功能存在昼夜依赖(PMID 31109762、PMID 33751050);④全血流式检测 p-STAT1 的方法已建立(PMID 40265078、PMID 23809143);⑤PBMC 昼夜钟基因表达可测且模式已被刻画(PMID 16096739、PMID 17412523);⑥人血转录组可用于推断昼夜相位(PMID 28218891)。
部分验证: ①进食与睡眠会改写 PBMC 基因表达读数(PMID 33518404、PMID 23440187)——但改写的量级相对月相成分的比值无标定;②月经周期对免疫功能的影响有 meta 分析支持(PMID 37309068、PMID 36195867)——但 28 天与 29.5 天成分在同一批样本中的可分离性未测;③人类近月节律有真实实证(PMID 35567349、PMID 38598621、PMID 42380750)——靶组织仍未落到免疫细胞。
待验证: ①人类 PBMC(或分选单核细胞)中 STAT1 总量、磷酸化水平与比值的 29.5 天表达成分,均无研究执行;②该成分的相位是否在个体间自持、是否与昼夜相位解耦(”零件 vs 传令兵”的判据尚无数据);③月相成分幅度相对个体内波动(PMID 28786508)与方法学变异(PMID 27810706)的相对大小;④”平人基线是否本身随月相漂移”这一问题在免疫指标上从未被检验。
相关阅读
- 近月钟靶蛋白的哺乳动物同源扫描
- 在人血里找 29.5 天的钟:难点不在采血,在让昼夜先让开
- 人类近月钟的分子实体:对照环节虫通路扫描哺乳动物同源基因的月相表达
- 人体内源近月振荡器:月亮时钟是否刻在基因里
- 研究池:生物频率密码
证据等级: B
参考文献: PMID 29996098, PMID 32477351, PMID 31109762, PMID 33751050, PMID 27818261, PMID 40265078, PMID 23809143, PMID 16096739, PMID 17412523, PMID 33518404, PMID 23440187, PMID 37309068, PMID 36195867, PMID 22155200, PMID 40806232, PMID 37558883, PMID 28786508, PMID 27810706, PMID 20876817, PMID 25326247, PMID 28218891, PMID 38598621, PMID 42380750, PMID 35567349
免责说明:本文属传统知识与科学文献的整理、对照与研究设计讨论,仅供理解与参考,不构成医疗建议,不能替代专业诊断与治疗。文中涉及的效应量、频率区间与暴露结论多为研究报告层面的推断,个体情况差异很大;如有身体不适或健康疑虑,请就医并由专业医师判断。
Identifying the 29.5-day component of STAT1 in PBMCs: Is it a peak or a messenger?
Evidence level: B
Among the three target proteins, STAT1 is the only one with robust evidence in mammals. However, precisely because it is so “active,” reading its lunar phase component in human blood is far more difficult than reading clock genes—the challenge lies not in sampling density, but in distinguishing whether the signal read reflects a component or a messenger.
Main Text
In the previous article, we compiled a list: among the target proteins of the leech lunar clock, proctolin lacks a direct homolog in vertebrates, the ependymin family is too variable to find a clear counterpart, and only STAT1 stands out—it has solid literature support for “clock-related rhythmic function” in mammals (PMID 29996098, PMID 32477351). It therefore becomes the most logical starting point. This article addresses: it is also one of the most easily misinterpreted.
Pitfall One: PBMCs Are Not a Single Cell Type, But a Pool
Peripheral blood mononuclear cells (PBMCs) are a mixture of monocytes, T cells, B cells, and NK cells. The baseline STAT1 expression levels differ inherently among these cell types, and their proportions within the blood are dynamic—immune cell subpopulation trafficking and function exhibit clear circadian dependencies (PMID 31109762), regulated by inter-organ communication networks (PMID 33751050). The consequence is: if the lunar phase truly alters the ratio of monocytes to lymphocytes, the average STAT1 signal across the entire PBMC pool will shift, while nothing changes within each individual cell. You are measuring a “changed recipe,” yet reporting it as “STAT1 expression has a lunar component.” Without subpopulation sorting or single-cell sequencing, this confounding factor cannot be resolved.
Pitfall Two: p-STAT1 Is a Ratio; Both Endpoints Must Be Reported Separately
Phosphorylation is one of the central mechanisms for circadian control (PMID 27818261), and the functional state of STAT1 is typically read via the phosphorylation ratio at Tyr701 and Ser727 sites. However, the term “ratio” hides a trap: an increase in p-STAT1/STAT1 may stem from enhanced phosphorylation (increasing the numerator) or reduced total protein (decreasing the denominator). These represent two distinct regulatory logics: one suggests “the pathway is turned up,” while the other implies “protein is being degraded.” Reporting “STAT1 activity has a lunar rhythm” conflates these two possibilities. The necessary approach is to report three distinct lines: total protein spectra, phosphorylation spectra, and ratio spectra. Whole-blood flow cytometry for p-STAT1 already has a clinical-grade protocol (PMID 40265078), and earlier signal spectra detection has been validated on neonatal samples (PMID 23809143)—technically feasible, yet those protocols aim to fix sampling times for patient comparison rather than treat time as an independent variable.
Pitfall Three: How Many Known Factors Rewrite It?
STAT1 is an effector constantly rewritten by external inputs, not a silent clock waiting to be read. At least the following factors influence it: feeding (post-prandial PBMC gene expression exhibits diurnal differences, PMID 33518404), sleep (sleep deprivation rewrites the expression amplitude of the entire human blood transcriptome, PMID 23440187), gender and sex hormones (systematic reviews and meta-analyses support the impact of the menstrual cycle on immune function and inflammation, PMID 37309068, PMID 36195867; the relationship between immunity and the menstrual cycle was previously reviewed, PMID 22155200; mechanisms for gender differences in autoimmune diseases are still being elucidated, PMID 40806232), and inter-individual variability (single-cell responses to viruses vary drastically across populations, PMID 37558883). Layered upon this are significant intra-individual fluctuations in healthy immune markers (PMID 28786508) and methodological variations across multiple immune assays (PMID 27810706). Therefore, the answer to “how large a lunar component exceeds noise” must be established beforehand.
The Core Tension: One Spectral Peak, Two Distinct Identities
Assume you have completed sampling across three consecutive cycles at fixed circadian times, stratified by gender, and used JTK_CYCLE or RAIN (PMID 20876817, PMID 25326247) to identify a 29.5-day component of STAT1 total abundance. This still does not answer the central question of this article: Is STAT1 a “part of the near-lunar clock” or a “messenger for lunar signals”?
Both identities produce the same peak on the periodogram, yet the criteria for validation differ completely. If it is a messenger (downstream effector), it should follow other lunar rhythms—you should examine its phase relationship with other lunar readings. If it is a part (internal timer), it should possess autonomy—its phase should be self-sustaining across individuals, decoupled from circadian phases, and continue running even in the absence of external lunar markers. Saying “STAT1 has a lunar rhythm” covers both possibilities simultaneously, amounting to saying nothing. This is the criterion we establish here: not “whether there is a peak,” but “where the phase originates and what drives it.” Human blood transcriptomes can already infer circadian phases (PMID 28218891); extending this inference to lunar phases remains untested.
Minimum Viable Protocol
Do not aim for a single step: ① Perform the study first on sorted monocytes (avoiding the mixed pool trap); ② Report total, phosphorylated, and ratio lines separately; ③ Include fasting status, fixed circadian times, and record sleep and menstrual cycle phases as covariates; ④ Use individuals as their own controls with random-effects models, allowing the “lunar component” to be directly compared against “intra-individual noise”; ⑤ In the female cohort, both the 28-day and 29.5-day components must be simultaneously fitted (PMID 38598621 argues for an endogenous lunar-timing system driving the ovarian cycle, PMID 42380750 argues for the existence of a human near-lunar clock from evolutionary and physiological perspectives, PMID 35567349 provides real near-lunar empirical evidence in human melatonin).
The Problem of a Reference Frame
The Su Wen · Pulse and Essence establishes a rigid rule for diagnosis: “Diagnosis should always be conducted at dawn, before Yin Qi moves and Yang Qi disperses, before food is ingested, before meridians are full, before collaterals are unbalanced, and before Qi and Blood are chaotic; thus, one may detect pathological pulses.” The logic is clear: Why must measurement occur at a fixed time? Because readings at other times are rewritten by “Yin Qi movement, Yang Qi dispersion, food ingestion, Qi and Blood chaos”—what you read is not disease, but a recent event. Modern practice involves fasting and fixed blood draw times.
However, this rule governs only one grid within a day. This article aims to supplement it: This reading also bears another ruler, scaled at 29.5 days.
More critically is the concept of “Ping Ren” (healthy person). The Su Wen · Ping Ren Qi Xiang Lun states “Ping Ren refers to one without illness,” using the pulse of a healthy person as the standard against which all deviations are calculated. But what if the STAT1 baseline of even a “Ping Ren” is itself traversing a 29.5-day circle? Then the reference frame is moving—every measured “abnormality” might simply be the ruler swinging. All measurement presupposes a stable reference frame. Traditional Chinese medicine treats “Ping Ren” as the reference; modern medicine treats healthy control baselines as the reference. Yet neither of these reference frames has been verified for their own stability.
Evidence and Limitations
Verified: ① STAT1 is a direct effector of the mammalian circadian clock-immune crosstalk (PMID 29996098, PMID 32477351); ② Phosphorylation is a central mechanism for circadian control (PMID 27818261); ③ Immune cell subpopulation trafficking and function exhibit circadian dependencies (PMID 31109762, PMID 33751050); ④ Whole-blood flow cytometry for p-STAT1 has an established protocol (PMID 40265078, PMID 23809143); ⑤ Circadian clock gene expression in PBMCs is measurable and its patterns have been characterized (PMID 16096739, PMID 17412523); ⑥ Human blood transcriptomes can be used to infer circadian phases (PMID 28218891).
Partially Verified: ① Feeding and sleep rewrite PBMC gene expression readings (PMID 33518404, PMID 23440187)—but the magnitude of rewriting relative to the ratio of lunar components lacks calibration; ② The impact of the menstrual cycle on immune function is supported by meta-analysis (PMID 37309068, PMID 36195867)—but the separability of 28-day and 29.5-day components within the same sample batch remains untested; ③ Real empirical evidence for human near-lunar rhythms exists (PMID 35567349, PMID 38598621, PMID 42380750)—but the target tissue has not yet been localized to immune cells.
To Be Verified: ① No study has executed the 29.5-day expression component of STAT1 total abundance, phosphorylation levels, and ratios in human PBMCs (or sorted monocytes); ② Whether the phase of this component is self-sustaining across individuals and decoupled from circadian phases (no data yet for the “part vs. messenger” criterion); ③ The relative magnitude of the lunar component amplitude compared to intra-individual fluctuations (PMID 28786508) and methodological variations (PMID 27810706); ④ The question of whether the “Ping Ren baseline itself drifts with the lunar phase” has never been tested in immune markers.
Related Reading
- Mammalian Homolog Scan for Near-Lunar Clock Target Proteins
- Finding a 29.5-Day Clock in Human Blood: The Challenge Lies Not in Sampling, But in Making Circadian Rhythms Yield
- Molecular Entities of the Human Near-Lunar Clock: Contrasting Leech Pathways with Mammalian Homolog Gene Lunar Expression
- Endogenous Human Near-Lunar Oscillator: Is the Lunar Clock Carved into Genes?
- Research Pool: Biological Frequency Codes
Evidence Level: B
References: PMID 29996098, PMID 32477351, PMID 31109762, PMID 33751050, PMID 27818261, PMID 40265078, PMID 23809143, PMID 16096739, PMID 17412523, PMID 33518404, PMID 23440187, PMID 37309068, PMID 36195867, PMID 22155200, PMID 40806232, PMID 37558883, PMID 28786508, PMID 27810706, PMID 20876817, PMID 25326247, PMID 28218891, PMID 38598621, PMID 42380750, PMID 35567349
Disclaimer: This article compiles traditional knowledge and scientific literature for comparison and research-design discussion only. It is not medical advice and cannot replace professional diagnosis or treatment. Effect sizes, frequency ranges and exposure conclusions cited here are largely inferences at the research-report level, and individual circumstances vary widely. If you have any discomfort or health concerns, please consult a physician.
Leave a Reply