Investigating the efficacy of ergothioneine to delay cognitive decline in mild cognitively impaired subjects: A pilot study
研究麦角硫因延缓轻度认知障碍受试者认知能力下降的功效：一项初步研究

ET, L-ergothioneine-d₉ (ET-d₉), L-hercynine, L-hercynine-d₉, S-methyl ET were provided by ERGOLD (formerly Tetrahedron, Paris, France). Encapsulated ET (25 mg/capsule Ergoneine^®, synthetic ET >99.5% purity, GMP-certified, Lot no. 128020) and placebo (99% microcrystalline cellulose, 1% magnesium stearate) were also provided in coded bottles by ERGOLD. HPLC grade methanol (A454-4) and acetonitrile (A998-4) were purchased from Fisher Chemical (US). Formic acid (98%; 8841) was purchased from Lancaster (England). All other reagents were purchased from Sigma-Aldrich (US) unless otherwise specified.
ET、L-麦角硫因-_{d 9} （ET-d₉）、L-海角氨酸、L-海角氨酸-d₉、S-甲基 ET 由 ERGOLD（原 Tetrahedron，巴黎，法国）提供。ERGOLD 还提供封装的 ET（25 毫克/粒麦角因 ^®，合成 ET >99.5% 纯度，GMP 认证，批号 128020）和安慰剂（99% 微晶纤维素，1% 硬脂酸镁）也装在编码瓶中。HPLC 级甲醇（A454-4）和乙腈（A998-4）购自 Fisher Chemical（美国）。甲酸（98%;8841）购自兰开斯特（英国）。除非另有说明，否则所有其他试剂均从 Sigma-Aldrich（美国）购买。

This randomized, double-blinded, placebo-controlled study (ClinicalTrials.gov identifier: NCT03641404) was conducted at the Investigational Medicine Unit, National University Hospital, Singapore in adherence with Good Clinical Practice (GCP). MCI subjects, 60–90 years of age were randomly assigned (GraphPad Prism Randomizer) to receive either placebo or ET (25 mg) capsules (identical capsule appearance, coded by manufacturer), administered orally three times a week, blinded to both subjects and administrators for the entire duration of the study (Figure 1). Participants visited the study site every 4 weeks (for a total of 14 visits) where they were provided with sufficient capsules to last until the subsequent visit and underwent safety and compliance monitoring. Blood samples were collectedl at baseline and quarterly (visits 1, 4, 7, 10, 14) for clinical safety assessment and biomarker analyses). Neuro-cognitive assessments were conducted biannually (visits 7 and 14). Baseline cognitive assessments were carried out during cohort screening. At the trial's conclusion, subjects underwent a final assessment, followed by a follow-up call one month later to monitor for any delayed adverse effects. The study adhered to ethical standards according to the Declaration of Helsinki and was approved by the Singapore National Healthcare Group, Domain Specific Review Board under protocol number 2017/00982 and Singapore Health Sciences Authority, Clinical Trial Certificate number CTC1800036.
这项随机、双盲、安慰剂对照研究（ClinicalTrials.gov 标识符：NCT03641404）在新加坡国立大学医院研究医学部进行，遵守良好临床规范 （GCP）。60-90 岁的 MCI 受试者被随机分配（GraphPad Prism Randomizer）接受安慰剂或 ET（25 mg）胶囊（胶囊外观相同，由制造商编码），每周口服给药 3 次，在整个研究期间对受试者和管理人员不知情（ 图 1）。参与者每 4 周访问一次研究地点（总共 14 次访问），在那里他们获得了足够的胶囊，以持续到随后的访问，并接受了安全性和合规性监测。在基线和每季度（访问 1、4、7、10、14）收集血样，用于临床安全性评估和生物标志物分析）。每半年进行一次神经认知评估（第 7 次和第 14 次访问）。在队列筛选期间进行基线认知评估。试验结束时，受试者接受了最终评估，然后在一个月后进行随访，以监测任何延迟的不良反应。该研究遵守了赫尔辛基宣言的伦理标准，并获得了新加坡国家医疗保健集团、特定领域审查委员会的批准，协议编号为 2017/00982，新加坡卫生科学局的临床试验证书编号为 CTC1800036。

MCI subjects from the Diet and Healthy Ageing study (DaHA),³⁶ Ageing in a Community Environment Studies (ACES),³⁷ or Community Health Intergenerational Study (CHI)³⁸ cohorts (comprised of community-dwelling elderly individuals from the Western regions of Singapore) were contacted (prior consent had been given to be contacted for future studies) for their potential interest to participate in the study. Given the slow identification rate of MCI subjects, consenting subjects were recruited in staggered enrollment. The recruited volunteers were screened (within a 6-month period of enrollment) using a cognitive assessment battery (comprising the Singapore modified Mini-Mental State Examination; SM-MMSE³⁹ to assess global cognitive function together with a neurocognitive battery to assess other cognitive domains in detail) and identified as MCI by a panel of two senior consultant psychiatrists (EHK, RM) and the cohort PI (LF) according the consensus Petersen's criteria.⁴⁰ General health conditions, including height, weight, blood pressure, age, medical history, dietary intake, were also assessed. Blood and urine samples were collected for biomarker and clinical safety monitoring (full blood count and renal and liver function tests). Inclusion criteria focused on ethnically Chinese individuals aged 60–90 years with MCI. From >1500 subjects screened in the three cohorts, 121 MCI subjects meeting the study criteria were invited to participate. Participants were required to be independent, capable of understanding the study's demands and of giving informed consent, and free from severe health issues or terminal illnesses. Exclusion criteria included mushroom allergies, history of severe cardiovascular complications, hematological conditions such as anemia, history of mental or psychiatric illnesses, history of drug or alcohol abuse, concurrent involvement in other studies requiring compound intake in the past 6 months, or any other underlying conditions deemed unsuitable by the investigator. Participants were asked to avoid the use of supplements or Traditional Chinese Medicines and excessive mushroom consumption for the duration of the study, and to report other concurrent medications.
来自饮食与健康老龄化研究 （DaHA）、³⁶ 社区环境老龄化研究 （ACES）、³⁷ 或社区健康代际研究 （CHI）³⁸ 队列（由来自新加坡西部地区的社区老年人组成）的 MCI 受试者被联系（事先同意联系他们进行未来的研究）以了解他们参与研究的潜在兴趣。鉴于 MCI 受试者识别率较慢，同意受试者以交错招募方式招募。招募的志愿者使用认知评估组合（包括新加坡改良的简易精神状态检查;SM-MMSE³⁹ 评估整体认知功能以及神经认知电池以详细评估其他认知领域），并由两名高级顾问精神科医生（EHK、RM）和队列 PI （LF） 组成的小组确定为 MCI，根据共识 Petersen 的标准。 ^{还评估了 40} 种一般健康状况，包括身高、体重、血压、年龄、病史、饮食摄入量。收集血液和尿液样本用于生物标志物和临床安全性监测（全血细胞计数以及肾和肝功能测试）。纳入标准侧重于 60-90 岁患有 MCI 的华裔个体。从三个队列中筛选的 >1500 名受试者中，邀请了 121 名符合研究标准的 MCI 受试者参加。参与者被要求独立，能够理解研究的要求并给予知情同意，并且没有严重的健康问题或绝症。 排除标准包括蘑菇过敏、严重心血管并发症史、贫血等血液系统疾病、精神或精神疾病史、药物或酒精滥用史、在过去 6 个月内同时参与其他需要复合摄入的研究，或任何其他潜在疾病研究者认为不合适。参与者被要求在研究期间避免使用补充剂或中药和过量食用蘑菇，并报告其他并发药物。

Subjects were required to visit the study site every 4 weeks for safety and compliance monitoring. During the study visits, subjects were required to return any unconsumed capsules from the previous batch and provided with a new batch. An administration and adverse event log kept by the subject was checked for evidence of study compliance and safety. In addition to monitoring weight and blood pressure, fasting blood was collected every 3 months to assess safety by monitoring any changes in full blood counts, fasting glucose, and liver and renal function tests. Serum and sodium tubes were sent to the National University Hospital Referral Laboratories (Singapore) for blood parameter analysis including hematological profile, liver panel (albumin, total bilirubin, aspartate transaminase (AST), alanine transaminase (ALT), alkaline phosphatase (ALP)), renal panel (sodium, potassium, urea, creatinine), and fasting glucose. Those failing to adhere to the administration schedule in the first 3 months or demonstrating abnormal blood parameters were withdrawn from the study.
受试者被要求每 4 周访问一次研究地点以进行安全和依从性监测。在研究访问期间，受试者被要求退回前一批未消耗的胶囊，并提供新批次。检查受试者保存的给药和不良事件日志，以寻找研究依从性和安全性的证据。除了监测体重和血压外，每 3 个月采集一次空腹血，通过监测全血细胞计数、空腹血糖以及肝肾功能测试的任何变化来评估安全性。将血清和钠管送往国立大学医院转诊实验室（新加坡）进行血液参数分析，包括血液学特征、肝脏检查（白蛋白、总胆红素、天冬氨酸转氨酶 （AST）、丙氨酸转氨酶 （ALT）、碱性磷酸酶 （ALP））、肾检查（钠、钾、尿素、肌酐）和空腹血糖。那些在前 3 个月内未能遵守给药计划或表现出血液参数异常的人被退出研究。

Fasting venous blood samples were collected in 1x K₂-EDTA spray-coated vacuum tubes (Becton Dickinson, US), 1x clot activator-treated vacuum tubes (Becton Dickinson, US) and 1x sodium fluoride-treated plasma tubes (Becton Dickinson, US) for whole blood/plasma, serum, and fasting plasma glucose measurement, respectively. Plasma samples were separated by centrifuging the whole blood samples in EDTA tubes at 1000 × g for 15 min. The samples were stored at −80 °C in 250 μl aliquots supplemented with 2 μl of 10 mM butylated hydroxytoluene (antioxidant) and 1 μl of 25 mM indomethacin (cyclooxygenase inhibitor) to prevent ex vivo autooxidation.
空腹静脉血样收集在 1x K 2-EDTA 喷涂真空管（Becton Dickinson，美国）、1x 凝块激活剂处理的真空管（Becton Dickinson，美国）和 1x 氟化钠处理的血浆管（Becton Dickinson，美国）中，分别用于全血/血浆、血清和空腹血糖测量。通过将 EDTA 管中的全血样本以 1000 × g 离心 15 min 来分离血浆样本。将样品储存在-80°C 的 250μl 等分试样中，补充有 2μl10mM 丁基化羟甲苯（抗氧化剂）和 1μl25mM 吲哚美辛（环氧合酶抑制剂），以防止离体自氧化。

Cognitive functions were assessed during screening, at visit 7 (26 weeks) and at visit 14 (52 weeks) using SM-MMSE modified for Chinese subjects,³⁹ Rey Auditory Verbal Learning Test (RAVLT), Digit Span (WAIS-III UK version), Color Trials Tests (CTT), Block Design (WAIS-III UK version), Semantic Fluency, Symbol Digit Modality Test (SDMT), Boston Naming Test and Clinical Dementia Rating Scale (CDRS). Z-scores of the test results were calculated using the age- and education-adjusted local norms.⁴¹
在筛选期间、第 7 次访视（26 周）和第 14 次访视（52 周）使用针对中国受试者修改的 SM-MMSE、³⁹ Rey 听觉语言学习测试 （RAVLT）、数字跨度（WAIS-III 英国版）、颜色试验测试 （CTT）、块设计（WAIS-III 英国版）、语义流畅性、符号数字模态测试 （SDMT）、波士顿命名测试和临床痴呆评定量表 （CDRS）。测试结果的 Z 分数是使用年龄和教育调整的当地标准计算的。⁴¹

Plasma samples were analyzed following previously optimized methods.⁴² Briefly, 10 μl of plasma was mixed with 100 μl methanol containing ET-d₉ and hercynine-d₉ as internal standards. The samples were mixed and incubated at −20°C for 2 h followed by centrifugation. The supernatants were dried under a stream of nitrogen gas, resuspended in 100 μl of dH₂O, and transferred to a salinized glass insert in a sampling vial (Agilent Technologies, USA) for analysis by LC-MS/MS on the Agilent 1290 UPLC and 6460-QQQ MS (Agilent Technologies, USA).
按照先前优化的方法分析血浆样品。⁴² 简而言之，将 10 μl 血浆与含有 ET-d₉ 和 hercynine-d₉ 的 100 μl 甲醇混合作为内标。将样品混合并在-20°C 下孵育 2 h，然后离心。将上清液在氮气流下干燥，重悬于 100 μl dH₂O 中，然后转移到采样瓶（Agilent Technologies，美国）中的盐渍玻璃插入物中，在 Agilent 1290 UPLC 和 6460-QQQ MS（Agilent Technologies，美国）上进行 LC-MS/MS 分析。

ELISA kits for human tumor necrosis factor-α (TNF-α, ab181421), interleukin-18 (IL-18, ab215539), brain-derived neurotrophic factor (BDNF, ab212166), and protein carbonyls (ab238536) were purchased from Abcam. ELISA kits for neurofilament light-chain (NfL, E-EL-H0741) were from Probioscience Technologies. All samples were measured in duplicate, according to the manufacturer's protocols. All the ELISA assays were colorimetric, with absorbances read at 450 nm (BioTek Synergy H1). Concentrations were calculated using appropriate calibration curves.
人肿瘤坏死因子-α（TNF-α，ab181421）、白细胞介素-18（IL-18，ab215539）、脑源性神经营养因子（BDNF，ab212166）和蛋白质羰基（ab238536）的 ELISA 试剂盒购自 Abcam。用于神经丝轻链的 ELISA 试剂盒（NfL，E-EL-H0741）来自 Probioscience Technologies。根据制造商的协议，所有样品均一式两份进行测量。所有 ELISA 测定均采用比色法，吸光度读数为 450 nm （BioTek Synergy H1）。使用适当的校准曲线计算浓度。

Our original sample size calculations (conducted by JL) were based on composite cognitive outcomes from a standard neuropsychological assessment battery (from the Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability).⁴³ Based on these cognitive outcomes, 44 MCI subjects per arm were required for a statistical power of 80% to detect a z-score difference of 0.3 over the one-year intervention, using a conventional alpha of 0.05. Given an estimated attrition rate of 20%, 106 subjects were needed for the study. However, given the complications in recruitment (see Discussion), and failure to hit the intended target, we have chosen to assess these data as a pilot study which demonstrates the safety and potential efficacy of ET.
我们最初的样本量计算（由 JL 进行）基于标准神经心理学评估组合（来自芬兰预防认知障碍和残疾的老年干预研究）的综合认知结果。⁴³ 基于这些认知结果，每组需要 44 名 MCI 受试者才能获得 80% 的统计功效，以检测一年干预期间 0.3 的 z 分数差异，使用常规 alpha 为 0.05。鉴于估计流失率为 20%，该研究需要 106 名受试者。然而，考虑到招募的复杂性（参见讨论）以及未能达到预期目标，我们选择将这些数据作为一项试点研究进行评估，以证明 ET 的安全性和潜在有效性。

The statistical analysis adopted an intention-to-treat approach that included data from all participants, including the early withdrawals, to preserve randomness and provide a conservative estimate of the treatment effects. Data are expressed as mean ± standard deviation unless specified, n = 11 for the ET-supplemented arm and n = 8 for the placebo arm. Paired mixed-effect analysis, which deals with missing data, was used for paired comparison among study time points, followed by Dunnet's pairwise comparison against baseline with the following annotation: ***p < 0.001; ****p < 0.0001. Outliers were identified by ROUT method (Q = 0.1). Statistical analysis was conducted using GraphPad Prism (Version10).
统计分析采用意向治疗方法，包括所有参与者的数据，包括早期退出的数据，以保持随机性并提供治疗效果的保守估计。除非另有说明，否则数据表示为平均值±标准差，ET 补充组为 n = 11，安慰剂组为 n = 8。处理缺失数据的配对混合效应分析用于研究时间点之间的配对比较，然后是 Dunnet 与基线的成对比较，并附有以下注释：***p < 0.001;p < 0.0001。通过 ROUT 方法识别异常值 （Q = 0.1）。使用 GraphPad Prism（版本 10）进行统计分析。

More than 1500 subjects from the three community cohorts were screened, of which 121 identified MCI subjects meeting the study criteria were contacted for interest to participate in the study. However, only 19 were enrolled and 14 completed the study. The poor uptake of the study was in part due to COVID-19 related fears among the volunteers as the pandemic hit soon after the commencement of the study (also leading to two voluntary withdrawals). Of the five withdrawn, two were due to health complications unrelated to the study and three were due to voluntary withdrawal or noncompliance. Among the 14 completed subjects, some scheduled measurements were missed due to restrictions during the COVID-19 pandemic. The enrolled participants all demonstrated MCI at baseline (as diagnosed by the clinical assessment team) with a median CDR-SOB of 0.5 (p = 0.001, Wilcoxon signed-rank test against 0).⁴⁴ No significant differences were observed in baseline plasma levels of ET and its metabolites, hercynine and S-methyl ET, between the treatment groups (Table 1). Baseline plasma ET levels demonstrated no significant correlation with Mini-Mental State Examination (MMSE) scores or age (Figure 2(a) and (b)). No significant differences were observed in mean plasma ET levels between males (872 nM) and females (828 nM) (Figure 2(c)). Similarly, levels of the ET metabolites showed no age or gender differences.
筛选了来自三个社区队列的 1500 多名受试者，其中联系了 121 名已确定符合研究标准的 MCI 受试者，以了解他们有兴趣参与该研究。然而，只有 19 人入组，14 人完成了研究。该研究的接受率低，部分原因是志愿者对 COVID-19 的担忧，因为研究开始后不久就出现了大流行（也导致了两次自愿退出）。在五名退出者中，两项是由于与研究无关的健康并发症，三项是由于自愿退出或不依从。在完成的 14 名受试者中，由于 COVID-19 大流行期间的限制，一些预定的测量被错过了。登记的参与者在基线时均表现出 MCI（由临床评估团队诊断），中位 CDR-SOB 为 0.5（p = 0.001，Wilcoxon 符号秩检验对 0）。⁴⁴ 治疗组之间 ET 及其代谢物 hercynine 和 S-甲基 ET 的基线血浆水平没有观察到显着差异（ 表 1）。基线血浆 ET 水平与简易精神状态检查（MMSE）评分或年龄没有显着相关性（ 图 2（a） 和（b））。男性（872 nM）和女性（828 nM）之间的平均血浆 ET 水平没有观察到显着差异（ 图 2（c））。同样，ET 代谢物的水平没有年龄或性别差异。

At baseline, the average plasma levels of ET, hercynine, and S-methyl-ET of the study population were at 855 nM, 27 nM, and 7 nM, respectively (Figure 3). In subjects with ET supplementation, the ET level increased almost 5-fold to 3960 nM by visit 7 and continued to increase linearly to 6860 nM by visit 14 (Figure 3(a)). Hercynine and S-methyl-ET, also increased over time, albeit to a smaller extent, and reached 125 nM and 25 nM at visit 14, respectively (Figure 3(b) and (c)). No significant changes in levels of the three compounds were observed in the placebo group throughout the study.
在基线时，研究人群的 ET、hercynine 和 S-甲基-ET 的平均血浆水平分别为 855 nM、27 nM 和 7 nM（ 图 3）。在补充 ET 的受试者中，到第 7 次访视时，ET 水平增加了近 5 倍，达到 3960 nM，到第 14 次访视时继续线性增加至 6860 nM（ 图 3（a））。Hercynine 和 S-甲基-ET 也随着时间的推移而增加，尽管程度较小，并在第 14 次访视时分别达到 125 nM 和 25 nM（ 图 3（b） 和 （c））。在整个研究过程中，安慰剂组没有观察到三种化合物的水平发生显着变化。

ET administration did not cause significant changes in whole blood parameters, liver function markers, renal function markers, or fasting glucose levels (Table 2). Participants in both ET and placebo groups recorded a lower total white blood cell count compared to baseline at visit 7, both of which recovered subsequently. The reasons for this anomaly are unclear but values were all still within the expected range for their age.
ET 给药不会导致全血参数、肝功能标志物、肾功能标志物或空腹血糖水平发生显着变化（ 表 2）。与第 7 次就诊时基线相比，ET 组和安慰剂组的参与者记录的白细胞总计数较低，随后均恢复。造成这种异常的原因尚不清楚，但数值仍在其年龄的预期范围内。

We evaluated the cognitive performance of the subjects using a neuro-cognitive assessment battery consisting of 10 test components, at visit 7 and visit 14 (6 and 12 month, respectively). Baseline scores obtained during the subject screening process (visit 0) within 6 months prior to the commencement of our study, were considered the reference levels. SDMT tests were not conducted during prescreening. At the baseline, no significant differences in the Z-scores of all NCA components were observed between placebo and ET-supplemented groups. Following ET intake, an increase in Z-scores was observed in RAVLT assessments (immediate and delayed recalls), which evaluates learning ability and memory (Figure 4(a) and (b)). In contrast, no significant increase was observed in Z-scores across all NCA components, in the subjects given the placebo. A decrease in the Z-score of CTT1 was seen at visit 7 after ET intake but returned to baseline Z-scores by visit 14 (Figure 4(e)). Block design scores, which assess visual-motor coordination and problem-solving skills, declined over time in both ET and placebo groups (Figure 4(g)). Although not measured at baseline, Z-scores of SDMT (written format) in the ET group showed a trend to increase from visit 7 to visit 14 (Figure 4(i)).
我们在第 7 次访问和第 14 次访问（分别为 6 个月和 12 个月）时使用由 10 个测试组件组成的神经认知评估组合评估了受试者的认知表现。在我们的研究开始前 6 个月内在受试者筛选过程（访视 0）期间获得的基线分数被视为参考水平。预筛选期间未进行 SDMT 测试。在基线时，安慰剂组和补充 ET 组之间所有 NCA 成分的 Z 分数没有显着差异。摄入 ET 后，在评估学习能力和记忆的 RAVLT 评估（立即和延迟回忆）中观察到 Z 分数增加（ 图 4（a） 和 （b））。相比之下，在给予安慰剂的受试者中，所有 NCA 成分的 Z 分数没有观察到显着增加。在摄入 ET 后的第 7 次访视时观察到 CTT1 的 Z 评分下降，但在第 14 次访视时恢复到基线 Z 评分（ 图 4（e））。评估视觉运动协调和解决问题能力的块设计分数在 ET 组和安慰剂组中都随着时间的推移而下降（ 图 4（g））。尽管未在基线时测量，但 ET 组中 SDMT（书面格式）的 Z 分数显示出从第 7 次访问到第 14 次访问的增加趋势（ 图 4（i））。

No significant differences in the plasma biomarkers were observed at baseline between ET and placebo arms. Following ET consumption in the MCI subjects, no significant changes in the plasma levels of NfL (a biomarker of neuronal injury) was seen over the 12-month study duration, with a trend to decreasing levels but this was not significant (Figure 5(a)). Conversely, a significant increase in plasma NfL was observed in the placebo group by visit 14. One data point was identified as an outlier by Prism likely due to technical error and hence the subject was completely removed from the analysis. Both ET and placebo consumption did not affect the plasma levels of BDNF, TNF-α, and IL-18 as markers of neurological function and inflammation, and protein carbonyls, as an index of oxidative protein damage (Figure 5(b)–(e)).⁴⁵
在基线时，ET 组和安慰剂组的血浆生物标志物没有观察到显着差异。在 MCI 受试者中服用 ET 后，在 12 个月的研究期间，NfL（神经元损伤的生物标志物）血浆水平没有显着变化，水平有下降的趋势，但这并不显着（ 图 5（a））。相反，在第 14 次就诊时观察到安慰剂组的血浆 NfL 显着增加。Prism 将一个数据点确定为异常值，可能是由于技术错误，因此该主题被完全从分析中删除。ET 和安慰剂的摄入均不影响血浆 BDNF、TNF-α 和 IL-18 的水平，作为神经功能和炎症的标志物，以及作为氧化蛋白损伤指标的羰基蛋白水平（ 图 5（b）–（e））。⁴⁵