Chee et al. (2021) have recently claimed that supplementing older male adults (70 years; n = 7, 3 of whom were on statins medication) with a daily carnitine and protein formulation that also contained 44 g of sugars in conjunction with twice-weekly exercise training sessions at 50% VO2max over 25 weeks would increase their muscle total carnitine stores. Subsequently, the higher muscle total carnitine levels were hypothesised to (1) improve whole-body insulin sensitivity and (2) increase whole-body fat oxidation during a moderate-intensity exercise undertaken after training. The reference was made against a similar age and medicated small group of male adults who received a similar drink formulation (but without carnitine) and exercise training.

The authors acknowledged there was no basis for upholding the claim that carnitine supplementation would improve resting insulin-stimulated whole-body or skeletal muscle glucose disposal. Furthermore, the authors could not show any significant mean group differences in energy expenditure, nor in the rates of plasma appearance or disappearance during two supposedly identical exercise tests—1 h of exercise at 50% VO2max—undertaken before and after carnitine/placebo supplementation. Conversely, the authors upheld the claim that carnitine supplementation in older male adults would increase whole-body fat oxidation, predominantly in the form of intermyofibrillar lipids (IMCL). Nevertheless, the evidence called in to support the latter claim does not stand up to detailed scrutiny, is circumstantial at best or is missing, and may convey an unsubstantiated message to the public. In fact, the personal interpretation of the present commentator, which will be detailed later, is that carnitine supplementation in conjunction with bi-weekly training sessions for 25 weeks increased CHO oxidation rather than that of IMCL.

Let us assume for a moment that the authors’ claim that a 20% rise in fat/IMCL oxidation would occur with carnitine supplementation and training is correct. Then, an extra ~40 (220–180) J/kg lean body released from fat/IMCL would have contributed to the energy expenditure in the carnitine treated group during the 60 min exercise test at 50%VO2max. Also, assuming a total lean mass of 50 kg (table 1; Chee et al., 2021) and that 1 g of triacylglycerol generated 39.4 kJ through oxidation (authors’ conversion factor), then an additional (40 × 60 × 50)/39,400 or 3 g of fat/IMCL would have been burnt during each exercise session. However, an increase in fat oxidation in the treated group should have occurred earlier rather than exclusively during the exercise test undertaken at the end of the training. If we assume generously that the additional fat oxidation with carnitine loading started from the first week of training, then a total of 150 g of fat/IMCL (3 g fat × 2 sessions per week × 25 weeks) would have been oxidised over 25 weeks (or <1 g fat/daily on average). In line with these calculations, the data displayed in table 1 and figure 5d (Chee et al., 2021) show no change in any regional fat content across all subjects irrespective of group or time. Equally, this minute amount of fat, which could have certainly not been captured by a DEXA scan, would have also been easily masked by the effects on the whole-body composition by the additional 44 g of sugars that all subjects had to ingest daily for 25 weeks. Overall, the claim that total carnitine would increase fat oxidation by 20%, predominantly in muscle IMCL, during the exercise test would have been insignificant when translated to an absolute value. It is also worth remembering that the reported increase in fat oxidation during the exercise test was derived from data recorded from a male cohort where six out of fourteen were on statins medication, a drug well-known to interfere with whole-body fat handling.

The males in the treated group appeared to store primarily 22% more muscle total carnitine than in the control group, even before supplementation (figure 1a; 3rd vs 1st column; Chee et al., 2021). However, a control male with the lowest muscle total carnitine content (10 mmol/kg dm) of all males enrolled in the study may have contributed to the sizable difference between the groups at the baseline. At the end of the training, it was equally unexpected to notice a marked decline in muscle total carnitine once again in another control male (figure 1a, left panel; Chee et al., 2021). Given the small number of males in each group, these two control males were, therefore, most likely to have acted as leverage points to biasedly increase the mean difference between the treated group and control at 25 weeks, thereby raising the chance of declaring a false-positive finding (figure 1b; Chee et al., 2021).

The authors also state that the values of the carnitine forms reported in the study cover the main three forms of carnitine: free, short- and long-chain acylcarnitine. However, as figure 1 (Chee et al., 2021) demonstrates, the reported values in the published paper represent free and acetylcarnitine values only. Therefore, the contribution of the long-chain acylcarnitine form, which is not trivial, is missing.

From a biochemical perspective, the present assertion that carnitine supplementation could have contributed to an increase in fat/IMCL also seems to be in collusion with in vitro experiments, which have reported that the Michaelis-Menten constant (K_m) of CPT1—reputed to be the main limiting step in fat oxidation—for free carnitine is 0.4 mmol l⁻¹ (McGarry et al., 1983). Since the lowest muscle-free carnitine content reported in the present study (~10 mmol kg⁻¹ dm or ~4 mmol l⁻¹ intracellular water; values computed from figure 1b; Chee et al., 2021) was well above (~10-fold) the reported K_m value for free carnitine, this would have made the muscle CPT1 enzyme kinetics independent of that of carnitine concentration. Therefore, any further increases in muscle-free carnitine would have been irrelevant to achieving the maximum capacity of muscle to oxidise fat.

The authors seemed to consistently and unjustifiable assign changes in muscle total carnitine, rather than free carnitine, the leading role to support their upheld hypothesis. This consideration is unwarranted since it is the free carnitine form that facilitates the transport of long-chain acylcarnitines across the mitochondrial membrane. Equally unjustifiable was conveying the belief that the calculated increase in whole-body fat oxidation could have been accounted for by a rise in muscle IMCL oxidation. It was not surprising that the muscle-free carnitine mean differences among groups or the main effect of free carnitine on muscle IMCL use were not statistically different (figure 1 and table 2, respectively; Chee et al., 2021), and the amount of additional IMCL burnt during the exercise test post-carnitine supplementation was indeed so negligible.

It is interesting to note that obese people (Harper et al., 1995) and chronic high fat dietary intake (Constantin-Teodosiu et al., 2019), conditions that are associated with an increase in circulating insulin levels to make up for an increase in insulin resistance, are inherently associated with higher total muscle and liver carnitine levels compared with subjects who are leaner. However, it remains to be established whether the natural rise in muscle-free carnitine stores with increased fat availability is due to (i) a cellular response in the face of an increase in fat handling or (ii) due to the higher content in free carnitine by the dietary fat or in combination with the higher-than-normal insulin levels.

The rate of muscle acetylcarnitine accumulation is a well-accepted marker of the flux through pyruvate dehydrogenase complex (PDC) reaction, the enzyme that limits the rate of carbohydrate oxidation (Constantin-Teodosiu et al., 1991). The most remarkable change in the pre–post training muscle acetylcarnitine accumulation was recorded in the treated group (figure 1b, right panel; Chee et al., 2021). Additionally, the lowest difference in the pre–post training muscle lactate accumulation, a well-accepted anaerobic metabolism marker, was also recorded in the treated group (figure 3c; Chee et al., 2021). In line with the latter observation, the lowest blood plasma levels, a marker of the amount of lactate produced and released by contracting muscle, were recorded in the treated group (figure 4b; Chee et al., 2021). Collectively, these data would suggest that the flux through PDC reaction in the treated group post-training was the highest of all conditions.

All in all, the results of three straight biochemical measurements (increased rates of muscle acetylcarnitine accumulation matched by reduced rates of muscle lactate accumulation and circulating blood levels) along with a significant increase in the respiratory exchange ratio during the hyperinsulinaemic-euglycemic clamp (RER; figure 5c; Chee et al., 2021) point to an increase in the oxidative CHO contribution, rather than that of fat, to the total energy expenditure in the carnitine treated group with training. Despite this firm evidence, along with the finding that energy expenditure during exercise was similar across treated/control groups (figure 2a; Chee et al., 2021), the authors claimed that more fat/IMCL was oxidised in the treated group than in the control during the exercise test. This assertion/calculation relied exclusively on the results of a multi-step convoluted gas chromatography mass-spectrometry (GA-MS) method.

The post-/pre-training fold changes in gene expression of selected muscle transcripts involved in fatty oxidation (ACAT1) and IMCL turnover (DGKD and PLIN2) were significantly greater in the carnitine group than in the control group (table 3; Chee et al., 2021). However, the reported differences, albeit significantly different, were vastly low to assign them a physiological significance. Interestingly, the muscle-specific CPT1B isoform gene expression in the control group was higher than in the carnitine group (table 3; Chee et al., 2021). This finding would advocate for better FFAs handling in control males than in the carnitine treated males. In addition, this observation would be another critical point at undoing the authors’ argument that only carnitine supplementation, rather than training per se, could increase the mitochondrial fatty acid oxidation turnover. This seems to have been overlooked by the authors.

We critically examined here Chee et al.'s claims that by supplementing daily a small group of older male adults with a carnitine protein formulation, in conjunction with twice-weekly exercise training sessions over 25 weeks, muscle total carnitine stores would increase and would thereby improve whole-body insulin sensitivity and increase whole-body fat (in the form of IMCL) use/oxidation during a subsequent bout of moderate exercise. While the authors correctly rejected the first claim, the present commentator could not find grounds based upon the papers’ results that carnitine supplementation truly increased either (1) muscle-free carnitine availability or (2) muscle IMCL oxidation during a subsequent moderate-intensity exercise in older male adults. Thus, the doubts of the commentator are based on the earlier presented calculations and are against the authors’ persistent perspective that total carnitine had the role in the marginal increase in fat oxidation, which was unreasonably accounted for by the rise in muscle IMCL use.

In conclusion, the data reported by Chee et al. are very debatable, and the ‘conclusions’ should be carefully viewed/interpreted considering the shortcomings presently addressed. Before exploring alternative carnitine loading and exercise strategies in older male adults, rather than in the general population, as the authors alluded to, the public would need to be reassured about the validity of the effects of carnitine supplementation on muscle fat oxidation during exercise in older male adults.

奇等人。( 2021 ) 最近声称，与每周两次的运动训练课程相结合，每天补充含有 44 克糖的肉碱和蛋白质配方的老年男性成人（70 岁；n = 7，其中 3 人服用他汀类药物）在 50% VO2max 超过 25 周会增加他们的肌肉总肉碱储存。随后，假设较高的肌肉总肉碱水平 (1) 提高全身胰岛素敏感性和 (2) 在训练后进行的中等强度运动期间增加全身脂肪氧化。该参考是针对接受类似饮料配方（但不含肉碱）和运动训练的相似年龄和药物治疗的一小群男性成年人进行的。

作者承认，没有依据支持补充肉碱可以改善静息胰岛素刺激的全身或骨骼肌葡萄糖处理的说法。此外，在补充肉碱/安慰剂之前和之后进行的两个假设相同的运动测试（以 50% VO2max 进行 1 小时运动）期间，作者无法显示能量消耗的任何显着平均组差异，也没有显示血浆出现或消失率的显着差异。相反，作者支持老年男性补充肉碱会增加全身脂肪氧化的说法，主要以肌原纤维间脂质 (IMCL) 的形式存在。然而，支持后一种说法的证据经不起详细审查，充其量只是间接的或缺失的，并可能向公众传达未经证实的信息。事实上，本评论员的个人解释（将在后面详述）是，补充肉碱结合每两周一次的 25 周训练会增加 CHO 氧化，而不是 IMCL。

让我们暂时假设作者的说法是正确的，即补充肉碱和训练会使脂肪/IMCL 氧化增加 20%。然后，在 50%VO2max 的 60 分钟运动测试期间，从脂肪/IMCL 释放的额外约 40 (220–180) J/kg 瘦肉体将有助于肉碱治疗组的能量消耗。此外，假设总瘦体重为 50 kg（表 1；Chee 等人，2021) 并且 1 克甘油三酯通过氧化产生 39.4 kJ（作者的转换系数），然后在每次锻炼期间会额外燃烧 (40 × 60 × 50)/39,400 或 3 克脂肪/IMCL。然而，治疗组脂肪氧化的增加应该更早发生，而不是仅仅在训练结束时进行的运动测试期间发生。如果我们慷慨地假设额外的脂肪氧化从训练的第一周开始，那么总共 150 克脂肪/IMCL（3 克脂肪 × 每周 2 次 × 25 周）将在 25 周内被氧化（或平均每天 <1 克脂肪）。根据这些计算，表 1 和图 5d 中显示的数据（Chee 等人，2021) 无论组别或时间如何，所有受试者的任何区域脂肪含量都没有变化。同样，这一微量的脂肪肯定不会被 DEXA 扫描捕获，也很容易被所有受试者每天必须摄入的额外 44 克糖对全身成分的影响所掩盖。 25 周。总体而言，在运动测试期间，总肉碱会使脂肪氧化增加 20%，主要是肌肉 IMCL，当转化为绝对值时，这一说法是微不足道的。还值得记住的是，所报告的运动测试期间脂肪氧化的增加来自男性队列记录的数据，其中 14 人中有 6 人服用他汀类药物，这是一种众所周知的干扰全身脂肪处理的药物。

即使在补充之前，治疗组中的男性似乎比对照组多储存了 22% 的肌肉总肉碱（图 1a；第 3 列与第 1 列；Chee 等人，2021 年）。然而，在参与研究的所有男性中，肌肉总肉碱含量最低 (10 mmol/kg dm) 的对照男性可能导致了基线时各组之间的巨大差异。在训练结束时，同样出乎意料的是，另一名对照男性的肌肉总肉碱再次显着下降（图 1a，左图；Chee 等人，2021 年））。鉴于每组中的雄性数量较少，因此，这两个对照雄性最有可能充当杠杆点，在 25 周时有偏见地增加了治疗组和对照组之间的平均差异，从而增加了宣布错误的机会- 阳性结果（图 1b；Chee 等人，2021 年）。

作者还指出，研究中报告的肉碱形式的值涵盖了肉碱的三种主要形式：游离、短链和长链酰基肉碱。然而，如图 1（Chee 等人，2021 年）所示，已发表论文中报告的值仅代表游离和乙酰肉碱值。因此，缺少重要的长链酰基肉碱形式的贡献。

从生物化学角度看，本断言肉碱可能的增加脂肪贡献/ IMCL也似乎是串通的体外实验中，已经报道，米氏常数（ķ_米的）CPT1-被誉为是脂肪氧化的主要限制步骤——游离肉碱为 0.4 mmol l ^-1 (McGarry et al., 1983 )。由于本研究中报告的最低无肌肉肉碱含量（~10 mmol kg ^-1  dm 或~4 mmol l ^-1细胞内水；从图 1b 计算的值；Chee 等人，2021 年）远高于（~10 -fold) 报告的K _m游离肉碱的值，这将使肌肉 CPT1 酶动力学独立于肉碱浓度。因此，无肌肉肉碱的任何进一步增加都与实现肌肉氧化脂肪的最大能力无关。

作者似乎一致且不合理地认为肌肉总肉碱的变化，而不是游离肉碱的变化，这是支持他们所坚持假设的主导作用。这种考虑是没有根据的，因为它是促进长链酰基肉碱跨线粒体膜运输的游离肉碱形式。同样不合理的是传达了这样一种信念，即计算出的全身脂肪氧化的增加可能是由肌肉 IMCL 氧化的增加引起的。毫不奇怪，组间游离肉碱的平均差异或游离肉碱对肌肉 IMCL 使用的主要影响没有统计学差异（分别为图 1 和表 2；Chee 等人，2021 年）)，并且在补充肉碱后的运动测试期间燃烧的额外 IMCL 的量确实可以忽略不计。

有趣的是，肥胖人群（Harper 等人，1995 年）和长期高脂肪饮食摄入量（Constantin-Teodosiu 等人，2019 年），这些情况与循环胰岛素水平增加相关，以弥补增加与较瘦的受试者相比，胰岛素抵抗与较高的总肌肉和肝脏肉碱水平相关。然而，随着脂肪可用性的增加，无肌肉肉碱储存的自然增加是否是由于（i）面对脂肪处理增加时的细胞反应，或（ii）由于游离肉碱含量较高，仍有待确定。肉碱通过膳食脂肪或与高于正常水平的胰岛素水平相结合。

肌肉乙酰肉碱积累的速率是通过丙酮酸脱氢酶复合物 (PDC) 反应的通量的公认标志，该酶限制碳水化合物氧化的速率（Constantin-Teodosiu 等，1991）。治疗组记录了训练前后肌肉乙酰肉碱积累的最显着变化（图 1b，右图；Chee 等人，2021 年）。此外，在治疗组中也记录到训练前后肌肉乳酸积累的最低差异，这是一种广为接受的无氧代谢标志物（图 3c；Chee 等人，2021 年））。与后一种观察结果一致，治疗组记录了最低血浆水平，这是肌肉收缩产生和释放的乳酸量的标志（图 4b；Chee 等人，2021 年）。总的来说，这些数据表明训练后处理组中通过 PDC 反应的通量是所有条件中最高的。

总而言之，三项直接生化测量的结果（肌肉乙酰肉碱积累率的增加与肌肉乳酸积累率和循环血液水平的降低相匹配）以及高胰岛素 - 正常血糖钳夹（RER）期间呼吸交换率的显着增加；图 5c；Chee 等人，2021 年）指出，在接受训练的肉碱治疗组中，氧化性 CHO 对总能量消耗的贡献增加，而不是脂肪的贡献增加。尽管有这个确凿的证据，并且发现运动期间的能量消耗在治疗组/对照组之间相似（图 2a；Chee 等人，2021 年）)，作者声称在运动试验期间，治疗组比对照组氧化了更多的脂肪/IMCL。这种断言/计算完全依赖于多步复杂气相色谱质谱 (GA-MS) 方法的结果。

与对照组相比，肉碱组中涉及脂肪氧化 (ACAT1) 和 IMCL 转换（DGKD 和 PLIN2）的选定肌肉转录本的基因表达的训练后/训练前倍数变化显着大于对照组（表 3；Chee 等人., 2021 年）。然而，报告的差异，尽管有显着差异，但非常低，无法将其分配为生理意义。有趣的是，对照组的肌肉特异性 CPT1B 亚型基因表达高于肉碱组（表 3；Chee 等人，2021）。这一发现将提倡在对照雄性中比在肉碱处理的雄性中更好地处理 FFA。此外，这一观察结果将是推翻作者的论点的另一个关键点，即只有补充肉碱，而不是训练本身，才能增加线粒体脂肪酸氧化周转。这似乎被作者忽略了。

我们在这里批判性地研究了 Chee 等人的说法，即通过每天向一小群老年男性补充肉碱蛋白配方，结合 25 周内每周两次的运动训练课程，肌肉总肉碱储存量会增加，从而在随后的适度运动中提高全身胰岛素敏感性并增加全身脂肪（以 IMCL 的形式）使用/氧化。虽然作者正确地拒绝了第一个主张，但本评论员无法根据论文的结果找到理由，即在随后的中等强度运动中补充肉碱确实增加了 (1) 无肌肉肉碱的可用性或 (2) 肌肉 IMCL 氧化在老年男性中。因此，

总之，Chee 等人 报告的数据。非常值得商榷，考虑到目前解决的缺点，应该仔细查看/解释“结论”。在探索老年男性的替代肉碱负荷和运动策略之前，而不是在一般人群中，正如作者所暗示的那样，公众需要确信补充肉碱对老年人运动期间肌肉脂肪氧化的影响的有效性。成年男性。