Oral 3‐hydroxybuturate ingestion acutely lowers circulating testosterone concentrations in healthy young males

Ketone bodies, such as 3‐hydroxybutyrate (3‐OHB), have been frequently used by endurance athletes, such as cyclists, to enhance performance and recovery and are recognized for their health benefits and therapeutic effects for decades. Testosterone is a potent regulator of red blood cell production. Evidence suggests that ketone bodies can increase the production of erythropoietin, which stimulates red blood cell production. Therefore, we investigated whether an acute increase in 3‐OHB levels affects testosterone levels in healthy young men. We studied six healthy, young male participants who fasted overnight and were tested twice: (i) after drinking 37.5 g of Na‐D/L‐3‐OHB dissolved in 500 mL of distilled water (KET), and (ii) after drinking 500 mL of placebo saline water (0.9% NaCl) (CTR). During the KET trial, 3‐OHB levels increased to approximately 2.5 mM. Testosterone levels decreased significantly by 20% during KET compared to 3% during CTR. A simultaneous increase in luteinizing hormone was observed in KET. We observed no changes in other adrenal androgens, such as androstenedione and 11‐keto androgens. In conclusion, an acute increase in 3‐OHB levels decreases testosterone levels. Concomitantly, an increase in luteinizing hormone was observed. This suggests that 3‐OHB may counteract some of the beneficial effects of endurance training. Further studies, involving larger sample sizes and performance outcomes, are required to fully understand this phenomenon.

trained athletes, demonstrating an improvement in cycling distance during a 30-min time trial and suggesting that 3-OHB may be a superior fuel to carbohydrates. 3 Over the past decades, ketone bodies, in particular 3-OHB, have demonstrated therapeutic benefits for various disease conditions, such as heart failure and dementia. [4][5][6] Due to its unique properties as a potent metabolic fuel and signaling molecule, 3-OHB is considered a metabolite with significant clinical implications. 1 In addition to being a substrate for oxidation, ketone bodies possess signaling properties and 3-OHB acts as a signaling molecule through the hydroxycarboxylic receptor 2 (HCAR2). 7 In rodents, signaling capabilities of 3-OHB have been shown in the pituitary gland. 8 We have recently demonstrated that an acute infusion of 3-OHB in healthy individuals increased erythropoietin production, which could potentially explain some of the reported benefits for endurance training. 9,10 This was supported by another recent study showing increased postexercise erythropoietin concentrations after 3-OHB monoester ingestion. 11 Testosterone is crucial for erythropoiesis, hemoglobin concentration, and hematocrit. 12 Testosterone is also a well-documented stimulator of muscle protein synthesis with an anabolic effect on muscle mass. 13 Testosterone synthesis is a complex process involving multiple steps from cholesterol, through androstenedione, to the functional steroid hormone testosterone. 14 Androstenedione can also undergo irreversible conversion to 11-βhydroxyandrostenedione (11OHA4), which further equilibrates through 11-ketoandrostenedione (11KA4) to 11-ketotestosterone (11KT). The keto-testosterone intermediate, 11KT binds to and activates the androgen receptor as do testosterone. 14 Another way to increase 11KT levels is the irreversible conversion of testosterone to 11-βhydroxytestosterone (11OHT) and then to 11KT, both of which are catalyzed by CYP11B1. 14 The intersection fields in the pathways are rather unexplored between testosterone, keto-testosterone intermediates and 3-OHB.
Therefore, we investigated the potential effects of an acute oral 3-OHB dose on testosterone and ketotestosterone intermediates to determine whether 3-OHB regulates testosterone, a potent inducer of erythropoiesis and performance.

| Study design
We utilized blood samples collected from a previously published randomized, controlled, crossover study involving six healthy young men 15 The participants underwent two separate sessions in randomized order (www.rando mizer.org) after an overnight fast, where they consumed either (i) 3-OHB (KET) or (ii) saline (CTR). Blood samples were drawn at baseline and at fixed time points throughout the 270-min study duration. During the KET session, participants consumed a beverage consisting of 36.5 g Na--D/L-3-OHB (Gold Biotechnology) dissolved in 0.5 L distilled water, while during the CTR session, they consumed an equal volume of an unflavored and salty placebo beverage containing saline (NaCl 0.9%) at t = 120 min. The study sessions commenced at 08:00 a.m. and were separated by a washout period of at least 2 weeks.

| Blood sampling
During the study, participants remained in a supine position and blood samples were collected hourly, centrifuged, and stored at a temperature of −80°C until analysis. The analysis of all blood samples was conducted at Aarhus University Hospital's Department of Clinical Biochemistry, which is accredited by the Danish Accreditation Fund (DANAK).

| Ethics
The study was conducted after approval by the local ethics committee (journal nr. 1-10-72-157-16) and registered at www.clini caltr ials.gov (ID number: NCT02917252) before enrolment. All participants gave oral and written informed consent prior to participation.

| Statistical analysis
The normality of the data was confirmed by examining the QQ-plots. Unless otherwise stated, the results are reported as means ±SE of the mean (SEM). Paired t-tests and two-way repeated measure ANOVA analysis were used to examine statistical associations between the two trial days when appropriate. A p value <0.05 was deemed statistically significant. No power calculation was performed specifically for the blood parameters included in the present manuscript as this was a post hoc analysis. Graphs and statistical analyses were performed using SigmaPlot 14.

| RESULTS
As previously reported, the six participants had a median age of 23 (range 21-29) years and a mean BMI of 24.5 ± 1.9 kg/m 2 . 15 All six participants completed both study days and were included in all analyses. Oral 3-OHB ingestion (KET) significantly increased plasma 3-OHB concentrations to 2.5 mmoL/L and remained around 0.1 mmoL/L during placebo (CTR). 15 In this human clinical study conducted on healthy, regularly trained males, we observed an acute and significant reduction in testosterone levels following 3-OHB ingestion. Testosterone dropped 20% during KET and 3% during CTR, time × intervention, p < 0.001 ( Figure 1). We noted a trend toward an increase in LH levels that did not reach statistical significance (time × intervention, p = 0.06; Figure 1). Neither FSH nor SHBG showed any changes during the study period or intervention ( Figure 1).
We further examined the testosterone precursors and did not find any differences in androstenedione levels whereas we observed a slight decrease in DHEAS in the KET period ( Figure 2).
Lastly, we investigated keto-testosterone intermediates and were unable to detect changes in our material for 11-KA4, 11-OHA, 11-KT, and 11-OHT ( Figure 2). However, they all decreased throughout both study days ( Figure 2). For individually plotted data, see Figures S3 and S4.

| DISCUSSION
In this clinical study involving healthy young men, we observed a rapid decline in testosterone levels after the ingestion of 3-OHB salt compound orally. We observed a trend toward an increase in LH levels, which may be a response to negative feedback from the pituitary gland. Contrary to our primary hypothesis, our findings suggest that testosterone does not play a significant role in the increase of erythropoietin levels observed after the administration of exogenous 3-OHB and our original hypothesis is thereby rejected.

| Testosterone
In healthy adult males, testosterone levels tend to decrease in response to calorie intake following an overnight fast, without any notable changes in SHBG or LH. 16,17 Similarly, adolescent males in pre-early puberty and mid-late puberty experience a reduction in testosterone levels after consuming a meal containing glucose and protein, which is also associated with a decrease in LH. 18,19 A recent meta-analysis of ketogenic diets used in obese patients suggested, that testosterone levels may increase during ketogenic diets. The study included 111 patients, and no other hormone levels from the hypothalamic-pituitary-gonadal axis were reported. 20 That increase in testosterone was strongly correlated to weight loss. 20 A study examining the effects of a long-term ketogenic diet consisting of low carbohydrate and high lipid intake found an increase in total testosterone concentrations after 11 weeks, while free testosterone concentrations remained unchanged. Hemoglobin levels were also found to be unaffected. 21 The authors did not report SHBG, LH and FSH.
There has been sparse literature characterizing circadian fluctuation in male testosterone concentrations. One study from WJ Bremner and colleagues reported in 1983 a decline in testosterone from the early morning hours onwards at some 2.5 ng/mL corresponding to some 8.5 nmoL/L (28.8%) measured by radio immune assay. 22 Our group have more recently published data on testosterone levels and fluctuations in the early morning hours demonstrating a decline of some 2 nmoL/L (8.7%) measured by LCMS-MS. 23 In the present study, we used the same method (LCMS-MS) and report a slight circadian decline around 3% which do not explain the 20% decline in testosterone reported in the intervention group. We do also believe that circadian variation would affect the two groups equally. From our investigation, we report a decline in the levels of testosterone after the consumption of the 3-OHB that contains calories, together with LH alterations, while the testosterone and LH levels remained unchanged in the placebo group. Databases of RNA and protein presence in different tissues confirm the presence of HCAR2 RNA and protein in male gonadal tissue, including Leydig cells, suggesting the possibility of a direct signaling effect from 3-OHB after ingestion. 24 From the study by Fu SP and colleagues, a signaling effect on the pituitary gland initially inhibiting LH cannot be ruled out either. 8

| Exercise
Even slight improvements in performance can be highly beneficial for endurance athletes such as cyclists. However, findings regarding 3-OHB and exercise performance are somewhat divergent. In trained cycling athletes, 3-OHB has been demonstrated as an optimal body fuel source, leading to enhanced performance. 3 In another human trial, the effects of 3-OHB (AcAc diester-compound) ingestion were tested on cyclist performance in an "all-out" time trial cycling setup, and a significant reduction in performance was found for the AcAc-diester compared to diet coke intake. 25 It is important to note that the cyclists experienced severe gastrointestinal issues after consuming the AcAc-diester compound used in the study, which may have affected the results. 26 Further studies have shown improved fat oxidation during steady-state exercise, while others have demonstrated impaired performance in shortterm, high-intensity training when consuming 3-OHB (salt). 27 Collectively, evidence could therefore indicate that supplementation of 3-OHB may have metabolic benefits during submaximal intensity exercise, whereas the acute ergogenic effect when exercising at high intensity is still questionable. 28 Improved recovery between training sessions and competitions is another essential key to achieve success in sport. In that perspective, it has reported that supplementation with 3-OHB (mono-ester compound) in the recovery phase right after exercise stimulates muscle protein synthesis in vivo as well as in vitro 29 and testosterone availability is important for this muscle protein synthesis. 30 With regards to recovery, Holdsworth and colleagues demonstrated higher glycogen synthesis in the presence of 3-OHB (mono-ester) during the recovery period. 31 It is worth noting that different 3-OHB compounds are used as mentioned above, and the effects of 3-OHB on acute performance may not solely depend on the increase in 3-OHB levels but could also be influenced by the specific compound (salt, monoester, diester) used and their possible side effects, which, in turn, may alter performance. We did not monitor performance in our study and choose 3-OHB salt as we have the most experience using that compound.

| Erythropoiesis
In a long-term perspective, erythropoiesis does play an essential role for endurance performance as hemoglobin transports oxygen to the working muscle. 28 In previous studies, it has been shown that erythropoiesis plays an essential role in both performance and training. Our research group demonstrated an increase in erythropoietin following an acute intravenous 3-OHB (salt) infusion, which could potentially raise hemoglobin and hematocrit levels if erythropoietin remained elevated over an extended period. 9 However, a 28-day daily ingestion of 3-OHB (mono-ester) did not result in any significant changes in hemoglobin in healthy individuals. 32 In the current study, we observed a significant 20% reduction in circulating testosterone levels after a single oral ingestion of 3-OHB (salt), with a subsequent increase in LH. As testosterone is known to stimulate erythropoietin 33 and low testosterone is associated with a higher risk of low hemoglobin, the final product of erythropoiesis, 34 it is reasonable to assume that a different mechanism is involved in the relationship between 3-OHB, testosterone, and erythropoietin. A direct signaling effect on the male gonadal glands via the HCAR2 cannot be ruled out, particularly given the presence of HCAR2 receptors in the testes. 24 In this context, we have observed that individuals with common gene variants in the SCOT gene, who are exposed to lifelong elevated 3-OHB levels, experience decreased hemoglobin and hematocrit levels (unpublished data).

| DHEAS and keto-testosterone
DHEAS is a byproduct of testosterone formation derived from sulphation of dehydroepiandrosterone (DHEA). 14 Our study showed decreased levels of DHEAS during KET. However, the very small difference of 0.2 umol/l barely plays a clinical role. In a 12-week intervention study involving women with polycystic ovary syndrome, a ketogenic diet led to decreased levels of both testosterone and DHEAS, along with a moderate amount of weight loss. 35 The keto-testosterones including 11-KA4, 11-OHA, 11-KT, and 11-OHT did not exhibit any changes in our study but declining over time during both study days. It is plausible that the duration of our study, limited to 150 min after the ingestion of the drink, might be inadequate to detect changes in keto-testosterones. Additionally, the production of keto-testosterones heavily relies on the availability of androstenedione, which could be compromised in our study due to the lower levels of testosterone. 14 All blood samples were drawn and stored at -80°C until analysis 5 years later. The samples were not dried during storage. The storage time until analysis has previously been shown not to affect hormone analysis of broad variance for up to 22 years of storage. 36 Therefore, we believe that the measurements in the present study are valid.
The finding that 3-OHB may influence testosterone is novel and may affect the way 3-OHB is used in training and sports such as cycling. As mentioned above, testosterone is important for muscle synthesis 30 and a declining testosterone may counteract some of the positive muscle related aspects of 3-OHB ingestion such as improved glycogen synthesis. 31 As this study is a very shortterm study, longer trials are needed to confirm whether testosterone remains affected over time or if it is only a pulsatile phenomenon. Though, the single dose of 3-OHB we used probably mimics how 3-OHB is used in sports quite well.
In conclusion, a single dose of 3-OHB acutely decreases testosterone more than would be expected from circadian fluctuation and does therefore not seem to play a role in the elevated EPO levels detected following 3-OHB administrations. The lower testosterone levels were associated with a dynamic LH response that shortly dropped and then increased back to baseline levels in the hours following 3-OHB consumption. The effects on physical performance and long-term effects on other health perspectives await future clinical trials.

| Perspectives
In this study, we demonstrate that the consumption of a single 3-OHB (salt) drink leads to a greater reduction in testosterone levels in healthy young males than what would typically occur due to circadian fluctuations. This finding suggests a potential drawback in the use of 3-OHB in sports and nutrition, considering that testosterone plays a crucial role in skeletal muscle development and overall performance, also in the acute setting. It is important to note that the effects of different compounds, such as 3-OHB salt/ester/diester, and 3-OHB concentrations may vary, and further research is required to fully understand their distinct actions.
Additionally, certain medical conditions, like polycystic ovarian syndrome characterized by excessive testosterone levels, might benefit from the testosterone-lowering effects of 3-OHB. Individuals with such conditions often experience metabolic dysfunction, including increased basal state lipolysis. There might be a dual benefit in reducing testosterone levels and lowering free fatty acid levels in this patient group.
Lastly, the precise mechanism underlying the decline in testosterone levels is not yet fully understood. It remains to be determined whether the signaling through HCAR2 primarily affects LH in the pituitary gland, subsequently impacting testosterone, or if it primarily influences testosterone directly in the gonadal glands. In either case, a similar hormonal reduction might occur for estradiol since the HCAR2 receptor is present in the female gonadal glands. Further investigations are needed to address these questions and shed light on the exact mechanisms involved.