Metabolism and Disposition of 2-Hydroxy-4-Methoxybenzophenone, a Sunscreen Ingredient, in Harlan Sprague Dawley Rats and B6C3F1/N Mice; a Species and Route Comparison

Abstract

1. 2-Hydroxy-4-methoxybenzophenone (HMB) is a common ingredient in personal care products and used as an UV stabilizer. In these studies, disposition and metabolism of [¹⁴C]HMB in rats and mice was assessed following single gavage administration (10, 100, or 500 mg/kg), single IV administration (10 mg/kg), or dermal application (0.1, 1, 10, or 15 mg/kg).

2. Following gavage administration, [¹⁴C]HMB was well absorbed and excreted mainly in urine (39–57%) and feces (24–42%) with no apparent difference between doses, species or sexes. Distribution of HMB in tissues was minimal in rats (0.36%) and mice (<0.55%).

3. Distribution of HMB following dermal application was comparable to that following gavage administration; no differences between doses, sexes, or species were observed but absorption varied between dose vehicles. Light paraffin oil had the highest absorption and excretion (98% of the HMB dose absorbed).

4. In rats, HMB slowly appeared in the systemic circulation (T_max ~2–6 h) and had poor bioavailability (F% < 1).

5. Urine metabolites for both species and all routes included HMB, HMB-glucuronide, 2,4-dihydroxybenzophenone (DHB), DHB-glucuronide, and DHB-sulfates, and novel minor dihydroxy metabolites including 2,5-dihydroxy-4-methoxybenzophenone.

6. In vitro hepatic metabolism in mice differed from human and in vivo metabolism especially for phase II conjugates.

1. Introduction

2-Hydroxy-4-methoxybenzophenone (HMB), also known as benzophenone-3 or oxybenzone, is used as an ultraviolet (UV) filter in personal care and sunscreen products. Sunscreen products in the USA are regulated as over-the-counter drugs and as an active sunscreen ingredient, the US FDA allows HMB concentrations up to 6% (U.S. FDA, 2018). HMB is also used in plastics as a stabilizer; Approximately 1 million pounds of HMB are manufactured or imported per year into the USA (NTP, 2006). It was reported that approximately 97% of the US general population (> 6 years of age) is exposed to HMB (CDC, 2012; Cottart et al., 2010). HMB has been detected in human urine at concentrations up to 18 ng/mL as unmetabolized HMB and up to 10.0 ng/mL for its metabolite, benzhydrol. (Calafat et al., 2008; ITO et al., 2009). HMB was detected in breast milk samples up to 121.40 ng/g (Schlumpf et al., 2010), and in placental tissue up to 9.8 ng/g (Vela-Soria et al., 2011). Various concentrations of HMB have also been detected in the environment (Ekpeghere et al., 2016; Kim and Choi, 2014) and biota (Balmer et al., 2005; Díaz-Cruz and Barceló, 2009; Fent et al., 2008).

Acute toxicity of HMB has been reported to be low with the oral LD₅₀ > 12.8 g/kg in rats (Lewerenz et al., 1972a) and the dermal LD₅₀ > 16.0 g/kg in rabbits (Cosmetic Ingredient Review, 1983). Wistar rats provided a diet containing 0.5 and 1.0% HMB via feed for 13 weeks displayed depressed growth, leukocytosis, anemia, reduced organ weights, and renal toxicity (Lewerenz et al., 1972b). HMB was a nonirritant to skin at concentrations from 4 to 100% and demonstrated little mutagenic activity in albino rabbits (NTP, 1992). F344/N rats and B6C3F1 mice administered HMB for 2- and 13-weeks via feed (3125–50,000 ppm) or topical application (1.25–20 mg in acetone or lotion) had increased liver and kidney weights; at the highest oral dose, there was a decrease in sperm density (NTP, 1992). In a reproductive assessment by continuous breeding study in CD-1 mice, dietary administration up to 50000 pm HMB in feed was associated with lower dam and litter weights, and poor pup survival. These findings were collectively attributed to HMB-induced systemic toxicity (NTP, 1992). The reproductive and developmental toxicity associated with HMB exposure was believed to occur via endocrine disruption (Bae et al., 2016; French, 1992; Schlecht et al., 2004; Schlumpf et al., 2001). HMB and/or metabolites have been reported to interact with steroid hormone receptors (Fent et al., 2008; Heneweer et al., 2005; Schlecht et al., 2004; Sieratowicz et al., 2011; Watanabe et al., 2015) and HMB exposure has been associated with estrogen-dependent diseases and probable birth outcomes in humans (Kunisue et al., 2012; Philippat et al., 2012; Wolff et al., 2008).

Absorption, distribution, metabolism, and excretion (ADME) and toxicokinetic (TK) studies are crucial in the interpretation of toxicology data. While there are some ADME data for HMB available in the literature for rats (male Sprague Dawley (SD), F344/N) and mice (male B6C3F1) there are neither disposition nor metabolism data in female rats or female mice. HMB is rapidly absorbed regardless of dose route in male rats and piglets (El Dareer et al., 1986; Jeon et al., 2008; Kadry et al., 1995; Kasichayanula et al., 2007; Okereke et al., 1993). Limited studies following oral (3 to 293 mg/kg) and IV (4.63 mg/kg) administration of [¹⁴C]HMB to male F344/N rats showed excretion of 64–67% and 23–42% in urine and feces, respectively within 72 h. (El Dareer et al., 1986). After dermal application to F344/N rats, approximately 39% and 22% of the applied dose was excreted in urine and feces (El Dareer et al., 1986). In the same study, approximately 1% of the oral and IV doses remained in all surveyed tissues and in gastrointestinal (GI) contents at 72 h, with extensive excretion of radioactivity in bile (37% within 4 h) following IV administration (4.46 mg/kg). In another study, following oral administration of HMB (100 mg/kg) in SD rats, the liver contained the highest amount of free and total HMB, followed by kidney, and testes (Kadry et al., 1995).

HMB has been reported to be metabolized in vivo and in vitro to both Phase I and Phase II metabolites including: 2,4-dihydroxybenzophenone (DHB), 2,3,4-trihydroxybenzohenone (THB), 2,2’-dihydroxy-4-methoxybenzophenone (2,2’-DHMB), glucuronide conjugates of HMB and DHB and a sulfate conjugate of a mono-hydroxylated HMB (El Dareer et al., 1986; Fediuk et al., 2012; Jeon et al., 2008; Kasichayanula et al., 2005; Nakagawa and Suzuki, 2002; Okereke et al., 1993). Oral administration of HMB (100 mg/kg) to male SD rats and male B6C3F1 mice showed species-dependent metabolism (Kadry et al., 1995; Okereke, C. S., 1994). It was reported that the absorption in mice (0.36 h) was faster than rats (0.71 h). The plasma concentration versus time profiles were by a one-compartmental model for mice and two-compartmental model for rats, respectively. DHB was the major metabolite detected only in rat plasma, while trace amounts of THB and DHMB were found in plasma of both species (Okereke, C. S., 1994). In the same study, DHB was the major metabolite in all examined tissues of the rat, and significantly higher accumulation of HMB and DHB in the liver, kidneys, intestine, and testes of the rat compared to the mice was reported. Following perinatal dietary exposure to HMB (0–30000 ppm) in Harlan Sprague Dawley (HSD) rats, free (unconjugated) HMB and DHB, and total (combined conjugated and unconjugated) HMB, DHB, THB, and 2,5-DHMB were reported in plasma (Mutlu et al., 2017).

Due to the potential for significant human exposure and the lack of adequate toxicity data, the National Toxicology Program (NTP) is evaluating the toxicity of HMB. In this study, we investigated the ADME of HMB in Harlan Sprague Dawley (HSD) rats and B6C3F1/N mice (the two rodent models used in NTP studies) following gavage administration and dermal application, the routes relevant to human exposures. The study design is given in Table 1. Male and female Harlan Sprague Dawley rats and B6C3F1/N mice were administered a single dose of [¹⁴C]HMB via gavage (10, 100, or 500 mg/kg), or dermally (0.1, 1, 10, or 15 mg/kg). Dermal dose was formulated in multiple vehicles including a lotion vehicle to mimic exposures in humans via sue of sunscreens and other consumers products. The doses selected were within 0.1 and 0.001 of oral LD₅₀ (> 12.8 g/kg) and dermal LD₅₀ (> 16.0 g/kg). In limited studies, plasma TK of HMB and bioavailability was determined following a single gavage administration. Limited studies were conducted following IV administration to interpret gavage and dermal data (Table 1). In addition, metabolism of HMB was evaluated in vitro utilizing rat, mouse, and human hepatocytes.

Table 1.

Study design of [¹⁴C]HMB in Harlan Sprague Dawley rats and B6C3F1/N mice.

Species	Route	Sex	Dose (mg/kg)	Study Duration (h)	End Point
Rat	Oral Gavage	male	10, 100, 500	24a, 72	Dose response
		male	100	2, 72	Tissue Distribution
		female	100	72	Sex Difference
		male/female	10	24	Toxicokinetic Study
	Dermal	male	10b, c, d, e,f	72	Route Difference
		male	0.1b	72	Tissue Distribution
		female	1c, 15c	72	Sex difference
	IV	male	10	72	Route Difference
		male	10	6	Excretion in Bile
		male/female	10	24	Toxicokinetic Study
Mouse	Oral Gavage	male	100, 500	72	Dose Response
		female	100	72	Sex Difference
	Dermal	male	1f, 10f	72	Dose Response
		male	10c, f, g	72	Route Difference
		female	10c, f	72	Sex difference
	IV	male	10	72	Route Difference

^aDisposition of radioactivity 24 h of following oral gavage administration of 10 mg/kg [¹⁴C]HMB in male HSD rats

^bLight Paraffin Oil,

^cLotion,

^dCoconut oil,

^e1:1 Ethanol:Coconut Oil,

^fEthanol,

^gAcetone

2. Materials and Methods

2.1. Chemicals and Reagents

HMB (CAS # 131-57-7, purity 98%, lot # 05928TH) was purchased from Sigma-Aldrich, Inc. (St. Louis, MO) and the identity was confirmed by nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry (MS). [¹⁴C]HMB (lot# 433106–0748-A-20090223-JK, specific activity 74.8 mCi/mmol, 1mCi/mL, label on the unsubstituted phenyl ring) was from Moravek Biochemicals, Inc. (Brea, CA). The radiochemical purity of [¹⁴C]HMB was determined to be 99.1% by high performance liquid chromatography (HPLC) and liquid scintillation spectrometry (LSS).

Ultima Gold scintillation cocktail was purchased from Packard Instrument Company, Inc. (Meriden, CT). Alkamuls EL-620 was purchased from Rhodia (Cranbury, NJ). Methanol was purchased from VWR International, LLC (Radnor, PA). DHB, 2,2′-DHMB, THB, β-Glucuronidase from E. coli and H. pomatia were purchased from Sigma-Aldrich Company, Inc (St. Louis, MO). 2,4- DHMB was custom synthesized by Richman Chemical Custom Solutions (Gwynedd, PA). Carbo-Sorb E and Permafluor E+ were purchased from Perkin Elmer (Shelton, CT). Heparin was from Abraxis Pharmaceutical Products (Schaumburg, IL) and Euthasol was from Delmarva Laboratories, Inc. (Midlothian, VA). Chlorhexidine 2% was purchased from VetOne (Meridian, ID). All other reagents were purchased from commercial sources.

2.2. HPLC with Radiochemical Detection

Agilent (Santa Clara, CA) 1100 HPLC was coupled to a β-RAM radiochemical detector (500-μL lithium glass, flow-through, solid flow cell detector, IN/US Systems; Tampa, FL) and a Varian Inertsil C8 column (5 μm, 250 × 4.6 mm) (Varian, Palo Alto, CA) at ambient temperature. Mobile phases were 0.1% TFA in water (Solvent A) and 0.1% TFA in acetonitrile (Solvent B).

2.2.1. Method A:

Solvents A and B were run at a flow rate of 1.5 mL/min using the following gradient: 0% B to 100% B in 30 min and held for 5 min.

2.2.2. Method B:

Solvents A and B were run at a flow rate of 1.0 mL/min using the following gradient: 0% B to 30% B in 5 min, then to 60% B in 30 min, then to 100% B in 5 min.

2.3. Liquid Chromatography-Mass Spectrometry (LC-MS)

Agilent 1200 HPLC was coupled to an API 5000 triple quadrupole MS (Applied Biosystems, Foster City, CA) or an API365 with EP10+ upgrade and hot-surface induced desolvation (HSID) interface was operated in positive ion mode using multiple reaction monitoring (MRM) or full scan (100 to1000 m/z) mode. Mobile phases were Solvent A, Solvent B, Solvent C (0.1% formic acid in water) and Solvent D (0.1% formic acid in acetonitrile).

2.3.1. Method C:

Solvents C and D were run at a flow rate of 1.0 mL/min on a Varian Inertsil C8 column using the following gradient: 0% B to 30% B in 5 min, then to 60% B in 30 min, then to 100% B in 5 min, then back to 0% B in 5 min at ambient temperature. The MS operating conditions for API365 and API500 were: CAD = 8 and 12, CUR = 12 and 25, Ion Spray = 5500 and 5500 V, Temperature = 300 and 250 °C, Declustering Potential = 21 and 80, Entrance Potential =10 and 10, Collision Energy = 25 and 50, Cell Exit Potential = 12 and 21, Dwell Time = 50 and 50 msec, respectively. Both instruments were operated in MRM mode.

2.3.2. Method D:

Solvents A and B were run at a flow rate of 0.8 mL/min on a Phenomenex Luna (3μm C18 column, 50 × 2.0 mm) using the following gradient: isocratic run at 60% B for 1 min, then from 60% B to 100% B in 2 min, held at 100% B for 1 min at 20 °C. The MS operating conditions were; HMB Q1 = 229.0 m/z, Q3 = 151.1 m/z, CAD = 12, CUR = 30, Gas 1 = 35, Gas 2 = 40, Ion Spray Voltage = 5500, Temperature = 450, Declustering Potential = 54, Entrance Potential = 14, Collision Energy = 27, Cell Exit Potential = 15; DHB (Internal Standard (IS)) Q1 =245.0 m/z, Q3 = 151.1 m/z, other parameters same as HMB above.

2.3.3. Method E:

The method is the same as Method C except that the MS (Applied Biosystems API365) was operated in full scan mode with a scan window of 100 to 1000 m/z and the operating conditions were; ion spray voltage, 5500 V; source temperature, 500 °C.

2.4. Metabolism of HMB in Hepatocytes

Cryopreserved male (Lot # A75257) and female (Lot # A75258) Sprague Dawley rat hepatocytes (harvested from rats obtained from Charles River, Raleigh, NC, ~8 weeks old) and fresh hepatocytes from male and female B6C3F1 mice (harvested from ~9 week-old mice obtained from Taconic, Germantown, NY) were purchased from Alive, LLC (Albuquerque, NM). Cryopreserved male (Lot # Hu4171) and female (Lot # Hu0617) human hepatocytes (pool of 10 donors), cryopreserved hepatocyte recovery media, cryopreserved hepatocyte maintenance supplement pack containing cell maintenance cocktail-B and 10 mM dexamethasone, and Williams medium E without phenol red were purchased from CellzDirect, Inc. (Austin, TX).

Hepatocytes were thawed and resuspended according to the suppliers’ protocol. Hepatocyte concentrations were determined by a Bright-line Hemacytometer (Reichert Technologies, Depew, NY). The final concentration of HMB was 1 μM (10 μCi, final concentration of 0.5% in dimethylsulfoxide) and a final cell concentration of 1 × 10⁶ cells/mL. To assess the metabolite formation, [¹⁴C]HMB (2 μM) in Williams medium E was incubated in triplicate with male and female cell suspensions (containing 1 × 10⁶ viable cells) and with heat-inactivated cells in duplicate (viability <1%) for each time point (15, 30, 60, 120, and 240 min). At each time point, aliquots of 100 μL were removed and added to 100 μL of ice-cold acetonitrile to stop reactions and lyse cells. The samples were frozen and kept on dry ice for a minimum of 10 min following collection. Next, samples were thawed and centrifuged in a refrigerated (4°C) microfuge (Beckman Coulter, Brea, CA) at 14,000 × g for 5 min. The supernatant was transferred to amber glass vials and stored at −80°C until analysis. Samples were analyzed by Method B and C.

2.5. Animals

The studies were conducted at Lovelace Biomedical and Environmental Institute and approved by their Institutional Animal Care and Use Committee. Animals were housed in facilities that are fully accredited by U.S. Department of Agriculture, Office of Laboratory Animal Welfare, and the AAALAC International. All animal studies were carried out in accordance with the “Guide for the Care and Use of Laboratory Animals” (National Research Council, 2011). Adult Harlan Sprague Dawley (HSD) male and female rats, including jugular vein, were purchased from Harlan Laboratories, Inc. (Dublin, VA). Bile duct cannulated adult HSD rats were also purchased for the toxicokinetic study. Adult B6C3F1/N male and female mice were purchased from the NTP colony at Taconic (Germantown, NY or New Hyde Park, NY). The animals were quarantined up to 14 days prior to use in the study and were randomized into dosing groups. Animals were 8–10 weeks old at the time of dosing.

Animals were acclimatized to metabolism cages for ~ 24 h prior to dosing. Jugular vein cannula and bile duct animals were acclimatized for 6 days prior to dosing. During acclimation and following dosing, animals were housed individually in all-glass metabolism chambers that provided for separate collection of urine, feces, CO₂, and expired volatile organic compounds (VOCs), with the exception of bile-duct and jugular vein cannulated animals. These animals were individually housed, and cages lined with paper until necropsy. A 12 h light/dark cycle was maintained during the quarantine and study periods. The room temperature was maintained at 18 to 26°C and relative humidity was maintained within 30–70%. Animals were provided irradiated certified NTP 2000 diet (Zeigler Brothers, Inc., Gardeners, PA) and municipal water (Kirtland Air Force Base, Albuquerque, NM) ad libitum. Feed was analyzed by the manufacturer and water is analyzed annually.

2.6. Dose Formulations and Administration of HMB

The stability and homogeneity of the dosing formulations were verified for 72 h at 4 °C by Method A. The concentration of [¹⁴C]HMB present in each dosing solution was determined by LSS analysis from three aliquots of the dosing solution taken pre- and post-dose.

Oral (corn oil) and IV (water:ethanol:alkamuls (3:5:2 v:v:v) dose formulations contained a mixture of HMB and [¹⁴C]HMB. The target oral doses were 10, 100, or 500 mg [¹⁴C]HMB/kg body weight. Oral doses were administered via intragastric gavage in a dose volume of 5 mL/kg (rat) or 10 mL/kg (mouse). The target IV dose was 10 mg [¹⁴C]HMB/kg body weight. IV doses were administered in a dose volume of 1 mL/kg (rat) or 4.5 mL/kg (mice) into the tail vein.

Several vehicles were formulated at 10 mg/kg [¹⁴C]HMB to evaluate HMB dermal absorption and disposition. These formulations were prepared in ethanol (Sigma # 493546, 200 proof USP), ethanol:coconut oil 1:1, coconut oil (SAFC Sigma# W530155, virgin, certified organic), light paraffin oil (light Mineral Oil Sigma# 330779), and lotion (olive oil:emulsifying wax, and water 15:15:70).

Approximately 18 h prior to dermal dosing, an area no less than 4 cm² for both rats and mice was clipped to minimize any trauma to the skin including abrasions or cuts. If any cuts or abrasions were observed, that animal was replaced. The target dosing area was identified and cleaned with alcohol. Either a foam isolator (3M, St. Paul, MN), or a metal mesh stainless steel isolator (Shandon Lipshaw Micro Cassette #331 positive clip, LabX Scientific Marketplace), was used to protect the dermal dosing site from grooming or other undesired contact. Prior to dosing, the dermal isolator was adhered to the skin using methylacrylate cement (The Original Super Glue Corporation/Pacer Technology, Rancho Cucamunga, CA). Dermal doses in rats were applied in a dose volume of 0.5 to 1 mL/kg using a 2.5-mL Luer lock gas-tight syringe (Hamilton) equipped with needle attached for all formulations except the olive oil:emulsifying wax:water formulation (lotion), which was applied in a dose volume of ~100 μL per rat using a 100 μL Wiretrol (Drummond Scientific Company, Broomall, PA). Dermal doses in mice were applied in a dose volume of approximately 2 mL/kg using a 0.5-mL Luer lock gas-tight syringe (Hamilton) equipped with needle attached. Animals were manually restrained and material delivered dropwise onto the dose site skin through spaces in the dermal isolator. Post application, a breathable linen protective cover was placed over the foam isolator; no cover was needed for the metal device.

2.7. Collection and Analysis of Biological Samples

Urine and feces were collected separately from each animal and cooled with dry ice during the collection period (at 0, 4, 8, 12, 24, 48, and 72 h for urine, at 24, 48, and 72 h for feces). The metabolism cages were rinsed with water at the end of each collection interval (or with water followed by ethanol after terminal collection only). The resulting rinsates were combined by time point. Feces were homogenized with an equal volume of water. Samples were stored at −20°C in the dark until analyzed. Expired volatile organic compounds (VOCs) were collected at 0–4, 4–8, 8–12, and 12–24 h after dosing by passing the air from the metabolism chambers through two traps containing isopropanol to trap expired VOCs, and then through two traps containing 1 N NaOH in H₂O to trap expired CO₂. All traps were cooled on wet ice and analyzed separately. Traps were changed at 4, 8, 12, 24, 48, and/or 72 h depending on the experiment.

At the termination of each study group, animals were euthanized with a sodium pentobarbital-based euthanasia solution (390 mg sodium pentobarbital/kg and 50 mg phenytoin sodium/kg) by intraperitoneal injection, and blood was collected via cardiac puncture. The following tissues and organs were collected: adrenals, brain, lung, heart, spleen, pancreas, kidneys, testes or uterus and ovaries, liver, thyroid, thymus, small intestine, small intestine contents, cecum, cecum contents, large intestine, large intestine contents, urinary bladder, and urinary bladder content (urinary bladder content was added to the last urine collection), and samples of adipose (perirenal, reproductive), muscle (hind leg, trapezius), and skin (ears). Contents from the stomach, small intestine, large intestine, and cecum were combined by animal and weighed. GI tract tissues were rinsed with deionized water and the rinsate was pooled with all the combined contents from the GI tissues for each animal. The remaining carcasses were weighed and digested with 2 N ethanolic sodium hydroxide. All samples were stored at −20°C.

At the termination of the dermal studies, following euthanasia and blood collection, the dermal isolator (and glue attached) was removed, wiped with gauze that was soaked in chlorohexidine soap, and then placed into a petri dish. The glue was removed from the dermal isolator by soaking in toluene. All excess glue was removed from the skin and placed in a labeled vial. The dermal site was wiped with chlorohexidine. All wipes were stored separately prior to extraction. The area was then dried with a clean gauze pad. The horny layer of the skin was removed using 20 pieces of tape consecutively along the cleaned dose site. Wipes and tape were extracted with toluene. The dermal site skin was excised and dissolved. All samples were stored at −20°C.

Bile was collected from the bile duct via surgically pre-implanted bile duct cannulae at 15–60 min intervals throughout the study. Pre-dose samples were also collected to determine background levels.

Triplicate aliquots of urine, cage rinse, and each breath trap solution were transferred directly in Ultima Gold scintillation cocktail and analyzed for radioactivity content by LSS. Triplicate aliquots of the fecal and GI content mixtures, and aliquots of tissues (~50–150 mg each) were combusted in a Packard 307 biological sample oxidizer, mixed with Carbo-Sorb E and Permafluor E+, and collected into vials for LSS.

The carcass and dose site skin, which were dissolved in 2N ethanolic sodium hydroxide prior to analysis, were neutralized with nitric acid, bleached with hydrogen peroxide, aliquots weighed into scintillation vials containing Ultima Gold scintillation cocktail, and analyzed for total radioactivity by LSS. The bleached samples were protected from light for at least 6 h prior to LSS. For dermal studies, the radiochemical content of the dermal site, wipes, glue, and rinsates was determined and defined as the unabsorbed dose.

Samples were analyzed for total radioactivity by a Packard Model 3100TR LSS. Samples of urine, feces, and breath (cage air drawn through breath traps) were collected during the acclimation period and analyzed for radiochemical content to determine the background radioactivity from the metabolism cages. In addition, blanks composed of Ultima Gold cocktail were prepared to correct for background radioactivity. For determination of total [¹⁴C]HMB in tissues, the total weight of rats was assumed to be comprised of 7.4% blood, 4.3% plasma, 7.0% adipose, 40.4% muscle, and 19% skin, and the total weight of mice was assumed to be comprised of 4.9% blood, 2.8% plasma, 7.0% adipose, 38.4% muscle, and 16.5% skin (Brown et al., 1997).

2.8. [¹⁴C]HMB-derived Radioactivity in Liver

Irreversible binding of [¹⁴C]HMB-derived radioactivity was evaluated in liver tissue collected from male rats at 2, 24, and 72 h post gavage administration at 100, 10, and 100 mg/kg. The following procedure was performed as previously described by Masson et al. (Masson et al., 2010). Aliquots of liver (~50 mg) were homogenized with 1 mL of 1:1 methanol:water in a 1.5 mL Eppendorf tube. The sample was centrifuged at ~ 16,000 g for 5 min and the supernatant was collected. The pellet was homogenized with 1 mL of 3:1 dichloromethane:methanol, and the supernatant was collected as described above. This two-step extraction procedure was repeated three times and extracts were analyzed by LSS. If any radioactivity was detected in the third extract, a subsequent fourth extraction was performed. The remaining pellet was oxidized and analyzed for radioactivity by LSS.

2.9. Metabolite Profiling and Identification

Urine composites were prepared per time point and dose group by combining urine in a ratio proportional to the total volume collected for individual animals. Urine samples were centrifuged at 11,200 g for 5 min to remove particulates. Urine (100 μL) was incubated at 37°C overnight (~18 h) with 30 μL of β-glucuronidase (2000 U in 100 mM phosphate buffer pH 6.8 for E. coli-derived enzyme, in 1.0 M acetate buffer pH 5.0 for H. pomatia-derived enzyme). Following hydrolysis, 3X volume of ice-cold acetonitrile was added to the sample, centrifuged and the supernatant collected for analysis. For the acid hydrolysis, equal volumes of urine and 12N HCl were combined and heated at 105°C for 90 min. For metabolite profiling, unhydrolyzed and hydrolyzed urine samples were analyzed by Method B and/or C. Metabolite identities were inferred from the MRM transitions used in Method C, analysis of β-glucuronidase incubations of urine, and by comparison of LC-MS with authentic standards prepared in methanol. The list of predicted MRM transitions and metabolite descriptions are shown in Supplementary Table 1.

2.10. Toxicokinetic Studies

Following gavage (10 mg/kg) and IV (10 mg/kg) administration of [¹⁴C]HMB in male and female rats, blood (0.3 mL) was collected directly from the cannulated jugular vein port using a heparinized syringe and blunt tip needle at 0 (pre-dose), 5, 15, 30 min, and 1, 2, 4, 6, 8, 10, 12, and 24 h. Canulae were verified for patency upon receipt and maintained by saline flush during quarantine prior to use. Sterile saline with anticoagulant was flushed into the cannula after blood sampling to prevent the blood from clotting. Immediately after collection, blood was centrifuged (1300 g, 2 to 8°C, 10 min), plasma separated, and plasma was stored frozen at −80°C until analysis.

The HMB LC-MS bioanalysis method was developed and validated in SD rat plasma for linearity, accuracy and precision of the method over the concentration range of 0.5–100 ng/mL. Low, mid, and high HMB concentration plasma quality control (QC) samples at 5, 75, and 200 ng/mL were evaluated for precision and accuracy of the method. Linear regression of 1/x weighing was used to calculate analyte peak area response to IS (DHB) peak area ratio. Analysis of rat plasma by LC-MS detected negligible amount of DHB allowing its use as an IS without interference. The correlation efficient was >0.98, the precision (relative standard derivation, RSD) was ≤4.3% and accuracy (relative error, RE%) was ≤± 13.6. Sensitivity of the method was estimated by analyzing a set of five rat plasma QC samples prepared at the lower limit of quantitation (LLOQ) of 0.5 ng/mL for HMB. The limit of detection (LOD) was confirmed by preparing five samples at 0.1 ng/mL.

Solvent standards of HMB (10–2500 ng/mL) were prepared in acetonitrile. Matrix standards (1–250 ng/mL) were prepared by spiking 90 μL of plasma with 10 μL of the appropriate solvent standard. IS (DHB) was prepared at 240 ng/mL in acetonitrile. Plasma samples or matrix standards (100 μL) were spiked with 200 μL of the 240 ng/mL DHB IS solution. Samples were vortexed and centrifuged at 19,000 g for 10 min. The supernatants were analyzed by LC-MS Method D to determine levels of HMB.

Plasma concentration vs. time data was analyzed by noncompartmental analysis using WinNonlin v. 5.1 (Pharsight, Cary, NC). The following toxicokinetic parameters were determined: terminal elimination rate constant (λ_z), terminal elimination half-life (t_1/2), area under the plasma concentration vs. time curve extrapolated from time zero to infinity (AUC_inf), maximum concentration achieved (C_max), time to maximum concentration (T_max), systemic clearance (Cl), apparent volume of distribution at steady state (V_ss), and mean residence time (MRT). Mean absorption time (MAT) was calculated as MAT = MRT_{oral gavage}-MRT_iv. Bioavailability (F) was estimated for the extravascular routes of administration relative to the IV as a reference route according to the following equation:

$$\text{F }=\left({\text{AUC}}_{\text{inf}(\text{PO})}•{\text{ Dose}}_{\text{iv}}\right)/\left({\text{AUC}}_{\text{inf}(\text{iv})}•{\text{ Dose}}_{\text{PO}}\right)$$

For all samples that report mean and SD, the statistics were determined using Microsoft Excel 2003, 2007, or 2010.

3. Results

3.1. Disposition of [¹⁴C]HMB Following Gavage and IV Administration and Dermal Application in HSD Rats

Disposition of [¹⁴C]HMB following single gavage administration in male (10, 100, 500 mg/kg) and female (100 mg/kg) rats is given in Table 2. HMB-derived radioactivity was excreted mainly in urine (including cage rinse) (53–58%) and feces (38–42%) within 72 h. The cumulative excretion is shown in Table 2 demonstrating steady excretion of radioactivity over time. Dose expired as VOCs or CO₂ was low (< 1%) 24 h following administration of 100 mg/kg in male rats; hence, VOCs and CO₂ were not collected in remaining groups.

Table 2.

Distribution and cumulative excretion of radioactivity up to 24 and 72 h following a single oral or 72 h following IV administration of [¹⁴C]HMB to male and female Harlan Sprague-Dawley rats

Sample	Cumulative percent of dose recovereda
	Oral Gavage					IV
	Male 10 mg/kgb	Male 10 mg/kgc	Male 100 mg/kgc	Male 500 mg/kgc	Female 100 mg/kgc	Male 10 mg/kgc
Urine (cage rinse)	42.1±6.50 (9.60±0.35)	44.4±5.9 (9.71±1.29)	47.5±5.3 (9.83±3.34)	41.3±6.3 (11.7±2.8)	41.4±9.1 (16.0±4.7)	47.3±7.7 (18.7±16.2)
Feces	25.3±6.80	42.4±1.29	38.6±8.2	41.0±3.8	37.5±9.4	27.8±6.4
Volatile organics	0.04±0.01	-	0.13±0.02	-	-	0.08±0.03
CO2	0.45±0.06	-	1.14±0.12	-	-	0.68±0.07
GI content	5.82±0.96	-	0.46±0.54	-	-	0.77±1.29
Carcass	0.03±0.05	-	-	-	-	0.59±0.35
Tail	-	-	-	-	-	1.27±2.61
Tissues	3.10±0.45	0.38±0.05	0.48±0.08	0.36±0.05	0.40±0.17	0.52±0.29
Total dose recovered	81.7±12.8	97.1±2.8	99.9±4.7	94.6±2.6	95.6±2.5	97.3±12.8

^aAll Values expressed as mean (± SD), N=5,

^b24 h,

^c72 h,

In male rats, the radioactivity in tissues increased with the increasing dose following gavage administration (Supplementary Table 2); in general, liver had a higher tissue:blood ratio (TBR) (2.27–4.93) than kidney (1.26–3.53) at 72 h post dosing. Total radioactivity in tissues at 2, 24, and 72 h following gavage administration of 100 mg/kg in male rats were 27.5, 3.1, and < 0.5%, respectively, suggesting that HMB was distributed to the tissues, however it was not retained in tissues (Table 2). There were no apparent sex differences in disposition of [¹⁴C]HMB in male and female rats following gavage administration. Total dose recovered was > 94%.

The pattern of disposition 72 h following a single IV administration of 10 mg/kg in male rats was similar to that following gavage administration; 66%, 28%, and < 1% of the administered dose was recovered in urine (including cage rinse), feces, and tissues, respectively (Table 2). In bile-cannulated animals, following IV administration of 10 mg/kg, approximately 10% of the dose was recovered in bile after 6 h (Supplementary Table 3).

Dermal absorption of HMB formulated with several vehicles were evaluated. The absorbed dose varied depending upon vehicle (Figure 1). Following application of 10 mg/kg [¹⁴C]HMB in male rats, the percent dose absorbed in all vehicles was high (64–80%) except in the lotion vehicle where the absorption was moderate (46%) (Table 3). The percent dose absorbed was similar following application of 0.1 (73%) or 10 mg/kg (80%) [¹⁴C]HMB formulated in paraffin oil. The absorption of [¹⁴C]HMB was lower in female rats (30%, 15 mg/kg) than in male rats (46%, 10 mg/kg) following application of [¹⁴C]HMB in a lotion vehicle. The absorbed dose was excreted mainly via urine (including cage rinse) (18–48%) and feces (15–22%) with ~ 3–10% of the absorbed dose remaining in tissues (Table 3).

Figure 1.

Total unabsorbed and absorbed dose 72 h following dermal administration of 10 mg/kg [¹⁴C]HMB formulated in selected vehicles to male Sprague Dawley rats. Absorbed dose is defined as total of all radioactivity recovered in excreta and tissues. Unabsorbed dose is defined as total radioactivity recover in applicator, tape/glue, and dose site swipes.

Table 3.

Distribution and cumulative excretion of radioactivity up to 72 h following a single dermal application of [¹⁴C]HMB to male and female Harlan Sprague-Dawley rats

Sample	Cumulative percent of dose recovereda
	Male, Paraffin Oil 10 mg/kg	Male, Lotion 10 mg/kg	Male, Coconut Oil 10 mg/kg	Male, Ethanol:Coconut Oil 10 mg/kg	Male, Ethanol 10 mg/kg	Male, Paraffin Oil 0.1 mg/kg	Male, Ethanol 1 mg/kg	Female, Lotion 15 mg/kg
Urine (cage rinse)	36.1±4.5 (12.0±3.3)	11.9±5.0 (5.96±2.44)	26.4±2.1 (5.66±2.14)	28.7±3.5 (8.38±1.79)	24.5±3.9 (9.51±3.18)	28.7±4.9 (8.08±2.31)	24.1±2.6 (6.93±1.86)	11.4±3.6 (4.70±1.42)
Feces	21.1±3.4	21.7±26.9	15.3±1.9	21.8±5.5	21.8±2.8	21.1±2.1	21.3±4.4	8.84±1.21
GI Content	1.56±0.66	2.45±0.61	2.38±0.95	2.50±0.65	1.24±0.32	2.35±0.68	0.73±0.17	2.00±0.48
Tissues	6.02±2.22	2.79±0.72	8.31±6.83	8.18±2.83	7.39±2.09	9.74±8.74	4.16±3.40	1.64±0.69
Total absorbed doseb	79.8±6.9	45.8±35.5	64.3±8.2	72.5±6.0	67.8±3.2	72.5±8.5	53.4±4.7	28.9±4.10
Total dose site skin	0.11±0.08	1.58±0.72	0.43±0.23	0.22±0.09	0.28±0.145	0.37±0.14	54.3±4.7	2.68±2.62
Total unabsorbed dosec	1.52±0.42	51.7±23.0	10.7±2.9	4.41±1.81	16.2±4.5	6.66±1.72	22.4±2.8	56.0±7.20
Total dose recovered	79.8±7.0	99.1±13.3	72.5±5.3	76.9±6.7	82.7±5.5	79.5±7.0	76.3±5.4	87.6±2.70

^aAll Values expressed as mean (± SD) for N=5,

^bTotal absorbed dose is the total administered dose recovered in excreta, GI content, carcass, and tissues.

^cTotal unabsorbed dose is the total administered dose recovered in dose site wipes, tape/glue, and applicator.

3.2. Disposition of [¹⁴C]HMB Following Gavage and IV Administration and Dermal Application in B6C3F1/N Mice

Disposition of [¹⁴C]HMB following single gavage administration in male (100 and 500 mg/kg) and female (100 mg/kg) mice is given in Table 4. In male mice, HMB was excreted mainly in urine (including cage rinse) (40–41%) and feces (24–39%) within 72 h. The cumulative excretion is shown in Table 4 demonstrating steady excretion of radioactivity over time. Excretion in VOCs, CO₂, and tissues accounted for ~ 0.8%, 5.5%, and 0.25% of the administered dose, respectively. The tissues with the most radioactivity in males were thymus and thyroid in both 100 and 500 mg/kg dose groups (Supplementary Table 4).

Table 4.

Distribution and cumulative excretion of radioactivity up to 72 h following a single oral and IV administration of [¹⁴C]HMB to male and female B6C3F1/N mice

Sample	Cumulative percent of dose recovereda
	Oral Gavage			IV
	Male 100 mg/kg	Male 500 mg/kg	Female 100 mg/kg	Male 10 mg/kg
Urineb (cage rinse)	17.5±14.0 (23.9±10.6)	18.2±22.8 (21.5±8.7)	4.24±4.35 (29.7±7.0)	4.13±6.45 (12.2±4.4)
Fecesb	38.7±2.8	24.3±8.0	23.9±8.8	29.2±6.9
Volatile organicsc	0.83±0.20	-	0.73±0.07	-
CO2c	5.51±0.87	-	15.6±21.3	-
GI contentb	0.06±0.05	0.05±0.02	0.06±0.04	0.11±0.04
Foodb	-	-	-	11.5±7.4
Carcassb	1.53±0.20	0.70±0.48	0.12±0.16	0.20±0.10
Tailb	-	-	-	1.12±0.59
Tissuesb	0.25±0.05	0.55±0.42	0.16±0.04	0.28±0.07
Total recovered	88.9±21.5	69.0±15.2	76.0±28.6	58.8±10.7

^aAll Values expressed as mean (± SD) for N=5 for oral gavage and N=3 for IV,

^b72 h,

^c24 h

The disposition was similar in female mice 72 h following a single 100 mg/kg gavage administration of [¹⁴C]HMB with ~34 and 24% in urine and feces, respectively, and slightly higher recovered radioactivity in CO₂ (~15%) with the total radioactivity remaining in the tissues ~0.16% (Table 4) which was slightly lower than males (~0.25–0.55%). The total radioactivity recovered in the 500 mg/kg dose group for male mice was lower (~69%) than 100 mg/kg groups for male (~89%) and female mice (~76%), which may or may not be sex related.

Disposition following a single 10 mg/kg IV administration of [¹⁴C]HMB in male mice was similar to that following a single gavage administration with HMB being excreted mainly in urine (~16%) and feces (~29%) after 72 h (Table 4). The rest of the radioactivity was recovered in tissues (~0.28%) and in the food (~11.5%), most likely due to urine contamination in the food resulting in a total recovery of ~58% of the dose. While no VOCs and CO₂ samples were collected, the potential loss in the recovery was most likely due to expired ¹⁴CO₂.

Dermal absorption of HMB was evaluated following formulation of [¹⁴C]HMB in several different vehicles and the data are given in Table 5. The total absorbed dose for males at 10 mg/kg varied amongst the vehicles with the highest absorbed dose observed in ethanol (~64%) followed by acetone (~57%), and lotion (~37%) (Table 5). The 1 mg/kg dose in ethanol was similar to the 10 mg/kg dose in lotion with ~41% total absorbed dose and recovery in feces, urine and cage rinse and tissues were within 10% of the 10 mg/kg lotion group. When normalized for total recovery (Supplementary Figure 1), the distribution of [¹⁴C]HMB radioactivity in tissues and excreta following dermal application to male mice was similar between the vehicles at 10 mg/kg with the exception of acetone showing higher tissue levels. HMB absorption in female mice following dermal application at 10 mg/kg lotion or 10 mg/kg ethanol was similar to that seen in males. The unabsorbed dose in female mice was ~41% with majority of radioactivity recovered in urine and feces at a 10 mg/kg dose in lotion (Table 5).

Table 5.

Distribution and cumulative excretion of radioactivity up to 72 h following a single dermal application of [¹⁴C]HMB to male and female B6C3F1/N mice

Sample	Cumulative percent of dose recovereda
	Male, Lotion 10 mg/kg	Female, Lotion 10 mg/kg	Male, Ethanol 1 mg/kg	Male, Ethanol 10 mg/kg	Female, Ethanol 10 mg/kg	Male, Acetone 10 mg/kg
Urine (cage rinse)	6.40±4.77 (7.00±1.75)	3.62±2.46 (12.5±3.4)	3.01±5.48 (11.8±4.1)	4.77±4.67 (15.8±3.4)	5.29±7.78 (15.2±2.4)	2.63±3.01 (16.2±4.61)
Feces	20.9±6.8	26.3±9.9	20.5±4.5	32.4±4.9	39.4±10.2	29.5±4.1
GI Content	0.58±0.15	0.71±0.11	0.30±0.15	0.72±0.19	0.34±0.07	0.57±0.21
Tissues	0.76±0.10	1.18±0.25	0.67±0.07	0.78±0.19	0.66±0.10	3.31±2.33
Total absorbed doseb	36.8±8.2	46.2±7.3	41.4±3.7	63.6±7.5	68.6±8.6	57.2±2.8
Total dose site skin	1.86±0.95	0.66±0.60	1.39±0.28	0.69±0.16	1.27±0.44	1.10±0.43
Total unabsorbed dosec	36.6±4.50	40.7±5.0	26.9±7.1	18.9±4.4	15.3±5.2	19.5±8.2
Total dose recovered	75.3±10.9	87.5±8.3	69.7±6.3	83.2±6.0	85.2±7.6	77.8±9.0

^aAll Values expressed as mean (± SD) for N=5,

^bTotal absorbed dose is the total administered dose recovered in excreta, GI content, carcass, and tissues.

^cTotal unabsorbed dose is the total administered dose recovered in dose site wipes, tape/glue, and applicator

3.3. Profiling and identification of HMB metabolites in vivo

Composite urine samples were analyzed by Method B to generate radiochemical profiles. Representative radiochromatograms are shown in Figures 2A–D. The radiochromatograms were very similar with no apparent species, sex or route differences except minimal metabolism observed following dermal application in acetone in mice. Up to 9 peaks were detected in the urine with one major peak at ~14 min (Peak 8) in all urine samples. Minor peaks 14 and 15 were not observed in radiochemical gavage profiles of male rats and mice at 500 mg/kg and of female mice and rats at 100 mg/kg (Figure 2C). Additionally, following IV administration (10 mg/kg, 6h), radiochemical profile of bile from male rats was analyzed and found to be similar to rat urine profiles with Peak 8 being the major metabolite in the bile (Supplementary Figure 3).

Figure 2.

A) HPLC radiochromatogram of male rat urine collected 0–72 h following 100 mg/kg single oral gavage administration of [¹⁴C]HMB, B) HPLC radiochromatogram of male rat urine collected 0–72 h following 10 mg/kg dermal administration of [¹⁴C]HMB, formulated in Coconut oil:Ethanol (1:1), C) HPLC radiochromatogram of female rat urine collected 0–72 h following 100 mg/kg single oral gavage administration of [¹⁴C]HMB, D) HPLC radiochromatogram of male mice urine collected 0–72 h following 100 mg/kg single oral gavage administration of [¹⁴C]HMB.

Urine and plasma metabolites from rats and mice following HMB administration by gavage, IV, or dermal application were identified by LC-MS (Method C). Peaks in the urine radiochromatograms (HPLC Method B) were also identified by LC-MS Method C, where identical chromatographic gradients and conditions were used. Commercially available standards and their MRM transitions were identified as shown in Supplementary Figure 4. Peaks 14 and 15 were identified as parent HMB and DHB, respectively, based on their retention times and MRM transitions in comparison to authentic standards. Other metabolites of HMB were tentatively identified by the MRM transitions. Approximate retention time, MRM transitions, and percent of total radioactivity for each peak derived from radioprofiles are shown in Supplementary Table 5. Plasma samples from male rats following 10 mg/kg single oral gavage administration (24h) were also analyzed by LC-MS Method C for metabolite identification. Five peaks, DHB-glucuronides (peak 3 and 10), HMB-glucuronide (peak 8), DHMB isomer (peak 14b), and HMB (peak 15), were identified based on their MRM transitions (Figure 3).

Figure 3.

LC-MS/MS total ion chromatogram of male rat sample following 10 mg/kg oral gavage administration of [¹⁴C]HMB (0 to 24 h Urine Composite Sample). MRM transitions: Peaks 3 and 10 = 391/137 (DHB Glucuronides), Peak 8 = 405/151 (HMB Glucuronide), Peak 14b = 245/151 (potential isomer of DHMB), and Peak 15 = 229/151 (HMB).

In addition to the MRM analysis, incubations of urine samples from male rats following oral administration (100 mg/kg single male rat and 500 mg/kg male rat urine pool) were performed with and without β-glucuronidase from E. coli and H. pomatia to evaluate and support metabolite identification. After incubation of urine from single male rat (100 mg/kg, gavage administration) with β-glucuronidase from E. coli, the radioactive peaks at 9.8 min and 14.6 min greatly diminished (Supplementary Figures 5A and B). Concomitantly, a peak at 28.7 min (DHB retention time) and the peak at 39.4 min (HMB retention time) grew in intensity. Analysis of the relative peak areas indicates that the radioactive peak at 9.5 and 14.6 min may be the glucuronide conjugates of DHB and HMB, respectively. Similar to the E. coli incubations, incubation of urine from single male rat (100 mg/kg, gavage administration) with β-glucuronidase from H. pomatia (Supplementary Figures 5A and C) shifted the 9.5- and 14.5-min peaks to 28.7 and 39.4 min, the retention times of DHB and HMB, respectively. Additionally, the radioactive peak at the 12.2 min shifted and peak intensity at 28.7 min (DHM) increased after incubation with glucuronidase from H. pomotia but not E. coli. This suggests that the radioactive peak at 12.2 min may be a sulfate conjugate of DHB, as β-glucuronidase from H. pomatia contains some sulfatase activity, while β-glucuronidase from E. coli does not.

Additionally, hydrolysis of HMB conjugate metabolites were evaluated by incubation of pooled male rat urine aliquots (500 mg/kg, single gavage administration) under three different conditions; incubations with β-glucuronidase from E. coli., β-glucuronidase and sulfatase from H. pomatia, or hydrochloric acid. Incubation with hydrochloric acid resulted in a more complete deconjugation of glucuronide and sulfate metabolites of HMB than enzyme incubations. Both HMB and DHB, along with two DHMB isomers and THB were detected by LC-MS analysis in the acid hydrolyzed sample (Figure 4). A similar result was observed for female rat urine (100 mg/kg) composite sample following acid hydrolysis (Supplementary Figure 6). Extracted ion chromatograms for DHB, DHMB, and THB from the male rat urine post acid hydrolysis are shown in Supplementary Figure 7. Assignments were based on comparison to standards as well as evaluation of the MRM transitions. Deconjugation of rat or mouse urine via acid hydrolysis or with β-glucuronidase or β-glucuronidase/sulfatase preparations released three peaks with MRM transitions of m/z 245 →167. It is not possible to determine the relative position of hydroxylation with MS alone, however, when an authentic standard of 2,5-DHMB was co-injected with the deconjugated urine samples, one of the three peaks matched the MRM transition and retention time of the authentic standard. By inference, the two other peaks with an identical MRM transitions and spectra could be tentatively assigned as 2,3-DHMB and 2,6-DHMB. The position of second hydroxylation was confirmed to be on the same phenyl ring as in HMB by the presence of 167, and 106 fragments and absence of 151 and 121 fragments (Supplementary Figures 4 and 8). Parent DHMB was not seen in the chromatograms prior the acid hydrolysis (Figure 4A). While acid hydrolysis was not performed on the mouse urine samples (500 mg/kg single gavage administration), following incubation with β-glucuronidase from E. coli, peaks at ~ 9.7, 11.6, 12.9, and 15 min (identified as glucuronides) disappeared or were reduced (Data not shown). The peaks at ~ 27.1 and 36.9 min (DHB and HMB, respectively) intensified.

Figure 4.

LC-MS-MRM Chromatograms of Male Rat Urine Composite (500 mg/kg, oral gavage) A)Before Incubation, B) After Incubation with Hydrochloric Acid C) After Incubation with β-glucuronidase and sulfatase from H. pomatia, and D) After Incubation with β-glucuronidase from E. coli.

The peak at ~9.7 min from 500 mg/kg oral gavage group (male mice) had a molecular ion of 567.7 m/z, and had 2 fragments with 392.0 and 215.7, which corresponded to DHB di-glucuronide metabolite with loss of one and two glucuronide fragments, respectively. Additional metabolites at ~12.9, 13.3, and 14.3 min detected in the sample have been identified as either HMB+Oxygen+glucuronide (DHMB isomer) or HMB+Oxygen+sulfate. The site of the additional oxygen was assumed to be on the same ring as the methoxy substituent based the existence of fragment 167 m/z and lack of the fragment 151 m/z (Supplementary Figure 8A). This indicated that these metabolites were not 2,2’-DHMB or isomers of DHMB with hydroxylation on the ring opposite to methoxy substituent. The fragmentation spectra for the DHMB isomers 2,2’- and 2,5-DHMB, HMB, and other commercially available HMB like molecules are shown in Supplementary Figure 8B.

3.4. Metabolism of HMB in Hepatocytes

Metabolism of HMB was investigated in male and female SD rat, B6C3F1 mouse, and human hepatocytes in the presence of 1 μM HMB. Peaks in chromatograms were identified by retention time matching those in urine profiles following exposure to [¹⁴C]HMB (Figure 2). Peaks with similar retention time to HMB and DHB were observed along with several other peaks representing HMB glucuronide and DHB sulfate conjugates (Supplementary Figure 2). There were minor differences in the metabolic profiles of HMB between species and sexes. The DHB sulfate metabolite was only present in samples collected from female mouse and female rat hepatocytes. Minor metabolites such as HMB sulfate were only present in female mouse hepatocyte samples but not in male mouse samples. The differences in the relative amounts of some metabolites generated by male and female mouse hepatocytes suggests minor sex differences in metabolic profiles of HMB. In contrast, male and female human hepatocyte metabolite profiles were very similar to each other.

3.5. [¹⁴C]HMB-derived Radioactivity in Liver

To investigate whether the HMB-derived radioactivity is associated with liver macromolecules, liver from male rats administered 10 or 100 mg/kg [¹⁴C]HMB and sacrificed 2, 24, or 72 h post-dose was exhaustively extracted and the radioactivity remaining in the tissue pellet was determined. Any radioactivity detected in the tissue pellet was presumed to be due to binding of HMB-derived radioactivity to the tissue and macromolecules. While only 3% of the total radioactivity in liver was associated with macromolecules at 2 h, the percent increased with time (24 and 65% at 24 and 72 h, respectively). Total HMB equivalent bound per g of liver was 1.1, 0.2, and 1.2 μg HMB equivalents at 2, 24, and 72 h, respectively. The 24 h sample at 10 mg/kg had a lower equivalent (0.2 μg) than anticipated due to lower dose administered.

3.6. Toxicokinetic Studies

The plasma HMB concentration versus time profiles for male and female rats given 10 mg/kg HMB via gavage and IV administration are shown in Figure 5 and TK parameters are shown in Table 6. Following IV administration, both C_max and AUC were high and were higher in females compared to males; plasma elimination half-life in males and females was 4–6 h. Following gavage administration, HMB was absorbed slowly in both male and female rats with C_max reached within 2–6 h. Both C_max and AUC were higher (p<0.05) in males (8.5±4.0 ng/mL and 80±28 ng*h/mL) compared to females (2.9±1.1 ng/mL and 51±13 ng*h/mL). Plasma elimination half-life of HMB was longer in females (18.5±4.9 h) compared to males (6.4±2.4 h) (p<0.05). Oral bioavailability of HMB following a gavage single administration was low in both males (0.9±0.4 %) and females (0.3±0.1 %).

Figure 5.

Plasma free HMB concentrations vs time profile following 10 mg/kg single oral gavage or IV administration of [¹⁴C]HMB in male and female Sprague Dawley rats A-D) TK analysis of observed vs fit data using non-compartmental analysis.

4. Discussion

HMB is a common UV filter present in sunscreens and other personal care products, and thus has significant potential for human exposure following both oral and dermal routes. Here we report the effect of species, sex, route, and dose on ADME of HMB in HSD rats and B6C3F1/N mice. A limited study was also conducted to investigate the TK behavior of HMB following gavage administration. To the best of our knowledge, this is the first comprehensive study investigating the disposition of HMB in rats and mice following administration via multiple routes.

Our data demonstrate that following a single gavage administration, HMB was well absorbed in both HSD rats and B6C3F1/N mice of both sexes. In male rats, following gavage administration of 10, 100, and 500 mg/kg, 53–58% and 39–42% of the administered dose was recovered by 72 h in urine (and cage rinse) and feces, respectively with no apparent dose-related effect in excretion; the majority of the administered dose was recovered within 24 h demonstrating rapid excretion (Table 2). The pattern of excretion following a single IV dose of 10 mg/kg [¹⁴C]HMB (urine, 66%; feces, 28%) was similar to that following a similar gavage dose (urine, 54%; feces, 42%) in male rats. In addition, following a single IV dose, 10% of the administered dose was excreted in bile by 6 h (Supplementary Table 3). These observations support the conclusion that significant portion of the administered dose recovered in feces following gavage administration in rats is from the absorbed dose which was secreted to the intestine via bile. In male rats, following gavage administration of a 100 mg/kg, the dose recovered in exhaled CO₂ and VOCs were minimal; hence, in other rat groups, VOC and CO₂ collections were discontinued. Approximately 28% the administered radioactivity was recovered in tissues at 2 h demonstrating that [¹⁴C]HMB was well-distributed to tissues following gavage administration; however, residual radioactivity remaining at 24 and 72 h respectively were ~ 3% and 0.5%, suggesting that there is low potential for accumulation (Table 2). The disposition of radioactivity 72 h following gavage administration of a 100 mg/kg in female rats (urine, 57%; feces, 38%; tissues, 0.4%) was similar to that of males (urine, 57%; feces, 39%; feces, 0.5%) (Table 2). Findings from our studies are consistent with previously reported data in male Fischer 344 rats following a single oral administration (3.01, 28.1, 293, and 2570 mg/kg) and IV administration (4.63 mg/kg) of [¹⁴C]HMB (El Dareer et al., 1986). In that study, HMB was rapidly excreted in urine and feces (88–105%) within 72 h. Total amount of radioactivity retained in each tissue was reported to be <0.11% of the administered dose. The percent of dose in each tissue was similar across doses.

The pattern of disposition in male mice following gavage administration of a 100 and 500 mg/kg, in general, was similar to male rats. While excretion via urine (40–41%) and feces (24–39%) in mice was slightly lower than rats (urine, 53–57%; feces, 39–41%), male mice had higher proportion of the dose excreted as ¹⁴CO₂ than male rats (male mice, 6%; male rats, ~1% at 100 mg/kg) (Tables 2 and 4). The administered dose recovered as exhaled ⁴CO₂ was higher in female mice (~16%) compared to male mice (~ 6%) following a gavage administration of 100 mg/kg with concomitant reduction in the dose excreted in feces (female, ~ 24%; male, ~ 39%) demonstrating some sex difference in excretion of [¹⁴C]HMB in mice. Similar to rats, the residual radioactivity recovered in male and female mice at 72 h following administration was low (≤0.6%) suggesting that there is low potential for accumulation of HMB-derived moieties in tissues (Table 4).

TBRs were highest in the liver (3.5), kidney (2.1), and thyroid (4.3), in male rats at 72 h following a gavage dose of 100 mg/kg [1⁴C]HMB with most tissues having a TBR of < 1 (Supplementary Table 2). The high TBRs observed for bladder (4.3) and GI tissues (2–3.5) are likely due to contamination from urine and feces, respectively. Unfortunately, a direct comparison to female rats cannot be accomplished in our studies since all tissues were not examined in female rats. Similar findings were reported following oral administration of 100 mg/kg unlabeled HMB in SD rats where liver contained the highest amount HMB and metabolites, followed by kidney, and testes (Kadry et al., 1995). In male mice 72 h following administration of 100 mg/kg [¹⁴C]HMB, TBRs for the liver (3.4) and kidney (1.0) was similar to rats; however, unlike in male rats, TBRs were much higher in adrenals (153), thymus (30), and thyroid (95) (Supplementary Table 4). In female mice, TBRs were similar to male mice for the liver (male, 3.4; female, 2.6) and kidney (male, 1.0; female, 0.7); however, values for adrenals (6.5), thymus (1.5) and thyroid (30) were lower than male mice. Although the reason for this species and sex (mice only) difference is not clear at the present time, it may help explain any sex and/or species difference in target organ toxicity following oral administration of HMB.

We also evaluated the disposition of HMB following dermal application when administered in multiple vehicles, including a lotion vehicle to mimic consumer products such as sun screen formulations. In male rats, the highest absorption of HMB was observed in paraffin oil with ~ 73–80% of the applied dose of 0.1 or 10 mg/kg was absorbed within 72 h. The absorption of [¹⁴C]HMB when formulated in a lotion vehicle was lower with male and female rats absorbing 46 (10 mg/kg) and 29% (15 mg/kg) of the applied dose, respectively. Similar to rats, absorption of 10 mg/kg [¹⁴C]HMB in mice when formulated in a lotion vehicle was low with male and female mice absorbing 37 and 46%, respectively, of a 10 mg/kg dose. Following application of 10 mg/kg [¹⁴C]HMB in ethanol or acetone, male and female mice absorbed a higher dose (57–69%) comparted to when applied in the lotion vehicle (37–46%). HMB in light paraffin oil penetrated the skin better than coconut oil, similar to results observed previously (Benson et al., 2005) following application of finite dose of HMB across human epidermis in a range of vehicles (paraffin oil, oily cream, aqueous cream, 50:50 Ethanol:coconut oil, and coconut oil). The pattern of disposition of [¹⁴C]HMB-derived radioactivity in rats and mice following dermal application was similar to that following gavage administration with the majority of the absorbed dose was excreted in urine (rats, ~11–36%; mice, 3–6%) and feces (rats, 9–22%; mice, 21–39%), regardless of the vehicle used (Tables 3 and 5). The higher percent dose recovered in feces of mice compared to rats is likely due to contamination of feces with urine in mice and is a common observation in mass balance studies (Waidyanatha et al., 2019). The residual radioactivity in tissues following dermal application of [¹⁴C]HMB was higher in rats (1.6–9.7%, Table 3) than in mice (0.7–3.3%, Table 5) and was higher than following gavage administration (0.2–0.6%, Tables 2 and 4, 72 h) in both species demonstrating some species and route difference in tissue distribution of [¹⁴C]HMB.

HMB was extensively metabolized in rats and mice with a plethora of phase 1 and phase 2 metabolites identified (Figure 6). Urinary metabolites identified in rats were more than those reported in the literature (El Dareer et al., 1986; Okereke et al., 1993). In rats up to 14 metabolites were detected in urine covering demethylation, ring oxidation and conjugation with the major metabolite being an HMB- and DHB-glucuronides. Among other metabolites identified were sulfate conjugates of HMB, DHB, glucuronide and sulfate conjugates of ring-hydroxylated metabolites, DHMB isomers and THB. While the pattern of urinary profiles of rats and mice were similar, there were some minor differences between species following gavage administration of HMB (Supplementary Table 5). Similarly, the urinary profiles were similar to those following gavage administration in female rats and male and female mice following dermal application. The minimal metabolism observed following dermal application in acetone in male mice may be due to minimal absorption or first pass metabolism in liver. The metabolic profile of plasma in male rats also showed multiple peaks, including HMB, HMB glucuronide, DHB glucuronide, ring-hydroxylated HMB, although definite identification was not achieved for plasma metabolites.

Figure 6.

Postulated Metabolism of HMB in Rats and Mice

While the presence of 2,5-DHMB was reported following incubation of HMB with rat and liver microsomes (Kamikyouden et al., 2013), to the best of our knowledge, this is the first investigation reporting the detection of ring dihydroxylated metabolites of HMB (2,3-, 2,5- and 2,6-DHMB) in rodents. These metabolites, present as glucuronides and sulfate conjugates, were not readily detected in radiochromatograms but were detected by MS. Following deconjugation of rodent urine via acid hydrolysis or with β-glucuronidase or β-glucuronidase/sulfatase preparations, three DHMB isomer peaks observed. While one of the three peaks determined to be 2,5-DHMB, the other two peaks were tentatively assigned as 2,3- and 2,6-DHMB. Presence of 2,2’-DHMB, where dihydroxylation is in opposite rings, has been reported previously (Jeon et al., 2008; Okereke, C. S., 1994; Okereke et al., 1993) although we did not find this metabolite in urine from HMB-exposed animals.

The detection of ring di- and tri- hydroxylated metabolites in rodents may have some significance on the potential toxicity of HMB. THB, 2,3-, and 2,5-DHMB can be further oxidized to form reactive quinones which can subsequently bind with macromolecules such as protein and DNA to form adducts. Interestingly, we observed binding of [¹⁴C]HMB-derived radioactivity in the rat liver following a single gavage dose of [¹⁴C]HMB. While only 3% of the total radioactivity in the liver was associated with macromolecules at 2 h, at 24 and 72 h, 24 and 65% of the total liver radioactivity was found associated with liver macromolecules. This observation is interesting and may have some implications on the toxicity of HMB hence it needs further investigation.

Limited studies investigating the kinetics of HMB in male and female rats following gavage administration of 10 mg/kg showed some sex difference in the plasma TK of HMB with male rats (6.4 h) eliminating HMB faster than females (18.5 h). Although HMB is well-absorbed following gavage administration as has been demonstrated above using [¹⁴C]HMB, oral bioavailability of HMB was low (< 1%) in both males and females suggesting extensive first pass conjugation in the intestine and the liver following oral administration in rodents. This observation is consistent with the extensive metabolism of HMB observed in our investigation and by the others.

To compare across species, including humans, hepatocytes from rats, mice and humans were incubated with HMB. There were minor differences in the metabolic profiles. One of the major metabolites, possibly a DHB sulfate metabolite, was only detected in rat and mouse hepatocytes and relative amount was much greater in males than in females. Three minor metabolites were only present in the female mouse. Two metabolites were not present in the rodent urine chromatograms but were detected in both male and female human hepatocytes with MRM transitions indicating an HMB+glucuronide+O metabolite (12.7 min) and an HMB+O metabolite (26.0 min). These findings imply some species difference in metabolism of HMB.

5. Conclusion

Our data demonstrate that HMB is well-absorbed following gavage administration and moderately absorbed following dermal application in rats and mice and excreted extensively via urine and feces with minimal tissue retention. ADME properties of HMB were, in general, similar across rats and mice of both sexes with minor species, sex, and route differences. HMB was extensively metabolized following gavage administration with oral bioavailability <1% in male and female rats. Pathways of HMB metabolism include demethylation, ring hydroxylation, and conjugation. Detection of ring hydroxylated metabolites which can subsequently be converted to reactive quinone species may have some toxicological implications.