The Fate of Inhaled ¹⁴C-labelled PCB11 and its Metabolites In Vivo

Abstract

Background

The production ban of polychlorinated biphenyl (PCB) technical mixtures has left the erroneous impression that PCBs exist only as legacy pollutants. Some lower-chlorinated PCBs are still being produced and contaminate both indoor and ambient air.

Objectives

To inform PCB risk assessment, we characterized lung uptake, distribution, metabolism and excretion of PCB11 as a signature compound for these airborne non-legacy PCBs.

Methods

After delivering [¹⁴C]PCB11 to the lungs of male rats, radioactivity in 34 major tissues and 5 digestive matter compartments was measured at 12, 25, 50, 100, 200 and 720 min postexposure, during which time the excreta and exhaled air were also collected. [¹⁴C]PCB11 and metabolites in liver, blood, digestive matter, urine and adipose tissues were extracted separately to establish the metabolic profile of the disposition.

Results

[¹⁴C]PCB11 was distributed rapidly to all tissues after 99.8% pulmonary uptake and quickly underwent extensive metabolism. The major tissue deposition of [¹⁴C]PCB11 and metabolites translocated from liver, blood and muscle to skin and adipose tissue 200 min postexposure, while over 50% of administered dose was discharged via urine and feces within 12 h. Elimination of the [¹⁴C]PCB11 and metabolites consisted of an initial fast phase (t_½ = 9-33 min) and a slower clearance phase to low concentrations. Phase II metabolites dominated in liver, blood and excreta after 25 min postexposure.

Conclusions

This study shows that PCB11 is completely absorbed after inhalation exposure and is rapidly eliminated from most tissues. Phase II metabolites dominated with a slower elimination rate than the PCB11 or phase I metabolites and thus can best serve as urine biomarkers of exposure.

1. Introduction

Production of polychlorinated biphenyls (PCB) continues despite the 1979 phase out of PCB manufacturing (Grossman 2013). 3,3’-Dichlorobiphenyl (PCB11), a signature congener for the non-legacy PCBs, is still released as a by-product in the manufacture of pigment, which is to date the only known significant source of PCB11 (Hu and Hornbuckle 2009). It is estimated that 1.5 t of PCB11 was produced via diarylide yellow pigment in 2006 (Rodenburg et al. 2010). PCB11 contamination occurs in a wide range of environmental media including water, sediment, air, paint and consumer goods (Choi et al. 2008; Hu et al. 2008; King et al. 2002; Litten et al. 2002; Rodenburg et al. 2010) and PCB11 is an abundant atmospheric congener. Yet virtually nothing is known about its uptake, distribution and metabolism (Hu et al. 2013).

PCB11 is ubiquitous throughout the Chicago airshed (Hu et al. 2008) and indoors in both new and old buildings (Thorne PS, unpublished data). While outdoor levels of PCB11 reach 70-90 μg/m³ (Choi et al. 2008; Hu et al. 2008), levels indoors can be orders of magnitude higher (Frederiksen et al. 2012; Harrad et al. 2006; Heinzow et al. 2007). Due to the presence of PCB11 in air and low persistence in the food chain (Brown Jr. 1994), inhalation is the major route of exposure. Recently, we detected PCB11 in the plasma of over 60% of adolescent children and their mothers participating in the Airborne Exposures to Semi-volatile Organic Pollutants (AESOP) Study (Marek et al. 2013). This detection rate likely reflected current exposure since the low-chlorinated PCBs have little tendency to bioaccumulate. Extensive human exposure to PCB11 raises health concerns as this congener was shown to inhibit calcium sequestration and interfere with protein kinase C translocation, dopamine and thyroid hormone activities (Hansen, 1998).

The biological consequences of exposure to PCB congeners contained in manufactured technical mixtures from the 1930s to 1970s have been well explored in animal and population-based studies (ATSDR 2000). It is understood that these PCB congener mixtures induce a multiplicity of adverse health effects, but their behavior and mode of action differ among congeners (Safe 1993). Higher-chlorinated congeners are generally resistant to metabolism by cytochrome P450 (CYP) due to the lack of adjacent unsubstituted positions, while metabolism of lower-chlorinated congeners usually occurs readily to generate phase I hydroxylated metabolites (OH-PCB), and phase II metabolites after conjugation to more polar functional groups like sulfate and glucuronide. The biological half-lives (t_½) of congeners in rats range from a few hours (Hu et al. 2010) to months (Tanabe et al. 1981), dependent on their propensity for metabolism, exposure route (Diliberto et al. 1996) and tissue sequestration (DeVito et al. 1998). For some congeners, metabolism does not necessarily facilitate excretion or detoxication. Some OH-PCBs are retained strongly in blood due to their binding to thyroxin transporter, leading to disruption of thyroid hormone homeostasis (Bergman et al. 1994; Hansen 1998). A recent study identified PCB sulfates including PCB11 as high-affinity ligands for human transthyretin suggesting similar potential toxicities (Grimm et al. 2013). Methylsulfonyl metabolites (MeSO₂-PCB) are formed following the mercapturic pathway of glutathione-conjugated PCBs. Some MeSO₂-PCB with lower chlorination are subject to metabolic elimination, while many others persist in lung, liver and kidney, and are linked to organ-specific toxicities (Letcher et al. 2000).

The importance of inhalation exposure to the more volatile congeners has gained recognition. In an investigation of New York City school buildings contaminated by PCB-laden caulk and light ballasts, the estimated child exposure level exceeded the reference oral dose for Aroclor 1254, and over 70% of the dose was ascribed to inhalation (Thomas et al. 2012). The US EPA has struggled to develop an informative human health risk assessment for inhaled PCBs because there is little information on the uptake of PCBs from the lung and a paucity of data on metabolism, excretion, and dose-specific toxicologic effects (K. Thomas, US EPA, personal communication).

Our previous inhalation studies showed that although airborne PCBs were susceptible to fast clearance, the elimination rates differed substantially by target organ and by the substitution pattern of congeners (Hu et al. 2010). Selective uptake and elimination resulted in distinct congener spectra in body tissues after longer-term inhalation exposure (Hu et al. 2012). In particular, PCB11 was metabolized to its major hydroxylated metabolite, 3,3’-dichlorobiphenyl-4-ol (4-OH-CB11) rapidly in liver. Both PCB11 and 4-OH-CB11 were eliminated exponentially within hours postexposure (Hu et al. 2013). However, many uncertainties remained, including the pulmonary absorption efficiency, the distribution and excretion of the compound and specific biotransformation products. Therefore, we performed a mass balance study using intratracheal instillation of radiolabeled PCB11 that addresses the time course of uptake and distribution into 34 tissues, 5 body cavities and excreta, tracking the parent compound and four classes of metabolites.

2. Methods and Materials

2.1. Chemicals

¹⁴C-labelled PCB11 (specific activity: 30.7 mCi/mmol) was synthesized (Moravek Biochemicals, Brea, CA). High-performance liquid chromatography analysis with and without co-injected standard and further confirmation by gas chromatography–mass spectrometry revealed a radiochemical purity of 99.9% and a compound purity of 99%. [¹⁴C]PCB11 was dissolved in hexane and diluted with saline after adding Tween®80 as a surfactant. Sterile saline was added in five aliquots and the mixture was sonicated after each aliquot to form an emulsion. The final PCB11 solution for instillation contained 1% hexane and 0.1% Tween80 in saline. Sources of other reagents used in this study are listed in Supplement Material.

2.2. Animal Protocol

Animal protocols were approved by the Institutional Animal Care and Use Committee. Eight-week-old male Sprague-Dawley rats (Harlan, Inc., Indianapolis, IN) (261.7±4.8 g) were exposed to [¹⁴C]PCB11 (100 μL) by intratracheal instillation under anesthesia by 3% isoflurane concentration in air for 3 min using a precision vaporizer. The animals recovered within 2 min after the dosing and were serially euthanized 25, 50, 100, 200 and 720 min postexposure. Single animal was used for each time point except 50 min (n=2) and 200 min (n=3). To assure immediate capture of exhaled PCB11, one additional rat was euthanized 12 min postexposure and was dosed similarly with [¹⁴C]PCB11 by inserting a needle through the exposed tracheal wall while anesthetized animal was positioned within the exposure chamber. All animals were provided with food (sterile Teklad 5% stock diet, Harlan, Madison, WI) and water (via automatic watering system) ad libitum until exposure. Most exposures occurred four hours after starting the light phase of the 12-h light-dark cycle and the animals were not fed thereafter. The only exception was the 720-min rat, which was exposed four hours after the beginning of the dark phase and was provided food and water ad libitum in a metabolic cage. One of the 200 min-exposure rats received a dose of 22.5 μg, and one of the 50-min rats received a dose of 18.0 μg. The rest of rats received 35.6 μg [¹⁴C]PCB11.

During the postexposure period, most animals (except the 720-min rat) were housed in a 2 L glass chamber (Figure S1, S:Supplementary Data) and exhaled air, urine and feces were collected. Air was drawn through the chamber at 6.0 L/min into a cartridge containing fresh Amberlite XAD-2 polymeric absorbent resin (Supelco Analytical, Bellefonte, PA) to capture exhaled PCB11. Rats rested on a support screen above the bottom of the chamber, allowing separation of excreted urine from feces as well as preventing ingestion of excreta. Urine was continuously collected via drain tubes. The excreta and exhaled air were collected similarly from the metabolic cage for the 720-min rat.

Rats were euthanized with isoflurane followed by cervical dislocation. Whole blood was collected via cardiac puncture. Serum and erythrocyte samples were prepared. Retroperitoneal and mesenteric visceral fat and subcutaneous fat were sampled as adipose tissue. Superficial masseter, triceps, and biceps femoris were sampled for the muscular system. Technical replicates of specimens (n=2-6 per tissue or digestive matter, 50-150 mg) were collected individually for most tissue types in the animal (Table 1). The digestive matter was collected separately from five sections of the gastrointestinal (GI) tract and vigorously mixed with deionized water. Specimens were solubilized overnight before 100-300 μL 30% hydrogen peroxide decolorizer was added. Alkaline samples were treated with 100 μL glacial acetic acid to eliminate chemiluminescence. Radioactivity was then measured by liquid scintillation counting (Beckman LS 6500, Fullerton, CA) after adding 10 mL scintillation cocktail. Counting efficiency was determined from quench correction curves obtained from [¹⁴C]-toluene standard (PerkinElmer, Waltham, MA). This process resulted in 38-41 samples from each of 9 rats.

Table 1.

[¹⁴C] concentration (dpm/mg wet weight) and estimated percentage of the administered dose (mean and SE) in each specimen in rats (n=3) at 200 min postexposure to [¹⁴C]PCB11^a.

Specimen	Conc.	% of dose	Specimen	Conc.	% of dose	Specimen	Conc.	% of dose
Respiratory System			Liver and Digestive System			Urinary System
Trachea	62±5.0	0.06±0.01	Liver	27±4.0	2.8±0.42	Kidney	30±1.8	0.48±0.02
Bronchi	108	0.02	Tongue	--	--	Urine	400±41	13±1.8
Lung	23±4.4	0.22±0.04	Salivary gland	9.8±2.1	0.06±0.01	Reproductive System
Diaphragm	11±3.2	0.04±0.01	Esophagus	8.6±3.3	0.01±0.01	Testis	4.5±1.8	0.11±0.05
Exhalation	N/A	0.24±0.11	Stomach	35±31	0.34±0.29	Head of epididymis	44±8.0	1.04±0.09
Cardiovascular System			Duodenum	33±9.6	0.36±0.27	Seminal vesicles	1.9±0.9	0.02±0.01
Blood	7.7±0.9	1.14±0.13	Jejunum and ileum	131±69	2.2±0.4	Endocrine System
Serum	13±0.4	0.92±0.03	Rectal ampulla	247±91	0.78±0.29	Pituitary gland	14±2.9	0.02±0.01
RBC	4.3±1.3	0.32±0.09	Colon	25±4.6	0.17±0.03	Thyroid	5.9±3.0	0.00±0.00
Heart	7.1±0.8	0.05±0.01	Pancreas	13±3.8	0.06±0.01	Adrenal gland	--
Carcass and Brain			Digestive matter inside the lumen			Lymphatic System
Skin	33±9.6	13±4.0	Stomach	113±111	3.3±3.5	Thymus	2.9±0.6	0.01±0.00
Muscle	2.9±0.2	3.2±0.21	Duodenum	84±41	0.74±0.64	Spleen	4.9±0.6	0.02±0.00
Adipose	49±8.0	7.3±1.3	Jejunum and ileum	637±141	15±3.7	Lymph nodes	12±2.1	I.W.
Bone	0.8±0.4	0.14±0.08	Rectal ampulla	1026±280	21±4.6	Bone marrow	5.8	I.W.
Brain	1.9±0.3	0.03±0.01	Colon	94.5±8.8	4.2±2.0	Total % of dose		91±1.4

^aBold typeface indicates over 3% of dose.

2.3. Separation of the Parent Compound and the Metabolites

PCB11 and its metabolites were extracted from additional tissue samples and separated into fractions following the procedure in Figure S2. Tissues and digestive matter were homogenized in 2 mL hexane/acetone (1:1 vol/vol) twice and then were twice extracted with each of the following: 1.5 mL dichloromethane, 1.5 mL chloroform/methanol (1:1 vol/vol), 1.5 mL methanol/water (1:1 vol/vol), and 1.5 mL water. Rats euthanized at 12 and 100 min had sufficient urine for metabolite separation which was extracted similarly. Supernatants from extraction with organic solvents were washed twice with 0.5 mL water to isolate water-soluble compounds. Supernatants from the hexane/acetone extraction were partitioned as follows: OH-PCBs were extracted into the alkaline aqueous phase with 2 mL 0.5M potassium hydroxide in 50% ethanol and were washed with 3×400 μL hexane which was transferred back to the hydrophobic phase. The hydrophobic phase was further partitioned after rinsing with 400 μL 1M hydrochloric acid. The potential MeSO₂-PCB was separated from the parent compound with 0.5 mL anhydrous dimethylsulfoxide. The dimethylsulfoxide fraction was washed with 3×400 μL hexane and then combined with the hexane fraction containing PCB11. The radioactivity was measured for each isolated fraction from selected tissues and digestive matter specimens.

2.4. Data Analysis

The [¹⁴C] concentration for each type of tissue or digestive matter was obtained as the mean concentration (dpm/mg wet weight) of sampled specimen. The radioactivity of the whole organ was calculated from the concentration and the organ weight. One exception was the lung for which radioactivity was obtained as the sum of individual lobes. To estimate the radioactivity of the animal carcass, the concentration of skin, adipose tissue, muscle, bone and blood were measured, while the weights were estimated respectively as 16%, 6%, 44%, 7% and 6% of BW based on our dissection measurements and literature values (Calder 1984). Data for the rat receiving the lower dosage (22.5 μg and 18.0 μg) were multiplied by factor of 1.58 and 1.98, respectively to normalize to the dose administered to the other rats.

3. Results

3.1. Time Course of Distribution

The tissue distribution of [¹⁴C]PCB11 and metabolites changed markedly within hours after exposure (Figure 1). Radioactivity appeared in the blood circulation within 12 min postexposure, most significantly in muscle and liver. Elimination soon started as the bulk of radioactivity appeared in the lumen of the duodenum (25 min), moved through the lower GI tract (50-200 min) and was excreted in feces by 720 min postexposure. Urinary excretion showed significant contribution after 100 min postexposure. Meanwhile, [¹⁴C]PCB11 and metabolites decreased rapidly in blood, liver and muscle (12-100 min), whereas accumulation was seen in skin and adipose tissue and followed by slow elimination (200-720 min) . While 96% of the radioactivity was in tissues at 12 min, 37% was in tissues at 200 min and only 10% remained 720 min postexposure.

Figure 1.

Time course of radioactivity disposition in rats after lung exposure to [¹⁴C]PCB11. The color scale indicates the estimated percentage of dose in each type of specimen.

A closer look at the [¹⁴C] concentration at 200 min revealed that the radioactivity was detectable in all specimen types, yet the distribution was highly uneven (Table 1). At 200 min, Lung contained moderate concentrations, with the whole respiratory system accounting for less than 0.30% of the dose. In contrast, the highest concentrations were found in intestinal digestive matter. Voided urine contained a comparable concentration. Tissue concentrations were high in adipose tissue, head of epididymis and skin; intermediate in kidney and liver; and lower in other tissues except for a few sections of the proximal ileum (Table S1).

The variances of tissue concentrations and mass percentages in the triplicate rats necropsied at 200 min were very small (Table 1). Further, 57% of the administered dose had been discharged into the lower GI tract and urine by 200 min, indicating that additional rats should be studied at shorter intervals to capture the distribution and elimination. Therefore, considering the high reproducibility of the experiments and the extremely high cost and scant amount of [¹⁴C]PCB11 that was synthesized, we decided to investigate responses in individual animals at shorter time points postexposure (12, 25 and 100 min). The small variances from duplicate rats at 50 min also justified our decision (Table S2).

The time course of the dynamic distribution of [¹⁴C]PCB11 and metabolites among 41 tissues and body compartments (categorized to 9 groups) is presented in Figure 2 and Table S3. The exposure chamber was washed with water followed by hexane/acetone to measure the residues on the wall. The radioactivity in the water rinse accounting for 0.35% of the dose (median, range 0.00-2.49%) likely came from water-soluble urinary metabolites of [¹⁴C]PCB11. Thus, it was combined with the urine results. The organic solvent rinse of the chamber contained 0.04% of the dose (range 0.01-0.06%), representing the least polar parent [¹⁴C]PCB11 which adhered to glass surfaces. This was combined with the air data.

Figure 2.

Mass percentage of administered dose in tissues, digestive matter and excreta at six time points after lung exposure to [¹⁴C] PCB11. Colors delineate organ systems and bar patterns indicate tissue types as shown at the bottom.

The total recovered radioactivity averaged 91±3.5% of the administered dose, indicating a successful mass balance (Figure 2 and Table S3). The exhaled breath consistently contained very little radioactivity (0.16±0.03% of the dose). The absorbed [¹⁴C]PCB11 and metabolites diffused into every tissue within 12 min postexposure. The rate and extent of uptake varied among individual organs – liver contained relatively more compounds than other tissues whose contribution reflected their tissue weight (e.g. muscle). Most tissues experienced major reduction in radioactivity 12 to 25 min postexposure, especially in respiratory system. The radioactivity was reduced by 70% in lung, over 90% in trachea and 50% in liver and muscle. After 25 min, most tissues maintained an insignificant portion of the radioactivity. Blood, kidney, liver and muscle continued a downward trend, although at a slower rate. The changes of skin and adipose tissue fluctuated until reaching a stable level after 100 min and then slowly declined during 200 and 720 min (Figure 2 and Figure S3). In contrast, large quantities of [¹⁴C]PCB11 and metabolites (30% of the dose) accumulated in the duodenum digestive matter, moved to jejunum and ileum content and continued to increase to until reaching a peak (44% of the dose) at 100 min. The majority then moved to large intestinal content (25% of the dose) at 200 min and to fecal excreta (35% of the dose) at 720 min. In addition, 28% of the dose was in intestinal content that was about to be excreted soon after 720 min.

3.2. Time Course of Elimination

Although some tissues did not carry significant quantities of radioactivity due to their low proportion of body weight (BW) (Figure 2), they contained exceptionally high [¹⁴C] concentrations (Figure 3A). At 12 min, the concentrations in the lung, trachea, and thyroid were all very high (197, 1582 and 247 dpm/mg, respectively). Other organs and tissues containing notable concentrations included liver, heart, pancreas, brain and diaphragm. These tissues were exposed to high internal doses for 25-100 min, potentially posing an increased risk for organ-directed toxicities.

Figure 3.

Time course of [¹⁴C] concentration change in most types of body tissues (A); urine and kidney (B); and sections of gastrointestinal tract and digestive matter (C and D), after lung exposure to [¹⁴C] PCB11. Error bars (shown only positive) represent standard deviation for 50 min (n=2) and standard error for 200 min (n=3).

The concentration decline of [¹⁴C]PCB11 and metabolites was not solely first order (Figure 3A). For example, the [¹⁴C] concentrations of several tissues (e.g., trachea, lung, salivary gland, heart and spleen) dropped markedly between 12 and 50 min with reduction of more than 80% and stabilized between 50 min and 200 min. Nevertheless, elimination in these tissues continued as lower concentrations were seen at 720 min. Two-phase exponential decay showed that initial elimination phases were extremely short with t_½ of 9-33 min while the subsequent phases were longer (up to 10 h, Table S2).

The urinary concentration had a dramatic 60-fold increase between 12 and 25 min and then stabilized (Figure 3B). However, urine collected from bladder at 720 min had only 15% of the concentration compared to urine excreted over the 720 min, indicating a downward trend. The cumulative excreted radioactivity increased as the urinary output increased with time. By 720 min, 18% of the dose had been discharged through urine (Figure 2). The [¹⁴C] concentration in kidney also peaked at 25 min and then declined (Figure 3B). Concentrations in small intestines peaked and dropped in the same pattern as the digestive matter (Figure 3C), while the concentrations of large intestines increased as the compounds accumulated in the digestive matter (Figure 3D). However, the total radioactivity in the GI tissue remained low, accounting for only 2.2 – 8.0% of the dose (Figure 2).

Our findings that PCB11 was eliminated rapidly via urine and feces suggested that metabolism of PCB11 played a major role in the clearance, leading us to investigate the metabolic profiles of the [¹⁴C]-labeled compounds.

3.3. Metabolism of PCB11

We hypothesized that potential metabolites of PCB11 would include OH-CB11s (phase I metabolites), phase II metabolites (sulfated and glucuronidated conjugates), and MeSO₂-CB11, however, only 4-OH-CB11 had been previously identified as a metabolite in vivo (Hu et al. 2013). To explore this hypothesis, we employed a series of solvents with increasing polarity that allowed separation of compounds based on their lipophilicity. PCB11, OH-CB11s and MeSO₂-CB11 were mainly extracted in hexane/acetone mixture and further separated (Bergman et al. 1992; Kania-Korwel et al. 2009). Recovery experiments using spiked authentic standards showed efficient extraction and good separation of PCB11, 4-OH-CB11, 4-OH-CB159 and 4’-MeSO₂-CB87 (Figure S4). Our experiment also showed that subsequent dichloromethane extraction increased the recovery of 4-OH-CB11 to 100%. None of the above compounds were detectable after dichloromethane extraction, indicating that they were separated from the more polar phase II metabolites. Our extraction system (Figure S2) was applied to investigate the time course of PCB11 and metabolite concentrations in liver, blood and intestinal matter. Although we call each fraction by the presumptive compound name, we note that separation might not be complete: We did not have authentic standards for the presumptive metabolites other than 4-OH-PCB11 and did not use other analytical techniques to identify compounds in tissue samples.

Change in the radioactivity of metabolites mirrored the total radioactivity although the rates varied largely among PCB11 and metabolites (Figure 4). In liver, PCB11 concentration declined rapidly from 44 to 6 dpm/mg and maintained this low level after 25 min. MeSO₂-CB11 followed the same trend at about half of the PCB11 concentrations, while OH-CB11 experienced a slower elimination. The level of phase II metabolites was surprisingly high even 12 min postexposure. The elimination, which did not start until 25 min postexposure, consisted of an initial fast and a subsequent slow phase (Figure 4A). The phase II metabolites remained the major components across all time points, after 12 min. The clearance of hepatic radioactivity appeared to be driven by the elimination of the parent compound before 25 min and then by the elimination of phase II metabolites. The dynamic change of metabolic profile in blood (Figure 4B) was very similar to that in liver.

Figure 4.

Time course of [¹⁴C] concentration change in fractions representing the parent PCB11 and its metabolites extracted from rat tissue and digestive matter after lung exposure to [¹⁴C]PCB11. Error bars represent standard deviation for 50 min (n=2) and standard error for 200 min (n=3).

The profile in jejunum and ileum matter (Figure 4C) was consistent with liver and blood in the overwhelming dominance of phase II metabolites, which rose dramatically from 25 to 50 min, and reached 1762 dpm/mg at 100 min and then quickly declined to 433 dpm/mg at 200 min. However, in spite of the extensive metabolism that occurred, a small amount of PCB11 was not metabolized but was excreted into large intestines (Table 3). In contrast, almost none of the parent PCB11 was detected in urine at 12, 100 or 720 min postexposure. The urinary radioactivity comprised only phenolic and phase II metabolites, suggesting that other lipophilic compounds were likely reabsorbed in the kidney (Table 2). It was also interesting that unlike liver, blood or jejunum and ileum matter, OH-CB11 accounted for 15-20% of radioactivity in urine.

Table 3.

Estimated radioactivity (nCi) of isolated fractions containing the putative group of compounds extracted from excreta, and digestive matter in large intestines, and adipose tissue postexposure to [¹⁴C]PCB11.

Putative Fraction	Urine			Feces	Digestive Matter in Large Intestines		Adipose Tissue
	12 mina	100 mina	720 mina	720 mina	Colon 200 min	Ampulla 200 min	200 min
PCB11	0	0	1	73	4	12	185
MeSO2-CB11	0	0	2	27	2	8	49
OH-CB11	1	48	129	606	67	229	8
Phase II metabolites	4	280	739	701	129	492	21
Unextractable	0	0	0	301	94	348	1

^aCumulative radioactivity excreted during the time since exposure.

Table 2.

Two-phase biological half lives of [¹⁴C]PCB11 and metabolites in organs and tissues that showed elimination after lung exposure to [¹⁴C]PCB11. Values were calculated by first-order kinetics. Phase 1 biological half lives (t_½-1) were estimated by using tissue [¹⁴C] concentrations at 12 min, 25 min and 50 min, while phase 2 half lives (t_½-2) were estimated by 50 min, 100 min, 200 and 720 min data.

Organ/Tissue	Phase 1	Phase 2
	t½-1	t½-2
Trachea	9 min	2.8 h
Thyroid	14 min	6.1 h
Lung	13 min	4.1 h
Liver	24 min	4.1 h
Heart	12 min	4.3 h
Pancreas	21 min	9.5 h
Brain	12 min	2.9 h
Diaphragm	18 min	4.3 h
Blood	33 min	4.6 h
Salivary gland	14 min	4.8 h
Spleen	15 min	7.5 h
Thymus	14 min	5.4 h
Muscle	14 min	6.4 h
Testis	17 min	4.3 h
Seminal vesicles	19 min	4.6 h

Metabolic analysis revealed that the major component in lung was parent PCB11 at 12 min. While PCB11 quickly declined to minimal levels, the levels of metabolites remained low showing no sign of retention (Figure 4D and Table S6). On the contrary, radioactivity peaked at 200 min postexposure in adipose tissue (7.3 % of the dose). Although the parent compound accounted for 70% of total radioactivity (Table 2), MeSO₂-PCB was also detected at significant levels. The retention of methylsulfone in adipose tissue was likely due to high lipophilicity, as its octanol-water partition coefficient is only slightly lower than parent PCB (Letcher et al. 2000).

4. Discussion

The primary objectives of this study were to: (1) determine the uptake efficiency of PCB11 in the lung; (2) assess the whole-body distribution; (3) determine the elimination kinetics in major organs; and (4) elucidate the excretion pathways and metabolic fate. We suggest that the investigation of PCB11 – a non-Aroclor congener that prevails in the atmosphere and is still being produced – can provide a suitable model for the biological fate of other prevalent airborne PCBs and improve our understanding of their current health risks.

The total radioactivity recovered in excreta, expired air, organs and the carcass demonstrated a [¹⁴C] recovery of 91±4%. The variability in radioactivity was likely due to: a) content in tissues that were not included in the calculation, such as bone marrow and lymph nodes due to their indeterminate weight; b) low radioactivity tissues that were not routinely measured (e.g., tail in which we found 0.25% of the administered dose at 12 min); and c) possible inaccuracy in proportion of BW for muscle and blood. In contrast to digestive matter, muscle and blood (plus most other tissues) had their highest [¹⁴C] concentrations at 12 min, which may explain why this time point had lower recovery.

We noticed that the digestive matter in stomach had an aberrantly high concentration of radioactivity at 50 min, associated with a high esophageal level. Similar results were seen in a biological replicate of this time point (Table S2). Metabolic analysis showed that the major component in stomach matter was the parent PCB11 (79%), raising the possibility that the excessive radioactivity was not from duodenum reflux but more likely oral uptake of PCB11. However, the concentration of the stomach tissue, the lung and the trachea were all comparable to the levels at other time points and were consistent with the time course trend (Table S4). Therefore, we speculated that PCB11 found in the stomach matter arose from swallowing after the mucociliary clearance of the [¹⁴C]PCB11. To further confirm this conjecture, we delivered the dose to the 12-min rat by the transtracheal instillation instead of intratracheal instillation. In this manner, the esophageal opening was bypassed and potential oral exposure from the injection was precluded. Nevertheless, we detected a high concentration in the esophagus, even higher than the 50-min rat, clearly showing that a small amount of the compound could be removed from the airway by mucociliary clearance and swallowed. The lack of downward trend of esophageal concentrations (Table S4) indicated that the esophageal concentrations came from mucociliary clearance pathway instead of diffusion from trachea. Assuming a tracheal mucous velocity of 1.75 mm/min in rats (Wolff 1985), the mucociliary escalator can transport inhaled debris about 21 mm in 12 min which is over 70% of the rat tracheal length (Tenney and Bartlett Jr. 1967). This provides further plausibility to this uptake pathway.

Uptake into respiratory bronchioles and alveoli has been considered highly efficient for lipophilic xenobiotics due to the high perfusion rate and close contact of air to capillaries in alveoli (Bend et al. 1985). Yet despite increasing recognition of the importance of the inhalation pathway, the pulmonary absorption fraction for PCBs remains unknown. In a study of PCBs in school buildings, an absorption fraction of 70% was assumed (Thomas et al. 2012). Our finding that the pulmonary uptake of PCB11 was nearly complete (99.8%) following intratracheal instillation was in line with our previous nose-only inhalation study (Hu et al. 2010). Compared to PCB11, the order-of-magnitude higher lipophilicity of high-chlorinated congeners and the rather comparable water solubility imply a more extensive uptake of higher-chlorinated congeners at equilibrium, as indicated by the octanol-air partition coefficient (Meylan and Howard 2005). Therefore, we propose that a pulmonary absorption fraction of 100% can be reasonably assumed for dose estimation and risk assessment of airborne PCB exposures.

Our study uniquely shows that body tissues take up PCB11 after inhalation exposure at a very fast rate with absorption phases for most tissues as short as a few minutes. This rate was much faster than those reported for orally administered congeners and Aroclors in mice, sheep and swine, wherein the serum concentrations did not reach their maximum until 2 to12 hours after a single oral dose (ATSDR 2000). Yet absorption after intravenous exposure showed comparable rates, with the lower-chlorinated congener concentrations peaking in rat muscle and liver within 15 min (Matthews and Anderson 1975). A slower uptake in the peripheral tissues (skin, adipose tissue, and head of epididymis) was found. Previous studies suggested the uptake of PCBs in adipose tissues depended on the type (subcutaneous vs mesenteric) and the preference varied by congeners (Saghir et al. 1999; Saghir et al. 2000). Similarly in our study, examination of tissues taken from multiple locations revealed uneven distribution in adipose tissue (37-86 dpm/mg, Table S5) and skin (25-127 dpm/mg) samples at 25 min, followed by redistribution and equilibration. Factors such as enterohepatic circulation or metabolite distribution into adipose tissue at later time points may account for the variance at early time points since at 200 min 30% of the radioactivity represented metabolites. In the above-mentioned study of intravenously injected monochloro- and dichlorobiphenyls, the fraction of metabolites in adipose radioactivity increased over time, suggesting a delayed distribution of metabolites compared to the parent compounds (Matthews and Anderson 1975). The uptake of [¹⁴C] in the head of epididymis was very similar to adipose tissue and its high concentrations were in line with previous studies suggesting a risk for male reproductive toxicity (Cai et al. 2013; Faqi et al. 1998).

The observation that the compounds were eliminated in a two-phase fashion in most tissue types was surprising. In two of our previous acute inhalation studies, we found single-phase first-order elimination for various low-chlorinated Aroclor congeners (Hu et al. 2010) including PCB11 (Hu et al. 2013). The exposure regimen in both studies was 2 h nose-only inhalation which precluded capturing the initial elimination. Single-phase elimination of dichlorobiphenyls was reported in rats after 5-day continuous oral exposure in contrast to the two-phase elimination of tetra- and higher-chlorinated congeners (Tanabe et al. 1981), yet the earliest sampling time in that study was 3 h postexposure. The two-phase elimination featured by fast and slow components has been widely found for orally- and intravenously-administered higher-chlorinated congeners whose initial t_½ last hours to a few days and subsequent t_½ last weeks and longer (Birnbaum 1983; Tanabe et al. 1981). Our study suggests that the inhaled lower-chlorinated congeners may follow a similar pattern, but at a faster rate. Interestingly, unlike other low-chlorinated PCBs (Hu et al. 2010), PCB11 was eliminated relatively faster in the brain than in other low lipid-content tissues (Table 2). One possible reason was binding of PCB11 or its metabolites to active transporters crossing blood-brain barrier. PCB11 sulfate is found capable of competing with thyroxine (T₄) for binding to its transporter transthyretin (TTR) (Grimm et al. 2013). As T₄ is mainly transported from brain to blood, this carrier-mediated mechanism which is generally much faster than simple diffusion (Banks 2012), may explain the fast elimination of PCB11 as opposed to ortho-substituted congeners that accumulated in the brain (Kodavanti et al. 1998; Saghir et al. 2000). On the other hand, potential binding of PCB or metabolites to proteins like hemoglobin (Tampal et al. 2003) may contribute to longer retention in the spleen possessing large storage of hemoglobin (Table 2).

Physiological differences between rats ant human should be noted, as rodents have higher relative CYP amount (Martignoni et al. 2006) and oxygen consumption (Tenney and Remmers 1963) per unit of body weight, and consequently faster elimination of xenobiotics. The differences in clearance can be significant for highly lipophilic compounds as human body contains a 3-4 times higher fraction of adipose tissue, leading to much longer biological half-lives (Sarver et al. 1997). Moreover, children exposed to PCBs may experience slower elimination than adults, as suggested by age effects in animal studies (Saghir et al. 1999). Thus elimination of PCB11 in human, especially children, may be slower than what we found in rats.

The excretory pathways of commercially manufactured PCBs after oral administration have been extensively studied while non-Aroclor congeners and pulmonary administration have not. Most orally-administered PCBs were eliminated mainly via bile and favor fecal excretion relative to urinary excretion (ATSDR 2000). Our finding that the majority of PCB11 and metabolites was found in digestive matter and/or feces agreed with this rule. The transit rates of GI tract content were found rapid in rat (Kirman et al. 2012), providing plausibility that PCB11 and metabolites could reach the rectum within a few hours if they were solely excreted via bile. However, it is possible that pathways other than bile excretion may contribute, for example, secretion by active transporters on intestinal wall as suggested with decabromodiphenyl ether in rat (Mörck et al. 2003) or reabsorption of deconjugated metabolites in enterohepatic circulation (Gustafsson et al. 1981; Roberts et al. 2002).

On the other hand, the chlorination of congeners likely influences the propensity for urinary excretion. In an early study, a decreasing proportion of the intravenously injected radioactivity was excreted in urine with increasing number of chlorines from one to six (Matthews and Anderson 1975). Our metabolic analysis showing no PCB11 or MeSO₂-CB11 in urine suggested that the metabolism to hydrophilic compounds may be a prerequisite for urinary excretion whereas unaltered PCBs have been identified in postexposure feces (Wehler et al. 1989) and also in small amounts in our study. For lower-chlorinated congeners that are more readily metabolized, the ability to utilize the urinary pathway was enhanced promoting more rapid excretion (as soon as 12 min) than the fecal pathway that required more than 200 min.

The metabolism of PCB11 occurred so rapidly that as early as 12 min postexposure. Unlike some other congeners (Lund et al. 1985; Stripp et al. 1996), methylsulfonyl PCB11 was not retained in lung. Instead, the radioactivity was mainly contributed by the phase II metabolites in blood circulation. The liver and the digestive matter in jejunum and ileum were also dominated by phase II metabolites, whereas in feces, there was a relatively greater representation of phase I metabolite (Table 3). This change could be attributed to (1) the macromolecular binding in the ampulla and colon digestive matter as shown by the high unextractable levels of [¹⁴C] in our study and in an earlier report (Bergman et al. 1992); (2) faster excretion of reabsorbed phase II metabolites via urine; and (3) enterohepatic circulation with cleavage of the conjugated functional groups. Nevertheless, the dominance of phase II metabolites was similarly found after intraperitoneal injection of 4-chlorobiphenyl (PCB3) (Dhakal et al. 2012), suggesting that phase II metabolites may be better blood or urinary biomarkers for assessment of human exposure to lower-chlorinated PCBs than the parent compound or phase I metabolites.

Twenty percent of [¹⁴C] remained bound to liver and blood cells 200 min postexposure (unextractable fraction, Table S4), likely attributable to cellular macromolecule adducts formed during metabolism of lower-chlorinated PCBs (Lin et al. 2000; Pereg et al. 2001; Tampal et al. 2003). The abundance and persistence of these genotoxic and mutagenic adducts (Lauby-Secretan et al. 2013) accentuates the concern for long-term health risks.

5. Conclusions

This manuscript is the first mass balance study to address uptake, distribution and metabolism of PCB11 and thus answers major environmental health questions, including: what uptake proportion should be used to calculate the reference concentration for risk assessment; do PCB11 and similar lower chlorinated congeners bioaccumulate; and what metabolites can best serve as biomarkers for exposure? Our study shows that PCB11 is 99.8% absorbed after lung exposure and distributed within 12 minutes to 41 body tissues, GI tract compartments and excreta. The majority of [¹⁴C]PCB11 and metabolites experienced fecal elimination driven by hepatic metabolism and biliary excretion, while a smaller fraction was excreted via urine. In most non-adipose tissues, the initial elimination phase of PCB11 was exceptionally short (t_½=9-33 min). PCB11 and OH-CB11s decayed to minimal levels within 25 min of exposure. Thus, phase II metabolites rather than PCB11 and OH-CB11s are better biomarkers of PCB11 exposure.

Highlights.

• ❖ We performed a mass balance study in rats after delivering [¹⁴C]PCB11 to the lung.

• ❖ The absorption of PCB11 in lung was rapid and complete (99.8%).

• ❖ 35% of the dose was excreted via feces and 18% via urine within 12 h postexposure.

• ❖ Half-lives of initial tissue elimination were ≤ 30 min except for skin and fat.

• ❖ Phase II metabolites dominated and decayed more slowly than other compounds.