TERATOLOGY 42:359-371 (1990)

Route-Dependent Pharmacokinetics, Distribution, and Placental Permeability of Organic and Inorganic Selenium in Hamsters CALVIN C. WILLHITE, VERGIL H. FERM, AND LAUREN ZEISE Department of Health Services, State of California, Berkeley, California 94704 (L.Z., C.C. W.); Department ofAnatomy, Dartmouth Medical School, Hanouer, New Hampshire 03756 (V.H3)

ABSTRACT Inorganic selenium (Se) salts (selenite and selenate oxyanions) and the organic selenoamino acids (selenomethionine and seleniferous grains) are teratogenic and embryolethal in domestic and wild birds. Selenium bioaccumulation has been held responsible for reproductive failure among waterfowl at the Kesterson Reservoir (California), the Ouray and Stewart Lake Wildlife Refuges (Utah), and the Carson Sink (Nevada). Anecdotal field and controlled laboratory reports have implicated Se exposure in mammalian embryotoxicity (including human), but developmental toxicity studies in hamsters failed to demonstrate an adverse response, except a t maternally toxic doses (Ferm et al., Reprod. Toxicol., in press). Uptake, distribution, and elimination of Se after a single bolus equimolar dose (60 pmol/kg) of selenate or selenomethionine by oral or intravenous administration were compared using day 8 pregnant hamsters. Intravenous selenate was eliminated ten times more rapidly from maternal plasma than oral selenate, but concentrated in liver, kidney, and placenta to the same degree. Intravenous (iv) L-selenomethionine achieved lower maximum circulating total [Sel, but it was eliminated more slowly than iv selenate. Larger areas under the plasma and peripheral tissue [Sel:time curve (AUC) after oral or parenteral selenomethionine than after equimolar selenate were consistent with previous studies in rodents and in humans. Embryonic [Se] plateaued at 3 nmol/g after selenate, but embryonic [Sel after selenomethionine continued to accumulate (80 nmol/g) as gestation progressed. The lack of a teratogenic response in hamsters a t doses of either selenate or selenomethionine less than those associated with maternal intoxication cannot be attributed t o lack of Se accumulation in early embryonic and placental tissue. Erosion of marine sediments along California's Coast Range initially concentrated selenium (Se) in the surface soils and shallow groundwaters of the western San Joaquin Valley. The naturally occurring Se is found in the -2,0, + 4, and + 6 oxidation states, and the net charge influences the environmental fate. The relatively immobile selenides and Se sulfides predominate in acid soils, the selenites (Se+4 as the HSe03-l and Se03-2 oxyanions) are sorbed and precipitated by iron oxides in neutral soils, and the highly mobile, water-soluble selenates (Se as SeO,) predominate in the alkaline soils of the San Joaquin, where a t high pH the selenates are not sorbed, even +

0 1990 WILEY-LISS, INC.

to clays (Frost and Griffin, '77; Korte et al., '76). Subsurface flow, agricultural tile drainage, and operational spill waters have redistributed the element such that environmental Se bioaccumulation has become evident, particularly a t the Kesterson National Wildlife Refuge, Merced County (Ohlendorf et al., '86; Hoffman et al., '88). Once in aquatic ecosystems, inorganic Se is

Received January 4, 1990; accepted May 7, 1990. This paper was presented in part at the 1989 Annual Meeting of the American Society for Pharmacology and Experimental Therapeutics (ASPET), Salt Lake City, Utah (Pharmacologist, 31:138, 1989).

360

C.C. WILLHITE ET AL.

converted to organic forms (Chau et al., '76), and aquatic and terrestrial plants incorporate Se in lieu of sulfur into the corresponding amino acids, notably selenomethionine, selenocysteine, and selenocystine. Although nonruminant animals are quite unable to synthesize selenoamino acids (Cummins and Martin, '67; Olson and Palmer, '76), selenomethionine is the major form in plant tissues (Olson et al., '70; Whanger, '86). Embryotoxicity and teratogenicity among domestic fowl (Palmer et al., '73) and aquatic birds (Heinz et al., '87) have been attributed to Se exposure. Ingestion of Kesterson's aquatic plants (Hoffman et al., ,881, controlled feeding (Heinz et al., '871, or direct injection of inorganic or organic Se into the avian air sac (Palmer et al., '73) resulted in malformations of the eye, brain, beak, and limbs. Selenomethionine exhibited greater bioavailability in rats (Whanger, '86) and chicks (Osman and Latshaw, '76; Humaloja and Mykkanen, '861, and it was a more potent teratogen in birds than the inorganic Se salts (Heinz et al., '87; Palmer et al., '73). Although the inorganic and organic forms of Se are confirmed teratogens for avian species, there are conflicting reports on their role in mammalian developmental biology (Clegg, '71; Scott, '73; Lo and Sandi, '80). Field (Rosenfeld,'47; Leipold et al., '74) and laboratory (Rosenfeld and Beath, '54; Schroeder and Mitchener, '71) investigations have implicated Se as a reproductive and developmental toxicant, but additional studies have demonstrated that the inorganic (Holmberg and Ferm, '69; Satoh et al., '81) and organic (Cekan et al., '85; Palmer et al., '73) Se compounds protect against the teratogenic actions of other agents. Oral (PO), intravenous (iv), or continuous parenteral (sc) infusion of selenite, selenate, or L-selenomethionine in pregnant hamsters failed to elicit a significant teratogenic response, save a t doses that produced concomitant, overt maternal intoxication (Ferm et al., '90). The uncertainty that accompanies interpretation and cross species extrapolation of animal teratogenicity testing data can be reduced when population pharmacokinetic parameters are taken into account (Nau, '87). In the case of the Se compounds, human absorption, metabolism, distribution, placental transfer, and elimination data have been published, which facilitates comparisons with the animal data. The objec-

tives of the present study were to obtain the time course for Se as selenate and L-selenomethionine, the forms that predominate in water and plants of the Kesterson area, using pregnant hamsters given a toxicologically relevant exposure and to compare those data with reports in other animals and in humans. The pharmacokinetic model and resulting parameters were compared for a single PO or iv administration of each Se compound using identical vehicle and equimolar dose in animals of equivalent gestational age. MATERIALS AND METHODS

All reagents were of analytical grade. Sodium selenate ~NazSeO,.lOH,Ol (ICN-K and K Laboratories) and L-selenomethionine [CH,-Se-CH,-CH,-CHNH,-COOHI (lot No. 667-0171,Sigma Chemical Go.)were dissolved in demineralized water just prior to administration. The L-selenomethionine was stored at -20°C in the dark. Sodium L7,Se]-selenate and L-[75Se]-selenomethionine were purchased from Amersham. Each radioactive compound was placed into its respective nonradioactive Se solution to produce a radiolabeled agent with final specific activity of 3-4 pCi/kg Se. Pregnant golden Syrian hamsters [Lak: LVG(SYR)] were purchased from the Charles River Breeding Laboratories (Wilmington, MA). The day following the evening of breeding was considered the first day of gestation. All animals were held in individual polycarbonate cages with pine shavings for bedding. The hamsters were fed laboratory stock rodent diet (RMH-3000; Prolab, Agway) containing 0.2 ppm total Se and were given demineralized drinking water ad libitum. Four to six pregnant hamsters were assigned on a random basis to each treatment group. Hamsters were given one of each of the Se solutions at 0.5 m1/100 g maternal body weight by either one of two routes of administration: a single oral intubation or a single injection into the sublingual vein under pentobarbital (ip) anesthesia at 9:00 AM on day 8 of pregnancy (Ferm, '67). This time corresponds to the presomite to early somite stage of development, and it was the stage used for the initial Se teratology study (Ferm et al., '90). The dose of each compound by either route selected for study was 60 pmol/kg, a quantity slightly less than the 5-day maternal iv LD,, for sodium sel-

ORGANIC AND INORGANIC Se IN HAMSTERS

enate in this species. Dams given oral Lselenomethionine were killed by inhalation of excess COz a t 1, 6, or 24 hr postadministration. Dams given either oral or parentera1 selenate or iv L-selenomethionine Se were sacrificed at 0.25, 1,4, or 24 hr postinjection. Maternal blood was collected by cardiac puncture using heparinized vacutainers, and the plasma and erythrocytes were separated after centrifugation. Maternal liver and kidney, placenta, and embryos were collected, weighed, and placed on ice. All embryos from each dam were pooled, and the results are expressed on a total litter weight basis. Radioassays were performed using a Beckman Biogamma 5500 spectrometer with discriminator module readings of 60 and 700 to maximize selectivity for the 75Se signal. Accumulated counts for each sample were such that the counting error was less than 2%. Total gram-atoms of Se were calculated b comparing the net sample counts with a ' 4 e standard of known specific activity. Tissue Se concentrations were calculated as Fmol Se/kg wet weight. A curve was fit to the tissue concentration:time data with a nonlinear regression program (Statistical Consultants, Inc., '86). The data were consistent with a one-compartment model with bolus input, no lag time and first-order elimination [C(t) = D/ V'-det'l where C is the tissue concentration at any time t, D is the initial dose (9 kmol/ animal), V is the volume, and K, is the apparent rate of elimination. Estimates were made for the half-time of elimination (t,,,), the area under the tissue concentration: time curve (AUC),the maximum tissue concentration (C,,,), and in the case of intravascular Se the apparent volume of distribution (Vd). Plasma and erythrocyte clearance and hepatic extraction ratio were calculated from the parenteral kinetic constants (Goldstein et al., '74). RESULTS

The shape of the plasma concentration: time curve following a bolus iv selenate dose (Fig. 1A) reflected initial rapid distribution to peripheral tissues. After initial distribution, a constant fractional loss of total Se per unit time occurred. Extrapolation to zero time yielded a C, estimate (Table 1)that would be obtained were Se distribution essentially instantaneous; the initial mea-

361

sured plasma [Sel (Fig. lA, 15 min) and the calculated value were in general agreement, given the standard deviation for the empiric data. The initial [Se] decline in maternal erythrocytes paralleled that in plasma, with agreement between calculated and observed (Fig. 1B) C, values. Over the 24 hr period after injection, the more rapid elimination from plasma compared to the erythrocytes resulted in a decline in the plasma/erythrocyte Se concentration ratio from 2.7 to 0.7. The highest Se concentrations occurred in the liver (Fig. 1C) and kidney (Fig. 1D). That the Se accumulated initially in liver and kidney was reflected in their relatively large AUC values; the placental AUC was the least of all tissues studied. The shape of the curve for placenta (Fig. 1E) was similar to that for plasma, with peak concentrations 50%those found in the circulation. The elimination rates for plasma and placenta were very nearly identical, with the liver exhibiting the slowest rate and corresponding long elimination half-time. The large apparent Vd (total body Selplasma Se) suggests that selenate was not confined to plasma water or extracellular fluid, but that it was distributed rapidly throughout total body water. The prolonged terminal elimination phase for the plasma appears to reflect Se movement from peripheral tissue to the central compartment. In contrast to all other tissues studied, total embryonic Se showed a prompt increase shortly after exposure, but the subsequent concentrations tended to plateau (Fig. 1F). At 15 min, the ratios of total Se in plasma and placenta to that in the embryo were 66 and 35, respectively. At 24 hr, these ratios had decreased to 1.5. The embryonic Se retention and the failure t o observe an elimination phase precluded calculation of kinetic constants and produced an infinite AUC value. Oral selenate was distributed to all tissues studied and the concentrati0n:time curves were consistent with first-order elimination (Fig. 21, except in the embryo. Oral selenate exposure was associated with lower C,, and large AUC values for all tissues compared to intravascular administration of an identical dose (Table 1). Liver AUC values were nearly identical for iv and PO dosing. The rates of Se elimination from the maternal circulation after oral dosing were only 0.1 that after iv injection, and tl,z for maternal tissues was much longer after

Hours

Hours

LOO

Route. IV Tlssue Liver

Route. Tissue

IV Kldnev

15

20

50

D 5

0

10

Hours 5:

lo 80 o\

4:

Route Tissue

IV Placenta

-

Route

IV

20

E

Sodium selenate I.V. Fig. 1. Tissue distribution and time course of 75Seafter a single iv injection of 60 pmolkg Na selenate in pregnant hamsters.The values represent the mean of five (0.25 and 1.0 hr) or four (4.0 and 24.0 hr) animals 5 S.D. following injection at 9:00 A M on day 8 of gestation.

25

ORGANIC AND INORGANIC Se IN HAMSTERS

363

TABLE 1. Maternal pharmacokinetic parameters of selenium as sodium selenate in vreenant hamsters' Tissue Route/parameter Plasma Erythrocyte Liver Kidney Placenta Intravenous C,, (pmolkg) 169 76 241 173 91 AUC (pmol.hrikg) 323 220 2,500 1,088 183 V, (ml) 530 K, (hr-l) 0.52 0.34 0.10 0.16 0.5 1.32 2.0 7.3 4.4 1.4 t,,2 (hrf 275 41 Clearance (gihr) Extraction ratio 0.59 Oral 32 16 14 104 59 C,, (Kmolikg) AUC (pmol,hr/kg) 620 460 2,490 1,910 430 K, (hr-') 0.05 0.03 0.04 0.03 0.03 13.4 20.6 16.5 22.3 20.6 t,,z (hr) 'Pregnant hamsters received a single iv or PO bolus of 11.3 mg/kg Na,SeO,~lOH,O (60 pmolikg) on the morning of day 8 of gestation, and tissues were collected to 24 hr after treatment. Parameters are as indicated in the text.

gavage than after iv injection. As was the case with intravascular selenate, plasma (Fig. 2A) and placental (Fig. 2E) [Sel decreased with time after an oral dose. Selenium concentrations in plasma were twice those in placenta. Again, the ratio of Se in plasma and placenta to that in the embryo declined over time, but the reduction was substantially less than after iv exposure. However, after either iv (Fig. 1F) or PO (Fig. 2F) selenate, embryonic [Sel increased and then plateaued at mean concentrations of -13 nmol/g wet weight. Peak plasma [Sel after iv selenomethionine (Table 2) was 44% that after an equimolar selenate injection (Table 1).Plasma [Sel failed to show the rapid decline (Fig. 3A) characteristic of selenate (Fig. lA), but instead showed a secondary peak a t 1hr after injection. Plasma and erythrocyte AUC after selenomethionine were 10 times that after equimolar selenate. Large AUC values for all peripheral tissues after iv L-selenomethionine were consistent, with placental [Sel some 50-fold that after equimolar selenate. Although Vd for selenomethionine Se was smaller than for selenate Se, the value still suggests distribution throughout total body water and it remains consistent with Se sequestration in liver, kidney, and other organs. As with iv or PO selenate, liver and kidney AUC values after selenomethionine reflected initial accumulation in those tissues; Se was eliminated over the course of the study such that, by 24 hr, residual selenate [Sel was 10-20% of initial concentrations, but residual hepatic (Fig. 3C) and renal (Fig. 3D) total Se after selenomethionine were 50% of peak values.

The shape of the kidney [Se]:time curve (Fig. 3D) paralleled that for plasma (Fig. 3A). Overall Ke values for all tissues were similar, but the rate of Se elimination from maternal plasma after iv selenate (Table 1) was 25 times that after iv selenomethionine (Table 2). Selenium clearance from the maternal circulation was slower after iv selenomethionine than after iv selenate. In marked contrast to all other tissues, embryonic selenomethionine [Sel continued to increase as gestation progressed (Fig. 3F). Embryonic [Se] a t termination of the study was 1,350% that a t 15 min after injection. Again, because no Se elimination from the embryo was observed, kinetic constants could not be calculated for the target tissue. Peak circulating [Se] occurred within 1-5 hr of a single oral dose of L-selenomethionine (Fig. 4A,B), and calculated plasma and erythrocyte C,, values (Table 2) were in general agreement with the measured values. Empirical Cmaxafter PO selenomethionine in liver (Fig. 4C) and kidney (Fig. 4D) were less than those after iv selenomethionine (Fig. 3C, D).Calculated oral CmaXwere greatest in liver and kidney, with liver and kidney AUC values less than those after iv selenomethionine (Table 2). Although placental C,, after PO selenomethionine was less than after iv selenomethionine, placental AUC was much larger and tIlzwas much longer after oral selenomethionine. This latter measure was associated with prolonged maintenance of elevated [Sel to near peak concentrations (Fig. 4E), in contrast to all other routes and Se forms studied (Fig. 1E). The elimination

364

C.C. WILLHITE ET AL.

PO

Route Tissue

M a t e r r i o l RHC

Route Tissue

Kidney

15

20

160 1

7 4

T

R o u t e . PO Tlssue- Liver

c

PO

l i

l O L - - T - - F - l - - 10

25

Hours

Route Tissue

1

R o u t e - PO Tissue P l a c e n t o

E

?

PO Embryo

-I F

l

Sodium selenate P.O. Fig. 2. Tissue distribution and time course of 75Se after a single PO intubation of 60 pmoYkg Na selenate in pregnant hamster. The values represent the mean of four (1.0 hr) or five (4.0 and 24.0 hr) animals 5 S.D. following intubation at 9:00 AM on day 8 of gestation.

ORGANIC AND INORGANIC Se IN HAMSTERS

365

TABLE 2. Maternal pharmacokinetic parameters of selenium as L-selenomethionine in Dreznant hamsters' Tissue Routeiparameter Plasma Ervthrocvte Liver Kidnev Placenta Intravenous C,, (pmolkg) 74 38 275 212 138 AUC (pmol.hr/kg) 3,208 2,043 7,506 7,666 8,425 V, (ml) 120 K, (hr-') 0.02 0.02 0.04 0.03 0.02 tIrz(hr) 29.9 37.4 18.9 25.1 42.4 2.4 0.005 Clearance ( g h r ) 0.42 Extraction ratio Oral 57 21 219 150 72 C,, (pmolikg) 37,780 AUC (pmol-hr/kg) 2,336 5,861 6,078 4,878 K- (hr-') 0.024 0.004 0.036 0.031 0.002 t,, (hr) 28.7 193 19.3 22.6 362 ~~~~

'Pregnant hamsters received a single iv or PO bolus of 11.7 m g k g L- selenomethionine (60 pmolkg) on the morning of day 8 of gestation, and tissues were collected to 24 hr after treatment. Parameters are as indicated in the text.

rate and prolonged erythrocyte t,,, reflected Se retention (Fig. 4B). Although embryonic [Se] after PO selenomethionine at the early time points (Fig. 4F) was similar to that after iv injection (Fig. 3F), embryonic [Sel at 24 hr after iv exposure was nearly threefold that after an equivalent oral dose. DISCUSSION

The results of the present study illustrate the high oral bioavailability of Se given in organic or inorganic form. The total absorption measured here (selenomethionine = 73%; selenate = 100%)was consistent with previous measures of 60-99% in humans and animals (Lo and Sandi, '80; Mutanen, '86; Patterson et al., '89). Intact selenomethionine is actively transported across the hamster small intestine, the uptake being identical to that for methionine (Spencer and Blau, '62); this is in contrast t,o simple diffusion for the absorption of inorganic Se at dietary levels (Mutanen, '86). The larger AUC values for plasma and peripheral tissues after selenomethionine than after selenate, regardless of the route of administration, were consistent with previous reports of increased Se retention in humans (Thomson et al., '78; Robinson et al., '78) and rats (Millar et al., '73; Whanger, '86) after selenomethionine than after inorganic Se. The time t o peak Se concentrations in human blood (3hr) after oral selenomethionine and their maintenance to 5 hr (Spencer and Blau, '62) were consistent with the hamster (Fig. 4A). The Se retention in maternal liver and kidney after either selenate or organic Se (Figs. 1-4) was consistent with Se distribution patterns in rats, mice, and sheep

given selenite, selenate, selenomethionine, or alfalfa grown in high Se media (Benard et al., '79; Lo and Sandi, '80). When pregnant heifers (Perry et al., '78; Killer et al., '84; Weiss et al., '84) or ewes (Hidiroglou et al., '69) were given oral or intramuscular sodium selenite, fetal [Se] was a t least as high as if not higher than (Koller et al., '84) that in maternal tissues. The ruminant studies are, nevertheless, complicated by the conflicting reports for biosynthesis of selenomethionine by rumen microflora from inorganic Se (Lo and Sandi, '80). Placental transfer of inorganic Se has been reported in rats (Westfall et al., '38; Wang, ,841, mice (Benard et al., '79; Matsumot0 et al., '77; Nishikido and Suzuki, '85; Yonemoto et al., '83; Xaum and Hopeu, '81), and dogs (McConnell and Roth, '641, and total [Sel tended to concentrate in fetal dophin liver (Itano et al., '84). In mice as with hamsters, the route of administration failed to influence tissue [Sel, as might be expected in that iv selenite (10 pmol/kg) increased maternal tissue [Sel but not that in fetus or placenta compared with sc injection of an identical dose (Nishikido and Suzuki, '85). Intravenous selenite (0.75 kmol/kg) produced higher [Sel in mouse placenta than in fetuses at 1hr after dosing (Yonemoto et al., '831, not unlike the situation with iv selenate in hamsters (Fig. lE,F). [75Se]selenomethionine concentrated in fetal mouse (Xaum and Hopeu, '81) and rat (Khait et al., '79) liver, bone marrow, and crystalline lens, with increased accumulation in fetal tissue compared with placenta as term approached. When pregnant rats were given either sodium selenite or selenomethionine

0

5

10

IV

Route Tissue

Maternal P l a s m

15

20

Route Tissue

Maternal

IV

Route Tissue

Kidney

25

Hours

Hours

Route Tissue

IV Liver

IV

50

C

0 0

T 7 - n m 7 - r r r m . T r T r r n - c - m

~

80

:

60

-rrT-,

E I 1

I

I I , !

I I , I I I

,

, I ,

- 8

lirrllli

-

F

S e Ie n o m e t h io n in e I .V. Fig. 3. Tissue distribution and time course of 75Seafter a single iv injection of 60 pmol/kg L-selenomethioninein pregnant hamsters. The values represent the mean of four (1.0 and 24.0 hr) or five (6.0 hr) animals S.D. following intubation at 9:00 AM on day 8 of gestation.

*

FBC

60 :

-

2

Route: PO Tissue: Maternal Plasma

i

50 1

m

i

:

Y

Route. PO Tissue: Maternal RBC

-

>40 f a r r -

Eo: ' 3 0 -: 2 I :

20 :

*l

A

lO,,,,.,",,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,l

15

10

14

20

25

Hours

240

200

Route PO Tissue. Liver

0'160

Y

s; -

E" 120

'

2 ' I a0 100:

Kidney

15

20

'r_

40

c l 50

PO

Route Tissue

1 20

15

10

25

-

0

Hours

90 :

10

Hours

-

Route: PO Tissue: Placenta 80: -

m

:

1

'

-

>70

I

-

a,

; -

2

:

2

: 50

E

-

:

; 4

0

1

I , I .

I I I

~

I 1 1 1 1

~

I-

Selenome t hionine P.O. Fig. 4. Tissue distribution and time course of 75Se after a single PO intubation of 60 pmol/kg L-selenomethionine in pregnant hamsters. The values represent the mean of four (1.0 and 24.0 hr) or five (6.0 hr) animals 5 S.D. following intubation at 9:00 AM on day 8 of gestation.

25

368

C.C. WILLHITE ET AL.

sc, higher [Se] appeared in the offspring of dams given the selenoamino acid (Shearer, '75), an observation similar t o the marked rise in embryonic [Se] with time after L[75Se]-selenomethionine found here (Figs. 3F, 4F) compared with the plateau in embryonic [Sel after oral or parenteral selenate (Figs. l F , 2F). In each controlled study, placental [Sel decreased and embryonic or fetal [Sel increased as gestation progressed. A number of authors have provided evidence for human transplacental passage of Se and have reported total [Sel in maternal blood, placenta, and term umbilical cord blood as part of clinical (Robkin et al., '73; Haga and Lunde, '78; Leffler and Nordstrom, '83; Korpela et al., '84; HyvonenDabek et al., '84) or epidemiologic (Baglan et al., '74; Creason et al., '76) investigations. Under normal conditions at term in humans, maternal blood, cord blood, and placenta exhibit equivalent total [Se]. In the only kinetic study of selenomethionine in pregnant humans, a secondary peak [75Selin maternal blood (associated with plasma proteinbound [75Se]-selenomethionine)occurred at 80 min after a single iv (10 pCi) bolus (Jandial et al., '761, similar to the secondary peak found in maternal hamster plasma a t 60 min after iv selenomethionine (Fig. 4A). Concentrations in human fetal cord plasma (protein bound + free) were essentially equivalent from 10 min to 24 hr after iv selenomethionine, but concentrations of free [75Sel decreased gradually. When the same total dose of [75Sel-selenomethioninewas infused for up to 3.5 hr, fetal cord blood [Sel was no different from that after an iv bolus (Jandial et al., '76). All the studies cited above along with the present protocol are difficult to interpret primarily because of the limited numbers of time points studied and the fact that total [Se] was measured. Goldstein et al. ('74) outlined the difficulties inherent in obtaining clinical and experimental data on xenobiotic transplacental transfer and the practice of sampling cord blood without distinguishing between umbilical venous and umbilical arterial blood. Because no methods are available for cannulation of the embryonic rodent circulation, systematic kinetic studies of rodent embryos or fetuses in utero necessitate single measurements at laparotomy and human rates of equilibration between maternal and cord blood obtained

at delivery require that data be "pieced together from a great many separate experiments" (Goldstein et al., '74). Thus the lack of serial samples from each animal or patient makes the comparisons semiquantitative. Circulating inorganic Se is associated initially with plasma protein, and with time it comes into combination with hemoglobin (Lo and Sandi, '80). Inorganic Se undergoes sequential methylation and reduction reactions in mammals, with an initial nonenzymatic glutathione reaction yielding selenodiglutathione. Hepatic, renal, and pulmonary biotransformation via glutathione and S-adenosyl-L-methionine produces dimethyl and trimethylselenides, the former normally being exhaled and the latter being the major urinary metabolite. The dimethylselenide and trimethylselenonium ions are said to represent detoxication products (Lo and Sandi, '80),and the measurement of total [Sel precludes interspecies comparisons of the rates and significance of Se metabolite formation. However, the present data suggest that, at the very high selenate doses (75-110 pmolkg) used in the teratology study that produced obvious maternal toxicity (Ferm et al., 'go), plasma protein binding was likely exceeded such that free selenate produced toxicity that would not otherwise occur at the lower physiologic or dietary levels. Human elimination t,,, have been computed for organic and inorganic Se administered either iv or PO using single, biphasic, or triphasic exponential decay. Waterlow et al. ('69) calculated that, over the first 24 hr after a single iv bolus of 0.1 pCi [75Sel-selenomethionine to male infants, 4.5% of the dose appeared in urine and 8.1% appeared in feces; oral selenomethionine showed increased fecal excretion, and a t least 78-88% of the dose was retained at 48 hr. Johnson ('77) calculated t,,, for iv [75Se]-selenomethionine for adult males as 75.2 days. After a single iv selenite bolus, Cavalieri and associates ('66) calculated t,,, = 65 days for the 75% of the dose not eliminated within the first 72 hr, and Jereb et al. ('75) estimated biphasic t,,, = 20 and 115 days for iv selenite. After a single oral selenite bolus in young women or men (Thomson and Stewart, '74; Patterson et al., '89), radioactivity appeared in plasma within 30 min and maximum urinary [Se] occurred within 2-4 hr of dosing,

ORGANIC AND INORGANIC Se IN HAMSTERS

patterns not unlike those observed in hamsters given oral selenate (Fig. 2A,D). Thomson and Stewart ('74) found three exponential elimination phases for young women, with tl,Z= 0.8-1.4, 4-14, and 92-143 days, and Janghorbani et al. ('84) reported biphasic elimination in young men, with tl,z = 2 and 162 days. Patterson et ale's ('89) estimates derived using a physiologically based pharmacokinetic model to account for enterohepatic Se circulation found four kinetically distinct plasma compartments (tlIz= 0.2-1, 3-8, 9-42, 200-285 hr), a liver and pancreas compartment, and a deep compartment. These authors concluded that, after 90 days, 35% of the original selenite dose was retained. Wang ('84) utilized a threecompartment physiologic model to compute maternal and transplacental oral (drinking water) selenate kinetic constants in adult and fetal rats, and with whole body (doseindependent) tlIz= 30 days, neonatal circulating [Se] declined only slightly, an effect consistent with the selenate plateau observed here (Fig. 2F). All the whole body exponential tlI2for humans (80-198 days) take into account Se-retaining peripheral tissues, and the plasma and peripheral tissue tlIz reported here (Tables 1, 2) reflect the 24 hr sampling protocol. Comparisons of interspecies kinetic and disposition data are to be facilitated by analyses of the Se toxicity and pharmacokinetic studies in nonhuman primates (Cukierski et al., '89; Tarantal et al., '89). In contrast to inorganic Se, selenomethionine is absorbed intact and incorporated randomly into protein, including that in mammalian cells (see Ferm et al., '90, for review), perhaps accounting for its prolonged retention in maternal and embryonic tissue. The present results illustrate route-dependent, rapid elimination of selenate compared with selenomethionine in the pregnant hamster. The time course for [Se] in maternal hamster plasma, its accumulation in liver and kidney, and placental transfer of both organic and inorganic Se are consistent with the results of studies in other animals and, within the limitations of the protocols, with results of pharmacokinetic studies in humans. The fact that iv or PO selenate or L-selenomethionine did not induce terata at doses that were not associated with maternal toxicity (Ferm et al., '90) cannot be attributed to failure of systemic absorption or lack of Se transpla&ntal transfer:

369

ACKNOWLEDGMENTS

This study was supported by U S . Department of Energy Contract No. B039924 (IMA 85-87088). Mention of a trade name, proprietary product, or specific equipment does not constitute a guaranty or warranty by the U.S. Department of Energy or the State of California. The conclusions expressed herein are those of the authors and do not necessarily represent those of the US. Department of Energy or the State of California. This study was also supported in part by EPA grant R810078 (V.H.F.). The contents do not necessarily reflect those views and policies of the EPA, nor does mention of trade names or commercial products constitute endorsement or recommendation for use. The authors acknowledge the able assistance of Dr. David P. Hanlon.

LITERATURE CITED Baglan, R.J., A.B. Brill, A. Schulert, D. Wilson, K. Larsen, N. Dyer, M. Mansour, W. Schaffner, L. Hoffman, and J . Davies (1974) Utility of placental tissue as a n indicator of the trace element exposure to adult and fetus. Environ. Res., 8:64-70. Benard, P., A.G. Rico, J.P. Braun, and V. Burgat-Sacaze (1979) Etude par autoradiographie macroscopique du passage transplacentaire du selenium chez la souris. Pharmacol. Dev., 89:461-468. Cavalieri, R.R., K.G. Scott, and E. Sairenji (1966) Selenite ['%el as a tumor-localizing agent in man. J . Nucl. Med., 7:197-208. Cekan, E., P. Slanina, K. Bergman, and B. Tribukait (1985) Effects of dietary supplementation with selenomethionine on the teratogenic effect of ionizing radiation in mice. Acta Radiol. Oncol., 24:459-463. Chau, Y.K., P.T.S. Wong, B.A. Silverberg, P.L. Luxon, and G.A. Bengert (1976) Methylation of selenium in the aquatic environment. Science, 192:1130-1131. Clegg, D.J. (1971) Embryotoxicity of chemical contaminants of foods. Food Cosmet. Toxicol., 9:195-205. Creason, J.P., D. Svendsgaard, J . Bumgarner, C. Pinkerton, and T. Hinners (1976 Maternal-fetal tissue levels of 16 trace elements in 8 selected continental United States communities. In: Trace Substances in Environmental Health-X. D.D. Hemphill, ed. University of Missouri Press, Columbia, p. 53-62. Cukierski, M.J., C.C. Willhite, B.L. Lasley, T.A. Hendrie, S.A. Book, D.N. Cox, and A.G. Hendrickx (1989) Thirty-day oral toxicity study of L-selenomethionine in female long-tailed macaques (Mucuca fuscicuZuris). Fund. Appl. Toxicot., 13:26-39. Cummins, L.M., and J.L. Martin (1967) Are selenocystine and selenomethionine synthesized in vivo from sodium selenite in mammals? Biochemistry, 6:31623168. Ferm, V.H. (1967) The use of the golden hamster in experimental teratology. Lab. Anim. Care, 17~452462. Ferm. V.H.. D.P. Hanlon. C.C. Willhite. W.N. Chov. and S.A: Book (1990) Embryotoxicity and dose-response

370

C.C. WILLHITE ET AL.

relationships of selenium in hamsters. Reprod. Toxicol., in press. Frost, R.R., and R.A. Griffin (1977) Effect of pH on adsorption of arsenic and selenium from landfill leachate by clay materials. Soil Sci. SOC.Am. J., 415357. Goldstein, A,, L. Aronow, and S. Kalman (1974) Principles of Drug Action: The Basis of Pharmacology, 2nd ed. Wiley, New York, pp. 199-210, 333-338. Haga, P., and G. Lunde (1978) Selenium and vitamin E in cord blood from preterm and full term infants. Acta Paediatr. Scand., 67:735-739. Heinz, G.H., D.J. Hoffman, A.J. Krynitsky, and D.M.G. Weller (1987) Reproduction in mallards fed selenium. Environ. Toxicol. Chem., 6:423-433. Hidiroglou, M., I. Hoffman, and K.J. Jenkins (1969) Selenium distribution and radiotocopherol metabolism in the pregnant ewe and fetal lamb. Can. J . Physiol. Pharmacol., 47t953-962. Hoffman, D.J., H.M. Ohlendorf, and T.W. Aldrich (1988) Selenium teratogenesis in natural populations of aquatic birds in central California. Arch. Environ. Contam. Toxicol., 17.519-525. Holmberg, R.E., and V.H. Ferm (1969) Interrelationships of selenium, cadmium, and arsenic in mammalian teratogenesis. Arch. Environ. Health, 18: 873-877. Humaloja, T. and H.M. Mykkanen (1986) Intestinal absorption of 75Se-labeled sodium selenite and selenomethionine in chicks: Effects of time, segment, selenium concentration and method of measurement. J. Nutr., 1I6:142-148. Hyvonen-Dabek, M., P. Nikkinen-Vilkki, and J.T. Dabek (1984) Selenium and other elements in human maternal and umbilical serum, as determined simultaneously by proton-induced x-ray emission. Clin. Chem., 30t529-533. Itano, K., S. Kawai, N. Miyazaki, R. Tatsuka, and T. Fujiyama (1984) Mercury and selenium levels at the fetal and suckling stages of striped dolphin, Stenella coeruleoalba. Agric. Biol. Chem., 48:1691-1698. Jandial, V., P. Henderson, and I. MacGillivray (1976) Placental transfer of radioactive selenomethionine in late pregnancy. Eur. J . Obstet. Gynecol. Reprod. Biol., (2295-300. Janghorbani, M., L.J. Kasper, and V.R. Young (1984) Dynamics of selenite metabolism in young men: studies with the stable isotope tracer method. Am. J . Clin. Nutr., 40:208-218. Jereb, M., R. Falk, B. Jereb, and C. Lindhe (1975) Radiation dose to the human body from intravenously administered 75Se-sodium selenite. J. Nucl. Med., 16: 846-850. Johnson. J.R. (1977) Whole bodv retention followine an iv injection of Se-75 as sefenomethionine. HGlth Phys., 33:250-251. Khait, S.E.. T.A. Norets, and V.S. Deeva (1979) Distribution of 75Se-methionine in the placenta and fetal body at different stages of embryogenesis. Meditsinskaia Radiol., 24(12):40-45. Koller, L.D., G.A. Whitbeck, and P.J. South (1984) Transplacental transfer and colostral concentrations of selenium in beef cattle. Am. J . Vet. Res., 45:25072510. Korpela, H., R. Loueniva, E. Yrjanheikki, and A. Kauppila (1984) Selenium concentration in maternal and umbilical cord blood, placenta and amniotic membranes. Intl. J. Vit. Nutr. Res., 54:257-261. Korte, N.E., J. Skopp, W.H. Fuller, E.E. Nievla, and B.A. Alesii (1976) Trace element movement in soils:

Influence of soil physical and chemical properties. Soil Sci., 122:350-525. Leffler, P., and S. Nordstrom (1983) Metaller i maternellt och fetalt blod undersukning av eventuella variationer i placentas barriar funktion. Statens Vattenfallsverk, Teknisk Rapport 71, 37 pp. NTIS DE83751427, ISS KHM-TR-71. Leipold, H.W., K. Huston, L.C. Hulbert, M. Guffy, and S.M. Dennis (1974) Congenital syndrome in Hereford calves with kyphoscoliosis, arthrogryposis and palatoschisis. Cornell Vet., 64:123-135. Lo, M.T., and E. Sandi (1980) Selenium: Occurrence in foods and its toxicological significance-a review. J . Environ. Pathol. Toxicol., 4:193-218. Matsumoto, N., S. Iijima, C. Tohyama, and C.C. Lu (1977) Placental transfer and distribution of z03Hg and 75Se among fetal mice in the same litter Teratology, 16t113. McConnell, K.P., and D.M. Roth (1964) Passage of selenium across the placenta and also into the milk of dogs. J . Nutr., 84:340-344. Millar, K.R., M.A. Gardiner, and A.D. Sheppard (1973) A comparison of the metabolism of intravenously injected sodium selenite, sodium selenate, and selenomethionine in rats. N.Z. J. Agric. Res., 16:115-127. Mutanen, M. (1986) Bioavailability of selenium. Ann. Clin. Res., 18:48-54. Nau, H. (1987) Species differences in pharmacokinetics, drug metabolism, and teratogenesis. In: Pharmacokinetics in Teratogenesis. H. Nau and W.J. Scott, Jr., eds. CRC Press, Boca Raton, FL, Vol. 1, pp. 81-106. Nishikido, N., and T. Suzuki (1985) Effects of gestational stage and injection route on the corporeal distribution and placental transfer of selenium in pregnant mice. Ind. Health, 23:95-106. Ohlendorf, H.M., D.J. Hoffman, M.K. Saiki, and T.W. Aldrich (1986) Embryonic mortality and abnormalities of aquatic birds: Apparent impacts by selenium from irrigation drainwater. Sci. Total Environ., 52: 49-63. Olson, O.E., E.J. Novacek, E.I. Whitehead, and I.S. Palmer (1970) Investigations on selenium in wheat. Phytochemistry, 9:1181-1188. Olson, O.E., and IS. Palmer (1976) Selenoamino acids in tissues of rats administered inorganic selenium. Metabolism, 25t299-306. Osman, M., and J.D. Latshaw (1976) Biological potency of selenium from sodium selenite, selenomethionine, and selenocystine in the chick. Poult. Sci., 55:987994. Palmer, I.S., R.L. Arnold, and C.W. Carlson (1973) Toxicity of various selenium derivatives to chick embryos. Poult. Sci., 52r1841-1846. Patterson, B.H., O.A. Levander, K. Helzisouer, P.A. McAdam, S.A. Lewis, P.R. Taylor, C. Veillon, and L.A. Zech (1989) Human selenite metabolism: A kinetic model. Am. J . Physiol., in press. Perry, T.W., R.C. Peterson, D.D. Griffin, and W.M. Beeson (1978) Relationship of blood serum selenium levels of pregnant cows to low dietary intake, and effect on tissue selenium levels of their calves. J. Anim. Sci., 46.562-565. Robinson, M.F., H.M. Rea, G.M. Friend, R.D.H. Stewart, P.C. Snow, and C.D. Thomson (1978) On supplementing the selenium intake of New Zealanders. 2. Prolonged metabolic experiments with daily supplements of selenomethionine, selenite and fish. Br. J. Nutr., 39:589-600. Robkin, M.A., D.R. Swanson, and T.H. Shepard (1973) Trace metal concentrations in human fetal livers. Trans. Am. Nucl. SOC.,17:97-98.

ORGANIC AND INORGANIC Se IN HAMSTERS Rosenfeld, I. (1947) Congenital malformations of eyes of sheep. J. Agric. Res., 75:93-103. Rosenfeld. I.. and O.A. Beath (1954) Effect of selenium on reproduction in rats. Proc. SOC.Exp. Biol. Med., 87t295-297. ~~. Satoh, H., T. Suzuki, T. Nobunaga, A. Naganuma, and N. Imura (1981) Effects of sodium selenite on distribution and placental transfer of mercuric mercury in mice of late gestational period. J . Pharmacol. Dyn., 4:191-196. Schroeder, H.A. and M. Mitchener (1971) Toxic effects of trace elements on the reproduction of mice and rats. Arch. Environ. Health, 23:102-106. Scott, M.L. (1973) The selenium dilemma. J. Nutr., 103: 803-810. Shearer, T.R. (1975) Developmental and post-developmental uptake of dietary organic and inorganic selenium into the molar teeth of rats. J. Nutr., 105r338347. Spencer, R.P., and M. Blau (1962) Intestinal transport of selenium-75 selenomethionine. Science, 136:155156. Statistical Consultants, Inc. (1986) PCNONLIN and NONLIN84: Software for the statistical analysis of nonlinear models. Am. Statist., 40.52. Tarantal, A.F., C.C. Willhite, B.L. Lasley, M.J. Cukierski, S.A. Book, and A.G. Hendrickx (1989) Developmental toxicity of L-selenomethionine in Mmmu fusciculuris. Teratology, 39:485. Thomson, C.D., C.E. Burton, and M.F. Robinson (1978) On supplementing the selenium intake of New Zealanders. 1. Short experiments with large doses of ~

~~

371

selenite or selenomethionine. Br. J. Nutr., 39:579587. Thomson, C.D., and R.D.H. Stewart (1974) The metabolism of [75Selselenite in young women. Br. J. Nutr., 32:47-57. Wang, J.H.C. (1984) The kinetics of transplacental transfer of selenium. PhD Dissertation, Northwestern University, Evanston, Illinois, 165 pp. Diss. Abstr. Intl. (B), 45:2091. Waterlow, J.C., J.S. Carrow, and O.J. Millward (1969) The turnover of [75Selselenomethioninein infants and rats measured in a whole body counter. Clin. Sci., 36: 489-504. Weiss, W.P., V.F. Colenbrander, and M.D. Cunningham (1984) Maternal transfer and retention of supplemental selenium in neonatal calves. J. Dairy Sci., 67:416420. Westfall, B.B., E.F. Stohllman, and M.I. Smith (1938) The placental transmission of selenium. J. Pharmacol. Exp. Ther., 64:55-57. Whanger, P.D. (1986) Some comparative aspects of selenite and selenomethionine metabolism. J. Am. Coll. Toxicol., 5:lOl-110. Xaum, C.E., and T.A. Hopeu (1981) Histotopographie der Radiopharmaka 75Se-selenmethionin und 75Se-selenit in kritischen Organen beim ausgewachsenen und embryonaten Tier. Radiobiol. Radiother., 22:697704. Yonemoto, J., A. Naganuma, T. Suzuki, and N. Imura (1983) Effects of vitamin E, glutathione and methylmercury on distribution and placental transfer of selenium in mice. Chemosphere, 12:1021-1029.