Anion binding with two polyammonium macrocycles of different dimensionality.

A comparative study of the binding of nitrate and sulfate with a polyammonium monocycle L(1), (3,6,9,17,20,23-hexaazatricyclo[23.3.1.1(11,15)]-triaconta-1(29),11,13,15(30),25,27-hexaene), and the corollary bicycle L(2), (1,4,12,15,18,26,31,39-octaazapentacyclo-[13.13.13.1(6,10).1(20,24).1(33,37)]-tetratetraconta-6,7,9,20(43),21,23,33(42),34,36-nonaene), is reported. Potentiometric studies indicated negligible binding for L(1) and nitrate, but high affinity was observed for sulfate (log K(H5L(SO4)/H5L-SO4) = 3.53(1), log K(H6L(SO4)/H6L-SO4) = 4.36(1)). Stronger binding was observed for the cryptand L(2) with both nitrate and sulfate (log K(H6L(NO3)/H6L-NO3) = 3.11(5), log K(H7L(NO3)/H7L-NO3) = 3.55(5); log K(H6L(SO4)/H6L-SO4) = 4.43(1), log K(H7L(SO4)/H7L-SO4) = 4.97(5)). Five crystal structures are reported: the nitrate (1) and sulfate (2) salts of L(1), the free base (3) of L(2), and the nitrate (4) and tosylate (5) salts of L(2). Structural results for L(1) indicate relatively planar monocycles with cis and trans orientations of the phenyl groups for 2 and 1, respectively, with the anions above and below the monocycle rings. For L(2), key features include an encapsulated water and intricate water network in 3, two encapsulated and four external nitrates and two external water molecules in 4, and six external tosylates with sulfonate groups pointing into the cavity and eight external waters in 5.

Trans-dioxocyclam (1,4,8,11-tetraazacyclotetradecane-2,9-dione) dihydrate and its NiII complex.

The crystal structures are reported of trans-dioxocyclam dihydrate, C(10)H(20)N(4)O(2).2H(2)O, a structural isomer of the well known cis-dioxocyclam, and of its novel Ni complex, (1,4,8,11-tetraazacyclotetradecane-2,9-dionato-kappa(4)N)nickel(II) dihydrate, [Ni(C(10)H(18)N(4)O(2))].2H(2)O, the first example of a transition metal complex of this ligand. Both molecules lie on crystallographic centres of inversion. The free ligand has two of its N atoms turned outwards from the ring and hydrogen bonded to water molecules. A major conformational change takes place in the complex in which the ligand binds in a trans tetradentate fashion, as suggested by the electronic spectrum. The nickel(II) ion is low spin, although the electronic spectrum of the complex in water indicates an equilibrium mixture of low-spin and high-spin species. The irreversible electrochemical oxidation of [NiL(1)] (L(1) is deprotonated trans-dioxocyclam, C(10)H(18)N(4)O(2)) in water occurs at a potential of 0.964 V [versus SHE (standard hydrogen electrode)], which is very similar to that for the Ni-cis-dioxocyclam complex.

Topologically constrained manganese(III) and iron(III) complexes of two cross-bridged tetraazamacrocycles.

A family of Mn3+ and Fe3+ complexes of 4,11-dimethyl-1,4,8,11-tetraazabicyclo[6.6.2]hexadecane (1) and 4,10-dimethyl-1,4,7,10-tetraazabicyclo[5.5.2]tetradecane (2) has been prepared by the chemical oxidation of the divalent manganese and iron analogues. The ligands are ethylene cross-bridged tetraazamacrocycles derived from cylam and cyclen, respectively. The synthesis and characterization of these complexes, including X-ray crystal structure determinations, are described. The structural evidence demonstrates that the tetradentate ligands enforce distorted octahedral geometries on the metal ions, with two cis sites occupied by labile ligands. Magnetic measurements reveal that the complexes are high spin with typical magnetic moments. Cyclic voltammetry shows reversible redox processes for the Fe3+/Fe2+ couples of the iron(III) complexes, while Mn3+/Mn2+ and Mn4+/Mn3+ couples were observed for the complexes with manganese(III). The manganese chemistry of 1 was studied in depth. The dichloro manganese(III) cation of 1 undergoes facile ligand substitution reactions at the labile, monodentate sites, for example substituting azide for chloride ligands. Air oxidation of the dichloro complex of Mn (1)2+ in basic solution does not give the expected mu-oxo dimeric product common to manganese. Instead, an unusual manganese(III)-OH complex has been isolated from this reaction and structurally characterized. A similar reaction under slightly different conditions gives a putative MnIII(OH)2 complex that metathesizes to MnIII(OMe)2 upon recrystallization from methanol.

Synthesis and properties of iron(II) and manganese(II) complexes derived from a topologically constrained pentadentate ligand.

The novel bicyclic pentadentate ligand 5-methyl-1,5,9,24,25-pentaazapentacyclo[7.7.7.5.5]pentacosane11,13,15,18(25),20,22-hexene (L1) has been synthesized. Because of its cross-bridged topology it exhibits a relatively rigid preorganized conformation especially appropriate to complex formation, as shown by the crystal structure of the monoprotonated ligand salt, HL1ClH2O [orthorhombic, P212121, a = 9.4405(5) A, b = 13.3617(5) A, c = 16.710(1) A]. The complexes of L1 with both iron(II) and manganese(II) have been characterized, including the crystal structures of [FeL1CH3CN][FeCL4] and [MnL1Cl][PF6] [monoclinic, P21/n, a = 10.0460(5) A, b = 19.237(9) A, c = 15.6254(8) A, beta = 95.97(2)degrees and a = 7.745(2) A, b = 22.786(4) A, c = 14.639(4) A, beta = 105.074(10)degrees respectively]. The manganese complex is high spin with mueff = 5.96 and theta = 2.5 +/- 0.8 cm(-1), indicating weak ferromagnetic interactions. The reactions of the complexes with tert-butyl hydroperoxide and hydrogen peroxide have been shown by ESR spectroscopy to produce the tert-butyl peroxyl and hydroperoxyl radicals, as evidenced by their spin adducts with the spin traps N,N-dimethyl-1-pyrroline-N-oxide and N-tert-butyl-phenyl-nitrone.

Binding of phosphate with a simple hexaaza polyammonium macrocycle.

A mixture of dihydrogen phosphate and phosphoric acid has been crystallized with a hexaprotonated 26-membered polyammonium macrocycle, 1,4,7,14,17,20-hexaazacyclohexacosane, as the counterion. The complex crystallizes in the monoclinic space group P2(1)/c with unit cell parameters of a = 10.006(2) A, b = 12.525(1) A, c = 19.210(2) A, beta = 102.91(1) degrees, and V = 2346.6(5) A3. The hexaprotonated macrocycle is located on a crystallographic center of inversion and is surrounded by eight phosphate anions. Six of the phosphates are dihydrogen phosphates (H2PO4-), and the other two are neutral phosphoric acid molecules. Intricate hydrogen-bonding networks, involving the anionic and neutral phosphates and the protonated macrocycle, dominate the crystal lattice. Potentiometric studies using NaCl as the supporting electrolyte indicate high formation constants for the triprotonated macrocycle, H3L3+, with PO4(3-) at pH approximately 9.5 (log K = 4.55(4)), for the tetraprotonated macrocycle, H4L4+, with monohydrogen phosphate, HPO4(2-), at pH approximately 8.0 (log K = 6.01(3)), and for ditopic complexes with H5L5+ and H6L6+ and dihydrogen phosphate, H2PO4-, at pH approximately 4.0 (log K = 6.16(6)) and pH approximately 2.5 (log K = 6.44(5)), respectively. The ditopic behavior in the simple polyazamacrocycle receptor is a somewhat unusual occurrence, as is the finding of phosphoric acid species in the crystal structure.

Lead screening among low-income children in Galveston, Texas.

The objective of this study was to report results of a lead-screening program for low-income children living in Galveston, Texas. We obtained blood lead by graphite furnace spectrophotometry on 1,571 children aged 6 months to 8 years. Nineteen percent of children had blood lead levels > or = 10 mcg/dL. Risk factors included African-American ethnicity, young age, and residence in old housing. Follow-up was accomplished in only 50% of children with low-level toxicity. Lead screening is an important public health measure in communities with old houses. For screening to be successful, caregivers need to devote additional effort to follow-up.

Sampling of low molecular weight iron by microdialysis following spinal cord injury.

We combined the use of desferrioxamine as an iron chelator, microdialysis sampling, and iron analysis by atomic absorption spectroscopy to measure extracellular levels of low molecular weight (LMW) iron in vivo in the spinal cord. Low molecular weight iron is free iron plus iron bound to small molecules. We show that the extracellular LMW-iron concentration is not increased significantly in situ by trauma to the rat spinal cord, suggesting that an extracellular elevation in LMW-iron at the site of injury is not the major initiating factor for lipid peroxidation following spinal cord injury.

Iron exacerbates aniline-associated splenic toxicity.

Our earlier studies have shown that aniline exposure in rats causes time- and dose-dependent accumulation of iron in the spleen, which may exacerbate aniline splenotoxicity by catalyzing free-radical reactions. The present studies were conducted to test whether aniline-induced splenic toxicity could be potentiated by iron overload. For 30 d male Sprague-Dawley rats received the following treatments: 0.5 mmol/kg/d aniline hydrochloride (AH) by gavage (AH group); 3% carbonyl iron-supplemented diet (IR group); 0.5 mmol/kg/d AH by gavage and iron-supplemented diet (AH + IR group); or no treatments (controls). Treatment-related significant increases in total iron, low molecular weight chelatable iron, lipid peroxidation, and protein oxidation were observed in the spleens of all the groups compared to control. However, these changes were much greater in the combined AH + IR group. The aniline-induced morphological changes in the spleen were consistent with our earlier observations, but were more pronounced in the AH + IR group. The increased toxicity, as evident from greater oxidative stress and morphological changes in the AH + IR group, suggests that iron potentiates the splenic toxicity of aniline.

Zinc deficiency and oxidative stress in the retina of pigmented rats.

PURPOSE: To determine the effect of moderate zinc deficiency on antioxidant defenses and measures of oxidative stress in the retina and retinal pigment epithelium (RPE) of Brown Norway Rats. METHODS: Twenty-four rats were housed individually and divided into three groups of 8 rats each. Group 1 was fed ad libitum a semipurified control diet formulated to contain 50 parts per million [ppm] total zinc; group 2 was fed ad libitum an identical diet but containing 5 ppm total zinc; and group 3 was pair-fed the control diet but restricted in amount to that consumed by group 2. Food intake was measured daily and the rats weighed weekly. After 6 weeks, the rats were killed and the following measurements were made: serum zinc, serum alkaline phosphatase, retinal zinc, RPE-choroid zinc, RPE-choroid catalase, liver metallothionein (MT), retinal MT, RPE-choroid MT, retinal catalase, and retinal thiobarbituric reactive substances (TBARS). RESULTS: The following showed statistically significant differences between groups 2 and 3, respectively: serum Zn (1216 micro/l versus 1555 microg/l, P < or = 0.01), serum alkaline phosphatase (3.75 U/mg versus 5.10 U/mg, P < or = 0.05), liver MT (4.3 microg/mg protein versus 16.7 microg/mg, P < or = 0.0001), RPE-choroid MT (1.3 microg/mg protein versus 2.2 microg/mg, P < or = 0.02), retinal MT (0.85 microg/mg protein versus 2.8 microg/mg, P < or = 0.05), and retinal TBARS (6.2 nM/mg protein versus 2.2 nM/mg, P < or = 0.05). CONCLUSIONS: The results show that retinal MT and RPE MT concentrations are very sensitive to intake of dietary zinc. The increase in retinal TBARS in group 2 indicates that moderate zinc deficiency increases oxidative stress to the retina. The results also suggest that MT is protective against lipid peroxidation of retinal membranes.

Polyatomics in zinc isotope ratio analysis of plasma samples by inductively coupled plasma-mass spectrometry and applicability of nonextracted samples for zinc kinetics.

Inductively coupled plasma-mass spectrometry (ICP-MS) is a powerful tool for both quantitative multielement analyses of inorganic elements and measurement of isotope ratios (IRs). The main disadvantage of this technique is the existence of polyatomic isobaric interferences at some key masses. Zinc has been investigated for such potential interferences in serum or plasma. The Zn isotopes, 66Zn and 68Zn, have no apparent interferences, but 32S1602 and 32S2 are isobaric with 64Zn. The possible effects of S and other major components of blood plasma-Na, K, Cl, P, Ca-on Zn IRs were investigated using a series of mineral solutions which simulated human plasma with respect to these elements. The mixture of all mineral elements interfered only with 64Zn (6.66 ng/mL) and 70Zn (8.51 ng/mL). Interferences to 66Zn, 67Zn, and 68Zn were minimal containing 0.90, 0.94, and 0.39 ng/mL, respectively. The copresence of Na or S shifted 35Cl16O2 (atomic mass 67 coming from Cl solution) to 35Cl2 which reduced the contribution to 67Zn. The hypothesis that Zn IRs obtained from plasma at various intervals after the intravenous administration of enriched 67Zn to humans would reflect those obtained after extraction of Zn was therefore tested. To compare the two pretreatment methods, "extraction" versus "nonextraction," specimens were collected from 10 human subjects at intervals of 5 min to 24 h postinjection, and in 4 subjects from 5 min to 9 d postinjection. Two separate aliquots of plasma from each time-point were dried and digested with hydrogen peroxide, and the residue dissolved in nitric acid. One specimen was subjected to zinc extraction using ammonium diethyldithiocarbamate chelate followed by back extraction into nitric acid. The matching aliquot received no further pretreatment. The normalized IRs obtained from 67Zn/66Zn and 67Zn/68Zn in both the "extracted" and "nonextracted" samples agreed well (r2 = 0.976 and r2 = 0.985, respectively) compared to those from other ratios (r2 = 0.838 for 67Zn/64Zn and r2 = 0.747 for 67Zn/70Zn). Considering the minimum possibility of isobaric interferences in plasma samples, 67Zn/68Zn obtained from "nonextracted" samples is sufficient for routine Zn kinetic analysis by ICP-MS.

Effects of repletion with zinc and other micronutrients on neuropsychologic performance and growth of Chinese children.

The knowledge that zinc is essential for growth and neuropsychologic performance and a report of zinc-responsive stunting in Chinese children prompted this project. This article summarizes findings from a 10-wk, double-blind, controlled trial of zinc repletion in 740 urban, 6-9-y-old first graders from low-income families in Chongqing, Qingdao, and Shanghai, People's Republic of China. Treatments were 20 mg Zn alone (Z), 20 mg Zn with micronutrients (ZM), and micronutrients alone (M). The M mixture was based on National Research Council guidelines. Nutrients that might interfere with zinc retention were excluded or given in lower amounts. Main outcomes were changes in neuropsychologic performance and knee height. Hemoglobin, serum ferritin, plasma and hair zinc, and whole blood and hair lead were also measured. Anemia was not common, and serum ferritin concentrations were usually within the range of normal. Mean baseline plasma zinc concentrations were marginal in children from Chongqing and Qingdao and normal in children from Shanghai. After treatment with ZM or M plasma zinc increased. Hair zinc tended to decrease after all treatments. Mean baseline whole blood lead concentrations were slightly below the limit considered excessive for children by the US Centers for Disease Control and Prevention. Neuropsychologic performance and growth were most improved after treatment with ZM. These findings were consistent with the presence of zinc and other micronutrient deficiencies.

A preliminary report: effects of zinc and micronutrient repletion on growth and neuropsychological function of urban Chinese children.

OBJECTIVE: Zinc is essential for growth and cognition of experimental animals. Past research found zinc repletion improved growth of stunted Chinese children. Therefore we measured effects of zinc repletion on growth and neuropsychological functions of children. DESIGN: Double-blind randomized controlled treatment trial. SETTING: Elementary schools in low income districts of Chongqing, Qingdao and Shanghai. SUBJECTS: Three hundred-seventy-two 6 to 9 year old first graders. INTERVENTIONS: Treatments were 20 mg zinc, 20 mg zinc with micronutrients, or micronutrients alone. The micronutrient mixture was based on guidelines of the US NAS/NRC. Treatments were assigned to classrooms of 40 or more children each, and administered by teachers 6 days per week for 10 weeks. MEASURES OF OUTCOME: Changes in knee height and neuropsychological functions. RESULTS: Zinc alone had the least effect on growth while zinc with micronutrients had the largest effect; micronutrients alone had an intermediate effect. Zinc-containing treatments improved neuropsychological functions, but micronutrients alone had little effect. CONCLUSIONS: The findings confirm the essentiality of zinc for growth of children, and show, for the first time, the essentiality of zinc for neuropsychological functions of children. In addition, the need for repletion of other potentially limiting nutrients in studies examining the effects of specific nutrients on growth and neuropsychological functions was confirmed.

Oxidative stress in the splenotoxicity of aniline.

Aniline-induced splenic toxicity is characterized by hemorrhage, capsular hyperplasia, fibrosis, and a variety of sarcomas in rats. Early biochemical events responsible for the observed effects are not known. To understand the mechanism(s) of aniline-induced splenic toxicity, single and multiple (four and seven) doses of 1 mmol/kg of aniline hydrochloride(AH) were given in rats. Apart from changes in the hematological parameters, these studies demonstrated that AH could induce lipid peroxidation and protein oxidation in the spleen, and significant increases were observed at four doses. Subsequently, a dose-response study of AH was performed. Male SD rats were given four doses each (one dose/day) of 0.25, 0.5, 1, and 2 mmol/kg of AH in water by gavage, while controls received water only. Animals were euthanized 24 hr following the last dose and tissues obtained. Spleen weight increased by 32 and 80% at 1 and 2 mmol/kg doses, respectively. Splenic lipid peroxidation showed dose-dependent increases of 24, 32, and 43% at 0.5, 1, and 2 mmol/kg, respectively. Protein oxidation in the spleen, quantitated by carbonyl content per milligram protein, showed 10, 28, and 27% increases at 0.5, 1, and 2 mmol/kg, respectively. Iron content in the spleen also showed dose-dependent increases of 72, 172, and 325% at 0.5, 1, and 2 mmol/kg, respectively. Dose-related histopathologic expansion of splenic red pulp was characterized by increasing vascular congestion (most pronounced at 2 mmol/kg), increased red pulp cellularity, erythrophagocytosis, and cellular fragmentation at 1 and 2 mmol/kg; iron deposition in red pulp also increased dramatically with dose. These studies establish that aniline induces lipid peroxidation and protein oxidation in the spleen and suggest that oxidative stress plays a role in the splenic toxicity of aniline.

Influence of zinc on selected cellular functions of cultured human retinal pigment epithelium.

Zinc is a necessary micronutrient, usually abundant in human RPE. Our study was undertaken to determine the effects of short-term, zinc deficiency on human retinal pigment epithelium (RPE) using a culture model of fetal human RPE cells. Human fetal RPE cells were isolated and cultured in Coon's modified Ham's F-12 medium. For zinc depletion studies, cells were cultured for 1 week in Chelex-treated Dulbecco's modified Eagle's medium containing low (0.25 microM) or physiologic (11 microM) total zinc concentrations as determined by flame atomic absorption spectroscopy. Protein synthesis was determined by incorporation of 35S-cysteine/methionine and labeled proteins analysed by polyacrylamide gel electrophoresis. Several cell parameters and enzymes were significantly reduced below control when cultured in low zinc: zinc content (40%), proliferation (63%), protein/well (50%), catalase activity (68%), alkaline phosphatase activity (61%), alpha-mannosidase activity (68%), and metallothionein (82%). No statistically significant decline was seen in acid phosphatase activity, superoxide dismutase activity, glutathione peroxidase activity and dexamethasone induction of metallothionein. Zinc repletion (100 microM, 1 h) increased catalase and alpha-mannosidase activities from 32% and 33% of control to 75% and 73%, respectively. Cycloheximide did not inhibit this short-term zinc-induced repletion of catalase or alpha-mannosidase. Protein synthesis in low zinc medium was depressed, but not significantly, as shown by incorporation of radiolabeled 35S-cysteine/methionine into newly synthesized proteins. The effects of zinc deficiency in cultured human RPE are selective. Adequate intracellular zinc was required for maximal activity of some enzymes. The dependence of catalase activity on zinc was not predicted and may help explain the observed decline in catalase activity seen with age in RPE. Our model of zinc deficiency should prove useful in elucidating the complex effects of zinc deficiency and repletion in human RPE.

Hematopoietic toxicity of linoleic acid anilide: importance of aniline.

The purpose of this study was to investigate the role of hydrolysis products of linoleic acid anilide (LAA), i.e., aniline and linoleic acid (LA), in the toxicity to the hemopoietic system, especially to the spleen. To achieve this, the parent compound (LAA) and its putative hydrolysis products, i.e., aniline or linoleic acid (LA), were given to male SD rats at equimolar doses (0.7 mmol/kg) in 0.25 ml mineral oil by gavage, daily, for 14 days. The controls received equal volumes of vehicle only. Five animals from each group were euthanized at Days 1, 7, and 28 following the last dose. At all time points, spleen weights increased in the LAA- and aniline-treated rats, but spleen to body weight ratios were increased only at Days 1 and 7 in these groups. No changes were observed in the LA-treated rats at any time point. RBC counts were decreased in the LAA and aniline groups at Days 1 and 7, whereas hemoglobin content was decreased by 20 and 13% in the LAA- and aniline-treated rats, respectively, only at Day 1. Methemoglobin content in the LAA and aniline groups also increased by 76 and 101%, respectively, at Day 1. Serum transaminases (AST and ALT) decreased in the LAA, aniline, and LA groups but the decreases were more consistent in the LA group. Serum IgA increased in the LAA and aniline groups only at Day 1. Splenic iron content was increased 381, 486, and 51% in the LAA-treated rats and 474, 491, and 58% in the aniline-treated rats at Days 1, 7, and 28, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Iron and zinc nutriture of premenopausal women: associations of diet with serum ferritin and plasma zinc disappearance and of serum ferritin with plasma zinc and plasma zinc disappearance.

The Second National Health and Nutrition Examination Survey found low iron and zinc intakes and low serum ferritin in many premenopausal women. Therefore food frequency history, serum ferritin, plasma zinc, plasma zinc disappearance, and zinc turnover rate were measured in healthy premenopausal women. Plasma zinc disappearance and turnover rate were determined by inductively coupled plasma mass spectroscopic analysis of the plasma ratio of zinc 67 to zinc 68 before and 30 to 60 minutes after intravenous injection of 67Zn. Food frequency predicted serum ferritin concentration in 38 women and plasma zinc disappearance in 19. Frequent red meat intake was associated with higher serum ferritin concentration and a "normal" plasma zinc disappearance. Frequent milk intake was associated with lower serum ferritin concentration and a rapid plasma zinc disappearance. After exclusion of an outlier, in 18 women serum ferritin concentrations were lower when plasma zinc was < 70 micrograms/dl (p < 0.03), and plasma zinc disappearance and turnover rate were higher when serum ferritin was < 20 ng/ml (p < 0.05). Serum ferritin concentration and plasma zinc disappearance constant were inversely and nonlinearly related (R2 = 0.777, p = 0.0003). These findings suggest that avoidance of red meat increases the risk of iron and zinc deficiencies, and low serum ferritin concentrations suggest the possibility of low zinc nutriture.

Sepsis-induced cascade of cytokine mRNA expression: correlation with metabolic changes.

The relationships between metabolic alterations and tissue-specific gene expression of tumor necrosis factor-alpha (TNF-alpha), interleukin 6 (IL-6), gamma-interferon (gamma-IFN), and interleukin 1 and serum levels of TNF-alpha and IL-6 before and after a live Escherichia coli septic challenge to rats were examined. From 0 to 2 h, serum glucose significantly decreased while plasma glucagon increased. By 8 h, plasma glucagon, serum insulin, and glucose appearance were significantly elevated. Gene expression of phosphoenolpyruvate carboxykinase increased 1 h after E. coli but by 4 h was significantly decreased. TNF-alpha mRNA (liver and spleen) and serum peptide levels peaked 1-2 h after the septic challenge and then decreased substantially by 6-8 h. Spleen IL-6 and gamma-IFN mRNA expression reached a maximum 4 h after E. coli challenge, whereas serum IL-6 levels were elevated by 2 h after injection of the bacteria. The increase in TNF-alpha mRNA and serum peptide levels correlated with the early fall in serum glucose and rise in plasma glucagon. Alterations in the rate of glucose appearance and plasma glucagon were observed later and coincided with the increased mRNA expression of IL-6 and gamma-IFN. Thus the metabolic alterations observed in the septic rat are associated with a complex cascade of several cytokines.

Chromium content of total parenteral nutrition solutions.

Chromium (Cr) present as contaminant was analyzed by flameless atomic absorption spectrometry in a variety of commercially produced solutions and additives commonly used in total parenteral nutrition (TPN) formulas. Total Cr likely to be administered unintentionally per day was estimated both by summing the Cr in appropriate volumes of each solution required for preparation of standard TPN formulas and by analyzing complete TPN solutions. Storage of TPN solutions in plastic bags for 14 days did not affect Cr concentrations. The amounts ranged from 2.4 to 8.1 micrograms/day for a high glucose formula and 2.6 to 10.5 micrograms for a high lipid formula. Amino acid solutions, especially when containing phosphate, or with phosphate salt additives and with lipid emulsions accounted for approximately 85 to 90% of the Cr found.