A high resolution PIXE measurement for blood plasma ultrafiltrate. Application to loosely bound copper.

A new technique for investigation of elemental concentrations in subfractions of blood plasma is presented. The method is composed of the ultrafiltration of plasma in the presence of ethylenediaminetetraacetic acid (EDTA) and the measurement of the elemental composition by proton-induced X-ray emission (PIXE). The blood samples were collected from both healthy persons and patients suffering from breast cancer. The main emphasis in this study was on the determination of loosely bound copper (Cu) in plasma subfractions containing substances with molecular mass under 10,000, but zinc (Zn) and iron (Fe) contents of these fractions were also determined. The detection limits obtained with this method for Cu, An and Fe were approximately 10 ppb (wet wt).

Early breast cancer, diet, and plasma copper fractions.

In a study of diet and early breast cancer, blood plasma copper has been analyzed by Proton Induced X-ray Emission analysis as both total copper (P-Cu) and that ultrafiltrable from plasma (P-edu-Cu) through membranes with a cut-off at molecular mass 10,000 after equilibration with disodium ethylene diamine tetraacetic acid (EDTA) at 4 degrees C. Ceruloplasmin (P-cer) was also measured using nephelometry of anticeruloplasmin monoclonal antibody-ceruloplasmin complexes. Dietary copper intake per day (D-Cu) was assessed over a five-day dietary record period and calculated from dietary components using a computer program. P-edu-Cu correlated significantly with both D-Cu and ceruloplasmin while P-Cu correlated only with ceruloplasmin. Further, ceruloplasmin did not significantly correlate to D-Cu. Hence, P-edu-Cu better reflects copper status than do P-cer or P-Cu as it relates to both the major copper enzyme in plasma and to daily copper intake. This may be important in drawing conclusions about the significance of copper in disease states where copper fractions other than ceruloplasmin may be most important owing, for example, to oxidative properties. Categorization as cancer or normal, by copper parameters (D-Cu, P-edu-Cu, P-Cu, P-cer), was studied in multiple correlation. In particular, the ratio P-cer/P-Cu and the ratio P-edu-Cu/D-Cu were significantly related to disease. Irrespective of age (pre- and post-menopausal), highly significant differences between normals and early stage breast cancer patients were seen with p < 0.0001 to p < 0.01. The precise role played by plasma and dietary copper fractions deserves continued attention in view of the present and earlier results in cancer studies.

Effects of preparation methods of biomedical samples on PIXE measurement sensitivity.

Instrumental sample preparation techniques for PIXE measurements of human blood serum and bovine muscle were investigated. The techniques studied were direct measurement of a drop, drying on a foil and on an Al cup, lyophilization on a foil, pellet pressing of dried material, microtome sectioning and biopsy preparation. The same serum and muscle were used in each procedure allowing the comparison of detection limits between different methods. The most beneficial technique for both samples was lyophilization and pellet pressing.

Proton-induced X-ray emission analysis of atherosclerotic plaques of the carotid bifurcation.

The trace elements of both calcified atherosclerotic plaques and plaque-free vessel walls of the carotid bifurcation from 31 autopsies were investigated using the proton-induced X-ray emission (PIXE) method. The trace elements studied were phosphorus (P), calcium (Ca), chrome (Cr), iron (Fe), copper (Cu), zinc (Zn), lead (Pb), selenium (Se), bromine (Br), strontium (Sr), and rubidium (Rb). All samples contained Fe and Zn. Mercury (Hg) was not detected in any of the samples studied. All plaque-free samples contained Cu and almost all Br and Ca, none Sr. All calcified atherosclerotic plaques contained Ca and almost all Br and Sr. The relative levels of Ca were higher in the calcified plaques than in the plaque-free vessel walls. The relative value of Ca in calcified and uncalcified samples was greatest in the group who had died because of cardiovascular disorders and smallest in the group who had died from other causes. There was a strong positive correlation between the Ca and Sr of the plaque samples and between the P and Br of the plaque-free samples.

Evidence for increased non-ceruloplasmin copper in early-stage human breast cancer serum.

We measured total serum copper and ceruloplasmin levels in pre- and postmenopausal Stage I and II breast cancer (BC) patients and omnivorous and vegetarian controls. The omnivorous groups included 14 premenopausal women [33 +/- 6 (SD) yrs] and 11 postmenopausal women (57 +/- 5 yrs), and the vegetarian groups were comprised of 12 premenopausal subjects (34 +/- 7 yrs) and 11 postmenopausal subjects (59 +/- 5 yrs). There were 13 premenopausal BC patients (39 +/- 7 yrs) and 10 postmenopausal BC patients (66 +/- 6 yrs). Fasting serum samples were taken on three consecutive days, typically four times in the year. Serum ceruloplasmin levels (g/l) were measured by nephelometry utilizing monoclonal antiserum, and total serum copper levels (mumol/l) were determined by proton-induced X-ray emission analysis. Premenopausal patients had higher serum copper levels than their controls (mean 18.7 vs. 16.6, p less than 0.03). For ceruloplasmin, the postmenopausal BC patients had significantly lower levels than pooled postmenopausal controls (0.309 vs. 0.370, p less than 0.001). The copper-to-ceruloplasmin ratio was significantly higher in the pooled cancer groups than in the pooled control groups (3.69 vs. 3.21, p less than 0.001), with similar patterns in both pre- and postmenopausal classes. This high serum copper-to-ceruloplasmin ratio in BC patients may reflect disordered copper metabolism in this disease, which could also have implications for the origin of, or the response to, the cancer process.

Trace element studies on whole human cerebrospinal fluid with external beam PIXE.

External beam PIXE analysis with a 2.4 MeV proton beam was used to determine the concentrations of K, Ca, Fe, Cu, Zn and Br in cerebrospinal fluid from patients having various disorders. The obtained total concentration ranges K 34,000-1,079,000, Ca 5300-81,300, Fe 40-1030, Cu 20-1650, Zn 15-1250 and Br 400-43,000 micrograms/kg are compared with the values given in the literature. In certain patients there were very high CSF bromine levels, but this was shown to be the result of taking medications presented as bromide salts. The possibility of using the method in clinical practice for CSF analysis is considered. The new method of preparing self-supporting films of the samples was used. This method was further optimized by investigating in detail the use of EDTA as a homogenizer.

A new rapid method for quantification of nitrogen in human serum employing the 14N(p,p' gamma)14N reaction: application to human pregnancy.

The total nitrogen concentrations in dried serum from 54 pregnant women and 17 newborn babies were determined by a new application of the 14N(p,p' gamma)14N reaction resonance at 3.9 MeV. The samples were bombarded in a He atmosphere by 4.1 MeV protons from a tandem Van de Graaff accelerator. The mean dry-weight nitrogen concentration in serum sampled during early pregnancy (6-12 weeks; 13.9 g per 100 g +/- 5.4%) was significantly higher than that in serum sampled during late pregnancy (38-42 weeks: 13.0 g per 100 g +/- 3.9%, p less than 0.001) and than that in serum taken from the umbilical cord (13.3 g per 100 g +/- 4.6%, p less than 0.01). The nitrogen levels measured using this rapid nuclear technique, applied for the first time to human serum analysis, agree well with parallel Kjeldahl analyses.

Selenium and other elements in human maternal and umbilical serum, as determined simultaneously by proton-induced X-ray emission.

Using PIXE (proton-induced X-ray emission), we simultaneously determined the concentrations of Se, Ca, Fe, Cu, Zn, Br, and Pb in blood serum from 56 pregnant women, 25 healthy controls, and 31 others with twin pregnancy or some complicating condition (diabetes, hypertension, epilepsy, hepatosis gravidarum, pre-eclampsia, small baby), and in cord-blood serum from 21 newborns. Pellets, pressed from the serum samples after addition of yttrium as an internal standard, mixing, and evaporating at 30 degrees C with or without reduced pressure (less than 1 kPa), were bombarded by 2.2 MeV protons from a Van de Graaff accelerator in the air and the induced X-rays collected by a Ge(Li) detector. Relative to mean Se values for early six- to 12-week pregnancy (0.045 ppm), those for 35-42 week pregnancy (0.028 ppm) were low (p less than 0.001). Umbilical cord blood serum showed even lower values (0.016 ppm, p less than 0.001)--findings in harmony with the incidence pattern of Keshan cardiomyopathy. Pb crossed the placenta; values for cord serum were not significantly different from those in pregnancy serum. Cu, Zn, Fe, and Ca showed the significant expected patterns in the different groups. Compared with the late-pregnancy controls, Fe was high in mothers of small-birth-weight babies (1.70 ppm, p less than 0.02). Br was high in pre-eclampsia (3.59 ppm, p less than 0.05) and mothers with twins (3.61 ppm, p less than 0.05).

In vitro measurement of phosphorus in small bone biopsies using 14 MeV neutron activation analysis.

In studying diseases of bone such as osteoporosis and renal osteodystrophy it is important to have an index of change in bone mineral content. For this purpose we describe a method for measuring the phosphorus content of 100--150 mg bone biopsies. Femoral and iliac crest biopsies from 6 cadavers were each paired with a standard bone of similar size and irradiated with 14 MeV neutrons for 10 minutes. A special rotor ensured the same activating flux across both sample and standard bone which were both wrapped in aluminium foil. The 28Al generated from bone phosphorus (31P(n,alpha)28Al) and the 27Mg generated from the aluminium foils (27Al(n,p)27Mg), which undergo beta-decay have useful gamma rays at 1779 keV and 844 keV respectively. These gamma rays were counted using a 110 cm3 Ge(Li) detector and a rotating frame permitting four 30 s counts of sample and standard alternately (interval for rotation 10 seconds) in identical geometries. Using this system the phosphorus content of the biopsies could be determined with a coefficient of variation of 4.6% using the standard bone as reference and 6% using aluminium foil as reference. In dense bone the P/dry weight ratio was 30--50% greater than in spongy bone. The clinical value of the method is discussed.

Elemental analysis of bone mineral by backscattering of alpha particles.

Backscattering of 2 MeV alpha particles has been applied for the first time to studies of the inorganic substances of human bone. Bone samples taken from the femoral shaft (dense bone) and iliac crest (spongy bone) of 18 cadavers were analysed for the calcium, phosphorus and oxygen content, after lyophilisation and heating for 1.5 h at 500 degrees C to remove water and organic material. The backscattering method produces a spectrum in which all the elements present in the sample except hydrogen can be seen at once. The reproducibility of the measurements was less than 2%. No previous studies were found in the literature of the measurement of elemental oxygen in bone samples. Ca/P ratios varied from 1.97 to 2.47, Ca/O ratios from 0.76 to 0.91 and P/O ratios from 0.34 to 0.41, the latter two being about 10--15% lower than expected if bone mineral were hydroxyapatite (Ca/O = 0.96 and P/O = 0.44). The results are consistent with previous work on Ca/P ratios but the low Ca/O and P/O ratios could not be explained without postulation of crystal or hydroxyl water in bone mineral. Support for the explanation was gained by differential gravimetric and thermal analysis.