Biol. Neonate 36 : 225 232 (1979)
Contents of Trace Elements in the Human Liver before Birth1 B. Meinet*, J.C. Bode*, W. Koenig** and F.-W. Richter** Fachbereiche Humanmedizin* und Physik**, Philipps-Universität, Marburg/Lahn
Key Words. Trace elements • Fetal/adult human livers Abstract. A multielement analysis was carried out in nine fetal livers (age: 23 weeks of gestation to term) in comparison to an adult control group (n = 37). 15 elements (phosphorus to lead) were analyzed by proton-induced X-ray emission (PIXE). Significant differences between the two groups, expressed as ratios (R) of fetal to adult liver values were observed for copper (R = 16), zinc (R = 2.0), iron (R = 1.5), calcium (R = 1.34), selenium (R = 1.22), manganese (R = 0.7), and molybdenum (R = 0.13). An interesting point seems to be the inverse relation of copper and molybdenum. The negative correlation of these two trace elements which is known from animal experiments has now been confirmed in the human liver.
It is well known that in human fetal liver the copper and zinc content is considerably higher than in the adult liver (23—25). Similar high cop per and zinc contents have been observed in fetal livers of animals of various species (18, 23). The aim of this study was to get information on the content of the main essential and toxic trace elements in the liver of mature or im mature newborns.
1 Supported by the ‘Deutsche Forschungsgemein schaft’.
Materials and Methods
Livers of 9 newborns (age range from 23 weeks gestation to term) have been analyzed. Cause of death in all cases was a maternal complication during late pregnancy or delivery. None of the 9 cases exhibited signs of malformation or inborn errors of metab olism. Further details are given in table I. For comparison, trace element contents in the liver of 37 adults have been determined. Postmortem samples were taken only when the histological diagno sis of the liver showed no significant pathological change. In addition, those patients were excluded from the study who had diseases which are known to influence the trace element contents in the liver (e.g., disease of the blood-forming system, metabolic dis eases). Data on sex and age distribution of the adult control group are given in table II.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 8/8/2018 4:22:41 AM
Introduction
226
Meinel/Bode/Kocnig/Rich ter
Table 1. Data on sex, age and weight of the members of the newborn group
Identifier1
Sex
Age,
Body
Body
weeks of
lengths cm
weight
Liver weight
g
g
gestation
Liver histology
103/772
30 38
40
2,080
32.1 64.7
52
2,990
9 9
23 24
29
420
17.8
zyanosis
22
-
14.8
zyanosis
174/77 183/77 184/77
141
partially autolysed normal
Table II. Age and sex distribution of the adult control group
Number of cases
n = 37
Age
mean = 69a
Sex
n ( d ) = 23 n (9)= 14
± SD= ± 11a n (d)/n = 62% n (9)/n = 38%
Trace element analysis was performed by protoninduced X-ray emission (P1XE) which allows a multi element analysis of most of the essential and toxic elements within one sample at the same time (5, 17). After homogenization of about 0.1 -1 g of liver tissue and diluting the homogenate with bidistilled water, 10 /al of a 1:10 dilution were pipetted on thin formvar foil (2 -5 Mg/cm2) and dried by infrared light there after. No metal instruments were used. The homoge nates were stored at 20 °C in polystyrol containers. Instruments and containers were cleaned by con centrated HNOj (65%) and bidistilled water. The
homogenate layers on the formvar foil were analyzed by observing the characteristic X-rays of the trace elements, by exposing the sample to a proton beam under vacuum conditions. To optimize the sensitivity in a larger region of atomic numbers two measure ments at 2 and 4 MeV proton energy were performed for each sample. The calibration of the experimental setup was checked by a NBS bovine liver standard. The agreement with the specified contents is better than 10%. Details of the method have been reported elsewhere (5, 17). It has been shown that the above-mentioned small sample masses can be taken as representative, at least for the main part of a single liver lobe (9, 14). In addition, it has been demonstrated that the variation of the trace element contents within one liver is negligible compared to the variation observed in livers from different individuals. Data analysis has been performed by a least squares fit of the total X-ray spectrum. The resulting trace clement masses are related to the dry weight sample mass by a special normalization, using the Bremsstrahlung background in the X-ray spectrum (5). Ab solute contents given on wet weight basis are calculat ed from the mean values of the ratio wet weight/dry
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 8/8/2018 4:22:41 AM
1 Marburg, Institute of Pathology. 2 Twins.
227
Contents of Trace Elements in the Human Liver before Birth
ppm
200f
%
ppm
Ca
% Mn
ppm %
1000-
Fe
ppmn
%
-200
-100
m 1
1-
t -50
Pi
Mo
If 1
S
60
-200
-
'/yyz
•
-100 ■ 4
p
-20 -
2
-50 m
-10
Pi
/ ////
•
i.O -
y.
% ' -400 Rb
* -
-5 •
-25
’i
Fig. 1. Trace element content in fetal livers in comparison to the range of values found within the adult control group. Left scale = Mass content in ppm (parts per million) wet -50 weight (log scale); right scale = deviation from the medial value of the adult control group in percent (log scale); • x = values for the single
fetal livers (x = tw ins),------ = medial values for the fetal livers; ------ bar = medial values for the adult control group and standard deviation (calculated for the logarithm of the single values); 0 = range of values for the adult control group as given in table III.
weight which was obtained from 100 measured liver specimens (wet weight/dry weight = 3.9). Statistical evaluations were performed using the Student’s t test (2). Significant deviations were as sumed if the p value was less than 1%.
Besides the enhancement of the contents of copper and zinc in fetal livers the small content of the essential element molybdenum in the fetal liver is the most remarkable result. It is only 13% of the value found in adult livers. The large SD given for Mo is mainly due to statistical errors, because the contents of Mo were decreased to levels near the detection limit. The element contents of the essential trace elements are shown separately for the fetal livers in figure 1 together with the range of values and the medial value found for the adult control group. Surprisingly, the iron content in the fetal livers of twins differed by nearly a factor of two (No. 103/77, 104/77, table I). Besides iron
Results
In table III, the mean values, relative standard deviations and the ranges of the mea sured element contents are shown for the fetal and adult livers. The ranges of element contents are given together with the percentage of values falling within the indicated range. In the adult control group the elements were found to be logarithmic normally distributed.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 8/8/2018 4:22:41 AM
m
Meinel/Bode/Koenig/Richter
228
Table III. Element content of the fetal and adult livers
Fetal liver (n = 9)
Element
P
S
Cl
K
x, ppm xH, ppm Range, ppm
3,152 2,853 2,083 4,233 26 100
2,349 2,324 2,075 2,706 8.5 100
2,515 2,474 2,193 2,922 10.6 90'
3,032 2,947 2,411 3.675 13.8 100
3,011 2,933 2,083 3,717 13.3 100
2,584 2,538 1,890 3,130 10.8 100
1,848 1,648 940 3,028 28 100
3,064 2,953 2,0304,020 15.7 100
v/Variance, % Percentage, within range Adult liver (n = 37)
x, ppm xH, ppm Range, ppm' x/Variance, % Percentage within range
Ratio xfeta|/xadult t test p, %
1.0
0.9 0.14
1.4 0.14
1.0
x = Mean value; xyj = mode value; p = probability that the difference between fetal and adult livers is not significant (values given only if p is less or about 1%). All values given are calculated on the basis of a logarithmic normal distribution. Contents are given on a wet weight basis.
the lead content also differed widely. It was about eight times higher in the liver of the twin who showed the lower iron content. But in the case of lead one has to consider the large variations within one liver (9). Also, the low CD values in fetal livers are worth mentioning. In 5 of 9 cases the cadmium content was below the detection limit of about 0.1 - 0.2 ppm (parts per million) wet weight.
the range. No significant age dependence has been observed for the contents of Fe, Zn, Se and Rb in adult human livers (15). Therefore, the values of our adult control group can be ta ken as representative for the hepatic trace ele ment content of adults, although the mean age of the subjects of this group is comparatively high. This holds at least for the above-men tioned elements. In agreement with this as sumption none of the trace elements analyzed exhibits a significant shift with increasing age.
Discussion
The trace element contents of the adult control group determined in this study are consistent with published data (3, 15, 22—24). This holds for the absolute values as well as for
Comparison o f the Element Content between Fetal and Adult Livers Copper The most significant difference in the trace element contents between the two groups has been found for copper (fig. 1; table IV) (24,
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 8/8/2018 4:22:41 AM
Adult Control Group
Contents of Trace Elements in the Human Liver before Birth
Fe
Cu
Zn
1.15 0.72 0 .3 2.44 60 100
467 338 175774 49 100
103 97 80 146 20 901
151 135 100 203 27 100
1.58 1.38 0.762.96 30 100
306 221 113 629 49 100
6.5 5.8 3.610.0 29 95
75 64 37 129 33 100
Ca
Mn
83 80 70107 14.7 90‘ 62 59 45 91 19 97 1.34 0.007
0.7 0.13
1.5 0.91
Br
Se
0.33 0.3 0.16 0.48 26 95 1.2 1.15
Mo
Rb
0.41 4.5 0.38 2.3 0.271.5 9.4 0.59 75 20 100 100
2.0 IO’ 5
16 C IO '7
229
Cd
Pb
3.6 3.3 2.5 5.2 25 100
0.1 0.04
Contents of Trace Elements in the Human Liver before Birth1 B. Meinet*, J.C. Bode*, W. Koenig** and F.-W. Richter** Fachbereiche Humanmedizin* und Physik**, Philipps-Universität, Marburg/Lahn
Key Words. Trace elements • Fetal/adult human livers Abstract. A multielement analysis was carried out in nine fetal livers (age: 23 weeks of gestation to term) in comparison to an adult control group (n = 37). 15 elements (phosphorus to lead) were analyzed by proton-induced X-ray emission (PIXE). Significant differences between the two groups, expressed as ratios (R) of fetal to adult liver values were observed for copper (R = 16), zinc (R = 2.0), iron (R = 1.5), calcium (R = 1.34), selenium (R = 1.22), manganese (R = 0.7), and molybdenum (R = 0.13). An interesting point seems to be the inverse relation of copper and molybdenum. The negative correlation of these two trace elements which is known from animal experiments has now been confirmed in the human liver.
It is well known that in human fetal liver the copper and zinc content is considerably higher than in the adult liver (23—25). Similar high cop per and zinc contents have been observed in fetal livers of animals of various species (18, 23). The aim of this study was to get information on the content of the main essential and toxic trace elements in the liver of mature or im mature newborns.
1 Supported by the ‘Deutsche Forschungsgemein schaft’.
Materials and Methods
Livers of 9 newborns (age range from 23 weeks gestation to term) have been analyzed. Cause of death in all cases was a maternal complication during late pregnancy or delivery. None of the 9 cases exhibited signs of malformation or inborn errors of metab olism. Further details are given in table I. For comparison, trace element contents in the liver of 37 adults have been determined. Postmortem samples were taken only when the histological diagno sis of the liver showed no significant pathological change. In addition, those patients were excluded from the study who had diseases which are known to influence the trace element contents in the liver (e.g., disease of the blood-forming system, metabolic dis eases). Data on sex and age distribution of the adult control group are given in table II.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 8/8/2018 4:22:41 AM
Introduction
226
Meinel/Bode/Kocnig/Rich ter
Table 1. Data on sex, age and weight of the members of the newborn group
Identifier1
Sex
Age,
Body
Body
weeks of
lengths cm
weight
Liver weight
g
g
gestation
Liver histology
103/772
30 38
40
2,080
32.1 64.7
52
2,990
9 9
23 24
29
420
17.8
zyanosis
22
-
14.8
zyanosis
174/77 183/77 184/77
141
partially autolysed normal
Table II. Age and sex distribution of the adult control group
Number of cases
n = 37
Age
mean = 69a
Sex
n ( d ) = 23 n (9)= 14
± SD= ± 11a n (d)/n = 62% n (9)/n = 38%
Trace element analysis was performed by protoninduced X-ray emission (P1XE) which allows a multi element analysis of most of the essential and toxic elements within one sample at the same time (5, 17). After homogenization of about 0.1 -1 g of liver tissue and diluting the homogenate with bidistilled water, 10 /al of a 1:10 dilution were pipetted on thin formvar foil (2 -5 Mg/cm2) and dried by infrared light there after. No metal instruments were used. The homoge nates were stored at 20 °C in polystyrol containers. Instruments and containers were cleaned by con centrated HNOj (65%) and bidistilled water. The
homogenate layers on the formvar foil were analyzed by observing the characteristic X-rays of the trace elements, by exposing the sample to a proton beam under vacuum conditions. To optimize the sensitivity in a larger region of atomic numbers two measure ments at 2 and 4 MeV proton energy were performed for each sample. The calibration of the experimental setup was checked by a NBS bovine liver standard. The agreement with the specified contents is better than 10%. Details of the method have been reported elsewhere (5, 17). It has been shown that the above-mentioned small sample masses can be taken as representative, at least for the main part of a single liver lobe (9, 14). In addition, it has been demonstrated that the variation of the trace element contents within one liver is negligible compared to the variation observed in livers from different individuals. Data analysis has been performed by a least squares fit of the total X-ray spectrum. The resulting trace clement masses are related to the dry weight sample mass by a special normalization, using the Bremsstrahlung background in the X-ray spectrum (5). Ab solute contents given on wet weight basis are calculat ed from the mean values of the ratio wet weight/dry
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 8/8/2018 4:22:41 AM
1 Marburg, Institute of Pathology. 2 Twins.
227
Contents of Trace Elements in the Human Liver before Birth
ppm
200f
%
ppm
Ca
% Mn
ppm %
1000-
Fe
ppmn
%
-200
-100
m 1
1-
t -50
Pi
Mo
If 1
S
60
-200
-
'/yyz
•
-100 ■ 4
p
-20 -
2
-50 m
-10
Pi
/ ////
•
i.O -
y.
% ' -400 Rb
* -
-5 •
-25
’i
Fig. 1. Trace element content in fetal livers in comparison to the range of values found within the adult control group. Left scale = Mass content in ppm (parts per million) wet -50 weight (log scale); right scale = deviation from the medial value of the adult control group in percent (log scale); • x = values for the single
fetal livers (x = tw ins),------ = medial values for the fetal livers; ------ bar = medial values for the adult control group and standard deviation (calculated for the logarithm of the single values); 0 = range of values for the adult control group as given in table III.
weight which was obtained from 100 measured liver specimens (wet weight/dry weight = 3.9). Statistical evaluations were performed using the Student’s t test (2). Significant deviations were as sumed if the p value was less than 1%.
Besides the enhancement of the contents of copper and zinc in fetal livers the small content of the essential element molybdenum in the fetal liver is the most remarkable result. It is only 13% of the value found in adult livers. The large SD given for Mo is mainly due to statistical errors, because the contents of Mo were decreased to levels near the detection limit. The element contents of the essential trace elements are shown separately for the fetal livers in figure 1 together with the range of values and the medial value found for the adult control group. Surprisingly, the iron content in the fetal livers of twins differed by nearly a factor of two (No. 103/77, 104/77, table I). Besides iron
Results
In table III, the mean values, relative standard deviations and the ranges of the mea sured element contents are shown for the fetal and adult livers. The ranges of element contents are given together with the percentage of values falling within the indicated range. In the adult control group the elements were found to be logarithmic normally distributed.
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 8/8/2018 4:22:41 AM
m
Meinel/Bode/Koenig/Richter
228
Table III. Element content of the fetal and adult livers
Fetal liver (n = 9)
Element
P
S
Cl
K
x, ppm xH, ppm Range, ppm
3,152 2,853 2,083 4,233 26 100
2,349 2,324 2,075 2,706 8.5 100
2,515 2,474 2,193 2,922 10.6 90'
3,032 2,947 2,411 3.675 13.8 100
3,011 2,933 2,083 3,717 13.3 100
2,584 2,538 1,890 3,130 10.8 100
1,848 1,648 940 3,028 28 100
3,064 2,953 2,0304,020 15.7 100
v/Variance, % Percentage, within range Adult liver (n = 37)
x, ppm xH, ppm Range, ppm' x/Variance, % Percentage within range
Ratio xfeta|/xadult t test p, %
1.0
0.9 0.14
1.4 0.14
1.0
x = Mean value; xyj = mode value; p = probability that the difference between fetal and adult livers is not significant (values given only if p is less or about 1%). All values given are calculated on the basis of a logarithmic normal distribution. Contents are given on a wet weight basis.
the lead content also differed widely. It was about eight times higher in the liver of the twin who showed the lower iron content. But in the case of lead one has to consider the large variations within one liver (9). Also, the low CD values in fetal livers are worth mentioning. In 5 of 9 cases the cadmium content was below the detection limit of about 0.1 - 0.2 ppm (parts per million) wet weight.
the range. No significant age dependence has been observed for the contents of Fe, Zn, Se and Rb in adult human livers (15). Therefore, the values of our adult control group can be ta ken as representative for the hepatic trace ele ment content of adults, although the mean age of the subjects of this group is comparatively high. This holds at least for the above-men tioned elements. In agreement with this as sumption none of the trace elements analyzed exhibits a significant shift with increasing age.
Discussion
The trace element contents of the adult control group determined in this study are consistent with published data (3, 15, 22—24). This holds for the absolute values as well as for
Comparison o f the Element Content between Fetal and Adult Livers Copper The most significant difference in the trace element contents between the two groups has been found for copper (fig. 1; table IV) (24,
Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 8/8/2018 4:22:41 AM
Adult Control Group
Contents of Trace Elements in the Human Liver before Birth
Fe
Cu
Zn
1.15 0.72 0 .3 2.44 60 100
467 338 175774 49 100
103 97 80 146 20 901
151 135 100 203 27 100
1.58 1.38 0.762.96 30 100
306 221 113 629 49 100
6.5 5.8 3.610.0 29 95
75 64 37 129 33 100
Ca
Mn
83 80 70107 14.7 90‘ 62 59 45 91 19 97 1.34 0.007
0.7 0.13
1.5 0.91
Br
Se
0.33 0.3 0.16 0.48 26 95 1.2 1.15
Mo
Rb
0.41 4.5 0.38 2.3 0.271.5 9.4 0.59 75 20 100 100
2.0 IO’ 5
16 C IO '7
229
Cd
Pb
3.6 3.3 2.5 5.2 25 100
0.1 0.04
