METAL IQN ANALYSIS BY EMISSEON

SPEC‘EROS‘COPY OF STARCH BLOCK

ELECTROPHQRETQCALLY SEPARATED
HUMAN SERUM PROTEINS

Thesis {or “we Dance of M. 5.
MECHFGAI‘I STATE UNIVERSITY
Mary Jean Long
i961

TH FSIS

C >—
LIBRA R y

Mfibjga :1 Sta Ce
mversity

 

 

 

MICHIGAN STME UT-QI‘JERSIT‘I’

EAST LANSING, MICHIGAN

 

ABSTRACT

METAL ION ANALYSIS BY EMISSION SPECTROSCOPY OF
STARCH BLOCK ELECTROPHORETICALLY
SEPARATED HUMAN SERUM PROTEINS

by Mary-'Jean‘Long

This study was undertaken to extend investigation in the metallic
elements associated with normal hurnanserum albumin and the alpha,
beta and gamma globulins. . Such new information is of valueiini con-
nection with .more completely understanding the nature and function of
metallo-protein complexes in normal blood serum.

Starch block electrophroesis was employed to separate the normal

serum protein fractions. The fractions were then eluted from the starch
segments with physiological saline solution and protein content identified
by measuring absorbancies with the Beckmana. DU spectrophotometer.
. These readings were plotted versus section number. Biuret determin-
ations were performed on the saline eluted proteins also, and readings
were plotted versus section number. The plots revealed the eluates to
be combined to give the respective blood» seruIn'protein fractions.

The results of this study lead to the major conclusion that emission
spectrosc0py applied to human serum protein fractions separated by
starch block electrophoresis shows promise of greater precisionand
accuracy in analysis for their metallic element content thanheretofore

achieved.

l‘v‘lETAL ION ANALYSIS BY EMISSION SPECTROSCOPY OF
STARCH BLOCK ELECTROPHORETICALLY
SEPARATED HUMAN SERUM PROTEINS

BY

Mary Jean Long

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Chemistry

1961

ACKNOWLEDGMENT

The author wishes to express appreciation and gratitude to:
(1) Dr. Hans A. Lillevik and the staff of the Chemistry Department
for guidance and counsel; (2) General Diagnostics Division of
Warner-Chilcott, Morris Plains, New Jersey, and the American
Society of Medical Technologists for the incentive and Scholarship
that started the pursuit of this work; (3) Dow Chemical Company,
Midland, Michigan, for a summer Fellowship which enabled the
author to carry this project to completion; and (4:; Dr. Robert Bo
Foy of the Edward W. Sparrow HOSpital, Lansing, Michigan, for
advice and supply of human serum specimens.

):< >',< >:< >:< ::: >:< ::< >§< >:< >',< 9,: >:< >§< a}: 3::

ii

VITA

The author was born March 28, 1914 in Duluth, Minnesota,
and her secondary education was completed in 1931 at East Grand
Rapids High School, East Grand Rapids, Michigan. In 1935 she
graduated from Michigan State University with a Bachelor of Science
degree. She was admitted to the School for Advanced Graduate
Studies of Michigan State University in September 1959 and has been
in attendance since.

Her employment experience has included that of Laboratory
Technician with Park, Davis and Company, Detroit, Michigan, and
as Medical Technologist 1954-1959 at Blodgett Memorial Hospital,
Grand Rapids, Michigan. She is a member of the American Society
of Medical Technologists, the Michigan Society of Medical Technolo-

gists and the Western Michigan Society of Medical Technologists.

iii

II.

III.

IV.

V.

BIBLIOGRAPHY . . . ........... ' ...........

TABLE OF CONTENTS

. INTRODUCTION ........................

HISTORICAL ..........................

A . Emission Spectroscopy ............

B. Electrophoresis ................

EXPERIMENTAL . ......................
A. Equipment ......................
B. Materials and Reagents . . . . ...........

C. Methods . . . ....................

DISCUSSION . . .....................
A. The Biuret Method of Protein Analysis versus DU

Absorbancies ....................

B. The Use of Three Milliliter Samples .......

C. Sponge Rubber as a Supporting Medium ......

D. Metal Ion Distribution in Serum Protein Fractions

SUMMARY ...........................

iv

Page

LA)

«Joe.»-

21

21
21
23
23

LIST OF TABLES

TABLE

II.

III.

IV.

VI.

VII.

VIII.

. Materials Sequence of Spectra, Medium Quartz

Spectrograph, Figure 8 . . . . . . . . .....
Metal Identification Data Obtained from Plate 1

Materials Sequence of Spectra, Medium Quartz
Spectrograph, Figure 8 . .

Metal Identification Data Obtained from Plate 2

. Metal Identification Data Obtained from Plate 3

Metal Identification Data Obtained from Plate 4

Metal Identification Data from Plate 1 by Visual
Comparison . .

Composite of Tables II, IV, V, VI, VII

Page

14

15

16

16

18

19

19

20

FIGURE

1a.

LIST OF FIGURES

E-C Pressure Plate Electrophoresis Apparatus .
E-C Power Supply . . .............

Comparison of Protein Analysis by E280 Absorbancy and
Biuret Method of Analysis . . ......

Comparison of Protein Analysis of E280 Absorbancy and
Biuret Method of Analysis .

Dispersion Curve. . . . . . . . . . ......

Biuret Protein Analysis of the Eluates of Starch Block
Sections When all Three Compartments of the E-C
Apparatus are Used. Each Compartment is graphed
separately . . ...... . . . . ......

Absorbance vs. Tube Number of Eluates of the Foam
Rubber Sponges After Electrophoresis of a Human
Serum Sample ....... . . . . . .

Biuret Protein Analysis of the Eluates of Starch Block
Sectionsvs. TubeNumber. . . . . . . . . . . . . .

Spectrographic Plates.

vi

Page

10

11
17

22

24

25
26

I. INTRODUCTION

One of the oldest and most important problems in biochemistry
is the study of protein-metal complexes. In all biological systems,
proteins are found in combination with or associated with metal ions.
Human blood serum is no exception. Here are found various metals
acting as cofactors in enzyme systems, copper forming a complex
with serum albumin and iron chelating with various protein fractions
(notably beta globulin), and calcium playing a role in the clotting
mechanism.

It was the purpose of this study to identify the metal ions of
human blood serum protein fractions after separation by electro-
phoresis on a starch block. Emission spectroscopy was the method

of metal identification employed.

II. HISTORICAL

A . Emission Spectroscopy

The emission spectrograph is often used in other than metal-
lurgical problems although that is one of its main applications. It has
been used extensively to determine the contents of alkali and alkaline
earth metals in plant and animal tissues, including such materials
as blood serum. The early workers in this field employed flame
methods. They ashed the material to be examined, folded it in a
filter paper and burned the entire packet in a flame while recording the
emissions on a photographic plate (37).

Le Duc (23) attempted spectroscopy on serum protein fractions
separated by paper electrophoresis. She reported that examination
of the Spectrophotographic plate showed no significant differences
between the eluted protein fractions, paper, and electrodes. She pre-
sumed that the cause of her failure was due to the minute quantities
of metals present in the 0. 2 m1. of serum applied to the paper strip.
Even 1 m1. of serum applied on heavier paper produced similar incon-
clusive results. Assuming that the cause of her failure was due to the
metal impurities in the paper, this study was carried out with spectro-
graphic analyses of protein fractions prepared by electrOphoresis on
a starch block.

Judd Lewis (37) in 1912 was the first to apply spectroscopy as a
regular means of analysis in industrial problems. By 1936 he described
a method for preparing samples of animal or vegetable matter for the
emission spectrograph. In 1928 Lowe (37) demonstrated success with
the use of carbon electrodes. He recommended placing the liquid
sample in the cavity of the lower electrode and included the arc
spectrum of the carbon electrode for comparison. Lowe's method was

adopted in this investigation.

B. Electrophoresis

Electrophoresis is defined as the migration of charged particles
in an electric field toward the oppositely charged electrode. Many
types of supporting media have been investigated. Tiselius after his
development of the moving boundary apparatus in 1933 demonstrated
that human blood serum separated by paper electrophoresis at pH 8. 6
in O. 05 to 0. 10 M Veronal buffer resulted in five apparently different
components (4).

Kunkel and Slater (22) in 1952 carried out a series of experi-
ments involving zone electrophoresis on various types of media in a
comprehensive search for an ideal support free of the disadvantages
of filter paper. Of all the materials studied, the starch block was
found to produce the most satisfactory results. The starch block was
made from raw, insoluble, unhydrolyzed starch granules. Around
1955 Smithies (31, 32) introduced the use of starch gel for zone electro-
phoresis. This starch was hydrolyzed with heat to form the gel,
which at the completion of the electrophoretic preparation could be
stained, washed and preserved similar to paper. Mitchell and
Herzenberg (25) successfully tried foam rubber sponges for the support-
ing material in the preparative electrophoretic separation of small
amounts of protein and pointed out that the advantage of this medium
is the ease with which the protein fractions could be recovered.

This study was commenced on paper to familiarize the investi-
gator with the technique and apparatus for serum protein separation,
continued on starch block for the major part of the investigation, and
for comparison one electrophoretic run was tried on foam rubber

sponges.

III. EXPERIMENTAL

A . Equipm ent

Electrophoresis Apparatus (Figure 1). -— The EwC Apparatus manu-

 

factured and sold by the E-C Apparatus Company, 538 Waln'-3;t Lane,
Swarthmore, Pennsylvania, was used. It is constructed of lucite and at
each end contains three separate compartments with platinum electrodes.
A water cooled lucite bed is provided to support the paper or starch
medium and a water cooled lucite upper plate is clamde over a pressure
pad. A 2000 volt power supply unit was also provided by the E-«C
Apparatus Company. Paper strips were first used in the apparatus to
observe the behavior of normal human serum samples under varying
conditions of pH, time and voltage. These strips were heavy 3MM
Whatman paper 46. 5 x 6. 5 cm. Foam rubber sponges 83xl3x13 mm.
were also used as supporting medium. Lucite strips of the dimensions
of the starch block-bed were made and clamps to hold them in position
obtained. These lucite strips, one, 45 x l. 5 x 0.6 cm. and two strips

6. 5 x 1. 5 x 0. 6 cm. , were necessary to subdivide the lucite bed into

smaller starch block sizes.

Beckman DU SpectrOphotometer, Power Supply and Hydrogen

 

Lamp. --The instrument manufactured by the Beckman Instrument
Company, Fullerton, California, was used to estimate the amount of

protein in saline elutes.

Beckman Model B Spectrophotometer. --This instrument was also

 

used to carry out the Biuret determinations on the above mentioned

protein solutions .

Bausch and Lomb Medium Quartz Emission Spectrograph. --

 

Emission Spectroscopy was carried out on this instrument made by

Bausch and Lomb Incorporated, Rochester, New York.

 

Figure 1. E-C Pressure Plate Electrophoresis unit: transparent
plastic with non wettable migration surfaces and platinum
electrodes; migration path direct-contact cooled; 2500 ml.

volume .

 

.Figure 1a. E—C Power Supply: input 110-120 v.; output adjustable
0-1000 v. at 200 ma. or less; two units can be connected

in series for 2000 v. output.

Bausch and Lomb Littrow Spectrograph. --This instrument was

 

also used to supplement the other spectrograph and is manufactured and

sold by the same company.

Arc and Spark Stand. --Electrode holders were carried on

 

porcelain insulators attached to rack and pinion movements by which
they were raised or lowered as indicated or desired. A voltage of 110

volts, 4. 5 amperes direct current was employed to supply the arc.

Special Graphite Spectroscopic Electrodes. --The electrodes used

 

in this investigation were obtained from the National Carbon Company,

Cleveland, Ohio.

' Kodak Spectrum Analysis Plates. --Number 1 Kodak Spectrum

 

Analysis Plates were used throughout the experimental work. They were
obtained from the Eastman Kodak Company, Rochester, New York. They

measured 4 x 10 inches and were 1 mm..thick.

Rocker Type Plate Developer. --The plate developer was also a

 

product of the Eastman Kodak Company.

B. Materials and Reagents

Starch. --The supply of starch used was labelled Fisher Potato
Starch No. 5-513 and obtained from Fisher Scientific Company, Fair
Lawn, New Jersey. It was purified by suspending 500 grams in 2 liters
of O. 1 M HCl, stirring well and allowing to settle. The supernatant was
decanted and the washing procedure was repeated six times with distilled
water or until the starch was neutral to pH-Hydrion paper indicating pH
7. 0. The final wash was made with glass distilled water. Before
pouring the starch into the lucite block-form the water was decanted and
the starch resuspended in an approximately equal volume of 0. 1 M

Veronal buffer of pH 8. 6.

Buffer. --The Veronal buffer used was prepared by dissolving

 

4. 00 g. sodium hydroxide pellets and 20. 26 g. Veronal in distilled water

and diluting to 1 liter.

Biuret Reagent. --The Biuret reagent was prepared by dissolving

 

1. 5 g. of crystalline cupric sulfate pentahydrate, and 6. 0 g. sodium
potassium tartrate in about 500 ml. distilled water in a 1 liter volumetric
flask. To this was added while swirling, 300 m1. freshly prepared,
carbonate free, 10% sodium hydroxide. The mixture was diluted to

1 liter with distilled water and stored in a polyethylene bottle.

Solutions for Deve10ping Plates. --Three solutions were used for

 

deve10ping the spectroscopic plates. The developer, Kodak K-19 was
made according to the directions furnished by the suppliers. The stop
solution was 1% acetic acid. The fixer solution was sodium bisulfite

dissolved in water to make a solution of 20%.

Physiological Saline. --The saline used to elute the protein from

 

the starch sections was prepared by dissolving 17 g. sodium chloride in

2 liters of distilled water.

Human Blood Serum Samples. --Serum samples were supplied

 

from the Clinical Laboratory of the Edward W. Sparrow Hospital,

Lansing, Michigan.

Digestion Solution. --The solution used to digest the protein and

 

non-protein nitrogenous materials in the samples preparatory for
spectroscopy, was made from 16 M nitric acid and 60% perchloric acid

in the proportion of two to one.

C . M ethods

Electrophoresis on Starch Block. --A wax paper box was prepared

 

by taking a piece of paraffin impregnated paper 57 x 10 cm. , folding it

in 6 cm.. at each end and leaving a base 45 cm. long. The sides were
folded in 18 mm. each leaving a bottom 6. 5 cm. wide. This paper box
was placed on the lower lucite bed. A 45 cm. long lucite strip was
placed along one side of the wax paper box for support. Two 6. 5 cm.
long pieces of lucite strip were placed on the inside of the box at the
ends to form square corners. Two extra lucite strips were placed
across the corners of the box where the long and the short lucite st rips
met and held in place with metal C-clamps.

The electrode vessels were filled with buffer and allowed to come
to the same level. Starch, previously purified, was suspended in 0.1 M
Veronal buffer pH 8. 6 and poured into the paper box-form to cast a
block about 0. 5 cm. thick. After the starch settled for ten minutes, the
supernatant buffer was blotted off with strips of absorbent paper cut to
size. The starch block was allowed to air dry until a test slit made with
a Spatula maintained its shape. This time varied from one to two hours
depending on the relative humidity. The cooling plate water supply was
not turned on because it was found that water condensing on the plate
from the atmosphere during humid weather kept the block wet thereby
delaying the drying process considerably.

When the barely moist block could be cut easily, a 1 cm. wide
transverse section was removed 10 cm. from the right edge of the lower
lucite plate at the point of sample application. This section of starch
was placed in a clean petri dish and 1 ml. of human blood serum for
electrophoresis was applied. The serum and starch were thoroughly
mixed. A stirring rod covered with tygon tubing and sealed at one end
to form a flat straight surface was used to transfer the starch serum
mixture quantitatively from the petri dish into the space from which the
section of starch was originally removed.

After the serum-starch mixture had been returned, the two metal

clamps and the lucite strips were removed, the ends of the paper box

torn off and contact with the electrode vessels established by means of
two paper strips previously moistened with buffer. The paper strips
were laid on the starch at each end of the block for a distance of five
centimeters and allowed to dip down into the buffer in the electrode
chambers. . The use of cellulose sponges provided for this purpose was
discarded since they caused the buffer to appear cloudy.

The entire block was now covered with a plastic sheet, the pressure
pad was placed on top of the sheet, and the top lucite cooling plate
clamped over the whole assembly. The power supply of the E-C Apparatus
was connected. The current was adjusted to 13 milliamperes under a
potential of 450 volts and allowed to run for twenty-four hours with the
cold water running at 180C. through the cooling plates.

At the completion of a run, the current was turned off and the upper
layers of pad and plastic film removed. Using a clean Sharp spatula,
the whole starch block was cut into 1 cm. transverse sections. Each
section was placed into a test tube containing 5 m1. physiological saline.
Each tube was thoroughly mixed by shaking, allowed to settle, its
supernatant poured into a 15 m1. centrifuge tube and centrifuged for five
minutes at nearly full speed. About 3-4 ml. clear supernatant was
decanted into a clean test tube and used for determining its absorbancy
at 280 mp. in the Beckman DU Spectrophotometer. The result gave a
measure of the protein content in each starch section. These solutions
were saved and then Biuret determinations performed on them. The
Biuret absorbancies were read on the Model B Beckman instrument at
540 mu. The two sets of absorbancy values were plotted on graph paper
against tube number to compare the results which are Shown in Figures
2 and 3.

From the graph (of Figures 2 and 3) the peaks representing the
five serum protein fractions were identified. The eluates could then be

combined into their various fractions or discarded according to the

10

 

 

 

 

 

 

 

 

 

 

Biuret UV
.6 '1
-. 20
I
'l
'I
I
:I
I.
II
I—. 15 II
I I
I I
I I
I I
I I
. 3 .. I l
I I DU
, I
. I
- 10 I I
. . I I
origin I I
I I
I I
I l
l l
N I \
To 05 MM I‘ \
\
I \
I I ————— -—B‘;' rkT
,' I
/I\ I l
,’/:y“\ ’1’ \\\\A \“‘/\“"/\J/1b ‘I
I"_ \ -I’\ I I z”‘\’ fr“
, " "T’Ls-“v' L I I3 0'3 ial a umi ‘1 Tube No.
0 10 20 30 40
Figure 2. Comparison of protein analysis by E280 Absorbancy

and Biuret Method of Analysis.

11

 

 

 

 

 

Biuret UV I
II
II
‘I
I
I
. I
.55_ I ‘
: I
I I
‘ I
I-.ZO ' I
l I
l I
I I
.45— , I!
' I
I I
I I
I I
' 15 I I
. I
‘ ~-
I
I
.30_ I
I
origin I
\ ,— I
" 10 I" \I I
I ,’ ‘. '
I‘ //' I \ ”I '
I\ I I X I, I If B.""r
I V I. I \ I I I. I,
I \ I | I\ I v
I I I ‘.-' ‘ ’l I ________ D._|
1‘ I I \-
'15—. . .,"“ \\\I I II
o‘\ /‘\a\ ’1‘
”I 05‘-" \I V ‘ ‘I ‘I/
V . I B Pl “bk. r... N-
0 10 20 30 40

Figure 3. Comparison of protein analysis by E280 Absorbancy
and Biuret Method of Analysis.

12

information obtained from the analytical results and the graph. Each
solution of combined eluate for each fraction was gently evaporated to
approximately 2 ml. on an electric hot plate. The method of Boyle,
Whitehead it a}. {5) for oxidizing the protein with a nitric acid-perchloric
acid mixture was used by adding ten ml. of the acid mixture to each
solution of concentrated protein fraction. The fractions were thus pre-
pared for Spectroscopy. They were stored in the refrigerator at 4

degrees C. between analyses.

Spectroscopy of Protein Fractions. --New electrodes were prepared

 

before each Spectrographic analysis (7). By placing 2-3 drOpS of the
concentrated protein fraction into the cup of the lower electrode and
passing a direct current arc, Spectra were photographed and developed.
Spectrographic studies were also made concurrently by using salts of

the metals of interest to the investigation and obtaining their Spectra on
the same plate. Thesseliiancctlrfded physiological saline; starch alone;

carbon, copper and iron electrodes; cobalt carbonate, manganese chloride,
zinc carbonate, magnesium oxide and magnesium carbonate,and calcium
chloride (Tables I, III).

The wavelengths of unknown lines were determined by marking the
plates while scanning each pair of Spectra with a magnifying lens, then
measuring the unknown lines and interpolating their respective wave-
lengths from a dispersion curve, Figure 4. The dispersion curve was
previously prepared by running Spectra of copper or iron, measuring
recognizable lines as found in the standard tables prepared by Harrison
(21) and plotting these measurements against their known wavelengths.
Marking was carefully done with waterproof India ink, always labelling
the Spectra in proper order and using the lines of the copper and iron
Spectra as reference standards. The exact points where a match occurred

was marked, or where a line appeared which was neither carbon, from

13

the electrodes, nor sodium, from the physiological saline. These lines

were then identified using Brode's tables (6). The results are Shown in

Tables 11, IV, V, VI, VII and VIII. Additional runs were made on

unoxidized protein and data compiled as before. This data is presented

in Tables V and VI.

14

Table 1. Materials Sequence of Spectra, Medium Quartz Spectrograph,
Figure 8,1 see page 26, Plate 1, Unoxidized Proteins.

 

 

Exposure Material Exposure Material
1 Cu electrodes 11 Alpha 1 globulin 2
2 Fe electrodes 12 Alpha 2 globulin l
3 MgO and MgCO3 13 Alpha 2 globulin 2
4 Zn CO3 14 Beta globulin l
5 Ca3(PO4)2 15 Beta globulin 2
6 Albumin 1 16 Physiological
_ saline
7 Albumin 2
. j 17 Carbon electrodes
8 Gamma globulin 1
18 C l t d
9 Gamma globulin 2 u e ec ro es
1 F l t d
10 Alpha 1 globulin 1 9 e e ec m 88

 

1Current 4 amps; exposure 40 sec.

15

Table 11. Metal Identification Data Obtained from Plate 1.

 

 

 

Observed Line Approximate
Protein Fraction at cm. Wave length Brode
Albumin l3. 3 2940 2940 Mn
13. 2 2935 2936 Fe
11. 55 2760 2766 Cu
Beta globulin 8. 7 2540 2543 Fe
11. 55 2760 2766 Cu
13. 2 2935 2936 Fe
13. 3 2940 2940 Mn
Alpha 1 and 2 globulins 10. 55 2715 2712 Zn
10.73 2720 2721 Ca
10. 97 2700 2698 Mg
11.4 2750 2750 Fe
11. 55 2760 2766 Cu
12. 65 2860 2852 Mg
13.2 2935 2936 Fe
13.25 2937 2937 Fe
13. 7 2980 2975 Mg
14.22 3040 3040 Fe
15.05 3180 3179 Ca
16. 25 3420 3417 Co
16. 4 3460 3455 C0
17.9 3855 3856 Fe
17.95 3860 3859 Fe
18. 5 4060 4059 Zn
Gamma globulin ll. 56 2760 2766 Cu
13. 67 2995 3000 Fe
14.87 3340 3345 Zn

 

16

 

 

 

Table 111. Materials Sequence of Spectra, Medium Quartz Spectrograph,
A Figure 84. see page 26, Plate 2, Oxidigederoteins.

Exposure Material Exposure Material
1 Cu electrodes 8 Albumin
2 Fe electrodes 9 Alpha 1 globulin
3 Carbon electrodes 10 Alpha 2 globulin
4 Starch 11 Beta globulin
5 Zn CO3 12 Gamma globulin
6 MgO and MgCO3 13 Starch
7 Ca3(PO4)z 14 Carbon electrodes

 

lCurrent 4-4. 5 amps, time 60 sec.

Table IV. Metal Identification Data Obtained from Plate 2.

 

 

Observed line Approximate

Protein Fraction at cm. Wave length Brode
Albumin 10. 7 2660 2661 Fe
14.1 3060 3060 Fe

15. 68 3300 3307 Cu

Beta globulin 10. 7 2660 2661 Fe
14. 35 3080 3079 Mn

15.68 3300 3307 Cu

Alpha 1 and 2 globulins 12. 8 2900 2901 Fe
13. 8 3020 3018 Zn

14.96 3198 3199 Fe

15. 1 3200 3200 Fe

15. 5 3260 3260 Co

15.68 3300 3307 Cu

16.1 3400 3405 C0

16. 6 3500 3495 C0

Gamma globulin 6. 2 2340 2341 C0
6. 25 2360 2363 Co

9. 5 2580 2576 Mn

10. 6 2660 2661 Fe

13. 05 2920 2923 Fe

(triplet) (triplet)
14.4 3100 3100 Fe
15. l 3220 3228 Mn

 

17

190 ..

180 -

170 _

160 L /

150 L /

140 - /

130 I.

120 _ /

110 I'

100 r

90..

Figure 4. Dispersion Curve; obtained by plotting

Angstrom units vs. mm. measured with a hand lens
and micrometer of known lines of the copper Spectrum.
70 p
60 r
50 r

40-

30+—

 

 

20 1
2000 2800 3600 X

18

Table V. Metal Identification Data Obtained from Plate 3.

 

 

 

 

Observed Line Approximate

Protein Fraction at cm. Wave length Brode
Albumin 9.45 2570 2572 Mn
11. 55 2760 2766 Cu
13. 2 2935 2936 Fe
13. 3 2940 2940 Mn
16. 35 3440 3440 Fe
17. 5 3740 3740.Zn
Alpha 1 and 2 globulins 9. 3 2560 2559 Co
10.55 2715 2712 Zn
10. 73 2720 2721 Ca
10. 97 2700 2698 Mg
11.4 2750 2749 Fe
11. 55 2760 2766 Cu
12. 65 2860 2852 Mg
13.2 2935 2936 Fe
13. 25 2937 2937 Fe
13. 7 2980 2981 Fe
14.22 3040 3040 Fe
15. 05 3180 3179 Ca
16. 25 3420 3417 C0
16. 4 3460 3455 C0
17. 9 3855 3856 Fe
17.95 3860 3859 Fe
18. 5 4060 4057 Zn
Beta globulin 8. 7 2520 2521 C0
9. 3 2560 2559 Co
11. 55 2760 2766 Cu
13.2 2935 2937 Fe
13.3 2940 2941 Fe
14. 9 3180 3179 Ca
16. 35 3440 3440 Fe
Gamma globulin 9. 3 2560 2559 Co
11. 56 2760 2766 Ca
13. 67 2995 3000 Fe
14. 87 3340 3345 Zn
16. 35 3440 3440 Fe

 

19

Table VI. Metal Identification Data Obtained from Plate 4.

 

Observed Line Approximate

 

Protein Fraction at cm. Wave length Brode
Albumin 1.9 2140 2139 Fe
2.7 2180 2178 Cu
3. 95 2240 2238 Cu
5.8 2330. 2331 Fe
8.7 2520 2522 Fe
9.8 2595 2592 Mn
12.45 2860 2862.5 Fe
triplet 2863.4 Fe
2863. 8 Fe
13.8 3025 3025 Fe
Alpha 1 globulin 8.75 2525 2527 Fe
10. 00 2620 2618 Mn
12.3 2860 2862 Fe
Beta globulin 8. 75 2525 2522 Fe
Gamma globulin 10. 0 2620 2618 Cu

 

Table VII. Metal Identification Data from Plate 1 by Visual Comparison

 

 

Protein Fraction Wavelength Brode
Alpha 1 and 2 globulins 2392 2392 Cu
2618 2618 Cu
2244 2346 Cu
2766 2766 Cu
3500 3502 C0
2442 2442 Fe
2824 2824 Cu
2411 2410 Fe
Gamma globulin 2824 2824 Cu
Albumin 2825 2824 Cu
2600 2599 Fe

 

20

Table VIII. Composite of Tables 11, IV, V, VI, VII.

 

 

Protein Fraction Fe Cu Mn Mg Co Zn Ca
Albumin 14 6 4 1
Beta globulin 7 3 2 2 1
Alpha 1 and 2 globulin 21 9 6 10 6 4
Gamma globulin 6 4 2 3 2

 

)1:
Incidence of lines of the Spectra of metal ions occurring in the Spectra

of the protein fractions.

21

IV. DISCUSSION

Several conclusions may be drawn from the results of these experi-

ments involving starch block electrophoresis and emission spectroscopy.

A. The Biuret Method of Protein Analysis versus E230 Absorbancies

AS demonstrated in Figures 2 and 3 the Biuret method of analysis
for humanrserum protein content is preferable when quantitative protein
distribution results are essential. The absorbancies at E280 are Sufficient
in the case of the starch sections to designate which eluted portions to
combine for each protein fraction, but examination of the graphs readily
demonstrates that these readings were not the true proportionate value
for the quantity of protein present in each fraction. This is verification
of a fact that such values represent the pr0portion of aromatic amino
acid residues rather than true protein content. However,’ Tomb, Souter,
and MacLagan (36) contend that (Spectrophotometric determination of
proteins at E210 is a reliable quantitative method if the protein to be
investigated does not contain large amounts of aromatic amino acids.

In human serum fractions one should take into account the proportions
of tyrosine and tryptophan particularly in relating this to actual protein

content.
B. The Use of Three Milliliter Samples

Although it is possible to run up to 3 m1. samples of serum for
electrophoresis by the starch block method, Figure 5 shows that for
accurate results with the E-C Apparatus, it is advantageous to use one
ml. sample at a time. The difficulty appears to be due to diffusion of
serum into the starch in the first strip (compartment one) while loading

the serum Onto the other strips, and diffusion of serum in strips one

22

    
  
   

 

 

  

 

Absorbancies
. 30.-
. 25 -
I
I
I
I
-Q-9~¢*. Compartment 1 i
- ___,_,__ - Compartment 2 I’
. 20 I' I
Compartment 3 .I’
I,
'I
I
II.
II
I.
. 15 p :1
I;
I II
"I
I
If I
u
'I
,‘o
. 10 — {I
, I
'I
I O
I I
U“
'0
.‘I
."
. o5 origin ‘I
Tube No.

 

 

Figure 5. Biuret protein analyses of the eluates of starch block sections when

callI three compartments of the E-C Apparatus are used. The absorbancies of
the eluates of each compartment are graphed separately. Three m1. of serum

were separated in one starch block etectrophoresis.

23

and two while loading strip three etc. The data from the Biuret determin-
ations made separately on the three strips are Similar but not precisely
alike. In other words, superimposable curves are not obtained because
of the time factor. Also, it was observed that the protein fractions
obtained by electrophoresis of human blood serum Specimens and after
concentration by gentle heating and refrigeration produced crystals.

This demonstrates that the starch block is a reliable means of prepara-
tive electrOphoreSis which produces protein fractions of conSlderable

purity and reproducibility.

C. Sponge Rubber as a Supporting Medium

It was found in this investigation that the use of sponge rubber as
a supporting medium for electrophoresis was less satisfactory than the
starch block. The starch block could be eluted of protein by one extraction
with physiological saline while the Sponges necessitated three extractions.
The protein distribution curves obtained by Biuret analysis of the saline
eluates from the starch block sections were more uniform than those
from the sponges. See Figure 6. Mitchell and Herzenberg (25) and
Davidson (10) cite success with the use of sponges and point out the ease
of recovery of the protein fractions. .Nevertheless the observations of
this study indicated starch to be more easily and readily extractable for

its protein content .

D. Metal Ion Distribution in Serum Protein Fractions

To definitely establish the presence of a metallic element in a
protein fraction, one Should observe at least three lines of the element's
Spectra in the resulting photographic plates of the protein fraction (6).

On this basis and from the data compiled in Table V111 and photographs

24

Absorbance

.12I-

.10“

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. 05 I- F\/albumin
I (3
C*2. ‘11
/
origin
_ L l 1 Tube No.
0 10 20 30 40

Figure 6. Absorbance vs. tube number of eluates of the foam rubber
Sponges after electrophoresis of a human serum sample.

defined: moms“ .m> maofloom M003 Joyoum mo moudsfim 93 mo mamafimnm GEN—own Houdfim .N oudmfim

Om ONI I oh 0
.02 00.58 _ ~ m a A x. a
d d

E

Gmwfino

 

awgflm

25

log:

 

:wouonnm .mME

26

 

 

 

 

 

Figure 8. Spectrographic Plates

. ‘ ‘«- 1" . a d - 2 4 3‘
g‘m“‘ 9 “I .b i“ V‘ is
' ' f ‘ '9 d a a a I , 0 \q
" I 'I II I HORN-III! Fe
. , I' O , 0. 0 g 0‘ | .
I III-umuo OI I) II II I III Ill'JlllIl III~IIIII IIII I IIIIIIIImII—I- Cu
- 9, I 9 II l- IIIIIII M_‘ _ C
, ‘ , .I . III I- -IIIIIII ‘Ifirlljynlum’ IIII-m'm-II— StarCh
. .- I “ll ,w-iu-mmmw-pmmumm u-III_ y-globulin
. - ’ 5
1." II I IIIIIII-_,.~I'.~I~I~II IIII_ I3' H
." '- "'W .. “ION-III mm- r-I III-ugI-Iun. III_ 02’ H
I I Hm 33mm: l' umpw_ 01" H
I I. I! -IIuI ' "pa-Imm- I IIIIIIII-Imo-«III_ Albumin
I . II I. mu- ell- mmm‘l— Starch
. I I. 'I. _‘ I I . II III . L j I CaCIZ
o ‘ 1’ «MI H‘flmfl'l I. ImmmII—I_ ZnCO3
'I, I II ..I I Im ' ‘ ' MgO "I" MgCO3
“‘3 I _°' w-tMIIIIOIIII' C
I'ftv-Illllillm-II ' , Fe
II ._ ~qu um.— .1 Cl I. |' m u I ”no: um IIIIIQII'III II Iu-mm_ Cu
PLATE I
II III II "II II 'I I II- III I, III .I I II III I I IIIIIIJ _ “In"; C11
'aIry .230! I J‘ ’1 but 199 1:70
. I :1 IIIIIIIIIIII IIIIIIIIIII ' Fe
_ I IIIIIIIIlIImIm_ C
I II III IIIIIIIIIIIIIIIII IIIIIqu..,II II IJIIIIIIIIIIII II IJIIIII— Starch
IIIIIIInIImII-I-IwmIIIIIIIIIIIIIII IIIIIIuIIII_ CaCO3
.IIIIII ..I . III! IIIII um 21 :II: .I_ III. IIIII MnCl3
. I I IIIIIIII' IIIIIIIIIIIIIII IIIIIII II IIHIIIIIIIIII I mum— ZnCO3
. .. _,, ‘ " .MgO
| - - 7}.) . I ‘ CaClz
' ' . ' , - ‘ - NaCl
I\ III I ||I| I I I I 'llI lll*1 I II'IIIIIIII_ Albumin
I II‘ ' III III I IIIIIII.III|| III {IIIIIIIII'yI IHI‘HI IIIII I III_ 0.1-globulin
I . .1 I .
. g . I'IIIIIIIIIIIII IIIII I IIIII III: II IIIIIII_ az- "
\‘Z... #37“. ‘ III 11”.“. I. I III: u III-ul— B- "
I II I . —. I II I_ 7- "
PLATE II

27

of Figure 8, it may be concluded:

1. The albuminfraction contains iron, copper and manganese.

2.. The beta globulin fraction appears chiefly associated with iron.

3. The alpha one and alpha two globulin fractions have iron, copper,

magnesium, cobalt, zinc and calcium present.
4. The gamma globulin fraction contains iron, manganese and
cobalt. ‘

Le Duc found calcium and magnesium present in all fractions,
iron in the beta globulin and copper and zinc in the gamma globulin (23).

Foy (16) in his research on metal proteins of normal and patho—
logical human sera found in all cases copper, iron, magnesium, calcium
and zinc in all five fractions.

The results of this work when compared to the preceding show
agreement in the finding of iron in all fractions. A difference is that
Le Duc and Foy both observed calcium to be present in all fractions
whereas the method of identifying the metal ions present by emission
spectroscopy demonstrated calcium in the alpha globulins and albumin
only. Foy reported copper present in all serum fractions, Le Duc found
it appearing only in the gamma globulin fraction. This work demonstrated
copper as occurring in the alpha globulins and albumin. They both also
found magnesium and zinc in all fractions. Whereas in this work the
presence of these two metal ions was seen only in the alpha globulins.

Experimental research along these lines should be further extended
to a quantitative basis. This can be accomplished by using an internal
standard as Vallee (38) did in his work with zinc metallo complexes of
enzymes. - It is also recommended that extending the list of metal ions
to include chromium, beryllium, nickel and strontium (11, 27, 1) would
be most revealing.

The possibilities of application of the emission spectroscopy

method to metallo-protein study as developed in this investigation are

28

numerous. After adequate normal blood serum samples are run
quantitatively to establish normal values for various metal ions in the
different serum fractions, many‘sera from pathological cases could
and ought to be similarly characterized. The results might lead to the
possibility that a definite pattern or change of content of certain metal
ions in certain protein fractions from various given pathological
conditions could be established. Meanwhile this experimental technique
may be regarded as a biochemical research tool to be utilized in such
application as the localization of enzyme activating metal ions of blood.

tissues or biological fluids.

29

V. SUMMARY

A technique for the investigation of metal ions present in human

blood serum protein fractions has been developed. This was achieved

by separating protein fractions by starch block electrophoresis and

qualitatively identifying the metal ions present in each fraction by

emission spectroscopy. The experimental technique has been described

in detail and the results presented with suggested applications.

Specific developments were as follows:

1.

The Biuret method of analysis is preferable to DU Spectro-
photometric analysis to obtain the protein content in saline

eluates .

. It is possible to run 3 ml° samples at one time, but for

accurate results it is advantageous to run one ml. sample

at a time.

The E-C Spparatus is suitable for preparative electrophoresis
of human serum protein and gives separation into fractions of

exceptional purity and crystallizability.

The starch block proved to be a more satisfactory supporting
medium than sponge rubber for electrophoretic separation of

human serum samples.

. Metal ion distribution by emission spectroscopy of the starch

block electrophoretically prepared fractions proved to show
that:
a. Albumin contained iron, copper and manganese,
b. Beta globulin contained iron,
c. Alpha one and alpha two globulins contained iron,
copper, magnesium, cobalt, zinc and calcium.
d. Gamma globulin contained iron, manganese and

cobalt.

10.

11.

30

BIBLIOGRAPHY

. Anson, M. L., Bailey, K., and Edsall, J. T. Advances in Protein

 

Chemistry, ”Zone ElectrOphoresis in Starch Gels and Its Application

 

to Studies of Serum Proteins, " XIV, pp. 65-111. New York:
Academic Press, 1956.

. Aronson, T. , and Gronwall, A. Electrophoretic Separation of Serum

Protein into Twelve Fractions. Scand. J. Clin. and Lab. Invest. 10.
p. 348 (1958).

. Ayers, G. H. Quantitative Chemical Analysis. New York: Harper

 

and Bros., 1958, p. 490.

. Block, R. J., Durrurn, E. L. and Zweig, G. Paper Chromatography

 

and Paper Electrophoresis. New York: Academic Press, 1955, p. 333.

 

. Boyle, A. J., Whitehead, T., Baird, E. J., Batchelor, T., Iseri,

L. T., Jacobson, S. D. and Meyers, G. B. Emission Spectroscopy
for the Determination of Na, K, Ca, Mg, and Fe in Plasma and
Urine, J. Lab. Clin. Med. _3_4_, pp. 625-30 (1949).

. Brode, W. R. Chemical Spectroscopy, second edition. New York:

 

John Wiley and Sons, 1943, pp. 59-81, 328-482.

. Brown, S. K. Precision of Preformed versus Laboratory Prepared

Graphite Electrodes, Appl. Spectroscopy 11, p. 157 (1959).

. Butler, J. A. V., Stephen, J. M. L. An Apparatus for Preparative

ElectrOphoresis, Nature 160, p. 469 (1947).

. Davis, D. R. and Budd, R. E. Continuous ElectrOphoretic Quantitative

Fractionation of Seru'm Proteins. J. Lab. Clin. Med. _5_3_,pp.. 958-65
(1959).

Davidson, H. M. Electrophoresis in Segmented Systems Composed
of Polyurethan Foam, Biochem. et Biophys. Acta _3_4_, pp. 67-8 (1959).

Del'va, V. A. The Detection of Micro Elements in CerebrOSpinal
Fluid of Man. Zhur. Neuropatol. i Psikhiatri 56,pp, 888-9 (1956).

Read in abstract only—-Chem. Abstrs.. VOL _5__}_I P. 14960i (1957).

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

31

Dodson, V. N.,.Haynes, T. P., Bierwaltes, W. H. Paper Electro-
phoresis of Serums of Apparently Normal Adults by a Standardized
Method. Am. J. Clin. Path. _3_1_,pp. 404-9 (1959).

Dubbs, C. A., Vivonica,.C. and Hilburn, J. M. Subfraction of
Human Serum Enzymes,Sc‘ience, 131,p,1529 (1960).

Ewing, G. W. Instrumental Methods of Chemical Analysis. New York:
McGraw Hill Publishing Company, 1954 , p. 109.

 

Fisher, B. Recent Contributions of ElectrOphoreSis to Clinical
Pathology: Am. J. Clin. Path. 2_3_, p. 246 (1953).

Foy, R. B. "The Binding of Metal Ions by Proteins in Normal and
Abnormal Human Blood Serum. " Ph. D. Thesis, Dept. of Chemistry,
Michigan State University, 1960, 81 numbered leaves.

Gordon, A. A., Kiel, B.,-Sebesta, K. Electrophoresis of Proteins
in Agar Jelly, Nature, l64,p. 498 (1943).

Gordon, M. J. The Charge on Cell Membranes. University of
Michigan Bulletin, pp.170~6 May 1959. .

Grabar, P. Advances in Protein Chemistry, "The Use of Immuno
Chemical Methods in Studies on Proteins. " Vol.~XIII. New York:
Academic Press, 1958, pp. 1-30.

 

Greenwood, C. T. Advances in Carbohydrate Chemistry, "Aspects
of the Physical Chemistry of Starchs. " Vol. XL New York:
Academic Press, 1956, pp. 336-385.

 

Harrison, G. R. Wavelength Tables. Massachusetts Institute of
Technology.

 

Kunkel, H. G. and Slater, R. J. vZone Electrophoresis in 3. Starch
Supporting Medium. Proc. Soc. Exper. Biol. and Med. 89p. 42
(1952).

Le Duc, H. M. "Equilibrium Dialysis, Moving Boundary and Paper
ElectrOphoresis Studies on the Binding of Human Serum Proteins
with Calcium, Magnesium, Iron and Copper Ions. " Ph. D. Thesis,
Department of Chemistry, Michigan State University, 1956, 77
numbered leaves.

24.

25.

26.

27.

28.

29.

30.

31

32.

33.

34.

35.

36.

32

Leutscher, J. A. Biological and Medical Applications of E1ectro-
phoresis, Physiol. Rev. 21, p. 621 (1947).

Mitchell, H. K. , and Herzenberg, L. A. The Use of Foam Rubber
as the Supporting Material in Preparative Electrophoresis, Anal.
Chem. 22, p. 1229 (1957).

Natleson, S. Micro Techniques of Clinical Chemistry. Springfield,
Illinois: C. C. Thomas Publishers, 1957, 484 pp.

 

Paixao, L. M. and Yoe, J. H. Spectrochemical Determination of
Mg, Cr, Ni, Cu, and Zn in Human Plasmas and Red Cells, Clin. Chim.
Acta 3p. 507, (1959).

Pert, J. H., Engle, R. E. Jr., Woods, K. R. and Sleisenger, M. H.
Quantitative Zone Electrophoresis in Starch Gel. J. Lab. Clin. Med.
53, p. 571-84 (1959).

Poulik, M. D. Starch Gel Immuno-electrophoresis, J. Immunology,
_8_2,p. 502 (1959).

Sandell, E. B. Colorimetric Determination of Trace Metals, Vol. III
New York: Interscience Publishers, Inc., 1944, pp. 7-8.

 

.Smithies, O. and Poulik, M. D. Two Dimensional ElectrOphoresis

of Serum Proteins, Nature. 177,p. 1033 (1956).

Smithies, 0. Zone ElectrOphoresis in Starch Gels, Biochem. J.
_6_1,pp. 629—641 (1955).

Stewart, R. D.,- Erley, D. s”. Skelly, N. E. and Wright, N.
Infrared Analysis of Blood Serum, Red Cells and Other Body Fuilds.
J. Lab. Clin. Med. 54, p. 644-55.

Strickland, R. D., Podleski, T. R., Gurule, F. T., Freeman, M.

~L. , and Childs, W. A. Dye Binding Capacities of Eleven Electro-

phoretically Separated Serum Proteins. Anal. Chem. 37. pp.1408-10
(1959). '—'

Tata, J. R. , Prealbumin as a complex in the Alpha Globulin Fraction
in Hmnan Serum, Nature. l83,pp. 877—9 (1959).

Tomb, M. P. , Souter, F. , MacLagan, N. F. SpectrOphotometric
Determination of Protein at 210 Microns, Biochem. J. 73, pp. 167-71
(1959L ‘—

37.

38.

39.

33

Twyman, F. Metal Spectroscopy, London: Charles Griffin '-
and Company, 1951, pp. 240,263.

 

Vallee, Bert L. Advances in Protein Chemistry, "Zinc and Metallo
Enzymes, " Vol. X. New York: Academic Press, 1955, pp. 318-378.

 

Wolfram and Thompson. The Chemistry of Starches. J. Am. Chem.
Soc., 18,p. 4116 (1956).