Downloaded from http://jcp.bmj.com/ on February 19, 2018 - Published by group.bmj.

com

J. clin. Path. (1957), 10, 161.

A METHOD FOR DEPROTEINIZATION OF BLOOD AND

OTHER BODY FLUIDS
BY
G. HUNTER
From Cowley Road Hospital, Oxford
(RECEIVED FOR PUBLICATION APRIL 26, 1956)

Deproteinization is a necessary step in many clear and colourless and remain clear on addition
procedures for the chemical analysis of body fluids. of sulphosalicylic acid.
A variety of reagents have been used for this
purpose, especially in blood analysis. Of these Deproteinization of Blood
perhaps tungstic acid and trichloracetic acid are pH Adjustment.-For this purpose 0.05 N-acetic
the most widely used to-day, and for this reason acid is used routinely, but other acids may be used
we may consider them a moment. Folin and Wu (see p. 163). One volume of plasma, cells, and whole
(1919) introduced tungstic acid to prepare protein- blood requires 1.0, 0.54, and 0.80 volume respectively
free filtrates for their system of blood analysis. of the acetic acid. Water is added to suitable dilu-
In this method it is usual to employ 667 m.Eq. of tion, e.g., 1 in 5 to 1 in 50, and the contents are mixed.
the acid to precipitate the proteins from a litre of Protein Precipitation.-When the volume of the
plasma or whole blood and twice this concentra- mixed solution is less than 10 ml. it is convenient to
tion to deproteinize the same volume of packed have it in a centrifuge tube. This is immersed for
erythrocytes. With the use of sodium tungstate three to five minutes in boiling water, then cooled,
and sulphuric acid equivalent amounts of Na and spun, and the supernatant fluid is transferred to a clean
SO4 are of course added to the filtrate. Tri- tube. With larger volumes of course the time of heating
chloroacetic acid is commonly used in a 5% (w/v) must extended so that the fluid is at 100' C. for
concentration in the filtrate, and it must be used two tobefour minutes.
in not less than half of this concentration. If the
blood is diluted 1 in 10 this means that the acid Deproteinization of Other Fluids
used equals 4 m.Eq./litre blood and the filtrate is
more than 3.0 N acid. Cerebrospinal fluid (C.S.F.) protein is satisfactorily
For such reasons tungstic and trichloroacetic precipitated with 0.7 volume of the acid at 1 in 10
acids are unsuitable deproteinizing agents in many dilution. In our experience protein is more easily
removed from C.S.F. by this method than by either
analytical methods, in procedures for the isolation tungstic or trichloroacetic acid.
of solutions for chromatography, and other Oedema fluids, nasal discharges, etc., have been
modern techniques. It therefore seems desirable found to require 0.4 to 0.7 volume of the acid, the
to have other means of deproteinization that will amount depending largely on the amount of NaHCO3
avoid the addition of gross amounts of extraneous and protein present. When the coagulation is not
matter to the filtrates, that can be rapidly applied, complete the solution should be adjusted to pH 5.3,
yield filtrates near neutrality, and cause a mini- at which point we have not failed to get good coagu-
mum distortion of the chemical picture of the lation. When means are not available for pH deter-
fluids concerned. The present communication is mination
1.0
a final adjustment with a drop or two of
M-acetic acid-sodium acetate buffer pH 5.3
an attempt in this direction. A method is de- is usually effective.
scribed for the preparation of protein-free
filtrates of blood, which is also applicable Deproteinization of Urine
to other body fluids. It depends on the adjust-
ment of the pH of the diluted fluid in question The composition and pH of urine are so variable
to a value suitable for the precipitation by heat that no rule can be followed. Our practice is to
adjust the pH of a given volume of urine with a
of the proteins present. Plasma, cells, and whole measured of 1.0 N-acetic acid to 5.2 to 5.3,
blood require about 50, 27, and 40 m.Eq. acetic using a pHamount meter and glass electrode, then diluting
acid/l. respectively for pH adjustment. The fil- as analytical requirements indicate before heating to
trates obtained after heating briefly at 1000 C. are precipitate the protein.
 Downloaded from http://jcp.bmj.com/ on February 19, 2018 - Published by group.bmj.com

162 G. HUNTER
Methods The pH of the filtrates is somewhat higher, 0.1
Non-protein nitrogen was determined on the blood to 0.2 pH unit, than that of the solutions before
filtrates shown in Table I by digestion with H2SO4, heating, due to loss of CO2.
addition of K2S20 to clear, followed by heating for Jaundiced sera yield colourless filtrates indi-
30 min., with subsequent nesslerization. cating that, of the pigments, at least bilirubin is
Acid required for the pH shift of Tables II and III adsorbed on the protein precipitate.
was determined by titrating 1 ml. plasma, diluted
with 9 ml. water, with 0.05 N-acetic acid to pH 5.3 Discussion
with the glass electrode; and 1 ml. cells, diluted with
19 ml. water, to pH 7.0 with the same acid. This method of deproteinization has proved
practicable for the determination of ergo-
TABLE I thioneine in blood (Hunter, 1949), poly-
NON-PROTEIN NITROGEN OF BLOOD DETERMINED ON vinyl pyrrolidone (plasmosan) in blood serum
CORRESPONDING FILTRATES OBTAINED WITH TUNG-
STIC ACID AND THE ACID-HEAT METHOD (Campbell and Hunter, 1953), and for the estima-
tion of isonicotinic acid hydrazide in blood serum
Blood Tungstic Acid Acid-heat (Hunter, 1955). It has now been found that fil-
No. Plasma Cells Blood Plasma Cells Blood trates from blood and other body fluids prepared
1
2
34
25
55
51
45
38
46
30
70
60
59
42
in this way are suitable for the determination of
3 36 80 60 41 88 71 Mg and Ca (Hunter, 1956).
4 19 35 26 35 45 50 The " heat and acetic acid test " has of course
5 32 44 42 55 74 64
long been used as a qualitative clinical test for
Results are expressed in mg./100 ml. protein in urine, but it is usually performed rather
TABLE II haphazardly, without control of pH, and to differ-
ESTIMATE OF ACID* REQUIRED TO SHIFT pH of 1 LITRE entiate protein from precipitated phosphates.
PLASMA TO 5 3 The method has not been suggested as
Protein, 70x 0-243 (Peters and van Slyke, 1931) = 17 m.Eq. a general and quantitative means for deprotein-
NaHCO5, at pH 5 3 (4 m.Eq. as NaHCO3)
Total by estimate
=24
=41
izing blood and other body fluids, though
Found in practice =40 there are a few observations in the literature on
* It is assumed that plasma contains 70 g. protein and 28 m.Eq.
the controlled use of acetic acid, e.g., by
NaHCO5I; 40 m.Eq. strong acid is equal to about 50 m.Eq. acetic Rimington (1940) to prepare filtrates of plasma
acid at pH 5-3. containing mucoprotein, by Solomon, Johnson,
TABLE 1II Sheffner, and Bergeim (1951) to prepare tissue
ESTIMATE OF AC1D REQUIRED TO SHIFT pH OF 1 LITRE
LAKED CELLS FROM 7-5 TO 7-0*
filtrates for chromatography, and by Ramsay
(1953), who used an acetic buffer at pH 5.0 to
About 2-5 m.Eq. acid required to shift 17 g. Hb through 1 pH unit remove protein in the course of the determination
(Peters and van Slyke, 1931). Hence acid required here is of Fe in plasma and serum. As far as I am aware
137x 25x0 5 =23-5 m.Eq.
17
To shift 16 m.Eq. NaHCO3 from pH 7-5 to pH 7 0
there has been no study of this method of
requires = 1-2 deproteinization.
Total by estimate = 24-7 The arbitrary finding that plasma requires the
Found in practice =27-0
addition of almost twice as much acid as an equal
* Ills assumed that cells contain 320 g. haemoglobin and 16 m.Eq. volume of packed cells was at first surprising in
NaHCO3. view of the much greater concentration of protein
Results and base in the cells. It was observed, however,
The filtrates obtained are clear and colourless that the protein-free filtrates from plasma or
and remain clear on addition of sulphosalicylic serum had a pH about 5.3, and the protein-free
acid. This is conclusive evidence that there is less filtrates from packed cells had a pH about 7.0 at
than about 2 mg. protein/100 ml. filtrate. The room temperature, pHs not far removed from the
xanthroproteic and biuret tests are faintly positive isoelectric points of denatured serum albumin and
in the filtrates as they are also in corresponding globulin and the isoelectric point of haemoglobin
tungstic acid filtrates and do not necessarily signify respectively. Postulating a normal plasma as con-
the presence of protein. For practical purposes taining 70 g. protein and 28 m.Eq. NaHCO3, the
the filtrates are therefore free from protein. amount of strong acid necessary to shift its pH to
Though free from protein, the non-protein 5.3 is calculable with fair accuracy, according to
nitrogen (N.P.N.) content of the filtrates from Table II, and agrees in a satisfactory manner with
plasma, cells, and whole blood is uniformly higher the amount found in practice. The NaHCO3
than the N.P.N. values obtained on corresponding present in plasma accounts for more than half the
tungstic acid filtrates, as shown in Table I. acid required.
 Downloaded from http://jcp.bmj.com/ on February 19, 2018 - Published by group.bmj.com

DEPROTEINIZATION OF BLOOD AND OTHER BODY FLUIDS 163

As might be expected the pH value of the acid ance of + 10% of the amount of acid recom-
used has some effect on the amount required in mended for use. This covers all but quite excep-
different pH regions. For example, there is little tional plasmas-more than a 20% change in
difference in the range pH 7.5-7.0, whether hydro- plasma protein or more than a 15% change in
chloric or acetic acid is used, but when the pH plasma bicarbonate.
has to be shifted to about 5.3, more of the weak Like arguments apply to the deproteinization of
acid is required. The following amounts (in ml.) cells and whole blood.
of different acids were required to shift 1 ml. of It may be noted that the loss of CO2 from blood
a diluted plasma to pH 5.3: Acetic 1.00, benzoic has no influence on the amount of acid required
0.88, lactic 0.80, perchloric, trichloroacetic and for pH adjustment of the whole blood, as it is a
hydrochloric 0.75, tartaric 0.85, sulphuric 0.80, volatile acid. However, plasma separated from
and citric 1.05. Similar values are obtained on blood with a low CO2 tension will have more
titration of bicarbonate solutions. On heating the chloride present and thus require less acid for pH
plasma solutions clear filtrates were obtained. It adjustment, and vice versa for plasma separated
should be noted, however, that there is a marked from blood with a high CO2 tension. Only under
rise in the pH of trichloroacetic acid filtrates owing extreme conditions does this chloride shift exceed
to the decomposition of the acid on heating. the range stated above.
In the case of cells the situation is quite differ- Despite the range of tolerance recognized here
ent, as seen from Table III. The effect of it should be emphasized that the amounts of acid
NaHCO3 is almost negligible over the relevant pH used must be measured precisely.
range, but to shift the haemoglobin over 0.5 pH The general directions are only applicable to
takes more acid than that required to shift the fresh bloods. On long standing, acid will
plasma protein over 2.0 pH. Again the acid cal- frequently develop in blood.
culated as required is near that found in practice, Fluids Other Than Blood.-The principles dis-
especially when a small amount of plasma with the cussed above for the precipitation of the proteins
cells is accounted for.
What remains surprising is that the proteins in from blood plasma would appear to apply to body
whole blood are effectively precipitated at about fluids in general. In the case of C.S.F. the pro-
pH 6.4. When a 1 in 10 dilution of plasma is tein is normally a negligible factor as a buffer, but
heated at this pH it will not coagulate well; the the total CO2 is somewhat higher than that of
supernatant will be opaque or very milky in plasma. Hence we might expect that unit volume
appearance. Likewise solutions of cells heated at will require about 0.7 volume of 0.05 N-acetic acid.
pH 6.4 give dirty red suspensions. Yet when both Cerebrospinal fluid also serves to show the sensi-
cells and plasma are present in about equal volume tivity of the process. For example, 1 ml. of a
the supernatant is clear and colourless. At pH C.S.F. containing 22 mg. protein /100 ml., with
6.4 the plasma proteins will be charged negatively 8.3 ml. water and 0.7 ml. acetic acid, showed a
and the haemoglobin positively, and it is suggested well-coagulated precipitate after heating.
on those grounds that coprecipitation is a major When the acid requirement of the fluid is un-
factor in the coagulation of the denatured protein known and its volume perhaps small, or pH deter-
from solutions of whole blood. minations are not readily made, it is suggested
that 0.6 volume of acid be tried, and if this fails
Range of pH Tolerance for Deproteinization.- to give a good coagulation a drop or two of acetate
It may be observed that pH 5.3 has been chosen buffer at pH 5.3 may be used. As noted above,
in Table II to calculate the buffering capacity of Ramsay has suggested the use of a similar buffer
plasma. This point was chosen as a mean of a at pH 5.0, but it may be calculated from his data
range pH 5.7-5.0, in which crystal clear filtrates that he is adding some 25 times more acetate than
of plasma are usually obtained. This range is of used in the present method.
considerable importance, as the present method In a case of nephrosis studied with Dr.
would not otherwise be practicable for different L. C. A. Nunn of the Pathology Laboratory, Stoke
bloods. Expressed in another way, it is found Mandeville Hospital, Aylesbury, the present
that 1 ml. diluted normal plasma requires 1.0 ml. method was found useful in the precipitation of
0.05 N-acetic acid for adjustment to the pH suit- urinary protein without the precipitation of poly-
able for deproteinization by heat. Were 0.9 ml. vinyl pyrrolidone (" plasmosan ") which had been
acid added, the pH in such a diluted plasma would administered to the patient to maintain oncotic
be near 5.7, and were 1. ml. acid added the pH pressure. It was thus possible to determine both
would be near 5.0. We thus have a range of toler- the protein and plasmosan excreted in the urine.
0
 Downloaded from http://jcp.bmj.com/ on February 19, 2018 - Published by group.bmj.com

164 G. HUNTER
Non-protein Nitrogen.-The non-protein nitro- The filtrates obtained are clear and colourless
gen values of the filtrates obtained ty the presentand show no clouding on the addition of sulpho-
method present a point of some interest. Its salicylic acid, but they contain more non-protein
elucidation awaits further investigation. However, nitrogen than corresponding filtrates obtained
it was stated by Benedict and Newton (1929) that with tungstic acid.
tungstic acid precipitates ergothioneine in blood, Some of the factors in the deproteinization pro-
and from similar findings the present method of cess are discussed and the advantages of such
deproteinizing (Hunter, 1949) was developed. It filtrates for analytical and other purposes are indi-
would appear that non-protein substances other cated.
than ergothioneine are involved, and it would no
doubt be of some interest to find out what they I am indebted to Miss M. Massey Stewart for assis-
are. tance. The work is part of a programme in the de-
It appearsunlikely that any proteolysis occurs, velopment of methods for the study of the blood-
even in plasma at pH 5.3, in the short heating
cerebrospinal-fluid barrier in association with Dr.
H. V. Smith, Reader in Medicine, Radcliffe
time. Extension of the heating time to 30 min. Infirmary.
was found to have no effect on the non-protein
REFERENCES
nitrogen. Benedict, S. R., and Newton, E. B. (1929). J. biol. Chem., 82, 5.
Summary Campbell, H., and Hunter, G. (1953). Lancet, 1, 197.
Folin, O., and Wu, H. (1919). J. biol. Chem., 38, 81.
A method is described for the deproteinization Hunter, G. (1949). Canad. J. Res. (E), 27, 230.
of blood and other body fluids, at the usual dilu- (1955). Brit. med. J., 1, 585.
tions of I in 5 to 1 in 50, which depends on the -(1956). Biochem. J., 62, 28P, 29P.
Peters, J. P., and van Slyke, D. D. (1931). Quantitative Clinical
addition of small amounts of acid followed by a Chemistry, Vol. 1, pp. 539 and 661. Williams and Wilkins,
brief period of heat at 1000 C. Baltimore.
The requisite amount of acid is calculable from Ramsay, W. N. M. (1953). Biochem. J., 53, 227.
Rimington, C. (1940). Ibid., 34, 931.
the buffer capacity, due mainly to NaHCO3 and Solomon, J. D., Johnson, C. A., Sheffner, A. L., and Bergeim, 0.
protein, and the pH shift involved. (1951). J. biol. Chem., 189, 629.
 Downloaded from http://jcp.bmj.com/ on February 19, 2018 - Published by group.bmj.com

A Method for Deproteinization

of Blood and Other Body Fluids
G. Hunter

J Clin Pathol1957 10: 161-164

doi: 10.1136/jcp.10.2.161

Updated information and services can be found at:

http://jcp.bmj.com/content/10/2/161.citation

These include:

Email alerting Receive free email alerts when new articles cite this
service article. Sign up in the box at the top right corner of
the online article.

Notes

To request permissions go to:

http://group.bmj.com/group/rights-licensing/permissions

To order reprints go to:

http://journals.bmj.com/cgi/reprintform

To subscribe to BMJ go to:

http://group.bmj.com/subscribe/