The influence of the injection of foreign proteins on the normal bactericidal activity of serum
11
612. 118. 223 : 578. 089.31 : 547. 96 THE INFLUENCE OF THE INJECTION OF FOREIGN PROTEINS ON THE NORMAL BACTERICIDAL ACTIVITY OF SERUM. SCOTT THOMSON. From the Department of Bacteriology, University of Edinburgh. SEVERAL workers (Walbum, 1921 ; Madsen, 1923 ; Steabben, 1925 ; Kalinin et al., 1935) have studied the effect of injections of various substances on the output of antibodies in immunised animals, and Mackie (1925) and Fleming (1927-28) have observed changes in the natural antibodies of animals similarly treated. In the present paper are recorded the results of a study of the effect of injections of various sera and of egg albumin upon the natural lytic substances in rabbit serum. According to Pettersson (1926, 1927-28) there are two groups of lysins, one of which (a lysin) involves the intervention of antibody and complement (Browning, 1927 ; Gordon and Wormall, 1928 ; Mackie and Finkelstein, 1930, 1931; Gordon and Carter, 1932; while the other (/3 lysin), though probably complex, is relatively heat stable. In general Gram-positive bacteria are sensitive to p lysins and Gram-negative to a lysins and the activity of /3 lysins is of a low order. It has been found that the effect of injections of certain sera and proteins is to cause a reduction in the activity of the a lysins of the serum and it is concluded that this reduction is due to an effect on the antibody-like fraction of these lysins. Methods. Bactericidal tests. The .technique adopted was that of Mackie and Finkelstein (1931). A constant volume (0.25 c.c.) of fresh serum was pipetted into each of six sterile tubes. The serum was used either undiluted or in a dilution of 1 : 2 in normal saline. In another six tubes decimal dilutions of a suspension of the organism under test were prepared in saline and added in 0.1 C.C. amounts to the serum. This gave a series of six tubes containing a constant amount of serum but varying amounts of bacteria. The mixtures were plated on agar immediately by a single stroke of a standard loop and again after four hours' incubation at 37" C. The density of bacterial emulsion from which the serial dilutions were made was chosen by trial so that the sixth dilution on plating would give no growth. The bactericidal activity was estimated by comparing the end-point of growth on immediate plating with that obtained on plating after four hours' incubation. For example a series which showed growth in all but the sixth tube (end-point 5) on immediate plating and in all but the fourth, fifth 131
-
Upload
scott-thomson -
Category
Documents
-
view
212 -
download
0
Transcript of The influence of the injection of foreign proteins on the normal bactericidal activity of serum
612. 118. 223 : 578. 089.31 : 547. 96
THE INFLUENCE OF THE INJECTION O F FOREIGN PROTEINS ON THE NORMAL BACTERICIDAL ACTIVITY O F SERUM.
SCOTT THOMSON. From the Department of Bacteriology, University of Edinburgh.
SEVERAL workers (Walbum, 1921 ; Madsen, 1923 ; Steabben, 1925 ; Kalinin et al., 1935) have studied the effect of injections of various substances on the output of antibodies in immunised animals, and Mackie (1925) and Fleming (1927-28) have observed changes in the natural antibodies of animals similarly treated. In the present paper are recorded the results of a study of the effect of injections of various sera and of egg albumin upon the natural lytic substances in rabbit serum. According to Pettersson (1926, 1927-28) there are two groups of lysins, one of which (a lysin) involves the intervention of antibody and complement (Browning, 1927 ; Gordon and Wormall, 1928 ; Mackie and Finkelstein, 1930, 1931; Gordon and Carter, 1932; while the other (/3 lysin), though probably complex, is relatively heat stable. In general Gram-positive bacteria are sensitive to p lysins and Gram-negative to a lysins and the activity of /3 lysins is of a low order. It has been found that the effect of injections of certain sera and proteins is to cause a reduction in the activity of the a lysins of the serum and it is concluded that this reduction is due to an effect on the antibody-like fraction of these lysins.
Methods.
Bactericidal tests. The .technique adopted was that of Mackie and Finkelstein (1931). A constant volume (0.25 c.c.) of fresh serum was pipetted into each of six sterile tubes. The serum was used either undiluted or in a dilution of 1 : 2 in normal saline. I n another six tubes decimal dilutions of a suspension of the organism under test were prepared in saline and added in 0.1 C.C. amounts to the serum. This gave a series of six tubes containing a constant amount of serum but varying amounts of bacteria. The mixtures were plated on agar immediately by a single stroke of a standard loop and again after four hours' incubation a t 37" C. The density of bacterial emulsion from which the serial dilutions were made was chosen by trial so that the sixth dilution on plating would give no growth. The bactericidal activity was estimated by comparing the end-point of growth on immediate plating with that obtained on plating after four hours' incubation. For example a series which showed growth in all but the sixth tube (end-point 5) on immediate plating and in all but the fourth, fifth
131
132 S. THOMSOiY
and sixth tubes (end-point 3) after incubation, was taken to have a bactericidal titre of 2, i.e. 5-3, the difference in the two end-points.
A 3 per cent. suspension of washed sheep's red cells was sensitised with 7 M.H.D. of haemolytic immune body from a rabbit (R.v.S. I.B.), and of this 0.5 C.C. was mixed with quantities of fresh serum ranging from 0.01 to 0-2 C.C. The mixtures were incubated at 37" C. for 16 hours and the amount of lysis noted. The hzmolytic activity was expressed as the reciprocal of the smallest fraction of a C.C. of serum which produced complete lysis of the cells.
Precipitins for horse serum proteins were detected by a ring test in which varying dilutions of horse serum were layered on a constant amount of undiluted rabbit serum. The highest dilution of horse serum producing a precipitate was recorded.
Complement titration.
Precipitin tests.
RESULTS. Experiments with horse and ox serum.
Following subcutaneous injections in rabbits of horse serum there was a fall in the bactericidal activity of the rabbit's serum towards B. typhosus and V . cholerm, usually to zero. This fall of a lytic activity was unaccompanied by any change in the /3 lysin as tested with Staphylococcus aureus. Table I details the results obtained in a rabbit which was injected subcutaneously on two occasions with 10 C.C. of horse serum.
TABLE I. Effect of injection of h,orse Serum on lytic activity of rubbit serum.
Expt. no.
~
638
kiactericidal activity. ::!$A! 5;:; Complement. Precipitins for horse serum.
__ ~ __._____~
1 10 r.c. 3 6 0 7 10 C . C . . . ... ... 8 0 0 20 512
2 4 1 3 4096 3 3 13 256
On the 8th day, the day after the second subcutaneous injection of horse serum, the bactericidal activity against B. typhosus and V . cholerce had completely disappeared. The rabbit's serum had not lost its ability to ac t as haemolytic complement and the rise in titre from 6 to 20 was not regarded as significant. Precipitins at this stage had reached a titre of 512. A week later the bactericidal activity of the serum had returned to the original levels. Precipitins at this stage were present to a titre of 4096 and were still demonstrable three weeks later to a titre of 256. Similar results were obtained in a large number of animals. Table I1 gives the results of three experiments where both a and /3 lytic activity were titrated with B. typhosus and Staphylococcus aureus respectively.
B A C T E R I O L Y S I N S OP N O R M A L S E R U M
TABLE 11.
Effect of horse serum iniection o n a and p lysins.
3 2 1 0 0 3
133
2 - > 1 2
2 > 4
Expt. 110.
858
860
842
Day of experiment.
1 3 6
10 17 33
1 3 6 8
10 17 26
1 3 6 8
10 18
Horse serum injected.
20 C . C . ... ... ... ...
20 C . C . ... ...
6 C . C . ... ... ...
6 C.C. ... ...
5 C . C . ... ...
Bactericidal activity.
B. typhosus. (U lysin).
2 3 1 0 3 3
2 2 1
0 2 3
3 3 1
0 0
...
...
Similar results were obtained following the serum (table 111).
TABLE 111.
lCffect of ox serum injection on lysins.
Expt. no.
86.5
Day of experiment.
1 3 7
11 18 26
Ox serum injected.
20 C . C . ... ... ... .. . ...
S. aureus. (/3 lysin).
1 1 1 1 1 1
1 1 1
1 1 1
1 1 1
1 1
...
1 . .
injection of ox
Bactericidal activity.
B. tmhosus. 1 S. aureus. I 1
The fall in a lytic activity towards B. typhosus was as marked with ox as with horse serum but there was little or no depression of ,f3 lysin for S. aureus. It is well established, however, that there are related antigens in ox and horse serum (Kolmer, 1923) and in one rabbit inoculated with ox serum in one of these experiments there was a precipitin titre for horse serum which reached 128 on the 8th day after injection and 256 on the 17th day.
JOURN. OF PATH.-VOL. XLVII. I 2
134 S. THOMSON
In further experiments seven rabbits were given single sub- cutaneous injections of horse (3) or ox (4) serum and it was found that the depression of a lytic activity occurred about the 6th-9th day and lasted for about a week, though in one instance it was still present on the 18th day.
The injection of horse or ox serum into the rabbit appears therefore to be followed by a diminution in the a lytic activity of the rabbit’s serum for B. typhosus and V . cholerot! but does not affect its ,8 lytic activity for S. aureus or its haemolytic complenient activity. Precipitins for the injected serum were present when the lytic activity was depressed but persisted after it had returned to normal.
Since a lysis depends upon the dual mechanism of a heat- stable antibody-like factor and heat-labile complement and /3 lysis is independent of complement, the possibility of some interference with complement was considered first. HaemoIytic complement, however, was not disturbed and the complement concerned in bactericidal action is closely allied to if not identical with haemolytic complement.
Dean (1931) has shown that the freshly separated serum of a rabbit previously injected with horse serum may show, when anti- bodies are beginning to appear, a fine precipitate due to the inter- action of antibodies with horse serum which has not disappeared from the circulation. This phenomenon is demonstrable only at a certain stage following the injection of horse serum and is dependent on the coexistence in the animal of an antigen and its homologous antibody. In the part,icular case considered here it is possible that antibodies t o the injected horse serum were appearing before the injected serum proteins had been completely eliminated from the rabbit’s system. The interaction between these two may have fixed or absorbed complement with a resulting fall in a lytic activity of the serum.
The experiment quoted in table I might seem to support this explanation, the absence of a lytic activity being noted on the day following the second injection of horse serum. It might be argued that the second dose of antigen (horse serum) was interacting with antibodies produced by the injection of the first dose. On the other hand in expt. 865 (table 111) the fall was noted on the 18th day after a single injection of horse serum.
Against the acceptance of this explanation is the fact that haemolytic complement as estimated with an R.v.S. I.B. maintained its activity. It might be supposed that an antigen-antibody reaction should have fixed the haemolytic as well as the bactericidal complement.
On the other hand there may have been some selective absorption of bactericidal complement. Neufeld and Htindel (1908), using an anticholera serum and V . cholerce, obtained a selective absorption of bactericidal
BACTERIOLYSINS OF NORMAL SERUM 1 3 5
complement at 0" C. and Muir and Browning (1908-09) treated normal serum with graded amounts of bacteria and found that bactericidal complement was absorbed before hzmolytic complement. Larger amounts of bacterial emulsion absorbed both bactericidal and haemolytic complement. I n all these experiments quantitative estimations were made and any procedure which interfered with one complement was able to interfere with the other if the treatment were intensified. Mackie and Finkelstein (1931) and Finkelstein (1933) have shown that cultures of bacteria readily yield a substance inhibitory to bactericidal action. They found in absorption tests with bacterial cultures that carefully washed growths had to be used because of this inhibitory substance. Some of the previous work on the separation of bactericidal and hEmolytic complement will require to be reviewed in view of these findings.
If, however, an in-vivo fixation of bactericidal complement occurs, then the phenomenon should be present for as long as antigen remains in the circulation. Jonesco-Mihaiesti (1911) detected horse serum in the circulation far 10 days following the last of a series of immunising injections and Opie (1923) found that horse serum was detectable in the blood for 7-9 days following an injection, but in certain cases where precipitins appeared slowly and to a low titre, the antigen might persist for as long as 19 days.
An antigen is removed more quickly from the circulation of immune than of normal animals (Culbertson, 1935). Accordingly
TABLE IV.
Effect of repeated injections of serum on lysins.
Expt. no. - 634
635
847
862
Day of Horse serum experiment.
-
1 6 7
13 27 36
1 6 7
13 17 28
1 7
10 14 25 39
1 3 6 8
10 18
injected.
10 C . C .
10 C . C . ... ... ... ...
10 C . C .
10 C . C . ... ... ... ...
5 C.C. 5 C.C. 5 C . C .
...
...
... 5 C . C .
...
... 5 C.C.
...
...
Bactericidal activity tB. tgphosecs).
3 1
0 3 2
3 4
0 2 3
3 0 0 0 0 3
3 3 1
0 0
...
...
...
Precipitins.
0 0
1,024 16,000 4,096
0 64
128 256
1,024
...
...
...
...
...
...
...
...
...
...
...
...
...
...
Complement.
13 10
24 20 25
10 40
16 10 20
...
...
...
. .
. .
...
...
...
136 S. THOMSON
the depression of a lysis might have been expected to be more transient following the second or third serum injection. Table IV shows the results of experiments where more than one injection of serum was given.
In experiment 847, no bactericidal power for B. typhosus was present on the 15th day following the 3rd subcutaneous injection of horse serum. Experiments 862, 634 and 635 showed no bactericidal activity towards B. typhosus for a week to ten days following the second injection of horse serum.
Table V shows the results of two experiments where the horse serum was injected intravenously.
TABLE V.
Effect of intravenous injection of horse serum on lysins.
Expt. no.
632
633
Day of Horse serum experiment. injected I.V.
1 5 C.C.
5 6 5 C . C .
... 1: 1 ...
1 5 C.C.
6 5 C.C.
7 5 C.C.
12 ~ ... 21 35 I :::
Bactericidal activity.
B. typhosus.
3 2
0 3
4
0 2 0 4
...
...
V . cholerce.
2 3
1 4
3
1 3 3 3
...
...
Precipitins.
0 0
512 1024
0
256 2048 8192 256
...
...
Depression of lysis occurred and in expt. 632 was present three days after the second intravenous injection of serum. In expt. 633 it was observed on the day following the second injection. Four days later recovery had occurred but after a further nine days, i.e. on the fifteenth day after the second intravenous injection of serum, there was a renewed effect. This second depression of activity might have been due to antibodies produced by the second injection of serum, but in both the animals the result was more definite than would have been expected if it was caused by in-vivo fixation of complement.
Experiments with other sera. In order to examine whether or not the depression of a lytic
activity is a general change following the injection of serum, small groups of rabbits (4 or 5) were injected with sheep, guinea-pig, fowl and duck sera. Cross reactions between sheep serum and anti-horse serum have been described but have not been encountered in these experiments. Guinea-pig, fowl and duck sera show no
BACTERIOLYSINS OF NORMAL SERUM 137
cross reactions with anti-horse serum. The results with sheep serum were not so constant as with horse serum but the same depression of a lytic activity did appear. A fall in bactericidal activity against B. aertrycke was also demonstrated, although not so uniformly nor so distinctly. There was no fall in p lytic activity for S. aureus.
Four rabbits received two injections of 10 C.C. of duck serum subcutaneously with an interval of a week. Three revealed no depression of lysin but the fourth showed complete absence of a lytic activity against B. typhosus on the 3rd day after the second injection.
No depression of a lytic activity was found following injections of fowl and guinea-pig serum. Four rabbits were tested with each.
No precipitins for horse serum were demonstrated.
Experiments with egg albumin. It was obviously of importance to discover if the phenomenon
followed the injection of foreign protein other than serum. For this purpose four rabbits were injected subcutaneously with egg albumin with the results shown in table VI.
TABLE VI. Effect of injection of egg albumin on lysins.
Espt. no.
93
94
96
97
Day of experiment.
1 7
10
1 7
10
1 7
10
1 10
Volume of 10 per cent. egg albumin injected.
10 C.C. 10 C.C.
... 10 C.C. 10 c.e.
...
10 C.C. 10 c.e.
... 10 ex.
...
Bactericidal activity (B. typhosus).
4
0
4
0
4
1
4 4
...
...
...
The depression of a lytic activity was again produced and would appear to be a general change following the parenteral injection of foreign protein.
Analysis of the depression. As no evidence was obtained to indicate that a disturbance of
bactericidal complement was responsible for the depression of a lytic activity, it remained to consider whether this was due to interference with the heat-stable antibody-like factor or t o an inhibitory process preventing the action of thermolabile complement on the sensitised bacterium.
138 S. THOMSON
Rabbit X was injected with horse serum, a depression of a lytic activity for B. typhosus being produced. Some of the serum of a normal rabbit Y with a bactericidal activity of 4 was inactivated by heating to 55" C. for half an hour. It was then tested for its ability to restore bactericidal activity to the serum of X, while serum X was tested for its effect, both fresh and inactivated, on the lytic action of fresh serum Y (table VII).
TABLE VII.
Reactivation of lytic serum.
1 Test no. ~ Rabbit serum used. I Bactericidal activity.
1 1 1 Y (fresh) 1 Y i55" C.) X (fresh)
X (fresh)+Y (55' C.) X (fresh)+Y (fresh) X (55" C.) +Y (fresh)
x ( 5 5 O C.)
From test no. 6 it would appear that the depression of lysis is not due to the presence of some substance which interferes with bactericidal activity in vitro, but that the lack of bactericidal power in the test animal is due to the absence of a heat-stable substance which can be supplied by adding heated serum from a fresh rabbit (test 5) . The restoration of lytic activity by inactivated normal serum could only have been due to the addition of either the heat- stable antibody-like factor or the heat-stable components of complement. No loss in haemolytic activity was observed with R.v.S. I.B. and the restoration of bactericidal activity by heated serum must have been due to the addition of the heat-stable antibody-like principle concerned in a bacteriolysis.
In order to test this further, rabbit A was injected with horse serum so as to depress the bactericidal activity of its serum against B. typhosus from 3 or 4 to 0. As before, serum from a normal rabbit (B) was inactivated by heating to 55" C. for half an hour (designated " B 55" C.,'). F'resh serum from rabbit B was absorbed with B. typhosus to remove the heat-stable antibody-like factor. For this absorption the growth from 6 agar plates was killed by heating to 65" C. for 1 hour. The emulsion was washed twice in 5 C.C. of saline. After centrifuging, the deposited growth was cooled to 0" C. before being resuspended in 3 C.C. of serum B, previously cooled to 0" C. Absorption was allowed to continue for 2 hours at 0" C. before centrifuging and the supernatant serum was designated " B absorbed." This procedure removes the anti- body but does not absorb complement if the correct amounts are
BACTERIOLYSINS OF NORMAL SERUM 139
a.
z 0 0 2 0
0 1
used. animals are recorded in table VIII.
The results of experiments with four different pairs of
b. C. d. ~ ~ _ _ _ -
3 3 2 0 0 0 1 0 0 2 1 2 0 0 0
1 0 0 3 1 a
TABLE V I I I .
Restoration of Eytic activity with fresh, absorbed, und heated serum. (Four experiments, a, b, c and d.)
Bactericidal activity. Serum. I
B fresh . B 55O C. . €3 absorbed . B 55" C. +B absorbed A (serum of rabbit previously injected with
A+B absorbed . . A+B 55" C. .
.
horse serum)
The absorbed serum, presumably containing only complement, was as a rule quite unable to restore the bactericidal activity to the serum of the test rabbit. The slight action in expt. b was not greater than that of the absorbed serum alone. If the phenomenon were due to interference with the heat-stable components of complement, the addition of absorbed serum should have restored the bactericidal effect as regularly and successfully as did the heated serum.
These results taken as a whole seem to be most reasonably interpreted as indicating that the depression of a lytic activity after injections of horse serum is due to interference with the heat-stable antibody-like component of normal serum. This interference is apparently due to a depression of the factor rather than to the production of something inhibiting its action.
This it failed to do.
DISCUSSION. In the rabbit a marked interference with the normal a lytic
activity of serum was produced by the parenteral injection of certain foreign sera and of egg albumin. This depression regularly followed injections of horse and ox serum but occurred also after injections of sheep and duck serum. It was not demonstrated with fowl or guinea-pig serum but with these only a small number of rabbits were tested. Prom the experiments, however, it would appear that some sera regularly produce the depression, others less regularly, and yet others either seldom or not at all.
The depression appears to be due to some change in the heat- stable antibody-like component of the serum. Interference with complement was apparently not the cause, because hzemolytic
140 S. THOMSON
complement (as estimated with sheep cells sensitised with a R.v.S. I.B.) remained active and the a lytic activity was restored by the addition of heated serum from a normal rabbit. Nor, apparently, is it due to interference with the heat-stable components of bactericidal complement, assuming this to be different froin hsmo- lytic complement, as absorbed fresh serum was unable t o restore the lost activity to the serum. The effect might have been due to a temporarily diminished production of antibody or to some inhibitory effect which in vitro was able to prevent the proper interaction between bacteria, antibody and complement. The addition of the serum under test, however, produced no inhibition of a fresh normal serum. The conclusion was reached, therefore, that the injected proteins produced a temporarily diminished production of the heat-stable antibody-like factor of the a lysin.
CONCLUSIONS.
1. Parenteral injection of foreign sera and egg albumin produces in the rabbit complete depression of the a lytic activity of the animal’s serum. This appears about the seventh day after an injection and lasts as a rule for a week.
2. The effect was demonstrated with horse and ox sera, less easily with sheep and duck sera and not a t all with fowl and guinea-pig sera.
3. There was no depression of the activity of hsmolytic complement or of /3 lysins.
4. The depression of a lytic activity is considered to be due to some depression of the antibody-like principle which is responsible for normal a lytic activity.
It was also produced with egg albumin.
It is a pleasure to acknowledge the encouragement given to this work The expenses of these experiments were met by Professor T. J. Mackie.
by a grant from the Earl of Moray Fund.
REFERENCES.
BROWNING, C. H. . . . . CULBERTSON, J. T. . . . . DEAN, H. R. . . . . .
FINEELSTEIN, M. H. . . . FLEMING, A. . . . . . . GORDON, J., AND CARTER, H. S. GORDON, J., AND WORMALL, A. JONESCO-MIHAIESTI, C. . . KALININ, W. S., SCHERES-
CHEWSKAJA, N. I., AND SELI- EOWA, R. E. . . . . .
1927. 1935. J . Imrnunol., xxviii. 279. 1931. Medical Research Council, A
system of bacteriology, vol. vi., p. 436.
Brit. Med. J. , ii. 978.
1933. this Journal, xxxvii. 359. 1927-28. 1932. this Journal, xxxv. 549. 1928. this Journal, xxxi. 753. 1911. G.R. Soc. biol., lxx. 429. 1935. Giorn. d. Batteriol. e. Immunol., xv.
681 (abstract in Bull. Hyg., 1936, xi. 327).
Proc. Roy. Soc. Med., xxi. 859.
B A C T E R I O L Y S I N S OF N O R M A L S E R U M 141
KOLMER, J. A. . . . .
MACKIE, T. J. . . . . . MACKIE, T. J., AND FINKEL-
STEIN, It. H. 2,
MADSEN, T. . . . . . . MUIR, R., AND BROWNING,
NEUFELD, F., AND HANDEL . OPIE, E. L. . . . . . . PETTERSSON, A. . . . . .
STEABBEN, D. B. . . . . WALBUM, L. E. . . . . .
C. H.
,*
1923. A practical text-book of infection, immunity and biologic therapy, 3rd edition, Philadelphia and London, p. 313.
1925. J . Hyg., xxiv. 176. 1930. Ibid., xxx. 1.
1931. Ibid., xxxi. 35. 1923. J . State Med., xxxi. 51. 1908-09. this Journal, xiii. 76.
1908. Arb. Gsndhtsamte., xxviii. 198. 1923. J . Immunol., viii. 55. 1926. 2. Immunitatsforsch., xlviii. 233. 1927-28 Ibid., liv. 292. 1925. Brit. J . E x p . Path., vi. 1. 1921. C.R. Soc. biol., lxxxv. 761.
