INFEcTION AND IMMUNITY, Sept. 1975, p. 647-655 Copyright ® 1975 American Society for Microbiology
Vol. 12, No. 3 Printed in U.S.A.
Blastogenesis as an In Vitro Correlate of Delayed Hypersensitivity in Guinea Pigs Infected with Listeria monocytogenes AMY M. FULTON, MEHER M. DUSTOOR, JOSEPH E. KASINSKI, AND ANDREW A. BLAZKOVEC* Department of Medical Microbiology, University of Wisconsin, Madison, Wisconsin 53706 Received for publication 6 May 1975
Randomly bred guinea pigs of both sexes were injected intracardially with one-half a 50% lethal dose of Listeria monocytogenes. When these animals were skin tested with 30 ,g of a water-soluble extract of sonically disrupted Listeria, animals had uniformly detectable levels of delayed-type hypersensitivity (DTH) 6 days after infection. Histological examination of skin test reaction sites, after fixation in Helly fixative and Giemsa staining, revealed a classical tuberculintype infiltrate consisting primarily of mononuclear cells with few polymorphonuclear cells. Many of the small vessels showed perivascular cuffing. When purified peritoneal exudate lymphocytes from these animals were cultured in vitro in the presence of various concentrations of Listeria antigen, it was found that the optimal antigenic dose for specific antigen-induced incorporation of [3H Ithymi-
dine varied for individual animals. In contrast to the early onset of uniformly detectable levels of in vivo DTH, in vitro lymphocyte blastogenesis was not uniformly demonstrable until 14 days postinfection and remained highly significant on days 21, 28, and 84 postinfection. At 7 days postinfection, lymphocytes from 7 of 17 animals were capable of undergoing significant blastogenesis. The Listeria antigen preparation was not mitogenic for peritoneal exudate lymphocytes from normal animals. It was found that no direct correlation exists between the in vivo levels of DTH and in vitro blastogenesis. Cell donors showing significant in vitro blastogenesis nevertheless were also skin test positive for most animals tested. Humoral antibody was found to play no significant role in the immune response of guinea pigs to a primary infection with Listeria monocytogenes.
Studies of acquired cellular resistance to Listeria monocytogenes in the mouse and rat suggest that the development of acquired cellular resistance is dependent upon the development of delayed-type hypersensitivity (DTH) (9, 10). Studies in this laboratory (5) showed that, in guinea pigs infected with one-half a 50% lethal dose of Listeria, bacterial multiplication occurs only during the first 24 h after infection. Listeria-infected animals are also resistant to a small challenge dose of Listeria given 48 h, 7 days, or 2 weeks after the primary infection. However, for this species DTH, as, detected by in vivo skin testing, is not uniformly detectable until day 5 or 6 postinfection, suggesting that the development of acquired cellular resistance is not dependent upon the development of DTH. Many workers have suggested that in vitro antigen-induced incorporation of [3H Ithymidine by sensitized lymphocytes correlates well with in vivo levels of DTH as determined by 647
skin testing and indeed may be a more sensitive measure of DTH (7, 11-13). This paper reports the results of studies designed to further characterize the immunological response of guinea pigs to a primary Listeria infection using four parameters: (i) onset and development of DTH, (ii) histopathology of the skin test reaction site, (iii) measurement of the in vitro Listeria antigen-induced incorporation of [3H ]thymidine by peritoneal exudate (PE) lymphocytes from Listeria-immune guinea pigs, and (iv) determination of the levels of humoral antibody after a primary Listeria infection. (The research described in this paper was submitted in part [by A.M.F. ] to the graduate school of the University of Wisconsin, Madison, in partial fulfillment of the requirements for the M.S. degree in Medical Microbiology.) MATERIALS AND METHODS Animals. Randomly bred albino guinea pigs of both sexes were obtained from local suppliers. The
648
FULTON ET AL.
animals weighed 500 to 700 g at the time of infection. Bacteria. L. monocytogenes was provided by D. W. Smith (Department of Medical Microbiology, University of Wisconsin, Madison). The culture was guinea pig-passaged four times, and a fresh isolate was recovered from the spleen of an infected animal. A subculture of the isolate was grown on a brain heart infusion (BHI) agar slant (Difco Laboratories, Detroit, Mich.) and used to inoculate 75 ml of BHI broth. The broth culture was incubated on a shaker at 37 C until the bacteria were in late log phase (optical density of 0.3 at 615 nm). The culture was then centrifuged at 1,000 x g for 15 min and resuspended in 75 ml of BHI broth. Samples (1 ml) of this suspension were stored at -70 C (frozen stock cultures). The 50% lethal dose for intracardially injected Listeria was 1.2 105. Immunization. BHI broth was inoculated with a BHI agar slant subculture of the frozen stock. When the broth culture was in late log phase, a portion of it was centrifuged at 1,000 x g for 15 min. The pellet was washed once and then diluted with sterile saline. Approximately one-half a 50% lethal dose in a total volume of 2 ml was injected intracardially into each guinea pig. The exact number of viable Listeria injected was determined by plate counts of the immunizing suspension. Antigen. The water-soluble extract of sonically disrupted Listeria was prepared by using a modification of the method of Hinsdill and Berman (6). Six liters of BHI broth (500 ml/flask) was inoculated with agar slant subcultures of the frozen stock and incubated on a shaker at 37 C. After 18 h, phenol was added at a final concentration of 5% (wt/vol), and the flasks were incubated for a further 8 h. The killed bacteria were collected by centrifugation at 7,000 x g for 30 min at 4 C and washed once with sterile saline and three times with sterile distilled water. The washed cells were suspended in enough distilled water to give a final volume of 50 ml and disrupted by sonic oscillation. The disrupted cells were centrifuged at 18,400 x g for 30 min at 4 C, and the supernatant was set aside. The cellular debris was suspended in 50 ml of distilled water and centrifuged as above. The two supernatants were combined and filtered (Whatman no. 1 filter paper). Samples (5 ml) were placed in small vials and lyophilized. The vials were sealed while under vacuum and stored in a desiccator at 4 C. Skin tests. Guinea pigs were injected intradermally on the shaved dorsal flank with 30 gg of the water-soluble extract in 0.1 ml of saline. The skin tests were read at 2, 4, 8, 12, 24, and 48 h. Cutaneous reactivity was measured by using three parameters: (i) degree of erythema, (ii) mean diameter of erythema, and (iii) changes in double skin thickness at the reaction site. The degree of erythema was scored arbitrarily as described previously (5). Induration was measured using a Schnelltester (System Kroplin, type A, 2T, H. C. Kroplin, Schliictern, Hessen, Germany). Each value given in the text represents the double skin thickness of the reaction site minus the mean of normal skin measured on either side of the test site. PE cells. PE were induced in guinea pigs by the intraperitoneal injection of 25 to 30 ml of sterile no. 31 x
INFECT. IMMUN. heavy paraffin oil (American Oil Co., Chicago, Ill.). Three days later the animals were killed by exsanguination via cardiac puncture. Lavage of the peritoneal cavity was carried out with a total of 300 ml of M-199 medium (Grand Island Biological Co., Grand Island, N. Y.). The PE cells were pelleted by centrifugation at 210 x g for 20 min. The pellet was resuspended in 15 ml of M-199 and centrifuged at 250 x g for 20 min. The cell pellet was resuspended in 8.0 ml of Eagle minimal essential medium (MEM; Grand Island Biological Co., Grand Island, N. Y.) containing 20% heat-inactivated normal guinea pig serum. Preparation of lymphocytes. Lymphocytes used for in vitro [3H ]thymidine assays were obtained from whole PE using the method of Rosenstreich et al. (15). PE cell suspensions were layered onto a column of rayon wool (Xerox cleaning absorbent, Xerox Corp., Rochester, N. Y.). The column was incubated at 37 C in an atmosphere of 5% CO2 for 20 min. The nonadherent cell population was eluted by passing 50 ml of MEM through the column. The eluate was centrifuged at 250 x g for 20 min. This procedure yielded an average of 1.5 x 107 cells per animal with an average viability of 90% as determined by trypan blue exclusion. Cell differentiation using Wright stain showed that 90 to 95% of the cells consisted of small lymphocytes. [3H]thymidine assay. Purified lymphocytes were diluted to a concentration of 5 x 106 cells/ml in Eagle MEM containing 100 U of penicillin per ml, 80 gg of streptomycin per ml, and 20% heat-inactivated normal guinea pig serum. The cells were dispensed in 0.1-ml aliquots in Linbro microtiter plates (model IS-FB-96-TC, Linbro Chem. Co., Inc., New Haven, Conn.). Various dilutions of Listeria antigen were added to the wells in 0.1-ml aliquots with final well concentrations of 2.0 and 0.2 mg/ml, and 20, 2.0, 0.2 and 0.02 Mg/ml. Cells incubated in MEM containing normal guinea pig serum but no Listeria antigen served as controls. A second set of specificity controls consisted of lymphocyte cultures prepared from guinea pigs 28 days after a primary Listeria infection and incubated in the presence of purified protein derivative (PPD; Parke-Davis, Co., Detroit, Mich.) in MEM. Plates were incubated at 37 C in an atmosphere of 5% CO2 for 72 h before adding 2 ACi of [3H ]thymidine (specific activity, 1.9 Ci/mmol, Schwarz-Mann, Orangeburg, N. Y.) in a volume of 0.05 ml of MEM containing penicillin and streptomycin. Cultures were incubated for 18 h and precipitated on glass fiber filters (Reeve Angel, Clifton, N. J.) using a multiple sample precipitator (Otto Hiller, Madison, Wisc.) washed with saline and fixed with 5% trichloroacetic acid. Filter strips were air-dried for 24 h and individual filter disks were placed in scintillation vials (Kimble Products, Toledo, Ohio) with 3 ml of scintillation cocktail consisting of 5.0 mg of 2,5-diphenyloxazole per ml and 0.4 mg of 1,4-bis-(5-phenyloxazolyl)benzene per ml in toluene (Packard Instrument Co., Inc., Downer's Grove, Ill.) adapted to dark and cold and counted for 10 min in a Packard Tri-Carb liquid scintillation spectrometer (Packard Instrument Co., Inc., model 3310, Downer's Grove, Ill.).
VOL. 12, 1975
649
BLASTOGENESIS IN LISTERIA-INFECTED GUINEA PIGS
Antibody titrations. Serum samples from Listeria- sharp drop in responsiveness is seen in animals immune animals were tested for the presence of tested 2 years after infection. At this time two of Listeria-specific antibody using a passive hemaggluti- five animals tested showed only trace skin nation test. Listeria antigen was coupled to sheep positivity. erythrocytes using the bisdiazobenzidine procedure (ii) Histopathology of skin test reaction recommended by Campbell et al. (2). A standard The histological appearance of skin test sites. reference antiserum pool prepared by hyperimmunization of guinea pigs with living and killed Listeria in reaction sites was evaluated at 4, 8, 12, 24, and complete Freund adjuvant served as a positive control 48 h after skin testing animals infected from 1 to for all titrations. The reference antiserum in addition 12 weeks earlier. In all instances, the cellular to all test and control serum samples were inactivated infiltrate was predominately mononuclear in by heating to 56 C for 30 min and then absorbed for 15 nature. At 8- and 12-h post-testing, however, min at room temperature with an equal volume of there were proportionately more polymorphonuwashed, packed sheep erythrocytes prior to microti- clear cells present than at the time of maximum tration using twofold serial dilutions. Titers are ex- skin reactivity, which routinely occurred at 24 pressed as the reciprocal of the highest serum dilution h. Figure 2A-C represents the histological feaat which hemagglutination was observed. Histological preparations. At varying time inter- tures of a 24-h skin test site elicited in an animal vals after skin testing, experimental and control infected 7 days previously. Many of the small animals were randomly chosen for sacrifice. Reaction vessels in Listeria-immune guinea pigs showed sites were removed for histology, fixed in Helly perivascular cuffing (Fig. 2A and B). Although fixative (100 ml of Zenker solution and 5.0 ml of 40% the majority of the infiltrating cells were indeed formaldehyde), embedded in paraffin, and sectioned mononuclear, occasional polymorphonuclear at 5 gm as described by Askenase (1). The sections cells were observed as indicated by arrows in were stained with Giemsa and observations were Fig. 2C. made by light microscopy. Lymphocyte blastogenesis. (i) Antigen Calculation of stimulation index (SI). The levels of specific antigen-induced lymphocyte transforma- dose response. Purified peritoneal lymphocytes tion were calculated by comparing the mean counts were taken from six guinea pigs 28 days after a per minute based on four replicate wells containing primary infection with Listeria. Figure 3 shows lymphocytes cultured in the presence of a given the blastogenic response of lymphocytes from antigen concentration with the mean counts per individual animals incubated with various conminute of four replicate wells containing lymphocytes centrations of Listeria antigen. The antigen cultured in the absence of antigen (controls). Due to concentration giving the maximum SI varies for the unequal variance obtained between groups, log10 transformations of all raw data were carried out prior to calculating the group SI as follows: mean log1O SI = =4.0 (log1O counts per minute in the presence of antigen)/ 3.0 (log10 counts per minute in the absence of antigen). Stimulation indexes were then plotted as antilogs to CD 2.0S. obtain real numerical values. :1.0 Statistics. To determine whether significant blastogenesis had occurred using cells from individual 25.0 animals, the mean value obtained for each set of experimental cultures was first compared to that of _ 20.0 control cultures using Student's t test. For comparing 32 15.0 group treatment mean SI values, preliminary analyses using the two-tailed F-test revealed unequal = 10.0 5.0 variance between group treatment means. To compare the means between treatment groups, it was therefore necessary to utilize the t test for unequal ° 25.0 variance (t). - 20.0 ~ ~
RESULTS Delayed hypersensitivity. (i) General trends. Figure 1 shows the cutaneous reactivity present 24 h after skin testing. Using all three parameters guinea pigs were shown to have uniformly demonstrable delayed hypersensitivity to Listeria antigen 6 days after infection. The response peaks at day 7 and remains fairly constant for the 12-week period of time during which skin testing was routinely carried out. A
~
~
~
~
~
~
.
.
Le t \0
=
5.0
14 28" U "730 21 DAYS AFTER INFECTION FIG. 1. Development and persistence of DTH in Listeria-infected guinea pigs. Responses 24 h after skin testing. Each point represents the mean of 6 to 10 guinea pigs. Bars indicate standard error.
1234561
A1..
-1,16.S'.p
,4
9.,~ ~~~~~ . ..
~10
-''
usf
fr3rt6 *.:
w~:
awi
.R "
I
_~~~~~t
~
e"
.I A.I
Ci
ve I.-
rr
IAO
_
0
FIG. 2. (A) Skin 24 h after intradermal injection of Listeria antigen. Hartley strain guinea pig infected 7 days earlier with Listeria monocytogenes. Area between dermis and musculus carnosus. Giemsa. x55. (B) Skin test reaction to Listeria antigen after 24 h as in (A). Giemsa. x220. Cellular infiltrate comprised primarily of mononuclear cells with perivascular cuffing evident. (C) Skin test reaction to Listeria antigen after 24 h in area adjacent to that shown in (A). Giemsa. x220. Cellular infiltrate comprised primarily of mononuclear cells with occasional polymorphonuclear cells indicated with arrows. 650
VOL. 12, 1975
BLASTOGENESIS IN LISTERIA-INFECTED GUINEA PIGS 50S
30
K
651
53.36
ANIMAL
I
I0 350 6 19
_'I. _ 30 _
20 z -
1
_
10
oI
I
2.0 210 200 LISTERIA ANT16EN (pg) FIG. 3. Listeria antigen-induced blastogenesis of PE lymphocytes. The dose-dependent responses of lymphocytes harvested from six guinea pigs 28 days after a primary Listeria infection. 0.02
each animal, with two animals responding maximally to concentrations of 20 ig/ml with SIs of 12.41 and 13.96 (Fig. 3). Lymphocytes from three animals responded maximally in the presence of 200 ig of Listeria antigen per ml with SIs of 1.22, 9.20, and 53.36. One animal responded maximally in the presence of 2.0 ug/ml with an SI of 12.96. Also note that a concentration of 2,000 ,ug/ml uniformly gives an SI of less than 1.0, indicating a cytotoxic effect at this high concentration. It was clear, therefore, that a wide range of antigen concentrations had to be employed to estimate the in vitro potential of lymphocytes to undergo blastogenesis. (ii) Specificity of the response. The potential of PE lymphocytes harvested from normal guinea pigs to undergo blastogenesis in the presence of Listeria antigen is shown in Fig. 4A. At no antigen concentration does the SI differ significantly from a value of 1.0 indicating that the Listeria antigen does not cause a nonspecific mitogenesis of nonsensitized cells. PE lymphocytes obtained from guinea pigs at 28 days after a primary Listeria infection were incubated in the presence of various concentrations of PPD to further confirm the specificity of Listeria antigen-induced blastogenesis. The SI of immune cells in the presence of all three PPD concentrations employed was not significantly greater than 1.0 (Fig. 4B). Peritoneal lympho-
(A)
T
T
T
I
i
0.02
0.2
2.0
20
LIST1RIA NT116N (pi)
200
zoo
10
10
PPD (pg)
FIG. 4. Specificity of antigen-induced blastogenesis of peritoneal lymphocytes. The response of normal guinea pig lymphocytes cultured in the presence of various concentrations of Listeria antigen (A). Response of lymphocytes harvested from guinea pigs 28 days after a primary Listeria infection and cultured in the presence of various concentrations of PPD (B). Each point represents the group mean of five or six guinea pigs based on the highest stimulation level for each animal obtained in dose-response experiments. Bars indicate standard error.
cytes from these same cell donors were shown to undergo blastogenesis in the presence of specific antigen, i.e., Listeria antigen dose-dependent
(Fig. 3). (iii) Development and persistence of the in
responses
652
FULTON ET AL.
INFECT. IMMUN.
vitro blastogenic response. The maximum < 0.01). It should be pointed out, however, that cells from only three of the five animals assayed mean SI for peritoneal exudate lymphocytes harvested from guinea pigs at various times at 2 years postinfection were shown to be after infection is shown in Fig. 5. Each point significantly positive. (iv) Delayed-type hypersensitivity, lymrepresents the geometric maximum mean SI standard error. As can be seen, there is little phocyte blastogenesis, and humoral antibody blastogenic response during the first week after production after a primary Listeria infection. infection. The SI does not differ significantly The immune response of individual guinea pigs from control animals during the first 7 days. after a primary infection with Listeria (Table 1) Although the maximum group mean SI at day 7 provides for direct comparison of (i) in vivo is not statistically significant, seven of the 17 DTH, (ii) in vitro blastogenesis, and (iii) huanimals tested at this time gave a statistically moral antibody production. Several generalizasignificant response when evaluated on an indi- tions become apparent upon examination of vidual basis, indicating that this may be the these data. There appears to be no direct correlation time postinfection at which the onset of blastogenic potential occurs. At 14, 21, and 28 days between the level of DTH and the SI of individpostinfection statistically significant maximum ual animals. In most instances when a significant SI was obtained, cell donors were also group mean SIs were obtained with values of 7.07 (P < 0.05), 7.76 (P < 0.01), and 12.30 (P < shown to be skin test positive. Several excep0.01), respectively. The peak maximum mean tions exist for animal no. 30 (28 days after infection) and animals no. 40, 42, and 43 (730 group SI of 12.88 is seen for the group of animals tested 12 weeks after the primary Listeria days after infection) which were skin test negainfection. The group of animals tested 2 years tive but whose cells did undergo significant after a primary infection still give a statistically blastogenesis in vitro. The reverse relationship, significant maximum mean group SI of 3.09 (P i.e., skin test positivity and a nonsignificant SI, occurred in 12 of the 44 animals tested. Of these 12 animals, seven were found in the first group 16.0 tested at day 7 postinfection. Data not recorded in Table 1 showed that the 14.0 five to six animals tested on each day prior to day 7 were sometimes DTH positive beginning on day 4; all were uniformly negative in terms of 12.0 SI. At 730 days after infection, only two of the five animals showed DTH but cells obtained 10.0 from three of these animals did show significant 8 .0
blastogenesis. In general, animals tested at 14, 21, 28, and 84 days postinfection were shown to be positive for both DTH and in vitro blastogenesis. In no instance was it possible to demonstrate appreciable, if any, humoral antibody production after a primary Listeria infection.
DISCUSSION The onset of uniformly demonstrable DTH after primary infection of the guinea pig with Listeria was shown to occur on day 7 postinfecDAYS AfTER INFECTION tion in the present study (Fig. 1). These results FIG. 5. Listeria antigen-induced blastogenesis of confirm the findings previously obtained in this PE lymphocytes. Onset and persistence of blastogenic laboratory (10). In the present study, it was potential of lymphocytes harvested from guinea pigs deemed necessary to histologically characterize at various time intervals after a primary Listeria the skin infection. Each point represents the group geometric to 48 h test lesion. Skin biopsies were taken 4 after skin testing, placed in Helly mean based on the highest stimulation level for each fixative, and stained with Giemsa according to animal obtained in dose-response experiments (0.02 to 2,000 Mg). Groups consisted of five to six animals the procedure recommended by Askenase (1). each, except for day 7 at which time a total of 17 Use of this histological procedure instead of the animals were tested. Bars indicate standard error. conventional fixation in formalin and staining
TABLE 1. Comparison of Listeria antigen (Ag)-specific lymphocyte blastogenesis and humoral antibody levels of guinea pigs after infection with L. monocytogenes Skin re-
activity Blastogenesis HuTime at 24 h after after skin .moa
~~~~~~~~~~~~~~~~
Time
nftec
tion (days)
7
14
21
28
84
730
Guinea test.in pigno.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
18 19 20 21 22 23 24 25 26 27 28
diameter erythema (mm)
29 30 31 32 33 34
9.0 5.5
Optimal concn
anti-
Counts/min + Ag
Counts/m
S
11.0 4.0 9.0 8.0 13.0 12.0 11.0 Neg' 9.0 9.5 Neg 9.0 7.0 3.5
24 55 41 216 18 296 49 632 41 446 31 1,558 ± 158 5,763 ± 472 85 7 64 14 381 17 63 11 103 24 266 31 104 29 335 67
241 ± 202 ± 189 ± 160 ± 129 + 344 + 266 + 813 + 764 + 62 + 127 + 229 + 71 + 92 + 135 ± 63 + 146 ±
21 40 31 4 35 21 32 46 37 12 43 30 15 10 32 14 86
0.95 1.19 1.50 1.35 2.29 1.84 1.68 1.92 7.54 1.38 0.50 1.67 0.89 1.12 1.97 1.65 2.30
13.5 6.0 10.0 10.5 8.5 9.0 10.0 7.5 13.5 9.0 10.5
20 20 20 20 0.2 2.0 200 200 200 20 20
6,724 ± 3,345 ± 213 ± 256 ± 569 ± 4,833 ± 3,427 ± 2,015 ± 838 ± 1,392 + 4,766 +
882 398 45 47 144 520 871 41 12 228 652
234 + 215 ± 203 ± 83 + 175 ± 683 + 600 + 311 + 249 ± 489 + 163 +
78 54 31 11 45 106 56 138 18 80 11
8.0
20 20 200 200 200 2.0
298 + 1,271 ± 38 + 3,257 ± 17,611 ± 5,212 +
35 371 12 207 2,107 319
24 + 91 + 46 + 354 + 330 ± 402 +
20 0.2 2.0 20 20
130 + 324 + 248 ± 17,569 + 22,841 ±
18 159 56 1,206 551
2.0 20 200 200 20
460 ± 312 ± 470 ± 360 ± 31 ±
56 67 56 50 8
Neg 7.0 6.0 9.0 9.0
35 36 37 38 39
12.5 15.0
40 41 42 43 44
Neg
15.0 10.0 9.5 9.5
Neg Neg 9.0
NSe NS NS NS NS
0 0
(Mg) 20 20 200 20 20 20 20 20 20 20 20 20 20 20 200 20 200
6.5
signifcance
body: titer'
229 240 284
8
NS P < 0.05 NS NS P < 0.05 NS P < 0.05
0 0 0 0 0 4 0 4 0 2 0 0 0 0
28.73 15.55 1.04 3.08 3.25 7.07 5.71 6.47 3.36 2.84 29.23
P < 0.01 P < 0.01 NS P < 0.02 P
Vol. 12, No. 3 Printed in U.S.A.
Blastogenesis as an In Vitro Correlate of Delayed Hypersensitivity in Guinea Pigs Infected with Listeria monocytogenes AMY M. FULTON, MEHER M. DUSTOOR, JOSEPH E. KASINSKI, AND ANDREW A. BLAZKOVEC* Department of Medical Microbiology, University of Wisconsin, Madison, Wisconsin 53706 Received for publication 6 May 1975
Randomly bred guinea pigs of both sexes were injected intracardially with one-half a 50% lethal dose of Listeria monocytogenes. When these animals were skin tested with 30 ,g of a water-soluble extract of sonically disrupted Listeria, animals had uniformly detectable levels of delayed-type hypersensitivity (DTH) 6 days after infection. Histological examination of skin test reaction sites, after fixation in Helly fixative and Giemsa staining, revealed a classical tuberculintype infiltrate consisting primarily of mononuclear cells with few polymorphonuclear cells. Many of the small vessels showed perivascular cuffing. When purified peritoneal exudate lymphocytes from these animals were cultured in vitro in the presence of various concentrations of Listeria antigen, it was found that the optimal antigenic dose for specific antigen-induced incorporation of [3H Ithymi-
dine varied for individual animals. In contrast to the early onset of uniformly detectable levels of in vivo DTH, in vitro lymphocyte blastogenesis was not uniformly demonstrable until 14 days postinfection and remained highly significant on days 21, 28, and 84 postinfection. At 7 days postinfection, lymphocytes from 7 of 17 animals were capable of undergoing significant blastogenesis. The Listeria antigen preparation was not mitogenic for peritoneal exudate lymphocytes from normal animals. It was found that no direct correlation exists between the in vivo levels of DTH and in vitro blastogenesis. Cell donors showing significant in vitro blastogenesis nevertheless were also skin test positive for most animals tested. Humoral antibody was found to play no significant role in the immune response of guinea pigs to a primary infection with Listeria monocytogenes.
Studies of acquired cellular resistance to Listeria monocytogenes in the mouse and rat suggest that the development of acquired cellular resistance is dependent upon the development of delayed-type hypersensitivity (DTH) (9, 10). Studies in this laboratory (5) showed that, in guinea pigs infected with one-half a 50% lethal dose of Listeria, bacterial multiplication occurs only during the first 24 h after infection. Listeria-infected animals are also resistant to a small challenge dose of Listeria given 48 h, 7 days, or 2 weeks after the primary infection. However, for this species DTH, as, detected by in vivo skin testing, is not uniformly detectable until day 5 or 6 postinfection, suggesting that the development of acquired cellular resistance is not dependent upon the development of DTH. Many workers have suggested that in vitro antigen-induced incorporation of [3H Ithymidine by sensitized lymphocytes correlates well with in vivo levels of DTH as determined by 647
skin testing and indeed may be a more sensitive measure of DTH (7, 11-13). This paper reports the results of studies designed to further characterize the immunological response of guinea pigs to a primary Listeria infection using four parameters: (i) onset and development of DTH, (ii) histopathology of the skin test reaction site, (iii) measurement of the in vitro Listeria antigen-induced incorporation of [3H ]thymidine by peritoneal exudate (PE) lymphocytes from Listeria-immune guinea pigs, and (iv) determination of the levels of humoral antibody after a primary Listeria infection. (The research described in this paper was submitted in part [by A.M.F. ] to the graduate school of the University of Wisconsin, Madison, in partial fulfillment of the requirements for the M.S. degree in Medical Microbiology.) MATERIALS AND METHODS Animals. Randomly bred albino guinea pigs of both sexes were obtained from local suppliers. The
648
FULTON ET AL.
animals weighed 500 to 700 g at the time of infection. Bacteria. L. monocytogenes was provided by D. W. Smith (Department of Medical Microbiology, University of Wisconsin, Madison). The culture was guinea pig-passaged four times, and a fresh isolate was recovered from the spleen of an infected animal. A subculture of the isolate was grown on a brain heart infusion (BHI) agar slant (Difco Laboratories, Detroit, Mich.) and used to inoculate 75 ml of BHI broth. The broth culture was incubated on a shaker at 37 C until the bacteria were in late log phase (optical density of 0.3 at 615 nm). The culture was then centrifuged at 1,000 x g for 15 min and resuspended in 75 ml of BHI broth. Samples (1 ml) of this suspension were stored at -70 C (frozen stock cultures). The 50% lethal dose for intracardially injected Listeria was 1.2 105. Immunization. BHI broth was inoculated with a BHI agar slant subculture of the frozen stock. When the broth culture was in late log phase, a portion of it was centrifuged at 1,000 x g for 15 min. The pellet was washed once and then diluted with sterile saline. Approximately one-half a 50% lethal dose in a total volume of 2 ml was injected intracardially into each guinea pig. The exact number of viable Listeria injected was determined by plate counts of the immunizing suspension. Antigen. The water-soluble extract of sonically disrupted Listeria was prepared by using a modification of the method of Hinsdill and Berman (6). Six liters of BHI broth (500 ml/flask) was inoculated with agar slant subcultures of the frozen stock and incubated on a shaker at 37 C. After 18 h, phenol was added at a final concentration of 5% (wt/vol), and the flasks were incubated for a further 8 h. The killed bacteria were collected by centrifugation at 7,000 x g for 30 min at 4 C and washed once with sterile saline and three times with sterile distilled water. The washed cells were suspended in enough distilled water to give a final volume of 50 ml and disrupted by sonic oscillation. The disrupted cells were centrifuged at 18,400 x g for 30 min at 4 C, and the supernatant was set aside. The cellular debris was suspended in 50 ml of distilled water and centrifuged as above. The two supernatants were combined and filtered (Whatman no. 1 filter paper). Samples (5 ml) were placed in small vials and lyophilized. The vials were sealed while under vacuum and stored in a desiccator at 4 C. Skin tests. Guinea pigs were injected intradermally on the shaved dorsal flank with 30 gg of the water-soluble extract in 0.1 ml of saline. The skin tests were read at 2, 4, 8, 12, 24, and 48 h. Cutaneous reactivity was measured by using three parameters: (i) degree of erythema, (ii) mean diameter of erythema, and (iii) changes in double skin thickness at the reaction site. The degree of erythema was scored arbitrarily as described previously (5). Induration was measured using a Schnelltester (System Kroplin, type A, 2T, H. C. Kroplin, Schliictern, Hessen, Germany). Each value given in the text represents the double skin thickness of the reaction site minus the mean of normal skin measured on either side of the test site. PE cells. PE were induced in guinea pigs by the intraperitoneal injection of 25 to 30 ml of sterile no. 31 x
INFECT. IMMUN. heavy paraffin oil (American Oil Co., Chicago, Ill.). Three days later the animals were killed by exsanguination via cardiac puncture. Lavage of the peritoneal cavity was carried out with a total of 300 ml of M-199 medium (Grand Island Biological Co., Grand Island, N. Y.). The PE cells were pelleted by centrifugation at 210 x g for 20 min. The pellet was resuspended in 15 ml of M-199 and centrifuged at 250 x g for 20 min. The cell pellet was resuspended in 8.0 ml of Eagle minimal essential medium (MEM; Grand Island Biological Co., Grand Island, N. Y.) containing 20% heat-inactivated normal guinea pig serum. Preparation of lymphocytes. Lymphocytes used for in vitro [3H ]thymidine assays were obtained from whole PE using the method of Rosenstreich et al. (15). PE cell suspensions were layered onto a column of rayon wool (Xerox cleaning absorbent, Xerox Corp., Rochester, N. Y.). The column was incubated at 37 C in an atmosphere of 5% CO2 for 20 min. The nonadherent cell population was eluted by passing 50 ml of MEM through the column. The eluate was centrifuged at 250 x g for 20 min. This procedure yielded an average of 1.5 x 107 cells per animal with an average viability of 90% as determined by trypan blue exclusion. Cell differentiation using Wright stain showed that 90 to 95% of the cells consisted of small lymphocytes. [3H]thymidine assay. Purified lymphocytes were diluted to a concentration of 5 x 106 cells/ml in Eagle MEM containing 100 U of penicillin per ml, 80 gg of streptomycin per ml, and 20% heat-inactivated normal guinea pig serum. The cells were dispensed in 0.1-ml aliquots in Linbro microtiter plates (model IS-FB-96-TC, Linbro Chem. Co., Inc., New Haven, Conn.). Various dilutions of Listeria antigen were added to the wells in 0.1-ml aliquots with final well concentrations of 2.0 and 0.2 mg/ml, and 20, 2.0, 0.2 and 0.02 Mg/ml. Cells incubated in MEM containing normal guinea pig serum but no Listeria antigen served as controls. A second set of specificity controls consisted of lymphocyte cultures prepared from guinea pigs 28 days after a primary Listeria infection and incubated in the presence of purified protein derivative (PPD; Parke-Davis, Co., Detroit, Mich.) in MEM. Plates were incubated at 37 C in an atmosphere of 5% CO2 for 72 h before adding 2 ACi of [3H ]thymidine (specific activity, 1.9 Ci/mmol, Schwarz-Mann, Orangeburg, N. Y.) in a volume of 0.05 ml of MEM containing penicillin and streptomycin. Cultures were incubated for 18 h and precipitated on glass fiber filters (Reeve Angel, Clifton, N. J.) using a multiple sample precipitator (Otto Hiller, Madison, Wisc.) washed with saline and fixed with 5% trichloroacetic acid. Filter strips were air-dried for 24 h and individual filter disks were placed in scintillation vials (Kimble Products, Toledo, Ohio) with 3 ml of scintillation cocktail consisting of 5.0 mg of 2,5-diphenyloxazole per ml and 0.4 mg of 1,4-bis-(5-phenyloxazolyl)benzene per ml in toluene (Packard Instrument Co., Inc., Downer's Grove, Ill.) adapted to dark and cold and counted for 10 min in a Packard Tri-Carb liquid scintillation spectrometer (Packard Instrument Co., Inc., model 3310, Downer's Grove, Ill.).
VOL. 12, 1975
649
BLASTOGENESIS IN LISTERIA-INFECTED GUINEA PIGS
Antibody titrations. Serum samples from Listeria- sharp drop in responsiveness is seen in animals immune animals were tested for the presence of tested 2 years after infection. At this time two of Listeria-specific antibody using a passive hemaggluti- five animals tested showed only trace skin nation test. Listeria antigen was coupled to sheep positivity. erythrocytes using the bisdiazobenzidine procedure (ii) Histopathology of skin test reaction recommended by Campbell et al. (2). A standard The histological appearance of skin test sites. reference antiserum pool prepared by hyperimmunization of guinea pigs with living and killed Listeria in reaction sites was evaluated at 4, 8, 12, 24, and complete Freund adjuvant served as a positive control 48 h after skin testing animals infected from 1 to for all titrations. The reference antiserum in addition 12 weeks earlier. In all instances, the cellular to all test and control serum samples were inactivated infiltrate was predominately mononuclear in by heating to 56 C for 30 min and then absorbed for 15 nature. At 8- and 12-h post-testing, however, min at room temperature with an equal volume of there were proportionately more polymorphonuwashed, packed sheep erythrocytes prior to microti- clear cells present than at the time of maximum tration using twofold serial dilutions. Titers are ex- skin reactivity, which routinely occurred at 24 pressed as the reciprocal of the highest serum dilution h. Figure 2A-C represents the histological feaat which hemagglutination was observed. Histological preparations. At varying time inter- tures of a 24-h skin test site elicited in an animal vals after skin testing, experimental and control infected 7 days previously. Many of the small animals were randomly chosen for sacrifice. Reaction vessels in Listeria-immune guinea pigs showed sites were removed for histology, fixed in Helly perivascular cuffing (Fig. 2A and B). Although fixative (100 ml of Zenker solution and 5.0 ml of 40% the majority of the infiltrating cells were indeed formaldehyde), embedded in paraffin, and sectioned mononuclear, occasional polymorphonuclear at 5 gm as described by Askenase (1). The sections cells were observed as indicated by arrows in were stained with Giemsa and observations were Fig. 2C. made by light microscopy. Lymphocyte blastogenesis. (i) Antigen Calculation of stimulation index (SI). The levels of specific antigen-induced lymphocyte transforma- dose response. Purified peritoneal lymphocytes tion were calculated by comparing the mean counts were taken from six guinea pigs 28 days after a per minute based on four replicate wells containing primary infection with Listeria. Figure 3 shows lymphocytes cultured in the presence of a given the blastogenic response of lymphocytes from antigen concentration with the mean counts per individual animals incubated with various conminute of four replicate wells containing lymphocytes centrations of Listeria antigen. The antigen cultured in the absence of antigen (controls). Due to concentration giving the maximum SI varies for the unequal variance obtained between groups, log10 transformations of all raw data were carried out prior to calculating the group SI as follows: mean log1O SI = =4.0 (log1O counts per minute in the presence of antigen)/ 3.0 (log10 counts per minute in the absence of antigen). Stimulation indexes were then plotted as antilogs to CD 2.0S. obtain real numerical values. :1.0 Statistics. To determine whether significant blastogenesis had occurred using cells from individual 25.0 animals, the mean value obtained for each set of experimental cultures was first compared to that of _ 20.0 control cultures using Student's t test. For comparing 32 15.0 group treatment mean SI values, preliminary analyses using the two-tailed F-test revealed unequal = 10.0 5.0 variance between group treatment means. To compare the means between treatment groups, it was therefore necessary to utilize the t test for unequal ° 25.0 variance (t). - 20.0 ~ ~
RESULTS Delayed hypersensitivity. (i) General trends. Figure 1 shows the cutaneous reactivity present 24 h after skin testing. Using all three parameters guinea pigs were shown to have uniformly demonstrable delayed hypersensitivity to Listeria antigen 6 days after infection. The response peaks at day 7 and remains fairly constant for the 12-week period of time during which skin testing was routinely carried out. A
~
~
~
~
~
~
.
.
Le t \0
=
5.0
14 28" U "730 21 DAYS AFTER INFECTION FIG. 1. Development and persistence of DTH in Listeria-infected guinea pigs. Responses 24 h after skin testing. Each point represents the mean of 6 to 10 guinea pigs. Bars indicate standard error.
1234561
A1..
-1,16.S'.p
,4
9.,~ ~~~~~ . ..
~10
-''
usf
fr3rt6 *.:
w~:
awi
.R "
I
_~~~~~t
~
e"
.I A.I
Ci
ve I.-
rr
IAO
_
0
FIG. 2. (A) Skin 24 h after intradermal injection of Listeria antigen. Hartley strain guinea pig infected 7 days earlier with Listeria monocytogenes. Area between dermis and musculus carnosus. Giemsa. x55. (B) Skin test reaction to Listeria antigen after 24 h as in (A). Giemsa. x220. Cellular infiltrate comprised primarily of mononuclear cells with perivascular cuffing evident. (C) Skin test reaction to Listeria antigen after 24 h in area adjacent to that shown in (A). Giemsa. x220. Cellular infiltrate comprised primarily of mononuclear cells with occasional polymorphonuclear cells indicated with arrows. 650
VOL. 12, 1975
BLASTOGENESIS IN LISTERIA-INFECTED GUINEA PIGS 50S
30
K
651
53.36
ANIMAL
I
I0 350 6 19
_'I. _ 30 _
20 z -
1
_
10
oI
I
2.0 210 200 LISTERIA ANT16EN (pg) FIG. 3. Listeria antigen-induced blastogenesis of PE lymphocytes. The dose-dependent responses of lymphocytes harvested from six guinea pigs 28 days after a primary Listeria infection. 0.02
each animal, with two animals responding maximally to concentrations of 20 ig/ml with SIs of 12.41 and 13.96 (Fig. 3). Lymphocytes from three animals responded maximally in the presence of 200 ig of Listeria antigen per ml with SIs of 1.22, 9.20, and 53.36. One animal responded maximally in the presence of 2.0 ug/ml with an SI of 12.96. Also note that a concentration of 2,000 ,ug/ml uniformly gives an SI of less than 1.0, indicating a cytotoxic effect at this high concentration. It was clear, therefore, that a wide range of antigen concentrations had to be employed to estimate the in vitro potential of lymphocytes to undergo blastogenesis. (ii) Specificity of the response. The potential of PE lymphocytes harvested from normal guinea pigs to undergo blastogenesis in the presence of Listeria antigen is shown in Fig. 4A. At no antigen concentration does the SI differ significantly from a value of 1.0 indicating that the Listeria antigen does not cause a nonspecific mitogenesis of nonsensitized cells. PE lymphocytes obtained from guinea pigs at 28 days after a primary Listeria infection were incubated in the presence of various concentrations of PPD to further confirm the specificity of Listeria antigen-induced blastogenesis. The SI of immune cells in the presence of all three PPD concentrations employed was not significantly greater than 1.0 (Fig. 4B). Peritoneal lympho-
(A)
T
T
T
I
i
0.02
0.2
2.0
20
LIST1RIA NT116N (pi)
200
zoo
10
10
PPD (pg)
FIG. 4. Specificity of antigen-induced blastogenesis of peritoneal lymphocytes. The response of normal guinea pig lymphocytes cultured in the presence of various concentrations of Listeria antigen (A). Response of lymphocytes harvested from guinea pigs 28 days after a primary Listeria infection and cultured in the presence of various concentrations of PPD (B). Each point represents the group mean of five or six guinea pigs based on the highest stimulation level for each animal obtained in dose-response experiments. Bars indicate standard error.
cytes from these same cell donors were shown to undergo blastogenesis in the presence of specific antigen, i.e., Listeria antigen dose-dependent
(Fig. 3). (iii) Development and persistence of the in
responses
652
FULTON ET AL.
INFECT. IMMUN.
vitro blastogenic response. The maximum < 0.01). It should be pointed out, however, that cells from only three of the five animals assayed mean SI for peritoneal exudate lymphocytes harvested from guinea pigs at various times at 2 years postinfection were shown to be after infection is shown in Fig. 5. Each point significantly positive. (iv) Delayed-type hypersensitivity, lymrepresents the geometric maximum mean SI standard error. As can be seen, there is little phocyte blastogenesis, and humoral antibody blastogenic response during the first week after production after a primary Listeria infection. infection. The SI does not differ significantly The immune response of individual guinea pigs from control animals during the first 7 days. after a primary infection with Listeria (Table 1) Although the maximum group mean SI at day 7 provides for direct comparison of (i) in vivo is not statistically significant, seven of the 17 DTH, (ii) in vitro blastogenesis, and (iii) huanimals tested at this time gave a statistically moral antibody production. Several generalizasignificant response when evaluated on an indi- tions become apparent upon examination of vidual basis, indicating that this may be the these data. There appears to be no direct correlation time postinfection at which the onset of blastogenic potential occurs. At 14, 21, and 28 days between the level of DTH and the SI of individpostinfection statistically significant maximum ual animals. In most instances when a significant SI was obtained, cell donors were also group mean SIs were obtained with values of 7.07 (P < 0.05), 7.76 (P < 0.01), and 12.30 (P < shown to be skin test positive. Several excep0.01), respectively. The peak maximum mean tions exist for animal no. 30 (28 days after infection) and animals no. 40, 42, and 43 (730 group SI of 12.88 is seen for the group of animals tested 12 weeks after the primary Listeria days after infection) which were skin test negainfection. The group of animals tested 2 years tive but whose cells did undergo significant after a primary infection still give a statistically blastogenesis in vitro. The reverse relationship, significant maximum mean group SI of 3.09 (P i.e., skin test positivity and a nonsignificant SI, occurred in 12 of the 44 animals tested. Of these 12 animals, seven were found in the first group 16.0 tested at day 7 postinfection. Data not recorded in Table 1 showed that the 14.0 five to six animals tested on each day prior to day 7 were sometimes DTH positive beginning on day 4; all were uniformly negative in terms of 12.0 SI. At 730 days after infection, only two of the five animals showed DTH but cells obtained 10.0 from three of these animals did show significant 8 .0
blastogenesis. In general, animals tested at 14, 21, 28, and 84 days postinfection were shown to be positive for both DTH and in vitro blastogenesis. In no instance was it possible to demonstrate appreciable, if any, humoral antibody production after a primary Listeria infection.
DISCUSSION The onset of uniformly demonstrable DTH after primary infection of the guinea pig with Listeria was shown to occur on day 7 postinfecDAYS AfTER INFECTION tion in the present study (Fig. 1). These results FIG. 5. Listeria antigen-induced blastogenesis of confirm the findings previously obtained in this PE lymphocytes. Onset and persistence of blastogenic laboratory (10). In the present study, it was potential of lymphocytes harvested from guinea pigs deemed necessary to histologically characterize at various time intervals after a primary Listeria the skin infection. Each point represents the group geometric to 48 h test lesion. Skin biopsies were taken 4 after skin testing, placed in Helly mean based on the highest stimulation level for each fixative, and stained with Giemsa according to animal obtained in dose-response experiments (0.02 to 2,000 Mg). Groups consisted of five to six animals the procedure recommended by Askenase (1). each, except for day 7 at which time a total of 17 Use of this histological procedure instead of the animals were tested. Bars indicate standard error. conventional fixation in formalin and staining
TABLE 1. Comparison of Listeria antigen (Ag)-specific lymphocyte blastogenesis and humoral antibody levels of guinea pigs after infection with L. monocytogenes Skin re-
activity Blastogenesis HuTime at 24 h after after skin .moa
~~~~~~~~~~~~~~~~
Time
nftec
tion (days)
7
14
21
28
84
730
Guinea test.in pigno.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
18 19 20 21 22 23 24 25 26 27 28
diameter erythema (mm)
29 30 31 32 33 34
9.0 5.5
Optimal concn
anti-
Counts/min + Ag
Counts/m
S
11.0 4.0 9.0 8.0 13.0 12.0 11.0 Neg' 9.0 9.5 Neg 9.0 7.0 3.5
24 55 41 216 18 296 49 632 41 446 31 1,558 ± 158 5,763 ± 472 85 7 64 14 381 17 63 11 103 24 266 31 104 29 335 67
241 ± 202 ± 189 ± 160 ± 129 + 344 + 266 + 813 + 764 + 62 + 127 + 229 + 71 + 92 + 135 ± 63 + 146 ±
21 40 31 4 35 21 32 46 37 12 43 30 15 10 32 14 86
0.95 1.19 1.50 1.35 2.29 1.84 1.68 1.92 7.54 1.38 0.50 1.67 0.89 1.12 1.97 1.65 2.30
13.5 6.0 10.0 10.5 8.5 9.0 10.0 7.5 13.5 9.0 10.5
20 20 20 20 0.2 2.0 200 200 200 20 20
6,724 ± 3,345 ± 213 ± 256 ± 569 ± 4,833 ± 3,427 ± 2,015 ± 838 ± 1,392 + 4,766 +
882 398 45 47 144 520 871 41 12 228 652
234 + 215 ± 203 ± 83 + 175 ± 683 + 600 + 311 + 249 ± 489 + 163 +
78 54 31 11 45 106 56 138 18 80 11
8.0
20 20 200 200 200 2.0
298 + 1,271 ± 38 + 3,257 ± 17,611 ± 5,212 +
35 371 12 207 2,107 319
24 + 91 + 46 + 354 + 330 ± 402 +
20 0.2 2.0 20 20
130 + 324 + 248 ± 17,569 + 22,841 ±
18 159 56 1,206 551
2.0 20 200 200 20
460 ± 312 ± 470 ± 360 ± 31 ±
56 67 56 50 8
Neg 7.0 6.0 9.0 9.0
35 36 37 38 39
12.5 15.0
40 41 42 43 44
Neg
15.0 10.0 9.5 9.5
Neg Neg 9.0
NSe NS NS NS NS
0 0
(Mg) 20 20 200 20 20 20 20 20 20 20 20 20 20 20 200 20 200
6.5
signifcance
body: titer'
229 240 284
8
NS P < 0.05 NS NS P < 0.05 NS P < 0.05
0 0 0 0 0 4 0 4 0 2 0 0 0 0
28.73 15.55 1.04 3.08 3.25 7.07 5.71 6.47 3.36 2.84 29.23
P < 0.01 P < 0.01 NS P < 0.02 P
