Characterization of the Pattern of Inflammatory Cell Influx in Chicks Following the Intraperitoneal Administration of Live Salmonella enteritidis and Salmonella enteritidis-Immune Lymphokines

Characterization of the Pattern of Inflammatory Cell Influx in Chicks Following the Intraperitoneal Administration of Live Salmonella enteritidis and Salmonella enter/f/d/s-lmmune Lymphoid nes MICHAEL H. KOGUT, EDWARD D. MCGRUDER,1 BILLY M. HARGIS,1 DONALD E. CORRIER, and JOHN R. DELOACH USDA, Agricultural Research Service, Food Animal Protection Research Laboratory, Route 5, Box 810, College Station, Texas 77845

1995 Poultry Science 74:8-17

phages derived from SE-immune chickens nor supematants from T lymphocytes We have shown that an increased from nonimmune chickens (NILK) were resistance to Salmonella enteritidis (SE) able to confer this protection (McGruder et organ infectivity in day-old and 18-d-old al, 1993; Tellez et al, 1993). This resistance chickens was conferred by the prophylac- was associated with a significant increase tic administration of supematants from in the number of circulating heterophils concanavalin A (Con A)-stimulated, SE- within 4 h of ILK injection (Kogut et al, immune T lymphocytes (ILK) (McGruder 1994a) followed by a decrease in the et al, 1993; Tellez et al, 1993), although number of circulating heterophils with a neither the supematants from macro- concomitant increase in the lamina propria thickness due to a marked infiltration of heterophils (Tellez et al, 1993). Significant increases in adherence, chemotaxis, and Received for publication June 6, 1994. phagocytosis of SE were found with those Accepted for publication September 1, 1994. 1 Texas Agricultural Experimental Research Station, heterophils isolated from ILK-injected chickens compared with the heterophils Texas A&M University, College Station, TX 77843. INTRODUCTION

8

Downloaded from http://ps.oxfordjournals.org/ at FU BerlinFB Humanmedizin on May 14, 2015

ABSTRACT We characterized the inflammatory cell influx in day-old chicks induced by the i.p. administration of live Salmonella enteritidis (SE) and lymphokines from concanavalin A-stimulated SE-immune T lymphocytes (ILK). An i.p. injection of ILK along with 5 x 103 cfu SE increased the survival rate of chicks 48 h later from 70% (ILK-treated controls) compared with 25% (saline-treated). The injection of both the ILK and live SE (but not formalinkilled SE) resulted in an increased influx of inflammatory heterophils into the peritoneum that peaked at 4 h after the injections with no increase in peritoneal macrophages. The heterophil accumulation was not influenced by polymyxin B, but was sensitive to heat treatment (100 C for 1 h) of the ILK, suggesting that lipopolysaccharide (LPS) did not contribute to the induced accumulation of heterophils. Treatment of the chicks with nordihydroguaiaretic acid or indomethacin did not abrogate the induced heterophil accumulation, suggesting that arachidonic acid metabolites were not involved in the SE/ILK-induced accumulation of peritoneal heterophils. The results from the current studies indicate that 1) ILK-mediated resistance to SE-induced mortality is mediated by a rapid influx of inflammatory heterophils to the site of infection; 2) live SE, during invasion, are vital for the site-directed migration of the heterophils; and 3) the mechanisms of induced heterophil accumulation are unknown but involve neither LPS nor arachidonic acid metabolites. (Key words: lymphokines, chicken, heterophils, inflammation, Salmonella enteritidis)

INFLAMMATORY HETEROPHIL ACCUMULATION IN PERITONEUM

9

Difco Laboratories, Detroit, MI 48232-7058.

Downloaded from http://ps.oxfordjournals.org/ at FU BerlinFB Humanmedizin on May 14, 2015

isolated from birds injected with either MATERIALS AND METHODS saline or NILK (Kogut et al, 1994b). After phagocytosis, the heterophils from the Experimental Animals ILK-injected chickens were also able to kill significantly greater numbers of SE more Qne-day-old Leghorn chicks (Hy-Line rapidly than did the heterophils from the W-36®) were obtained from a commercial saline-injected control birds. hatchery and randomly placed in electriClarification of the relationships be- cally heated commercial brooder batteries2 tween the administration of ILK, the SE located within a biological hazard isolation invasion process, and the induction of the unit on the research farm of the College of localized inflammatory response would be Veterinary Medicine, Texas A&M Univerinstrumental in defining the mechanisms sity. Upon arrival, paper liners from the responsible for the ILK-mediated host chick transport containers were cultured for resistance to SE organ infectivity. How- the presence of Salmonella using described ever, delineation of the local mechanisms procedures (Andrews et al, 1978). Chicks of ILK-mediated resistance is complicated were provided ad libitum access to water by technical difficulties in monitoring host and a balanced unmedicated corn-soybean antimicrobial activity of the chicken at the ration. The feed ration contained or excellular level in the intestine. To facilitate ceeded the levels of critical nutrients recomthe study of ILK-induced protection and mended by the National Research Council the inflammatory response in vivo, we (1984). Before use, the feed ration was made use of the distinctive properties of cultured for the presence of Salmonella the chicken peritoneum as a model for using a standard culture method (Andrews inflammatory cell recruitment. Chickens et al, 1978). Salmonella were not detected in have few resident phagocytic cells in the the feed or from the paper liners. peritoneal cavity and respiratory tract and, thus, depend upon the constant recruitment of phagocytic cells from the Infective Material bone marrow (Rose et al, 1974; Glick and A primary poultry isolate of SE from the Rosse, 1976; Ficken et al, 1986; Toth and National Veterinary Services Laboratory, Siegel, 1986). This approach is especially Ames, IA 50010, and approved by the suited for studying the regulation of the USDA Animal and Plant Health Inspection inflammatory cell responses elicited by Service for use in the authors' laboratory, both host cell products, including cytokine was selected for resistance to novobiocin production, and bacterial cell products (NO)-nalidixic acid (NA) and maintained during the in vivo host response. on nutrient agar. Medium used to culture The purpose of the research reported in the resistant isolate in experimental studies the present paper was to evaluate whether contained 25 /*g/mL NO and 20 ^ig/mL NA an i.p. injection of ILK and SE promote the to inhibit the growth of other bacteria. accumulation of inflammatory phagocytes Salmonella enteritidis challenge inocula were in the peritoneal cavity and to characterize prepared in sterile PBS. The viable cell the pattern of inflammatory cell influx. concentration of the inoculum was deterAdditionally, we evaluated whether mined by colony counts on brilliant green lipopolysaccharide (LPS) or arachidonic agar3 (BGA) plates. Where indicated, an acid metabolites stimulated inflammatory overnight culture of SE was killed by cell accumulation in vivo and examined adding .5% formalin and holding the culthe effect of SE/ILK-induced inflamma- ture at 4 C for an additional 24 h. The killed tion in the peritoneal cavity on the resis- bacteria were washed at least four times tance of the chicks to a lethal i.p. challenge with cold PBS to remove the formalin, of SE. suspended to concentrations of approximately 104 bacteria per milliliter, and then stored frozen at -20 C. The absence of viable SE in the suspension was confirmed by 2 Petersime Incubator Co., Gettysburg, OH 45328. plating on BGA plates. 3

10

KOGUT ET Al.

Immunization for Lymphokine Production

Ten 42-wk-old layer hens (Hy-Line W36®), determined to be sero-negative and feces-culture-negative for Salmonella, were immunized by oral gavage with 108 cfu of SE. On Days 14 and 21 postimmunization, the hens were challenged via gavage with the same dose of SE. Preparation of Spleen Cell Suspensions

Isolation of T Lymphocytes

The T lymphocytes were purified from the total spleen cell populations as described in detail previously (McGruder et al, 1993; Tellez et al, 1993). Preparation of Lymphokines

Lymphokines from the T lymphocytes of SE-immune chickens were prepared by incubating the nylon7wool-nonadherent cell suspensions (1 x 10 cells per milliliter) in RPMI 1640 medium containing 7.5 /xg/mL of Con A in 75-cm2 tissue culture flasks for 48 h at 37 C in a 5% COz incubator. After 48 h incubation, the supernatant fluids were collected, centrifuged at 2,000 xg for 15 min to remove all cells, and methyl amannopyranoside (40 /*g/mL) added to inactivate residual Con A. The supernatants were then concentrated fivefold by ultrafiltration using YM-10 membranes (10 kDa cutoff*) (McGruder et al, 1993; Tellez et al, 1993), filtered through a .45-/*m filter, and stored at -20 C until used. Enumeration of Inflammatory Peritoneal Phagocytes

The ability of SE alone, ILK alone, or SE and ILK together to induce the accumula-

4

Amicon Corp., Danvers, MA 01915. sShandon Scientific Co., Pittsburgh, PA 15275. 6 Curtin Matheson Scientific Co., Houston, TX 77251.

Experimental Design

Study 1. For the initial experiments, we evaluated the ability of ILK alone or in combination with SE to recruit inflammatory phagocytic cells into the peritoneal cavity. A series of four identical experiments were conducted over a 4-wk period. Day-old chicks were randomly divided into four treatment groups of 25 chicks each and received i.p. injections of: 1) pyrogen-free saline only (.5 mL, controls); 2) 103 cfu SE only (.1 mL); 3) .5 mL ILK only; or 4) .5 mL ILK followed 30 min later by 103 cfu live SE (.1 mL). At 4, 8, 24, 48, and 72 h postchallenge, five chicks from each group per time period were killed by COz asphyxiation, their peritoneal cavities lavaged with HBSS/EDTA, and the lavage samples were

Downloaded from http://ps.oxfordjournals.org/ at FU BerlinFB Humanmedizin on May 14, 2015

The preparation of the spleen cell suspensions from the SE-immune hens followed the procedure described previously (Kogut and Slajchert, 1992; McGruder et at, 1993; Tellez et al, 1993).

tion of inflammatory heterophils and macrophages was determined by injecting chicks i.p. with .5 mL of either pyrogen-free saline (control) or ILK with or without an i.p. challenge of 5 x 103 cfu SE. At appropriate time points, chicks were killed by C0 2 asphyxiation, and their peritoneal cavities lavaged three times with 2 mL of Ca2+- and Mg2+-free Hanks' balanced salt solution containing .1 M disodium EDTA and .25% bovine serum albumin (HBSSEDTA). The peritoneal exudates from individual chicks were placed into centrifuge tubes and maintained in an ice bath. The recovered total leukocyte numbers were counted with a hemacytometer. Separate samples (300 id,) were removed from each peritoneal cell suspension, and cytospin smears were prepared.5 Smears were airdried, fixed in methanol, and stained with the Hema 3 staining system.6 At least 200 cells on each slide were examined microscopically (l,000x magnification), and the proportions of macrophages, polymorphonuclear cells (PMN), and lymphocytes were scored. Because of the low numbers of eosinophils and basophils in chicks, all PMN cells counted were considered to be heterophils (Lucas and Jamroz, 1961). The number of inflammatory heterophils and macrophages recovered from each chick was determined and used to calculate the mean ± SEM inflammatory heterophils and macrophages per chick (five chicks per group in each experiment).

INFLAMMATORY HETEROPHIL ACCUMULATION IN PERITONEUM

7

Sigma Chemical Co., St. Louis, MO 63178-9916.

Study 4. We next evaluated the effects of cyclooxygenase and lipooxygenase inhibitors on the ILK/SE-induced accumulation of peritoneal inflammatory heterophils in chicks. Four identical experiments with each inhibitor were conducted over a 4-wk period. Day-old chicks were randomly divided into 10 treatment groups of five to seven chicks each and received i.p. injections of 1) pyrogen-free saline only (.5 mL, controls); 2) 103 cfu live SE only (.1 mL); 3) .5 mL ILK only; 4) .5 mL ILK followed 30 min later by 103 c fu HVe SE (.1 mL); 5) .5 mL ILK with a concomitant injection of indomethacin 7 (100 fig in .1 mL) followed 30 min later by 103 cfu live SE (.1 mL); 6) pyrogen-free saline (.5 mL) with a concomitant injection of indomethacin (100 fig in .1 mL); 7) 103 c fu live SE (.1 mL) with a concomitant injection of indomethacin (100 fig in .1 mL); 8) .5 mL ILK with a concomitant injection of n o r d i h y d r o guaiaretic acid (NDGA; 200 fig in .1 mL) 7 followed 30 min later by 103 cfu live SE (.1 mL); 9) pyrogen-free saline (.5 mL) with a concomitant injection of NDGA (200 fig in .1 mL); and 10) 103 c fu live SE (.1 mL) with a concomitant injection of NDGA (200 fig in .1 mL). Data from these replicate experiments were pooled for presentation and statistical analysis. Study 5. Lastly, we sought to determine what effect the ILK/SE-induced inflammatory heterophil accumulation might have on the host resistance to SE. We have developed a 48 h in vivo peritoneal mortality model for SE in which a LD 80 of 5 x 103 cfu live SE was used. Preliminary studies showed that 1 x 103 cfu SE was not lethal for day-old chicks when given i.p., whereas 5 x 103 cfu SE killed 80% of the chicks within 48 h. Day-old chicks were randomly divided into three treatment groups of 20 chicks each and received i.p. injections of: 1) pyrogen-free saline only (.5 mL, controls); 2) pyrogen-free saline followed 3 h later with an i.p. challenge of 5 x 103 cfu SE (.1 mL); or 3) .5 mL ILK followed 3 h later with an i.p. challenge of 5 x 103 cfu SE (.1 mL). Survival was determined after the injection of SE. The survival rate (number of chicks that survived/total number x 100%) was recorded 48 h after the infection.

Downloaded from http://ps.oxfordjournals.org/ at FU BerlinFB Humanmedizin on May 14, 2015

collected. Total and differential peritoneal leukocyte counts were determined as described above. Data from these replicate experiments were pooled for presentation and statistical analysis. Study 2. In order to evaluate whether the recruitment of inflammatory heterophils into the peritoneal cavity by ILK and SE was dependent upon the viability of the bacteria, a series of four identical experiments were conducted over a 2-wk period. Based on the results of the preliminary experiments in Study 1, we chose 4 h as the time of optimal accumulation of inflammatory heterophils in the peritoneum. Day-old chicks were randomly divided into six treatment groups of five to seven chicks each and received i.p. injections of: 1) pyrogen-free saline only (.5 mL, controls); 2) 103 c fu live SE only (.1 mL); 3) 103 cfu of formalin-treated (dead) SE only (.1 mL); 4) .5 mL ILK only; 5) .5 mL ILK followed 30 min later by 103 c n i live SE (.1 mL); or 6) .5 mL ILK followed 30 min later by 103 cfu dead SE (.1 mL). At 4 h postchallenge, all chicks from each group were killed by CO z asphyxiation, their peritoneal cavities lavaged with HBSS-EDTA, and the lavage samples collected. Total and differential peritoneal leukocyte counts were determined as described above. Data from these replicate experiments were pooled for presentation and statistical analysis. Study 3. In order to evaluate the contributions of LPS to SE/ILK-induced heterophil accumulation into the peritoneum, we compared the inflammatory heterophil response in the peritoneal cavity of chicks injected with: 1) ILK (.5 mL) + live SE (103 cfu in .1 mL); and that of chicks injected with either 2) live SE (103 cfu in .1 mL) preincubated 30 min with polymyxin B (10 fig) and ILK (.5 mL) preincubated 30 min with polymyxin B (10 fig); and 3) chicks injected with live SE (103 cfu in .1 mL) plus heat-treated ILK (100 C for 1 h; .5 mL). Dayold chicks were randomly divided into the treatment groups of five to seven chicks each. Total and differential peritoneal leukocyte data from three replicate experiments were pooled for presentation and statistical analysis.

11

12

KOGUT ET AL.

Statistical Analysis

A - • -

Control

i a *'

0

-- + -- SE

- * -

B

3

ILK

- • -

SE/ILK

B\

6

9

12 20

32

44

56

68

80

Time postinjection (h)

B - • -

Control

- +-

SE

-A-

ILK

--•-•

SE/ILK

RESULTS For our experiments, we evaluated the ability of SE alone, ILK alone, or SE and ILK together to recruit inflammatory cells into the peritoneum. As illustrated in Figure 1A, heterophil accumulation was enhanced (P < .0001) from 4 to 24 h after the i.p. injection of the combination of SE and ILK as compared with that of the chicks injected with pyrogen-free saline, SE alone, or ILK alone. Peak heterophil accumulation occurred at 4 h with a dramatic decline in the number of peritoneal heterophils over the next 20 h before returning to control levels by 48 h. Both SE alone and ILK alone recruited greater (P < .0001) numbers of heterophils by 4 h after injection than the pyrogen-free saline injection. The heterophil accumulation in these two groups remained relatively constant in numbers of cells through 24 h postinjection before returning to control numbers by 48 h. However, it should be noted that there were approximately 70 to 75% fewer heterophils in the peritoneum of the chicks injected with either SE alone (1.56 ± .14 x 106 cells per chick) or ILK alone (1.25 ± .18 x 106 cells per chick) as compared to the chicks injected with the combination of SE and ILK (5.09 ± .19 x 106 cells per chick). We found no differences for accumulation of

.00 ' 0

' 20

' 40

' 60

' 60

Time postinjection (h)

FIGURE 1. A) Time course of peritoneal heterophil accumulation after an i.p. injection of either pyrogen-free saline (control), 103 cfu live Salmonella enteritidis (SE), SE-immune lymphokine (ILK), or 103 cfu live SE and ILK (SE/ILK). B) Time course of peritoneal macrophage accumulation after an i.p. injection of either pyrogen-free saline (control), 103 cfu live SE, SE-immune lymphokine (ILK), or 103 cfu live SE and ILK (SE/ILK). Results are expressed as the mean ± SEM macrophages x 106 per chick (five chicks per group in each of four experiments). Symbols with different letters at each time point are significantly different (P < .0001).

inflammatory peritoneal macrophages among any of the treatment groups during the 72-h observation period of this study (Figure IB). We next evaluated whether bacterial viability was an important factor in the recruitment of the inflammatory heterophils. The results from these experiments are represented in Figure 2. Only the viable bacteria in combination with the ILK were able to stimulate an influx of inflammatory heterophils into the peritoneal cavity that was significantly higher

Downloaded from http://ps.oxfordjournals.org/ at FU BerlinFB Humanmedizin on May 14, 2015

Absolute heterophil numbers are presented as the mean of five chicks per group. Data for replicate experiments were pooled for the present studies. Differences among treatments for Studies 1 (2 x 2), 2 (2 x 3), 3 (2 x 2 x 2), and 4 (2 x 2 x 2) were determined by analysis of variance with factorial arrangement of treatments using the General Linear Models procedure of SAS Institute (1982). Significant differences were separated using Duncan's multiple range test (Schefler, 1969). Significance in SE-induced mortality was determined by comparing the number of surviving chicks versus the total number of chicks injected with SE in each group using the chi-square test of independence (Schefler, 1969). The mortality data was then expressed as the percentage survival (number of survivors per total chicks challenged in each group).

13

INFLAMMATORY HETEROPHIL ACCUMULATION IN PERITONEUM A I

A I

I Control

I Control

mm SE

B

B

T

.#.

c

TREATMENTS

\A

B

en

ILK

•

SE/ILK

•

H i I

SE/ILK +PB i SE/heat ILK

TREATMENTS

We were only able to recover a similar number of heterophils from the killed SE/ ILK chicks as were recovered from the chicks treated with ILK only (1.49 ± .24 x 10* cells per chick for the killed SE/ILK group and 1.89 ± .47 x 106 cells per chick for the chicks treated with ILK alone). In order to eliminate the possibility that contaminating LPS was responsible for the influx of inflammatory heterophils, we compared the recovery of peritoneal heterophils in chicks injected with either SE/ ILK, SE and ILK preincubated with polymyxin B (10 /*g), or SE and heat-treated ILK (100 C for 1 h). The results from these experiments are shown in Figure 3. Polymyxin B had no effect on the ability of the SE/ILK to recruit inflammatory heterophils, whereas heat treatment of the ILK reduced (P < .0001) heterophil accumulation in the peritoneal cavity. In fact, the number of heterophils recovered from the peritoneum of the chicks injected with SE plus heated ILK (1.35 ± .44 x 106 cells per chick) was nearly identical to the numbers of cells recovered from chicks injected with SE alone (1.71 ± .21 x 10* cells per chick). We also determined whether the release of arachidonic acid metabolites mediated

the influx of inflammatory heterophils into the peritoneal cavity by evaluating the effects of cyclooxygenase (indomethacin) and lipooxygenase (NDGA) inhibitors on the SE/ILK-induced accumulation of heterophils. Neither indomethacin nor NDGA, at the doses used here, induced a nonspecific inflammatory response in the peritoneum (Figure 4). Concomitant administration of either indomethacin or NDGA with SE/ILK had no effect on the accumulation of inflammatory peritoneal heterophils. Injection of either arachidonic acid metabolite inhibitor with SE or ILK had no effect on the accumulation of peritoneal heterophils that was observed in the birds given the SE or ILK alone (data not shown). Lastly, we determined what effect the peritoneal inflammatory heterophil accumulation had on host resistance to a lethal challenge with SE. The results are shown in Figure 5. An increase (P < .001) in the number of survivors was observed in the ILK-treated chicks as compared with the SE-infected controls. All chicks in the nontreated, challenged controls died by 3 d after Salmonella challenge; whereas,

Downloaded from http://ps.oxfordjournals.org/ at FU BerlinFB Humanmedizin on May 14, 2015

FIGURE 2. Recovery of peritoneal heterophils 4 h FIGURE 3. Effect of lipopolysaccharide (LPS) on after the i.p. injection of either pyrogen-free saline the induction of peritoneal heterophils. Chicks were (control), 103 cfu live Salmonella enteritidis (SE) (SE-L), injected i.p. with either pyrogen-free saline (control), 103 cfu formalin-killed SE (SE-D), SE-immune lym- 103 cfu live Salmonella enteritidis (SE), SE-immune phokine (ILK), 103 cfu live SE and ILK (SE-L/ILK), or lymphokine (ILK), 103 cfu SE and ILK (SE/ILK), 103 103 cfu formalin-killed SE (SE-D) and ILK (SE-D/ cfu live SE and ILK preincubated with 10 /ig ILK). Each bar represents the mean ± SEM heter- polymyxin B (SE/ILK + PB), or 103 cfu live SE and ophils x 10* per chick (five chicks per group in each ILK that was heated at 100 C for 1 h (SE/heat ILK). of four experiments). Bars with different letters are Chicks were killed 4 h later and their inflammatory significantly different (P < .0001). peritoneal heterophils recovered and quantified. Each bar represents the mean ± SEM heterophils x 106 per chick (five chickens per group in each of four experiments). Bars with different letters are signifithan seen in either ILK or SE-L controls. cantly different (P < .0001).

14

KOGUT ET AL. iijl]

Nonintected

ILK.inlected

20/20

A

1-

T

S 5

|

I «

""-

2

1

r

I

B

! • o

A A

D

D

D

LZZ3 SE 15/20

LZZ1 ILK

. .1 .

w

1 11 s 11

g

80

T

3 C ^ H

SE/ILK

isj&®ai SE/ILK+lndo I

I SE/ILK+NDGA

FIGURE 4. Effect of the administration of arachidonic acid metabolite inhibitors on the Salmonella enteritidis (SE) plus SE-immune lymphokine (SE/ ILK)-induced accumulation of peritoneal heterophils. Chicks were injected i.p. with either pyrogen-free saline (Control), 103 cfu live SE, ILK, 100 *tg indomethacin (Indo), 200 /tg nordihydroguaiaretic acid (NDGA), 103 cfu SE and ILK (SE/ILK), 103 cfu SE and ILK plus 100 ng indomethacin (SE/ILK + Indo), or 103 cfu SE and ILK plus 200 pg NDGA (SE/ ILK + NDGA). Chicks were killed 4 h later and their inflammatory peritoneal heterophils recovered and quantified. Each bar represents the mean ± SEM heterophils x 106 per chick (five chicks per group in each of four experiments). Bars with different letters are significantly different (P < .0001).

40 -

6/20

1 GROUPS

FIGURE 5. Effect of Salmonella enteritidis (SE) plus SE-immune lymphokine (SE/ILK)-induced peritoneal heterophil accumulation on the survival of chicks injected with a LD80 of SE. Groups of 20 chicks were injected i.p. with either pyrogen-free saline (noninfected), pyrogen-free saline followed 3 h later by 5 x 103 cfu live SE (infected control), or ILK followed 3 h later by 5 x 103 cfu live SE (ILK, infected). Mortality was determined for 48 h. The results are expressed as the percentage survival and the difference between infected and ILK-treated, infected chicks was significant (P < .001).

their cellular sources, and the accumulation of inflammatory cells involved in the ILK-mediated host inflammatory response against SE. no further mortality was observed in the The experiments described here are, to ILK-treated, challenged group through 5 d the authors' knowledge, the first direct after challenge (data not shown). evidence of a connection between the recruitment of heterophils to the site of DISCUSSION Sa/moneZ/a-induced inflammation and enhanced anti-Salmonella resistance. The inThe movement of polymorphonuclear jection of either ILK or live SE alone leukocytes (PMN) from the peripheral induced the accumulation of predominblood into damaged tissue is characteristic antly inflammatory heterophils in the of the acute stage of inflammation following infection by invasive bacteria. This peritoneum. However, approximately four recruitment of the circulating PMN into times more heterophils were recovered tissues during the inflammatory process is from chicks injected with both SE and ILK dependent upon the local production and than either alone. This synergy suggests release of chemoattractant mediators. The that each component is integral for the mediators involved in the cell recruitment induction of mediators responsible for phase of inflammation as well as the optimal site-directed migration of the cellular sources of the mediators that activated heterophils. regulate these events are poorly characterAlthough the exact mechanism(s) by ized in the chicken. As described in the which heterophils are recruited to the present paper, the chicken peritoneum peritoneum remains unknown, the local provides a unique environment with few generation of chemotactic factors appears resident phagocytic cells that can be used to be essential for the initiation and for the in vivo analysis of the production evolution of the process. Based on the and release of inflammatory mediators, results from the present experiments, one

Downloaded from http://ps.oxfordjournals.org/ at FU BerlinFB Humanmedizin on May 14, 2015

TREATMENTS

o a.

INFLAMMATORY HETEROPHIL ACCUMULATION IN PERITONEUM

Based on these findings and the results from our present experiments we have developed the following working hypothesis for the mechanism of heterophil recruitment and protection against SE organ invasion. Injection of the ILK induces a heterophilia and, either directly or indirectly, activates these cells. The invasion process of the SE is then the signal directing the activated heterophils to the site of bacterial invasion. The result is a significant reduction of bacteria into the extraintestinal organs due to the phagocytosis and killing of the activated heterophils. Further experiments are ongoing to confirm this hypothesis. In this paper, we present preliminary experiments that attempted to identify the mediators responsible for regulating the recruitment of heterophils into the peritoneum induced by both ILK and live

SE. The injection of formalin-killed SE along with the ILK did not result in an increase in the number of peritoneal heterophils above the number recovered from the ILK-injected controls, suggesting that the heterophil accumulation was not mediated by bacterial products such as LPS. Arachidonic acid metabolites, i.e., prostaglandins and leukotrienes, produced by the host during the inflammatory process have been shown to be involved in the site-directed migration of PMN (Larsen and Henson, 1983; Davies et al, 1984; Borsos and Leonard, 1990). However, in the studies presented here, treatment of the chicks with NDGA or indomethacin did not abrogate the induced heterophil accumulation, suggesting that arachidonic acid metabolites were not involved in the induced accumulation of peritoneal heterophils. Based on the data presented, it is also doubtful that LPS contributed to the SE/ ILK-induced inflammatory cell response that we have described. The failure of polymyxin B to block the heterophil accumulation and the heat-labile property of ILK, i.e., endotoxin is heat stable even after boiling (Matsumoto et al, 1987; Mims, 1987), likely eliminates the possibility of LPS contamination of the ILK. One can question whether the presence of LPS in the cell wall of SE was a mediator of the observed response; however, the formalinkilled SE in combination with ILK did not induce an influx of heterophils into the peritoneum (Figure 2). The conclusion that LPS had little contribution to the SE/ILKinduced accumulation of heterophils in the peritoneum is further supported by the chicken's relative insensitivity to the effects of low concentrations of LPS (Adler and DaMassa, 1978). Lastly, over the last few years, the concept of "preventive activation" of avian respiratory phagocytic cells as a means of promoting nonspecific defense of the respiratory tract against respiratory pathogens has been introduced (Toth et al, 1987, 1988, 1992). Our previous studies with the immunoprophylactic use of ILK (McGruder et al, 1993; Tellez et al, 1993; Kogut et al, 1994b) combined with the results presented here provide strong evidence that preventive activation may

Downloaded from http://ps.oxfordjournals.org/ at FU BerlinFB Humanmedizin on May 14, 2015

of the most important factors in the sitedirected migration of heterophils is the actual invasion process of bacteria (Figure 2). The penetration of the mucosal surface by invasive Salmonella induces an acute inflammatory response characterized by the local accumulation of PMN (Takeuchi and Sprinz, 1967; Candy et al, 1985; Pomeroy, 1988; Conlon and North, 1992; Porter and Holt, 1993; Tellez et al, 1993). In a series of elegant in vitro experiments, two independent groups of investigators d e m o n s t r a t e d that intestinal epithelial cell lines, in response to the adherence and invasion by Salmonella, secrete: 1) interleukin (IL)-8, a potent chemoattractant and activator of PMN, which induces the migration of the PMN to the intestinal epithelium (Eckmann et al, 1993b; McCormick et al, 1993); and 2) an unidentified chemotactic factor(s), which directs PMN movement across the epithelium to the site of invasion (McCormick et al, 1993). Moreover, neither noninvasive bacteria nor killed bacteria could induce the secretion of either substance; similar to our observations in the present studies. Furthermore, endothelial cells, fibroblasts, and epithelial cells in the local environment have been shown to secrete potent chemoattractants following stimulation with inflammatory cytokines (Larsen et al, 1989; Eckmann et al, 1993a; Schuerer-Maly et al, 1994).

15

16

KOGUT ET AL.

Kogut, M. H., and T. Slajchert, 1992. T lymphocytes induce protection in chickens against Eimeria tenella by the production of lymphokines. Immunol. Infect. Dis. 2:69-80. Larsen, C. G., A. O. Anderson, J. J. Oppenheim, and REFERENCES K. Matsushima, 1989. Production of interleukin8 by human dermal fibroblasts: an imAdler, H. E., and A. J. DaMassa, 1978. Toxicity of munocytochemical and fluorescent in situ endotoxins to chicks. Avian Dis. 23:174-178. hybridization study. Immunology 68:31-36. Andrews, W., P. Poelma, C. Wilson, and A. Romero, 1978. Isolation and identification of Salmonella. Larsen, G. L., and P. M. Henson, 1983. Mediators of inflammation. Ann. Rev. Immunol. 1:335-360. Pages 1-29 in: Bacteriological Analytical Manual. 5th ed. Association of Official Analytical Lucas, A. M., and C. Jamroz, 1961. Atlas of Avian Hematology. Agricultural Monograph 25. Chemists, Washington, DC. USDA, Washington, DC. Borsos, T., and E. J. Leonard, 1990. Physiologic functions of complement. Pages 152-165 in: Matsumoto, M., S. Matsubara, T. Matsuna, M. Tamura, K. Hattori, H. Nomura, M. Ono, and T. Immunophysiology. The Role of Cells and Yokota, 1987. Protective effect of human Cytokines in Immunity and Inflammation. J. J. granulocyte colony-stimulating factor on Oppenheim and E. M. Shevach, ed. Oxford microbial infection in neutropenic mice. Infect. University Press, New York, NY. Immun. 57:48-54. Candy, D.C.A., M. P. Osborne, S. J. Haddon, T. S. Wallis, W. G. Starkey, and J. Stephen, 1985. McCormick, B. A., S. P. Colgan, C. Delp-Archer, S. I. Miller, and J. L. Madara, 1993. Salmonella Salmonella. Pages 288-293 in: Infectious Diartyphimurium attachment to human intestinal rhoea in the Young. Strategies for Control in epithelial monolayers: transcellular signalling to Humans and Animals. Proceedings of an Intersubepithelial neutrophils. J. Cell Biol. 123: national Seminar. S. Tzipori, G. Barnes, R. 895-907. Bishop, I. Holmes, and R. Robins-Brown, ed. McGruder, E. D., P. M. Ray, G. I. Tellez, M. H. Kogut, Excerta Medica, New York, NY. D. E. Corrier, J. R. DeLoach, and B. M. Hargis, Conlan, J. W., and R. J. North, 1992. Early pathogene1993. Salmonella enteritidis (SE) immune sis of infection in the liver with the facultative cytokines: effect on increased resistance to SE intracellular bacteria Listeria monocytogenes, Francisella tularensis, and Salmonella typhimurium organ invasion in day-old Leghorn chicks. Poultry Sci. 72:2264-2271. involves lysis of infected hepatocytes by leukoMims, C. A., 1987. Pages 200-201 in: The Pathogenecytes. Infect. Immun. 60:5164-5171. sis of Infectious Diseases. 3rd ed. Academic Davies, P., P. J. Bailey, M. M. Goldenberg, and A. W. Press, New York, NY. Ford-Hutchison, 1984. The role of arachidonic acid oxygenation products in pain and inflam- National Research Council, 1984. Nutrient Requirements of Poultry. 8th rev. ed. National Academy mation. Ann. Rev. Immunol. 2:335-358. Press, Washington, DC. Eckmann, L., H. C. Jung, C. Schurer-Maly, A. Panja, E. Morzycka-Wroblewska, and M. F. Kagnoff, Pomeroy, B. S., 1988. Fowl typhoid. Pages 79-81 in: Diseases of Poultry. M. S. Hofstad, H. J. Barnes, 1993a. Differential cytokine expression by huB. W. Calnek, W. M. Reid, and H. W. Yoder, ed. man intestinal epithelial cell lines: regulated Iowa State University Press, Ames, LA. expression of interleukin 8. Gastroenterology Porter, R. E., and P. S. Holt, 1993. Effect of induced 105:1689-1697. molting on the severity of intestinal lesions Eckmann, L., M. F. Kagnoff, and J. Fierer, 1993b. caused by Salmonella enteritidis infection in Epithelial cells secrete the chemokine chickens. Avian Dis. 37:1009-1016. interleukin-8 in response to bacterial entry. Rose, M. E., and P. Hesketh, 1974. Fowl peritoneal Infect. Immun. 61:4569-4574. exudate cells: collection and use for the macroFicken, M. D., J. F. Edwards, and J. C. Lay, 1986. phage migration test. Avian Pathol. 3:297-304. Induction, collection, and partial characterization of induced respiratory macrophages of the SAS Institute, 1982. SAS® Users Guide: Statistics. SAS Institute Inc., Cary, NC. turkey. Avian Dis. 30:766-771. Glick, B., and C. Rosse, 1976. Cellular composition of Schefler, W. C, 1969. Enumeration data—Chi square and poisson distribution. Pages 86-95 in: Statisthe bone marrow in the chicken. Identification tics for Biological Sciences. Addison-Wesley of cells. Anat. Rec. 185:235-246. Publishing Co., Menlo Park, CA. Kogut, M. H., E. D. McGruder, B. M. Hargis, D. E. Corner, and }. R. DeLoach, 1994a. Dynamics of Schuerer-Maly, C.-C, L. Eckmann, M. F. Kagnoff, M. T. Falco, and F.-E. Maly, 1994. Colonic epithelial the avian inflammatory response to Salmonellacell lines as a source of interleukin-8: stimulaimmune lymphokines. Changes in avian blood tion by inflammatory cytokines and bacterial leukocyte populations. Inflammation 18: lipopolysaccharide. Immunology 81:85-91. 373-388. Kogut, M. H., E. D. McGruder, B. M. Hargis, D. E. Takeuchi, A., and H. Sprinz, 1967. Electron microCorrier, and J. R. DeLoach, 1994b. In vivo scope studies of experimental Salmonella infecactivation of heterophil function in chickens tions in the preconditioned guinea pig. II. following injection with Salmonella enteritidisResponse of the intestinal mucosa to the invaimmune lymphokines. J. Leukocyte Biol, (in sion by Salmonella typhimurium. Am. J. Pathol. press). 51:137-146.

be applicable to protection against invasive Salmonella in the intestine.

Downloaded from http://ps.oxfordjournals.org/ at FU BerlinFB Humanmedizin on May 14, 2015

INFLAMMATORY HETEROPHIL ACCUMULATION IN PERITONEUM

17

Tellez, G. I., M. H. Kogut, and B. M. Hargis, 1993. residing macrophages in the normal chicken. Avian Dis. 30:67-75. Immunoprophylaxis of Salmonella enteritidis (SE) infection by lymphokines in Leghorn chickens. Toth, T. E., P. Siegel, and H. Veit, 1987. Cellular defense of the avian respiratory system. Influx Avian Dis. 37:1062-1070. of phagocytes: elicitation versus activation. Toth, T. E., R. Curtiss, IE, H. Veit, R. H. Pyle, and P. Avian Dis. 31:861-867. B. Siegel, 1992. Reaction of the avian respiratory Toth, T. E, H. Veit, W. B. Gross, and P. B. Siegel, system to intratracheally administered avirulent 1988. Cellular defense of the avian respiratory Salmonella typhimurium. Avian Dis. 36:24-29. system: protection against Escherichia coti airsacToth, T. E., and P. B. Siegel, 1986. Cellular defenses of culitis by Pasteurella multocida-activated respiratory phagocytes. Avian Dis. 32:681-687. the avian respiratory tract: paucity of free-

Downloaded from http://ps.oxfordjournals.org/ at FU BerlinFB Humanmedizin on May 14, 2015