INFECTION AND IMMUNITY, OCt. 1990, p. 3183-3186 Vol. 58, No.

10
0019-9567/90/103183-04$02.00/0
Copyright © 1990, American Society for Microbiology

Inflammatory Effects of Salmonella typhimurium Porins

FRANCESCO GALDIERO,l* MARIA ANTONIETTA TUFANO,1 MARILENA GALDIERO,1
SANDRA MASIELLO,2 AND MASSIMO DI ROSA2
Istituto di Microbiologia, I Facoltad di Medicina e Chirurgia, Universita' di Napoli, Larghetto S. Aniello a Caponapoli,
2-80138 Naples,1 and Dipartimento di Farmacologia Sperimentale, Universita Napoli, 49-80131 Naples,2 Italy
Received 19 December 1989/Accepted 9 July 1990

The inflammatory activity of porins purified from Salmonella typhimurium has been investigated. Porins (0.3
to 30 ,ug) injected into the rat paw induced a dose-related edema that was not due to lipopolysaccharide
contamination and did not appear to be dependent on the activation of the complement system. The edema
induced by 30 ,ug of porins was comparable to that caused by 1 mg of carrageenin and was inhibited by
indomethacin (5 mg/kg) and dexamethasone (0.1 mg/kg). Porins (1 to 10 ,ug/ml) induced a concentration-
related release of histamine from rat peritoneal cells. These results are discussed in the light of the possible
pathogenic role of porins in infections.

The outer membrane of gram-negative bacteria contains
some major proteins that are called porins because of their
role in cellular permeability (14, 15). We carried out various
studies on the immunobiological properties of these pro-
teins. Salmonella typhimurium porins are toxic for macro-
phages in mice, and subtoxic concentrations of porins re-
duce phagocytosis and intracellular killing (27). Our studies
show that porins inhibit phagocytosis by activating the
adenylate cyclase system (4).
These proteins, therefore, induce the activation of the
complement system by acting both on the classic pathway
and on the alternative pathway (9); they bind to human
polymorphonuclear leukocytes, decreasing some of their
biological functions (29), and act as mitogens for B lympho-
cytes (24, 27). Furthermore, in rats the porins increase the
toxicity of cardiotoxic molecules (8) and damage renal
tubules (28).
Our results show that porins play a role as pathogenicity
determinants. These proteins, whether directly secreted by
bacteria (13) or derived from the lysis of the bacterial cells,
are resistant to the action of proteolytic enzymes of the host
and can therefore interact with host-mediated cellular and
humoral systems.
The aim of the present study was to examine the inflam-
matory properties of porins and the possible dependence of
porins on the activation of the complement system.
MATERIALS AND METHODS
Preparation of porins. S. typhimurium SH5014 served as
the source of porins. The method described by Nurminen
(16) was used to extract and purify the porins.
One gram of envelopes was treated with 2% Triton X-100
in 0.01 M Tris-HCl (pH 7.5) containing 10 mM EDTA; after
the addition of trypsin (10 mg/g of envelopes), the pellet was
dissolved in sodium dodecyl sulfate buffer (4% [wt/vol]
sodium dodecyl sulfate in 0.1 M sodium phosphate [pH 7.2])
and applied on an Ultragel ACA34 column equilibrated with
0.25% sodium dodecyl sulfate buffer. The fraction enriched
in protein, identified by A280, was extensively dialyzed and
checked by gel electrophoresis by the method of Laemmli
(11). All possible traces of lipopolysaccharide (LPS) were
identified by means of gel electrophoresis and staining with
*
Corresponding author.
3183
silver nitrate as described by Tsai and Frasch (26) and by
means of the Limulus amoebocyte lysate assay (Limulus
test) (25) with as controls, endotoxin-free water and S.
typhimurium SH5014 LPS extracted by the method of Ga-
lanos et al. (6) as described in a previous paper (7).
Preparation of LPS. LPS-R was isolated from S. typhimu-
rium SH5014 with phenol-chloroform-petroleum ether as
described by Galanos et al. (6). Briefly, liquid phenol (90 g of
dry phenol plus 11 ml of water-chloroform-petroleum ether
in volume ratio of 2:5:8) was added to 1 g of the dried
bacterial extraction mixture. After homogenization for 2
min, the bacteria were centrifuged and extracted twice. The
supernatant was filtered through filter paper and treated as
described by Galanos et al. (6).
LPS-S from S. typhimurium (Difco Laboratories) was also
used.
Production of acute inflammation. Edema of the hind paw
of male Wistar rats (140 to 160 g) was produced by injecting
0.1 ml of phosphate buffer containing 0.3, 3, or 30 ,ug of
porins. Groups of at least five rats were used, and the
volume of the paw was determined immediately after the
injection with a differential volume measuring instrument
(Basile, Milano) as previously reported (5). Subsequent
readings of the same paw were carried out every hour for 5
h and compared with the initial reading. In some experi-
ments the edema produced by injecting 0.1 ml of 1% lambda
carrageenin (Sigma Chemical Co.) in saline was also assayed
for comparative purposes.
Inflammation in decomplemented rats. In some experi-
ments paw edema was induced by porins in complement-
depleted rats.
To determine the drop in complement level of the animals,
three male Wistar rats were injected intravenously with
bovine serum albumin-anti-bovine serum albumin. The sol-
uble complexes were prepared by precipitating antibody
from the rabbit antiserum at the point of equivalence and
dissolving the precipitate in excess bovine serum albumin
(20 times the amount needed for precipitation at equiva-
lence) (4). Complement activity of the serum was evaluated
as described by Lachmann and Hobart (10).
Release of histamine and prostacyclin. Resident peritoneal
cells (approx 5% mast cells) were recovered from male
Wistar rats (200 to 250 g) as previously described (2). Cells
were washed and suspended in Tyrode solution (pH 7.2),
which contained 137 mM NaCl, 2.7 mM KCl, 1 mM CaCl2,
3184 GALDIERO ET AL.
A 8
FIG. 1. Protein patterns of porins preparation made from S.
typhimurium SH5014. Lanes: A, SH5014, 20 ,ug of protein in the
sample; B, molecular weight standards (phosphorilase b, 94,000;
albumin, 67,000; ovoalbumin, 43,000; carbonic anhydrase, 30,000;
trypsin inhibitor, 20,100).
1 mM MgCl2, 12 mM NaHCO3, 0.4 mM NaH2PO4, 5.6 mM
glucose.
Cell suspensions (final volume, 1 ml) were allowed to
equilibrate at 37°C in a metabolic shaker with gentle agita-
tion, and histamine release was initiated by the addition of
the following agents (final concentrations): porins (1, 3, and
10 ,ug/ml), concanavalin A (ConA; 1, 3, and 10 ,ug/ml),
compound 48/80 (0.1, 1, and 10 ,ug/ml).
Experiments with ConA were carried out in the presence
of phosphatidylserine (50 ,ug/ml).
The release was terminated after 10 min by the addition of
2 ml ice-cold Tyrode solution.
Cells and supernatants were recovered by centrifugation
(10 min, 150 x g, 4°C), and histamine concentrations in
solutions and cells were quantified fluorimetrically (19).
Histamine release was calculated as a percentage of the
total cellular content of the amine. All values were corrected
for spontaneous release occurring in the absence of the
inducer (approximately 5%).
Prostacyclin, measured as 6-keto-prostaglandin F/ot in the
supernatants of cell suspensions incubated with porins, was
determined by a specific radioimmunoassay after suitable
dilution in radioimmunoassay buffer without prior extraction
or purification (17).
Statistics. Comparisons between groups were done by the
Student t; statistical significance was considered to be P <
0.05.
RESULTS
The purity of the porins preparation is shown in Fig. 1.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis
showed two bands corresponding to porins with molecular
weight of 34,000 and 36,000. The LPS contamination was
evaluated on the same gel that was used to test porin
purification. The gel was divided in two halves: one was
stained with Coomassie blue and served for porin identifica-
tion, and the other was stained with silver nitrate for LPS
identification. With this procedure no detectable amounts of
LPS were found in the porins fractions. The detection limits
of the gel system were 1 ng with pure LPS of known
concentrations. The Limulus test revealed traces of LPS (10
pg/10 ,ug of porins [final concentration]).
INFECT. IMMUN.
4 I'I
I~~~~~~~~~
Im
U
I/ V
6
0
C J1
c
D2 :/ III -/
Il1 I --- I ---
1
0 _ 1 2 31 4 5
Time (b)
FIG. 2. Rat paw edema induced by 0.3 ,ug (O), 3 jig (-), or 30 jig
(0) of porins or by 1 mg of carrageenin (0). Each point represents
the mean + standard error of the values from at least five rats.
Rat paw edema. Porins injected into the rat paw (0.3, 3, or
30 ,ug) induced a dose-related edema; a peak response
occurred 2 to 3 h after the injection (Fig. 2). The curve for
edema induced by 30 ,ug of porins was still evident after 5 h
and was virtually superimposable on the curve for edema
induced by 1% lambda carrageenin (Fig. 2). The edema
induced by porins was not due to contamination by polysac-
charides (LPS).
Thus two types of such compounds (LPS-R from S.
typhimurium and LPS-S from Difco) injected into the paw
(0.3 or 30 ,ug) failed to induce any detectable edema.
The decomplemented animals showed an 80% reduction in
complement activity, which remained constant for approxi-
mately 12 h. The same degree of paw edema was observed
when porins were injected in complement-depleted rats (data
not shown).
The edema induced by porins was reduced by 50 to 60% in
animals treated subcutaneously with indomethacin (5 mg/kg)
or with dexamethasone (0.1 mg/kg) (Fig. 3).
Histamine and prostacyclin release. Porins (1, 3, or 10
,ug/ml) induced a concentration-related release of histamine
from rat peritoneal cells (Fig. 4). The maximum release
(about 90%) was observed after the cells were incubated for
10 min with 10 pug of porins; the effect was stronger than that
of ConA and about equally as active as that of compound
48/80. At a concentration of 1 ,ug/ml, porins induced a
weaker release as compared with those induced by ConA
and compound 48/80.
In contrast, porins at any concentration (1, 3, or 10 ,ug)
failed to modify the spontaneous release of prostacyclin by
rat peritoneal cells. This release remained in the range of
about 3 ,ug per 105 cells even at the highest porin concentra-
tion.
DISCUSSION
Our results show that porins are clearly endowed with
proinflammatory activity. In fact, when injected into the rat
VOL. 58, 1990 INFLAMMATORY EFFECTS OF S. TYPHIMURIUM PORINS
o
-a2 I/
0 0.1
0 1 3 4
Time (b)
FIG. 3. Effect of indomethacin and dexamethasone on the rat
paw edema induced by 30 ,ug of porins. Symbols: 0, control; 0,
indomethacin 5 (mg/kg) given subcutaneously 1 h before porins; A,
dexamethasone (0.1 mg/kg) given subcutaneously 2 h before porins.
Each point represents the mean ± standard error of the values from
at least five rats.
paw at concentrations between 0.3 and 30 ,ug, these porins
induce dose-dependent edema with long-lasting effects. The
highest dose injected (30 ,ug) induced edema comparable to
that induced by 1 mg of lambda carrageenin. The inflamma-
tion produced by the porins is sensitive to both steroid
(dexamethasone) and nonsteroid (indomethacin) anti-inflam-
matory drugs.
The in vitro studies carried out on peritoneal cells of the
rat show that the porins at concentrations between 1 and 10
,ug/ml are able to induce the release of histamine. At the
highest concentration tested, the porins had an effect that
was stronger than that of ConA and equal to that of com-
pound 48/80.
Furthermore, in vitro experiments have also shown that
porins do not seem to induce release of prostacycline, since
the concentrations of its stable metabolite, 6-keto-prosta-
glandin Fl-a, are not increased on incubation with porins.
Porin-induced inflammation may depend on the release of
histamine, even if we cannot completely exclude the partic-
ipation of arachidonic acid metabolites. In fact, our in vitro
results tend to exclude an increase of 6-keto-prostaglandin
F6-a and consequent prostacycline release, whereas the in
vivo results confirm both the prolonged duration of porin-
induced edema and its marked inhibition by indomethacin.
Porin-induced inflammation was also observed in decom-
plented animals; therefore it is unlikely that the activation of
the complement system plays a major role in the inflamma-
tion induced by porins.
Our results allow us to better understand previous data
dealing with the interaction between endotoxins and inflam-
matory cells (19, 21, 22). Asboe-Hansen and Glick (1) first
examined the in vitro response of rat peritoneal mast cell
3185
100
06-
S
I I I
1 10
jig/ml
FIG. 4. Effect of various concentrations of porins (0), ConA
(0), or compound 48/80 (0) on histamine release from rat peritoneal
cells. Each point is an average of two duplicate experiments.
suspensions to an Escherichia coli endotoxin prepared by
the method of Boivin. They were unable to detect any
influence of endotoxin on mast cells. Sandusky et al. (18)
obtained similar results with either E. coli 0127:B8 Boivin or
Westphal prepared endotoxin even at concentrations as high
as 1 mg/ml. Morrison and Betz (12) demonstrated that,
although endotoxins themselves were unable to stimulate rat
peritoneal mast cells, the isolated protein component of the
endotoxin was at a low concentration (1 to 10 ,ug/ml) and was
able to initiate the noncytotoxic secretion of vasoactive
amines.
Further experiments (18) demonstrated also that LPS from
Salmonella minnesota R595 enhanced the response of pe-
ripheral blood basophils obtained from allergic donors but
not of basophils from nonallergic donors.
On the basis of our experimental results, we hypothesize
that during natural infection bacteria accumulated in an
infectious focus could stimulate an inflammatory process by
way of products either secreted or released through lysis by
the bacterial cells.
In the case of gram-negative bacterial cells, the porin
concentration is approximately 2 x 105 molecules per cell,
covering an area of approximately 1.8 p.m2 or roughly a third
of the cell surface area. Therefore, lysis of about 107 cells is
sufficient for the release of porins in the range of amounts
that in our experiments exhibited an inflammatory effect.
This load is easily produced by bacteria. The activation
carried out by the porins in our experimental model there-
fore is consistent with the possibility of porin-induced in-
flammation occuring in gram-negative bacterial infections.
This evidence of proinflammatory activity by porins fur-
ther enlarges the field of action of protein-host interactions.
These proteins intervene in nonspecific host defenses (4, 9,
29) and, as a consequence, in the specific immunologic
response (27). Among surface components, therefore, porins
can be considered another surface molecule, in addition to
LPS, that may be an important inducer of biological activity
in interactions with the host.
3186 GALDIERO ET AL.
LITERATURE CITED
1. Asboe-Hausen, G., and D. Glick. 1958. Influence of various
agents on mast cells isolated from rat peritoneal fluid. Proc. Soc.
Exp. Biol. Med. 98:458-461.
2. Atkinson, G., M. Ennis, and F. I. Pearce. 1979. The effect of
alkaline earth cations on the release of histamine from rat
peritoneal mast cells treated with compound 48/80 and peptide
401. Br. J. Pharmacol. 65:395-402.
3. Cochrane, C. G. 1963. Studies on the localization of circulating
antigen-antibody complexes and other macromolecules in ves-
sels. I. Structural studies. J. Exp. Med. 118:489-501.
4. Di Donato, A., G. F Draetta, G. Illiano, M. A. Tufano, L.
Sommese, and F. Galdiero. 1986. Do porins inhibit macrophage
phagocyting activity by stimulating the adenylate-cyclase? J.
Cyclic Nucleotide Protein Phosphorylation Res. 11:87-97.
5. Di Rosa, M., J. P. Giroud, and D. A. Willoughby. 1971. Studies
of the mediators of the acute inflammatory response induced in
rats in different sites by carrageenin and turpentine. J. Pathol.
104:15-29.
6. Galanos, C., 0. Luderity, and 0. Wastphal. 1969. A new method
for the extraction of R lipopolysaccharides. Eur. J. Biochem.
9:245-249.
7. Galdiero, F., F. Gorga, C. Bentivoglio, R. Mancuso, E. Galdiero,
and M. A. Tufano. 1988. The action of LPS, porins and
peptidoglycans fragments on human spermatozoa. Infection
16:349-353.
8. Galdiero, F., M. A. Tufano, F. Rossi, W. Filippelli, F. Del
Gaudio, L. Berrino, and E. Marmo. 1986. Effect of Salmonella
typhimurium porins on the cardiovascular and renal apparatus.
J. Med. 17:285-297.
9. Galdiero, F., M. A. Tufano, L. Sommese, A. Folgore, and F.
Tedesco. 1984. Activation of the complement system by porins
extracted from Salmonella typhimurium. Infect. Immun. 46:
559-563.
10. Lachmann, P. J., and M. J. Hobart. 1978. Complement technol-
ogy, chapter Sa. In D. M. Weir (ed.), Handbook of experimental
immunology, 3rd ed. Blackwell Scientific Publications, Ltd.
11. Laemmli, U. K. 1970. Cleavage of structural proteins during the
assembly of the head of the bacteriophage T4. Nature (London)
227:680-685.
12. Morrison, D. C., and S. J. Betz. 1977. Chemical and biological
properties of a protein-rich fraction of bacterial lipopolysaccha-
rides. II. The in vitro rat peritoneal mast cell response. J.
Immunol. 119:1790-1795.
13. Morrison, D. C., S. J. Betz, and D. M. Jacobs. 1976. Isolation of
a lipid A bound polypeptide responsible for "LPS initiated"
mitogenesis of C3H/HeJ spleen cells. J. Exp. Med. 144:840-
846.
14. Nakae, T. 1976. Identification of the outer membrane protein of
Escherichia that produces trans membrane channels in recon-
INFECT. IMMUN.
stituted vesicle membranes. Biochem. Biophys. Res. Commun.
71:877-884.
15. Nakae, T. 1976. Outer membrane of Salmonella. Isolation of
protein complex that produces transmembrane channels. J.
Biol. Chem. 251:2176-2178.
16. Nurminen, M. 1978. A mild procedure to isolate the 34K, 35K
and 36K porins of the outer membrane of Salmonella typhimu-
rium. FEMS Microbiol. Lett. 3:331-334.
17. Salmon, J. A. 1978. A radioimmunoassay for 6-keto-prostaglan-
din F/a. Prostaglandins 15:383-397.
18. Sandusky, C. B., A. R. Johnson, and N. C. Moran. 1973. Effect
of endotoxin and anaphylatoxin on rat mast cells and chopped
rat lung. Proc. Soc. Exp. Biol. Med. 143:764-768.
19. Shore, P. A., A. Burkhalter, and V. H. Cohn. 1959. A method for
the fluorometric assay of histamine in tissues. J. Pharmacol.
Exp. Ther. 127:182-186.
20. Smith, T. F., M. Aelvoet, and D. C. Morrison. 1985. The effect
of bacterial lipopolysaccharide (LPS) on histamine release from
human basophils. Clin. Immunol. Immunopathol. 34:355-365.
21. Sultzer, B. M. 1968. Genetic control of leukocyte responses to
endotoxin. Nature (London) 219:1252-1254.
22. Sultzer, B. M. 1969. Genetic factors in leukocyte responses to
endotoxin: further studies in mice. J. Immunol. 103:32-38.
23. Sultzer, B. M. 1972. Genetic control of host responses to
endotoxin. Infect. Immun. 5:107-109.
24. Takada, H., T. Ogawa, F. Yoshimura, K. Otsuka, S. Kokeguchi,
K. Kato, T. Umernoto, and S. Kotani. 1988. Immunobiological
activities of a porin fraction isolated from Fusobacterium nu-
cleatum ATCC 10953. Infect. Immun. 56:855-863.
25. Thye Yin, E., C. Galanos, S. Kinsky, R. A. Bradshaw, S.
Wessler, 0. Luderity, and M. F. Sarmiento. 1972. Picogram-
sensitive assay for endotoxin: gelation of Limulus polyphenus
blood cell lysate induced by purified lipopolysaccharide and
lipid A from Gram-negative bacteria. Biochim. Biophys. Acta
261:284-289.
26. Tsai, C. M., and C. E. Frasch. 1982. A sensitive silver stain for
detecting lipopolysaccharides in polyacrylamide gels. Anal.
Biochem. 179:115-119.
27. Tufano, M. A., M. T. Berlingieri, L. Sommese, and F. Galdiero.
1984. Immune response in mice and effects on cells by outer
membrane porins from Salmonella typhimurium. Microbiolog-
ica 7:353-366.
28. Tufano, M. A., G. Capasso, P. Anastasio, D. R. Giordano, D.
Giordano, E. Galdiero, and N. G. De Santo. 1987. Clearance
studies on the renal action of porins extracted from Salmonella
typhimurium. Int. J. Pediatr. Nephrol. 8:193-198.
29. Tufano, M. A., R. Ianniello, M. Galdiero, L. De Martino, and F.
Galdiero. 1989. Effect of Salmonella typhimurium porins on
biological activities of human polymorphonuclear leukocytes.
Microb. Pathogen. 7:337-346.