Postinflammatory increase of absorption from peritoneal cavity into lymph nodes: Particulate and oily inocula

EXPERIMENTAL

AND

MOLECULAR

PATHOLOGY

43. 124-134 (1985)

Postinflammatory Increase of Absorption from Peritoneal Cavity into Lymph Nodes: Particulate and Oily Inocula’ SEYMOURLEVINE Pathology

Department,

New

York

Medical Valhalla, Received

College.? and Westchester New York 10595 December

County

Medical

Center,

28, 1984

One week after intraperitoneal injection of rats and mice with a chemical irritant, the abdominal cavity was in the healing phase of a sterile peritonitis. At this time, mineral oil or a particulate dye or metal injected intraperitoneally was absorbed into lymph nodes in much larger amounts than normal. Absorption of oil caused fourfold enlargement of nodes, distention of afferent lymphatics, and sometimes widespread oil embolization of lungs. Absorption was increased when the chemical irritant had bathed the entire peritoneum but not when it was limited to the lower abdominal cavity. The chemical peritonitis caused shrinkage, agglutination, fixation, and fibrosis of the greater omentum which thereby lost its ability to sequester particles. In the absence of omental scavenging, dye, metal, or oil accumulated under the diaphragm and penetrated the diaphragmatic lymphatics. In addition, enhanced absorption by proliferated subdiaphragmatic lymphatics may have been involved. Prior induction of a chemical peritonitis augments absorption into the lymphatics of certain aqueous and cellular as well as particulate and oily inocula. so it is likely to find many apphcations in experimental pathology. 0 1985 Academic press. Inc.

INTRODUCTION Materials injected into the peritoneal cavity can be absorbed into the blood stream or into the lymphatics. Recently we reported that lymphoid cells responsible for producing experimental allergic encephalomyelitis, graft-versus-host disease, and host-versus-graft reaction were more effective when inoculated into a peritoneal cavity recovering from a mild chemical peritonitis than after inoculation into a normal peritoneal cavity (Levine, 1985). These observations suggested that absorption of lymphoid cells via the lymphatic route was enhanced during the healing phase of a chemical peritonitis. In the present work, this suggestion has been confirmed by direct visual and histological observations of increased absorption of particulate and oily materials into the draining lymph nodes. Our observations suggest that the functional status of the greater omentum and related peritoneal surfaces is the chief determinant of the lymphatic absorption of these materials. In addition to revealing a new aspect of the pathophysiology of inflammation, the present studies offer a simple way to enhance the absorption of several different types of inocula into the lymphatic system, a potentially useful device for immunopathology, toxicology, oncology, and other areas of experimental pathology. METHODS

AND MATERIALS

Lewis rats of either sex, 180-300 g, from Harlan Sprague Dawley, Inc., Walkersville, Maryland, were maintained in hanging wire cages on Purina Rodent ’ Supported by a research grant from the National Multiple Sclerosis Society. * Address for reprint requests. 124

0014-4800/85 $3.00 Copyright All rights

0 1985 by Academic Press. Inc. of reproduction in any form reserved.

INCREASED

ABSORPTION

INTO

LYMPH

NODES

125

Laboratory Chow 5001 and tap water ad libitum. Female 30-g Swiss-Webster albino mice were kept in shoe-box type cages. Except as specified, all injections were intraperitoneal and were done without anesthesia. The rats were held upside-down, and a short-beveled, 3/s-in., 21-gauge needle penetrated the lower abdomen, just left of the midline. Sterile peritonitis was induced by any of several chemicals. Most often 50 ml/ kg of a 1:lOO dilution in saline of a commercial 5.25% sodium hypochlorite (NaOCl) solution (household bleach) was injected 1 week before the material to be assayed. Occasionally the NaOCl or the assay material was inadvertently injected into the cecum or retroperitoneally. The rats were routinely fasted the night before inoculation because this reduced the incidence of such errors. Control rats were given an equal volume of saline. Monastral Blue B was a commercial 3% suspension of phthalocyanine blue (Sigma Chemical Co., St. Louis, MO.) that was diluted 1:lOO in saline before injection. Carbonyl iron, type SF, kindly contributed by the GAF Corporation, Linden, New Jersey, was used as a 5% suspension in saline. It was composed of spherical particles, average 3.8~Frn diameter, known to be readily absorbed into draining lymph nodes (Levine and Sowinski, 1970). Several types of light mineral oil (liquid paraffin), usually Bayol F (Exxon Oil Co.), were used as received. The rats were killed by exsanguination while under ether anesthesia. Evidences of chemical peritonitis were noted and mediastinal lymph nodes were weighed fresh. Tissues were fixed in Bouin’s fluid, embedded in paraffin, sectioned, and stained with hematoxylin & eosin. Oil content of lymph nodes was scored by histologic estimates of oil vacuoles and oil-laden macrophages: 1 + , a few small foci; 2 + , many small foci; 3 + , large foci in parenchyma and sinusoids, but less than area of residual lymphoid tissue; 4 + , large foci occupying an equal or greater area than residual lymphoid tissue. RESULTS Chemical

Peritonitis

Rats killed 2 or 3 weeks after injection of NaOCl or other chemical irritants had adhesions involving spleen, stomach, liver, and diaphragm, rounding of the liver edge, fibrosis of the serosal surfaces of spleen, liver, intestines, and omentum. The omentum was thickened and contracted; it could no longer be extended into a thin delicate membrane. One or more of these gross findings were invariably present but the severity and distribution were very variable. No clinical signs were present and the rats appeared healthy. (Higher concentrations than those recommended here can cause bloody ascites.) Microscopically, much of the omental and other peritoneal surfaces was covered with a layer of fibrous connective tissue. Mediastinal lymph nodes were not enlarged and had no microscopic abnormalities except for hemorrhage in some sinusoids. Absorption of a Particulate Dye and Metal In two separate experiments, groups of four rats pretreated with NaOCl or saline were injected with 10 ml/kg of a 1:lOO dilution. of Monastral Blue suspension. When the controls were killed after 4 hr, most of the dye was in the omentum. There were only a few spots of dye on the parietal peritoneum. The mediastinal nodes and some of the abdominal nodes were lightly stained. In

126

SEYMOUR

LEVINE

contrast, the short, thick omentum of NaOCl pretreated rats was almost devoid of dye. There were numerous dye spots scattered over visceral and parietal peritoneum including the abdominal surface of the diaphragm. There was increased staining of mediastinal and abdominal nodes, and striking staining of the retrosternal lymphatic vessels. Rats sacrificed after 24 hr revealed the same striking effects of NaOCl pretreatment except dye was no longer seen in lymphatic vessels. Pretreatment of mice with 0.26% NaOCl gave similar results. An equally remarkable redistribution of inoculated metal powder was observed. Four control rats were pretreated only with saline and killed 10 days after intraperitoneal injection of 2 ml of a 5% suspension of carbonyl iron powder. Most of the iron was in the omentum, distributed in diffuse and finely punctate patterns and also agglomerated into small I- or 2-mm lumps. There was relatively little iron elsewhere in the abdominal cavity, but the mediastinal nodes were enlarged (average 0.28 g) and well pigmented by iron. Four other rats were pretreated with NaOCl and then given iron in the same way as the controls. The shrunken omentum of these rats had very little iron and none of the agglomerated masses that occurred in controls. Instead, there were widespread gray-black deposits of iron throughout the peritoneal cavity with especially conspicuous coating of liver, spleen, and intestines. The mediastinal lymph nodes were enlarged to the same extent as the controls (average 0.26 g) but they were much more pigmented due to increased absorption of iron. All these macroscopic observations on dye and metal distribution were confirmed in microscopic preparations. Absorption

of Mineral

Oil (Liquid

ParafJin)

As previously described (Levine and Sowinski, 1982), injection of 1 ml oil into control rats pretreated with saline or not pretreated, resulted in accumulations of oil-laden macrophages in omentum (Fig. I) and some other peritoneal surfaces

FIGS. 1-6. All figures are from rat tissues taken 2 weeks after intraperitoneal injection of mineral oil. All are stained by hematoxylin & eosin, x 70. FIG. 1. Omentum, saline pretreatment: overlying the omental fat and between the arrows there is a thick surface layer of oil-laden macrophages.

INCREASED

ABSORPTION

INTO LYMPH

NODES

127

but there was not much under the diaphragm (Fig. 3). Only a rare oil vacuole was discovered in the mediastinal nodes (Fig. 5). In contrast, rats pretreated with NaOCl had less oil on the omentum due to fibrotic surfaces and loss of surface area (Fig. 2). There was a striking increase of oil under the diaphragm and permeating the diaphragmatic lymphatics (Fig. 4). The lymph nodes in the mediastinum had absorbed large amounts of oil (Fig. 6). As early as 1 day after injection of 1 ml oil, the nodes contained many oil vacuoles although they were not yet enlarged. A few polymorphonuclear neutrophils were also present in these nodes. Five, eight, or fourteen days after injection, the nodes were enlarged to 0.15-0.45 g (a normal rat’s mediastinal nodes are less than 0.1 g). They contained innumerable oil vacuoles, often in large clusters, often distending the subcapsular and intermediate sinusoids (Fig. 6). Polymorphonuclear leukocytes were no longer present. Occasionally the oil infiltration was so extensive that cut sections of the node revealed a honeycomb pattern due to tiny oil “cysts,” visible to the unaided eye. Afferent lymphatic channels outside the lymph node capsule, like those in the diaphragm, were distended with oil (Fig. 6). Most of the oil was within macrophages and, after 14 days, some of these had developed into epithelioid cell granulomas. Some of the oil in lymphatic vessels was undergoing organization and incorporation into the wall. The accumulation of oil in the lymph nodes was so great that it was no surprise to find oil embolism in the lungs of some rats. Affected lungs were more than doubled in weight (3.0-3.6 g compared to 1. I - 1.4 g for two normal lungs in 200to 250-g male rats) and hemorrhagic or tan in color. Clusters of oil-filled macrophages and focal hemorrhages were numerous and widespread in these lungs and present in small numbers even in rats with normal lung weight. The magnitude of the changes in nodes and lungs had no relation to the time of sacrifice (5, 8,

FIG. 2. Omentum, NaOCl pretreatment: overlying the omental fat and between the arrows, there is a surface zone of fibrosis with no oil-laden macrophages (left) or only a few (right). The diagonal band of fibrous tissue in the fat layer may represent an adhesion due to agglutination of the omentum.

128

SEYMOUR

FIG.

3. Diaphragm,

saline

pretreatment:

LEVINE

there

is no oil on the peritoneal

surface.

or 14 days) or to the severity of the NaOCl-induced peritonitis, nor were they influenced by a fourfold increase in the dose of oil. Most of these experiments were done with Bayol-F mineral oil (Exxon Oil Co.) but similar results were obtained with three other commercial mineral oils (Marcol, from Humble Oil Co., and Draketex 50 and Drake01 6VR kindly contributed by Penreco, Butler, Pa.), and with Pristane (2,6,10,14-tetramethylpentadecane from Aldrich Chemical Co., Milwaukee, Wise.). Similar but less dramatic effects of NaOCl pretreatment were

FIG. 4. Diaphragm, NaOCl pretreatment: there are many oil-laden macrophages surface (below), as well as in greatly dilated lymphatic vessels (L) in the muscular thoracic surface (above).

on the peritoneal layer and on the

INCREASED

ABSORPTION

INTO LYMPH

NODES

129

FIG. 5. Lymph nodes, saline pretreatment: the subcapsular sinusoid (arrow) is inconspicuous and there are no visible oil drops.

found with I-bromohexadecane and with squalane (2,6,10,15,19,23-hexamethyltetracosane). Normal infant and weanling rats, treated only with saline at ages 2 and 5 weeks, respectively, absorbed very little of the injected oil, but adult mice absorbed a small amount. In all three instances, however, pretreatment with NaOCl caused a noteable increase in the size of the nodes and their content of oil (Table I).

FIG. 6. Lymph node, NaOCl pretreatment: the subcapsular sinusoid and afferent lymphatic outside the capsule are distended and some lumens contain oil-laden macrophages. The afferent lymphatic shows early fibrous organization of the oil embolus. There is a granulomatous response to the oil in some areas of the lymph node parenchyma.

130

SEYMOUR Enhanced Absorption

LEVINE

TABLE I of Oil following Chemical Peritonitis Mediastinal nodes Day +14

Subiects (N)”

Day -lh

Rats, 200 g

(12)

Rats, 5 weeks, 50 g

(16) (2) (6)

Rats, 2 weeks, 25 g

(3)

(6) Mice, 30 g

(3) (5)

Saline NaOCl Saline NaOCl Saline NaOCl Saline NaOCl

Day 0 Oil Oil Oil Oil Oil Oil Oil Oil

1 ml/rat 1 ml/rat 0.2 ml/IO 0.2 ml/IO 0.2 ml/10 0.2 ml/l0 0.2 ml/l0 0.2 ml/l0

g g g g g g

Weight (gjr

Oilc

0.05 0.22 0.04 0.11 0.08 0.20 0.01 0.03

L 4+ I+ 3+ 1+ 3-t 2+ 3+

k 2 2 2 f i * k

0.02 0.07 0.01 0.02 0.03 0.07 0.01 0.01

0 Age and average weights refer to Day - 7. b NaOCl, 50 ml/kg of 1: 100 dilution (I:20 for mice) or an equal volume of saline, injected 7 days before Bayol F mineral oil. both intraperitoneally. Several experiments are summarized. r Average moist weights & SD, 14 days after injection of oil. Oil content scored histologically, scale oto4+.

The first experiments to elucidate the mechanism for the enhanced absorption of oil involved the volume and concentration of the NaOCl used for pretreatment. In some of these animals, a very dilute Monastral Blue suspension was added (l/ 10,000, v/v) as a dye marker; a preliminary experiment showed that it had no effect on the ability of NaOCl to induce a peritonitis or on the subsequent absorption of oil. The usual 0.05% concentration of NaOCl was effective in producing peritonitis and in enhancing absorption of oil when injected in a volume of 100 or 50 ml/kg. It was somewhat less effective at 25 ml/kg and ineffective at 10 ml/kg. When this lowest volume of 10 ml/kg had its NaOCl content increased by a fivefold elevation of NaOCl concentration, its ability to produce peritonitis was restored, but the inflammatory changes in the omentum were only partly restored. The enhancement of absorption of oil was also only partly restored (Table II). Therefore, use of a large volume was more important than the concentration of NaOCl.

TABLE II Volume of NaOCl Determines Peritonitis and Subsequent Absorption of Oil

Pretreatment Day - 7” Saline NaOCI, NaOCl. NaOCl, NaOCI, NaOCI,

0.05% 0.05% 0.05% 0.05% 0.26%

50 ml/kg 100 ml/kg 50 ml/kg 25 ml/kg 10 ml/kg 10 ml/kg

Day 0

Peritonitish Day +14

Oil Oil Oil Oil Oil Oil

0 + + 2 0 +

Mediastinal nodes Weight (g) 0.04 0.17 0.23 0.11 0.06 0.09

2 ” 2 t k 2

0.01 0.03 0.07 0.06 0.02 0.03

Oil ? 4+ 4+ 3+ 1+ 2+

a Adult rats suspended in head-down position for 5 min after intraperitoneal injection, under ether anesthesia. Four rats each group. h Adhesions, perisplenitis, fibrosis and shrinkage of omentum, rounding of liver edge when sacrificed 14 days after intraperitoneal injection of 1 ml oil (21 days after NaOCl or saline).

INCREASED

ABSORPTION

INTO LYMPH

131

NODES

In saline pretreated control rats, the marker dye was mostly in the omentum, with very little staining of the abdominal wall. NaOCl pretreated rats had diffuse staining of the entire abdominal wall including the abdominal surface of the diaphragm, except this was almost absent in the rats given the lowest volume, even when the NaOCl concentration was increased to compensate. Presumably, the stained areas correspond to areas of NaOCl contact and injury. These experiments suggested that NaOCl was effective only when given in large volume because small volumes might fail to establish contact with some essential intraabdominal target. That target might be the omentum, the diaphragm, the abdominal wall, or some combination. In an attempt to ascertain the essential target, rats were given NaOCl while they were kept in vertical position, either head-up or head-down. When the animals were kept in either of these positions only during the injection of NaOCl and released immediately thereafter, all of them subsequently showed the usual enhancement of oil absorption. When the positions were maintained for 5 min with or without the aid of ether anesthesia, or maintained for 1 hr with the aid of intramuscular pentobarbital anesthesia, then striking differences were observed in the absorption of subsequently administered oil (Table III). The rats kept 5 or 60 min in the head-up position after NaOCl injection had no upper abdominal adhesions, rounding of liver, perisplenitis, shrinkage of omentum, or redistribution of oil to the abdominal surface of the diaphragm, and they had little or no enhancement of oil absorption into the lymph nodes. The rats kept 5 or 60 min in the head-down position after NaOCl had all the changes described for conventionally injected rats, including a spectacular increase of oil absorption. Repetition of this experiment with marker dye added to the NaOCl showed that 5 min in the head-up position was sufficient to confine the NaOCl effects and the dye staining to the lower abdomen (staining of lower abdominal wall, cecum, fatty tissue around gonads). Clearly, these areas were not the important targets. After 5 min in the head-down position, the marker dye had stained the upper abdominal wall and organs, including the diaphragm. These areas must have included the essential targets. Chemical peritonitis identical to that seen after NaOCl was produced with a recently developed sterilant, “Exspor,” a source of chlorine dioxide, kindly con-

TABLE III Position of Rat after NaOCl Injection Determines Peritonitis and Subsequent Absorption of Oil

Day -7” Saline NaOCl NaOCl Saline NaOCl NaOCl

of rath

Day 0

Peritonitis’ Day + 14

Head-down, 5 min. Head-down, 5 min. Head-up, 5 min. Head-down. 1 hr. Head-down, 1 hr. Head-up, 1 hr.

Oil Oil Oil Oil Oil Oil

0 + 0 0 + 0

Mediastinal nodes Weight (g) 0.05 0.29 0.06 0.09 0.19 0.10

k 0.01 IL 0.06 k 0.03 zlz 0.03 t- 0.02 27 0.03

Oil 0 4+ 1+ 0 3+ 1+

’ NaOCl 0.05%. 50 ml/kg, or equal volume of saline, intraperitoneally. Four adult rats each group. b Vertical position maintained after NaOCl or saline injection with the aid of ether for 5 min or pentobarbital for 1 hr. ’ Adhesions, perisplenitis, fibrosis and shrinkage of omentum, rounding of liver edge when sacrificed 14 days after intraperitoneal injection of 1 ml oil (21 days after NaOCl or saline).

132

SEYMOURLEVINE

tributed by the Alcide Corporation, Westport, Connecticut. Pairs of rats were pretreated with 50x or 100x dilutions of activated Exspor (50 ml/kg ip), and inoculated 7 days later with 1 ml Bayol F mineral oil. They developed enlarged mediastinal lymph nodes (0.20- and 0.27-g averages, respectively) due to abundant absorption of oil. A 200 x dilution did not cause enlargement (0.09 g) but considerable absorption of oil was demonstrated histologically. Control rats given only the oil had 0.12-g nodes and controls given only the Exspor had 0.07-g nodes, and none of these controls contained histologically detectable oil. In additional experiments, similar results were produced by hydrogen peroxide and by a surfactant in concentrations that produced a chemical peritonitis. DISCUSSION With no equipment more sophisticated than a needle and syringe, and with no reagents more exotic than household bleach (NaOCl), we have produced a striking increase in the lymphatic absorption of all categories of intraperitoneal inocula (Levine, 1985). The increased absorption has been demonstrated morphologically in the present work, and by immunopathologic effects in the case of lymphocytic inocula. This discovery could have been made in the nineteenth century. Therefore, it is difficult to be certain that these findings are original, but it seems unlikely that such a potentially useful discovery would have been published and then lost in the literature. It is more likely that no previous investigator produced a peritonitis with a suitable chemical at the right concentration, injected in a large volume and studied for effects on absorption at the correct time and with an appropriate test inoculum. Some further considerations on each of these prerequisites follow. NaOCl and chlorine dioxide are not the only chemicals active in our system, inasmuch as hydrogen peroxide was effective in other experiments (Levine, 198.5). These and other oxidizing agents are well known for producing inflammation with prominent effects on permeability (Steele and Wilhelm, 1967; Fontagne et al., 1973). NaOCl and H,O, were effective only within a limited range of concentrations, suggesting that necrosis due to excessive chemical action was counterproductive. The many chemical irritants that produce more necrosis than permeability effects might have been ineffective in our system. The greatest enhancement of absorption was found after pretreatment with NaOCl in large volume. The comments of Bolton in 1921, although directed to absorption of inocula, are equally appropriate to the distribution of irritant effects of NaOCl: “The fluid or suspension, which is injected into the peritoneum, must be in such an amount that it can flow freely about so as to reach the diaphragm in a thorough manner. If it is in small amount and particularly if injected into the hinder part of the peritoneum, it is held up by the intestines and other organs, and has to gradually find its way between them to reach the diaphragm . . .” Such a delay was likely to be particularly detrimental in our work because NaOCl is a reactive chemical whose effects are quickly dissipated and probably limited to sites of initial contact. The importance of large volume when the intent is to expose all peritoneal surfaces has recently been revived in the “belly bath” chemotherapy of ovarian carcinoma (Litterst et al., 1982). Investigators of inflammation have been very much concerned with permeability, but this concern was usually directed toward the exudation of fluid and

INCREASED

ABSORPTION

INTO

LYMPH

NODES

133

proteins from the blood stream into the peritoneal cavity, leading to edema or ascites when the absorptive capacity of the lymphatic vessels was exceeded (Casley-Smith, 1973). When the inflammation subsided, the edema or ascitic fluid was reabsorbed, which could be due to an increase in the rate of absorption, but this healing phase of inflammation received little attention. Menkin, for example, wrote many important papers on absorption from foci of inflammation, but perusal of about a dozen of his publications revealed no observations on inflammation of more than 4-days duration. When peritonitis was studied during its first day, absorption of foreign particles into the lymphatics was found unchanged or decreased (Bangham et al., 1953). During the first day of inflammation, absorption of certain diffusible dyes and carbohydrates was increased (Miller, 1938), but this absorption was directly into the blood stream and therefore irrelevant to the present study. Hurley (1983), however, noted that during healing “the bulk of the fluid and all the protein of the exudate is removed via lymphatic vessels.” His statement that “the behaviour of these vessels in inflammation is merely an amis very pertinent to the enhancement of plification of their normal function” absorption reported here. We were fortunate in that the initial direction of our work concerned the absorption of lymphoid cells, the delicacy of which impelled us to use a phlogogenic agent that left no toxic residue, a concentration and timing that might avoid drastic changes in the peritoneal cavity, and a volume that would leave no part of the cavity untouched. What is the mechanism by which NaOCl pretreatment causes enhancement of absorption into the lymphatics? Clearly, the function of the greater omentum was impaired by the chemical peritonitis. The inflamed omentum was no longer able to scavenge dye particles and large areas of its surface had lost the ability to hold oil drops. Surface fibrosis, loss or damage to the mesothelial cells, decreased surface area, and fixation to one small part of the abdominal cavity were all obvious contributors to loss of function. Decreased or retarded migration of blood monocytes through the surface fibrosis may have occurred also. When the omentum was unaffected by NaOCl (volume of pretreatment too small or use of head-up position), absorption was not enhanced. We recognize that contiguous areas of peritoneum may share some properties of the omentum. In any event, the loss of omental function removed the major obstacle between the injected inoculum and the diaphragmatic lymphatics, witness the accumulation of dye particles or oil on the abdominal surface of the diaphragm after NaOCl. The diaphragmatic lymphatics are known to be easily penetrated, (French et al., 1960) but the magnitude of the lymphatic distention by oil was astounding. Therefore, it seems possible that NaOCl had a second, and more direct, effect on the diaphragm, namely an increase in the number and/or permeability of its lymphatics. It is well known that lymphatics regenerate rapidly after injury (CasleySmith, 1973; McMaster and Hudack, 1934; Holub et a/., 1979), achieving supernormal numbers, and the increase of lymphatic channels was usually obvious in our material. However, it is difficult to prove the functional significance of this direct NaOCl effect in the presence of the shift of inoculum caused by the NaOCl inactivation of the omentum. In other studies, we have found that NaOCl enhanced the absorption into the lymphatic system of certain aqueous inocula and of lymphoid cells involved in cellular immunological reactions (experimental allergic encephalomyelitis, graft-

134

SEYMOUR

LEVINE

versus-host disease, host-versus-graft reaction) (Levine, 1985). Inasmuch as the prior establishment of a chemical peritonitis enhanced the effects of intraperitoneal inocula of particulate, oily, aqueous and cellular materials, it is likely to be a useful method in many areas of experimental pathology. The effects of augmented absorption of antigens, toxic materials, and tumor cells will probably be of considerable interest. ACKNOWLEDGMENTS We thank Richard manuscript preparation.

Sowinski

for

assistance

in the experiments

and

Ms.

Lorraine

Ostrubak

for

REFERENCES BANGHAM, A. D., MAGEE, P. N., and OSBORN, S. B. (1953). The effect of inflammation and other factors on the movement of radioactive glass particles from the peritoneal cavity. Brif. J. Exp. Pathal. 34, I- 11. BOLTON, C. (1921). Absorption from the peritoneal cavity. J. P&ho/. Bacterial. 24, 429-445. CASLEY-SMITH. J. R. (1973). The lymphatic system in inflammation. In “The Inflammatory Process” (B. W. Zweifach, L. Grant, and R. T. McCluskey, eds.). Vol. 2, pp. 161-205. Academic Press, New York. FONTAGNE, J., TIMSIT, J., GIROUD, J. I?, and LECHAT, P. (1973). Proprietes inflammatoires et antiinflammatoires de diverse substances oxydantes. Arch. Int. Pharmacodyn. Ther. 206, 242-252. FRENCH, J. E., FLOREY, H. W., and MORRIS, B. (1960). The absorption of particles by the lymphatics of the diaphragm. Quart. J. Exp. Physiol. 45, 88-103. HOLUB, M., JAROSKOVA. L., FISCHER, H., and VIKLICKY. V. (1979). Neoformation of lymphatics in the mouse omentum. Adv. Exp. Med. Biol. 114, 427-432. HURLEY, J. V. (1983). “Acute Inflammation,” 2nd ed., pp. 109- 117. Churchill Livingstone, Edinburgh. LEVINE, S. (1985). Post-inflammatory increase of pathogenicity of lymphocytic inocula in the peritoneal cavity. Proc. Sot. Exp. Biol. Med. 178, 462-467. LEVINE, S.. and SOWINSKI, R. (1970). Carbonyl iron: A new adjuvant for experimental autoimmune diseases. J. Zmmunol. 105, 1530-1535. LEVINE, S.. and SOWINSKI, R. (1982). Macrophages in the omentum: effects of drugs and pertussis vaccine evaluated by a simple weight assay. Exp. Mol. Pathol. 37, 208-216. LITTERST, C. L.. TORRES, I. J., ARNOLD, S., MCGUNAGLE, D., FURNER, R., SIKIC, B. I., and GUARINO, A. M. (1982). Absorption of antineoplastic drugs following large-volume IP administration to rats. Cancer Treut. Rep. 66, 147-155. MCMASTER, P. D., and HUDACK. S. S. (1934). The participation of skin lymphatics in repair of the lesions due to incisions and burns. J. Exp. Med. 60, 479-499. MILLER, R. G. (1938). The influence of inflammation on the absorption of substances of varied diffusibility. J. Exp. Med. 67, 619-641. STEELE, R. H., and WILHELM, D. L. (1967). The inflammatory reaction in chemical injury. II. Vascular permeability changes and necrosis induced by intracutaneous injection of various chemicals. Brit. .I. Exp. Pathol. 48, 592-607.