Diet, amyloid enhancing factor (AEF) and amyloidogenesis: an hypothesis.

At least two forms of amyloidosis, amyloid A (AA) and prion protein (PrP), can be transmitted by dietary ingestion of an agent(s) present in crude mammalian tissues. Although the incubation time for PrP or scrapie-induced diseases to develop in experimental animals extends over months or years, AA or secondary amyloidosis in mice is inducible within a week. In response to inflammatory stimuli we hypothesize that dietary factor(s) modulate the rate at which beta-pleated sheet fibrils accumulate in most forms of amyloidosis. The critical importance of precursor protein polymorphism, cell surface proteoglycans (PG), lipids and apolipoprotein metabolism has also been addressed in this hypothesis.

The acute phase response in apolipoprotein A-1 knockout mice: apolipoprotein serum amyloid A and lipid distribution in plasma high density lipoproteins.

In plasma, the bulk of apoSAA, a positive acute phase reactant protein, is transported in high density lipoproteins (HDL), especially HDLH (apoA1-rich HDL). In this study we tested whether apoA1 deficiency would adversely affect apoSAA concentration and lipid distribution in mouse plasma lipoproteins. Acute phase response (APR) was induced in C57BL/6J (apoA1+/+) and apoA1-knockout mice (apoA1-/-) by a subcutaneous injection of silver nitrate. The APR increased cholesterol concentrations in LDL of apoA1-/- mice and apoA1+/+ mice in a like manner. In contrast to apoA1+/+ mice, concentrations of cholesterol, phospholipids and proteins in both HDLL (1.063

A unique amyloidogenic apolipoprotein serum amyloid A (apoSAA) isoform expressed by the amyloid resistant CE/J mouse strain exhibits higher affinity for macrophages than apoSAA1 and apoSAA2 expressed by amyloid susceptible CBA/J mice.

CBA/J and other inbred strains of mice that express the amyloidogenic apolipoprotein serum amyloid A (apoSAA) apoSAA2, together with apoSAA1, are susceptible to amyloid A (AA) amyloidosis, whereas CE/J mice that express a single unique isoform, apoSAACEJ, are resistant. Studies indicate that CBA/JxCE/J hybrid mice that express apoSAA2 in the presence of apoSAACEJ are protected from amyloidogenesis. To define a mechanism by which expression of apoSAACEJ may protect from AA formation in the presence of apoSAA2, binding of recombinant apoSAA (r-apoSAA) isoforms, validated by N-terminal sequencing, to a murine macrophage cell line was investigated. Maximal specific binding occurred after incubation of radiolabeled apoSAA with IC-21 macrophages (1x105 cells/ml) for 30 min at 4 degreesC. The binding of 125I-r-apoSAA1, 125I-r-apoSAA2 and 125I-r-apoSAACEJ was specific and saturable, with an affinity (Kd) of about 2.8, 3.2 and 1.3 nM, respectively, and approximately 2-4x106 sites per cell. Competitive binding experiments indicate apoSAACEJ binds with higher affinity to macrophages than does either apoSAA1 or apoSAA2. We suggest that greater cellular affinity of apoSAACEJ compared to apoSAA2 may contribute to protection from AA amyloid in certain CBA/JxCE/J hybrid mice by interfering with interaction of apoSAA2 by macrophages and hence either membrane associated or intracellular degradation.

Catabolism of lipid-free recombinant apolipoprotein serum amyloid A by mouse macrophages in vitro results in removal of the amyloid fibril-forming amino terminus.

Serum amyloid A fibrils are formed when the normally rapid catabolism of the acute-phase reactant apolipoprotein serum amyloid A (apoSAA) is incomplete; thus amyloidosis may be viewed as a condition of dysregulated proteolysis. There is evidence that apoSAA is dissociated from plasma high-density lipoprotein (HDL) prior to fibril formation. The objective of this study was to investigate degradation of lipid-free apoSAA by tissue macrophages derived from amyloid-susceptible CBA/J mice in vitro. Peritoneal macrophages derived from untreated (normal) mice converted apoSAA (12 kDa) to a single 4 kDa C-terminal peptide while splenic macrophages converted apoSAA to 10, 7 and 4 kDa C-terminal peptides and a 4 kDa peptide that lacked the C- and N-terminal regions. Similar patterns of proteolysis occurred when peritoneal and splenic macrophages from amyloidotic CBA/J mice were used. Conditioned medium prepared from peritoneal, but not splenic macrophages, degraded apoSAA. Specific sites of cleavage indicated activity of cathepsin G- and elastase-like neutral proteases. The data indicate that lipid-free apoSAA can be degraded by secreted or cell-associated neutral proteases that are generated by macrophages to yield peptides that lack fibrillogenic potential.

Amyloid enhancing factor and dietary transmission in accelerated amyloid A amyloidosis.

The etiology and pathogenesis of amyloid A (AA) amyloidosis that may occur as an occasional complication of chronic inflammatory and infectious diseases are poorly understood. The preamyloid phase of experimentally induced AA amyloidosis can be greatly shortened in recipient animals by intravenous or intraperitoneal transfer of amyloid enhancing factor (AEF) when there is a concomitant inflammatory episode. AEF is an operational term applied to poorly characterized tissue extracts and increased AEF activity that precedes amyloid deposition. We now report that AA is rapidly formed in mice following oral administration of an AEF preparation that does not contain AA peptides. This finding indicates that a transmissible agent present in diet may be a contributory factor in amyloid fibril formation.

Polyarthritis and periostitis induced by Escherichia coli. Lipopolysaccharide injection in young male hamsters.

OBJECTIVE: To describe the clinicopathological manifestations of lipopolysaccharide (LPS) induced arthritis in the hamster and to compare its time of onset, duration, and severity with other forms of experimentally induced arthritis. METHODS: A preparation containing 30 microg LPS from Escherichia coli was injected subcutaneously for 5 to 21 days into young male hamsters (Mesocricetus auratus). Arthritis was quantified by measuring soft tissue swelling of affected joints with calipers. After decalcification, paraffin sections were cut and stained with hematoxylin and eosin, Giemsa, and azan. Acute phase reactant apolipoprotein serum amyloid A (apoSAA) levels were determined by ELISA. RESULTS: Symmetrical polyarthritis developed within 3 days and persisted for 14-21 days, provided the hamsters received daily LPS injections. Most prominent were lesions in the carpal-metacarpal joints of the front legs and in the tarsal-metatarsal joints of the hind legs. Animals in whom LPS injections were discontinued after 4 or 7 days recovered completely. Histological findings of exudative synovitis, periarticular soft tissue swelling, and juxtaarticular periostitis were associated with a sharp rise in serum titers of apoSAA. CONCLUSION: The unusually rapid onset of arthritis and periostitis in this experimental animal model suggests that its systemic manifestations were not mediated by a classical immune response, and may represent an "innate" response of targeted cells within the synovial membrane and periosteum to bacterial cell wall endotoxins.

In vitro amyloid fibril formation by synthetic peptides corresponding to the amino terminus of apoSAA isoforms from amyloid-susceptible and amyloid-resistant mice.

Specific proteins of the apolipoprotein serum amyloid (apoSAA) family that are synthesized in large quantities during the acute, early phase of inflammation can serve as the proteinaceous precursors for amyloid fibrils. To model fibrillogenesis in such inflammatory diseases, we have used electron microscopy and X-ray diffraction to examine the structures formed by synthetic peptides corresponding in sequence to the 11 amino-terminal amino acids of murine apoSAA1, apoSAAcej, and apoSAA2 and to the 15 amino-terminal amino acids of apoSAA2. This region is reported to be the major fibrillogenic determinant of apoSAA isoforms. Both in 1 mM Tris buffer and in 35% acetonitrile, 0.1% trifluoracetic acid (ACN/TFA), all of the peptides formed macromolecular assemblies consisting of twisted, approximately 40- to 60-A-thick ribbons, which varied in width from around 40-70 A (for 11-mer apoSAA2 in Tris) up to 900 A (for the other peptides). X-ray diffraction patterns recorded from lyophilized peptides, vapor-hydrated samples, and solubilized/dried samples showed hydrogen bonding and intersheet reflections typical of a beta-pleated sheet conformation. The coherent lengths measured from the breadths of the X-ray reflections indicated that with hydration the growth of the assemblies in the intersheet stacking direction was comparable to that in the hydrogen-bonding direction, and analysis of oriented samples showed that the beta-strands were oriented perpendicular to both the long axis and the face of the assemblies. These X-ray results are consistent with the ribbon- or plate-like morphology of the individual aggregates and emphasize the polymorphic nature of amyloidogenic peptides. Our findings demonstrate that X-ray diffraction measurements on vapor-hydrated or solubilized/dried versus lyophilized, amyloidogenic peptides are a good indicator of their fibrillogenic potential. For example, from the highest to the lowest potential, the peptides examined here were ranked as: Abeta1-28 > Abeta1-40 > apoSAA1 approximately apoSAAcej > apoSAA2 > Abeta17-42. Experiments in which the three different 11-mer apoSAA isoforms were solubilized in ACN/TFA and then combined as binary mixtures showed that the ribbon morphology was not affected but that the extent of hydrogen bonding in the assemblies was substantially reduced. Our observations on the in vitro assembly of apoSAA analogs emphasize that amyloid fibril formation and morphology depend on primary sequence, length of polypeptide chain, the presence of additional fibrillogenic polypeptides, and solvent conditions.

Apolipoprotein E and apolipoprotein A-1 knock-out mice readily develop amyloid A protein amyloidosis.

Studies have identified apolipoprotein E (apoE) ubiquitously in biochemically distinct amyloid deposits including amyloid A protein (AA) in secondary amyloidosis and amyloid beta protein (A beta) amyloid in Alzheimer's disease (AD). Apolipoprotein A-1 (apoA-1) has been identified in cortical plaques derived from the tissues of patients with AD. To determine if apoE is essential for and apoA-1 may be a factor in AA-amyloidogenesis we investigated induction of secondary amyloidosis in mutant C57BL/6J mice that lack either apoE or apoA-1. Induction of secondary amyloidosis in nonmutant C57BL/6J mice that are AA amyloid-susceptible were the AA positive control. Discreet deposits of AA amyloid were detected in the perifollicular regions of spleens derived from mutant and nonmutant strains. The findings clearly demonstrate that generation of AA fibrils can occur independently of apoE and ApoA-1 expression.

Polymorphism of acute-phase serum amyloid A isoforms and amyloid resistance in wild-type Mus musculus czech.

Until CE/J mice and their offspring were characterized as amyloid-resistant, all mice were thought to be amyloid-susceptible to multiple injections of azocasein or a single injection of silver nitrate following administration of amyloid enhancing factor. We now report, for the first time, that wild-type Mus musculus czech and F1 hybrids bred by crossing M. musculus czech with amyloid-susceptible CBA/J mice are also amyloid resistant. Based on the derived amino acid sequences of two serum amyloid A (SAA) cDNA clones, we describe two unusual SAA gene isoforms in M. musculus czech, one of which differs from four previously characterized acute-phase apoSAA isoforms at several amino acid residues. Our findings support the hypothesis that protection against amyloid fibril formation in wild-type M. musculus czech mice and their offspring is linked to apoSAA gene mutations (molecular motif).

Degradation of serum amyloid A in amyloid-susceptible and amyloid-resistant mouse strains.

Degradation of serum amyloid A (apoSAA) by resident peritoneal cells (RPCS) and conditioned medium (CDM), prepared with RPCS, from amyloid-susceptible CBA/J mice, amyloid-resistant CE/J mice and their amyloid-resistant CBA/J x CE/J F1 progeny was investigated in vitro. Serum amyloid A was derived from murine acute phase (AP) plasma and associated with high density lipoprotein (HDL). Degradation of apoSAA by RPCS and CDM from CBA/J mice was complete while degradation by RPCS and CDM from CE/J mice did not occur. Degradation of apoSAA by RPCS and CDM from CBA/J x CE/J F1 hybrid mice was indistinguishable from that by RPCS and CDM from the CBA/J parent. Intermediate fragments were not detected with either RPCS or CDM from CBA/J mice or CBA/J x CE/J F1 hybrid mice. Degradation of apoSAA was inhibited by phenylmethanylsulfonyl fluoride (PMSF) indicating that the enzyme, secreted into the fluid phase, was a serine esterase. Unlike apoSAA, HDL-associated apoA-1 remained intact. It was thus concluded that while selective degradation of HDL-associated apoSAA (apoSAA-HDL) by RPCS from the CBA/J and CE/J mice was significantly different, the genetic study did not support the hypothesis that there was direct linkage between impaired degradation of apoSAA-HDL in the CE/J mouse strain and protection against amyloid fibril formation. As amyloid resistance in CBA/J x CE/J F1 hybrid mice is not attributable to failure to express the amyloidogenic isoform apoSAA2, the study supports the original hypothesis that amyloid resistance may be linked to expression of apoSAAcej.

Amyloid enhancing factor is produced by rats and amyloid-resistant CE/J mice.

Amyloid enhancing factor (AEF) is an operational term applied to poorly defined extracts of amyloidotic or preamyloidotic tissues capable of shortening the induction time of amyloid deposition in recipient mice from 1 to 2 weeks to 48 to 72 hours. Its derivation has always left open the question of whether activity was dependent on the presence of amyloid fibrils or preamyloid fibril fragments. In these studies, we have assayed AEF activity in extracts of spleen and liver from azocasein-injected rats and CE/J mice that do not develop amyloidosis and, hence, cannot have amyloid A (AA) fibrils or fibril fragments in their tissues. Susceptibility to amyloid induction was compared in three strains of mice and three strains of rats by subjecting each group of experimental animals to multiple injections of azocasein. Spleens and livers were removed 24 hours after the last injection, and samples of all tissues were examined for amyloid deposits. AEF was extracted from the remainder of the tissues taken from amyloid resistant CE/J mice and Sprague-Dawley rats. Graded doses of the resulting tissue extract were given to naive Swiss-Webster (SW) recipient mice by i.p. injection concomitantly with subcutaneous injection of 0.5 ml 2% AgNO3. All tissues from both CE/J mice and rat donor animals were negative for amyloid by histologic examination of Congo Red stained samples, as were the AEF extracts. All recipient mice (six of six) given 600 micrograms of the CE/J-derived AEF developed large amyloid deposits in their spleens (mean severity 3.7 -/+ 0.3 SEM). Lower doses (200 micrograms protein) resulted in similar incidence of amyloid accumulation (in four of five), but quantitatively smaller amounts of amyloid protein were present. Doses of 100 micrograms decreased incidence (in one of five), whereas animals receiving 50 micrograms were all negative. AEF derived from rat tissues also induced high incidence of amyloid (in four of five) at high doses, although the amount of AA protein was less than in mice given equivalent amounts of CE/J mouse-derived AEF. Although 200 micrograms and 100 micrograms of rat AEF was effective (in two of five and one of five, respectively), 50 micrograms did not result in demonstrable amyloid deposition. The presence of AEF in tissues from azocasein-treated amyloid-resistant rats and CE/J mice excludes the possibility that AEF activity may be due to the presence of amyloid fibrils or fibril fragments in the donor tissue.

Linkage of protection against amyloid fibril formation in the mouse to a single, autosomal dominant gene.

Inbred strains of mice provide a model for studies of the pathogenesis of amyloid A (AA) amyloidosis. All susceptible strains of mice described to date codominantly express two serum amyloid A (apoSAA) isoforms, apoSAA1 and apoSAA2, of which only apoSAA2 serves as a precursor for amyloid fibrils. In previous studies, we have shown that the CE/J strain, which produces a single, novel apoSAA isoform, apoSAACE/J, is amyloid resistant. In the present study amyloid-resistant CE/J females were mated with amyloid-susceptible CBA/J males to produce F1 hybrid offspring which were then backcrossed to the parental CBA/J mouse strain. Amyloid susceptibility was determined in 30 backcrossed mice 72 h after injection of murine amyloid enhancing factor and silver nitrate. ApoSAA isoforms in plasma were separated by isoelectric focusing gel electrophoresis and visualized after immunoblotting with anti-AA antiserum. Amyloid A fibrils in spleen homogenates were denatured by formic acid and AA protein was quantified by ELISA using anti-mouse apoSAA antibodies. Values < 5 apoSAA equivalent units were considered negative. 13 mice expressed an apoSAA1 and apoSAA2 doublet characteristic of CBA/J mice, whereas 17 mice, expressed the apoSAACE/J isoform codominantly with apoSAA1 and apoSAA2. The correlation of amyloid resistance to expression of the apoSAACE/J isoform was absolute (17/17 were negative; mean score 2.6 +/- 0.17 [standard error of the mean] apoSAA equivalent units) and the correlation between amyloid susceptibility and the expression of apoSAA2/apoSAA1 was also striking (12/13 were amyloid positive; mean score 47.9 +/- 9.0 [standard error of the mean] apoSAA equivalent units (P < 0.001). This is not significantly different from the 50% segregation of apoSAA phenotypes expected for linkage to a single gene. These results indicate that a single gene governs apoSAACE/J expression and thus confers protection against amyloid deposition even in the presence of apoSAA1 and apoSAA2 isoforms and show for the first time that resistance to AA amyloidosis is a dominant trait governed by a single gene.

Characterization of the inbred CE/J mouse strain as amyloid resistant.

Inbred CE/J mice have been identified as extremely resistant to azocasein-induced amyloidosis relative to five commonly used inbred strains, A/J, CBA/J, C57BL/6J, C3H/HeN, and SJL/J. The enhanced amyloid resistance in CE/J mice seems to derive from the novel structure of the SAA gene family in CE/J mice, as determined by Southern blot hybridization analysis of SAA gene structure and isoelectric focusing analysis of acute phase SAA proteins in the six inbred strains. In CE/J mice, a single, novel SAA isoform of pI 6.15 is present, whereas in the other strains the amyloidogenic SAA2 isoform (pI 6.3) is codominantly expressed with SAA1 (pI 6.45). Two other inbred strains, PERU and IS/CAM, share common SAA specific HindIII DNA fragments with CE/J mice. Wild-derived Mus musculus mice differ from all of the inbred strains studied, both in SAA gene structure and in the pattern of SAA isoform production; two isoforms, one pI 6.15 and the other pI 6.3 (corresponding to SAA2), were codominantly expressed. Only the pI 6.15 isoform, not SAA1 and 2, was produced by CE/J mice in response to lipopolysaccharide, casein, silver nitrate, interleukin-1, or tumor necrosis factor; tumor necrosis factor was a weaker stimulus than interleukin-1 for the pI 6.15 isoform as it is for SAA1 and 2 production in the other inbred strains. This study provides a new line of evidence supporting the role of precursor structure as a determining factor in murine amyloid A amyloidosis and provides a valuable model for studies of amyloidogenesis.

Fish oil fatty acids and experimental arthritis.

Maintenance of mice on dietary regimens containing fish oil decreases severity of collagen-induced arthritis. Macrophages from fish oil fed animals had decreased omega-6 and significant amounts of omega-3 polyunsaturated fatty acids in membrane phospholipids and produced significantly less prostaglandins than macrophages from corn oil fed animals. Gender differences in both prostaglandin production and susceptibility to arthritis were noted.

Dietary n-3 fatty acids and arthritis.

We have evidence that dietary fish oil (FO) decreases severity of collagen-induced arthritis (CIA), changes the fatty acid composition of macrophage (M phi) membrane phospholipids, decreases M phi synthesis of prostaglandins (PGs), changes chemotactic ability of M phi s, and affects metabolism of acute phase proteins. Gender also has pronounced effects on susceptibility to CIA and M phi prostaglandin profiles. The mechanisms by which dietary n-3 fatty acids may act to alleviate symptoms of CIA, as well as interactions of dietary n-3 and n-6 fatty acids and gender are discussed. We suggest that the ability of FO diets to influence favourably the course of chronic inflammatory diseases is mediated via alterations in n-6 fatty acid metabolism and that intrinsic differences in n-6 fatty acid metabolism may account not only for our reported gender differences in incidence and severity of CIA, but also the well-documented sexual dimorphism in immune/inflammatory responses in general.

A fish oil diet reduces the severity of collagen induced arthritis after onset of the disease.

We have previously reported that compared to a corn oil diet a fish oil diet (5% by weight) fed to B10R.III mice before the induction of collagen induced arthritis markedly reduced disease severity. In this study we determine whether a fish oil diet could reduce the severity of collagen induced arthritis if begun after the arthritis was clinically apparent. Mice were initially fed either a fish oil or corn oil diet and immunized with bovine type II collagen 4 weeks later. At the onset of collagen-induced arthritis, half of the corn oil fed mice were switched to fish oil and arthritis assessed on a weekly basis. Four weeks after the diet change until killing 5 weeks later, the mice switched to fish oil developed much less severe arthritis than the corn oil fed controls. Thus the severity index of corn oil fed mice ranged between 9.4 and 7.1; the severity index of fish oil fed mice was between 6.8 and 4.3 while the mice switched to fish oil ranged between 7.2 and 5.6. Analysis of peritoneal macrophages 13 weeks after immunization showed that macrophages from fish oil fed mice incorporated eicosapentaenoic acid into phospholipids and produced less arachidonate products than corn oil fed mice. There was no difference between macrophages obtained from mice switched from corn oil to fish oil and those maintained on fish oil with respect to fatty acid composition of membrane phospholipids or prostaglandin profile. These results suggest that arthritis severity may be modulated after the onset of CIA by altering the PG profile of macrophages present at inflammatory sites.

Dietary fish oil modulation of macrophage amyloid P component responses in mice.

Arthritis-susceptible B10.RIII mice, maintained on either fish oil (FO) or corn oil (CO) diets (5% by weight), and amyloid-susceptible CBA/J mice fed chow diets were given 20 micrograms purified LPS by i.p. injection. Both strains of mice responded to LPS with a 20- to 30-fold increase in plasma amyloid P component (AP) levels. There were no differences in the response between males and females or between FO and CO treatment groups. The data demonstrated that cultured peritoneal macrophages (M phi) respond to LPS stimulation with increased secretion of AP. In contrast to plasma AP levels, the MO response to LPS stimulation, as measured by production of AP, was influenced by both gender and diet. Although M phi from both male and female mice on the CO diet and male mice on the FO diet responded similarly, those from female mice on the FO diet secreted only 25 to 35% as much AP as did the other three groups. There were no dietary effects on the LPS-induced serum amyloid A protein response nor was there detectable serum amyloid A protein produced by the M phi. These results demonstrate that unstimulated, resident peritoneal M phi secrete AP as a normal constituent and in increasing amounts in response to LPS stimulation.

A fish oil diet retards experimental amyloidosis, modulates lymphocyte function, and decreases macrophage arachidonate metabolism in mice.

Several recent reports have shown that diets in which the only source of fat was fish oil can modify the course of diseases with an inflammatory or immune component. In these experiments we examined the effect of a fish oil diet on experimental amyloidosis in mice. In most azocasein-treated mice, amyloid deposits were found in the spleen, varying from a trace to wide and contiguous perifollicular bands. The spleens of mice receiving fish oil had significantly less amyloid than did spleens of mice fed corn oil. A marked increase in spontaneous blastogenesis that occurred in azocasein-treated mice on corn oil was suppressed in azocasein-treated mice on fish oil. Substitution of the unsaturated fatty acids of corn oil with those of fish oil was also found to modify the prostaglandin profile of macrophages. This altered profile may change cellular immune function and/or enhance the processing of serum amyloid A to retard the induction of secondary amyloidosis in mice.

A fish oil diet inhibits amyloid P component (AP) acute phase responses in arthritis susceptible mice.

Amyloid P component (AP) bears close homology with C-reactive protein and behaves as an acute phase reactant in the plasma of mice but not in man. Our aim was to determine whether AP is influenced by diet, gender, and arthritis severity in a murine model of arthritis. B10.RIII mice were segregated according to gender and diet at 8 wk of age: the source of fat was either corn oil, fish oil, or beef tallow (5% by weight). Four weeks later, each mouse was immunized with 100 micrograms fetal bovine type II collagen, and the incidence and severity of arthritis was noted at weekly intervals. AP was measured by competitive ELISA in plasma taken 5 wk and 15 wk after immunization. AP levels were less in fish oil fed males and females. Under all conditions tested AP levels of females were greater than in males. There was a negative correlation between AP levels and the severity of arthritis. We conclude from these data that although AP levels cannot be used as indices of arthritis severity, there are significant dietary and gender effects on AP concentrations as long as 15 wk after immunization with type II collagen.

Gender differences in eicosanoid production from macrophages of arthritis-susceptible mice.

Collagen-induced arthritis (CIA) in rodents is an experimental animal model that shares many clinical and pathologic findings with rheumatoid arthritis in man. Our previous findings suggested that the amelioration of CIA in mice by a fish oil diet was associated with macrophage accumulation and metabolism of eicosapentaenoic acid and a subsequently altered prostaglandin (PG) profile. In these experiments, we examined the role of gender and found that macrophages from female arthritis-susceptible B10.RIII or B10.G mice synthesized more PG and thromboxane than macrophages isolated from the males. Compared with males, female mice had higher circulating anti-type II collagen antibodies but were less likely to develop CIA. Females, especially those on a fish oil diet, developed a much less severe disease than the males. This supports our hypothesis that the type and/or amount of eicosanoid produced from the macrophage may alter the course of experimentally induced arthritis.