- Email: [email protected]
Studies on the origins of biliary immunoglobulins in rats
GASTROENTEROLOGY
1984:87:173- 9
Studies on the Origins of Biliary Immunoglobulins in Rats RONALD J. MANNING, PATRICIA G. WALKER, LISA CARTER, PATRICIA J. BARRINGTON, and GRAHAM D. F. JACKSON School of Microbiology, Australia
The University
of New South
This study was designed to investigate the origins of biliary immunoglobuiins in rats. Natural antibodies to species of Lactobacillus and Staphylococcus were detected in blood, bile, and lymph, with predominant activity being associated with immunoglobulin A in lymph and bile and with immunoglobuiins M and G in blood. Six hours after thoracic duct cannuIation, serum immunoglobulins G, M, and A had fallen to SO%, 65% and 40% respectively, of the preoperative levels, whereas the level of total and specific immunoglobulin A in bile was reduced to 25% with no significant decline of immunoglobulin G. Thoracic duct occlusion caused a similar selective reduction in biliary immunoglobulin A. It is calculated that 50% of thoracic duct lymph immunoglobulin A entering’ the blood is secreted in bile. Synthesis of immunoglobulins M, G, and A in the liver was established by the detection of antibody-secreting cells. It is proposed that the major part of biliary immunoglobulin A is derived from intestinal lymphoid tissues and that a portion of the remainder, as well as part of biliary immunoglobulins M and G, results from local synthesis. Although rat bile is rich in secretory immunoglobulin (Ig) A (l),its exact origin is not known. The bulk of evidence, however, suggests that the immediate source of biliary secretory IgA is polymeric IgA in the circulation which is actively and selectively transported from blood to bile by the liver (2-4). This process, mediated by membrane-bound vesicles (5,6).occurs after polymeric IgA complexes with
Wales,
Kensington,
New South
secretory component expressed on the surface of hepatocytes (7,8). Not surprisingly, the origins of biliary IgG and IgM have not been considered insofar as the low levels (1)could be presumed to arise by passive transfer from serum. IgA in the circulation appears to be derived from &A-secreting cells in the intestinal tract (9) stimulated by intestinal antigens including the resident microbial flora (10).It reaches the blood via the mesenteric and thoracic duct lymph (TDL) which is enriched in IgA (e.g., ll,l2). Furthermore, it has been shown that 50°b of TDL IgA injected intravenously into rats is rapidly transported to bile (13).
In addition, sources of biliary immunoglobulins other than TDL have been proposed. First, synthesis in the liver itself has been suggested (14,15) and in humans up to 50% of the IgA may be of local origin (16).Second, recent studies on the appearance of specific antibodies in rat bile after injection of antigen have revealed the involvement of other lymphoid tissues in the production of IgA (17,18)and a role for the spleen in the transient appearance of biliary IgM (19). The present study has examined aspects of the sources of natural biliary immunoglobulins in the rat, confirming TDL as the major source of IgA. The detection of antibody-forming cells in the liver suggests local synthesis may contribute to biliary immunoglobulins. Materials
and Methods
Experimental Received November 7, 1983. Accepted February 13, 1984. Address requests for reprints to: Dr. G. D. F. Jackson, School of Microbiology, University of New South Wales, P.O. Box 1, Kensington, N.S.W. 2033, Australia. This work was supported by a grant from the National Health and Medical Research Council of Australia. 0 1984 by the American Gastroenterological Association 0016-5085/84/$3.00
Wales,
Animals
Male and female rats of the outbred Melbourne University Albino strain weighing between 200 and 250 g were used. Abbreviations used in this paper: GALT, gut-associated thoracic duct lymph.
lymphoid
AFC. antibody-forming tissue: Ig, immunoglobulin:
cells; TDL,
174
MANNING
GASTROENTEROLOGY Vol. 87, No. 1
ET AL.
Collection
of Body
Serological
Fluids
Serum. One- to two-milliliter volumes of blood were obtained by puncture of a tail vein. Terminal samples were obtained from the systemic circulation of anesthetized rats by severing the axillary artery, and from the portal circulation by puncturing the portal vein near the hilum of the liver. Sera were prepared by clotting and centrifugation and then stored at -zo”C. Bile. Bile duct cannulations were performed by Lambert’s method (20). Bile was collected into tubes containing 1 mg of sodium azide and stored at -20°C. Lymph. The thoracic duct was cannulated in the abdomen above the cisterna chyli by the method of Bollman et al. (21) as modified by Gowans (22). Lymph was collected and stored in the same manner as bile.
Thoracic Ligations
Duct
Ligation
were performed
as described
Determination of immunoglobulin concentrations in serum, lymph, and bile. The IgG, IgA, and IgM content of samples was determined by the method of Mancini et al. (24) using the antisera described above and the standards as described by Kenrick and Cooper (25). Agglutination tests for S. aureus and Lactobacillus antibodies. Serial twofold dilutions of serum, bile, lymph, and separated fractions of each were made in O.l-ml volumes in Dreyer tubes. The bacterial suspension (0.1 ml) of S. aureus or Lacotbacillus was then added to each dilution. After 2 h at 37°C the tubes were left at room temperature overnight. The titers recorded refer to the reciprocal of the highest dilution in which agglutination was visible with the aid of a hand lens. Fractions containing both IgA and IgM were pretreated for 1 h at 37°C with an equal volume of 0.2 M 2-mercaptoethanol before titration for antibody by the above method.
by Lambert
(20).
Detection
Bacterial
Suspensions
Staphylococcus aureus (University of New South Wales type culture collection No. 0485) was recovered from the lyophilized state and maintained on nutrient agar (Oxoid, London). A species of Lactobacillus (University of New South Wales type culture collection No. RSl3), which had been isolated from the small intestine of rats used in this study, was kindly provided by Associate Professor A. Lee. Cultures were prepared by inoculating supplemented brain heart infusion broth (23). For agglutinating suspensions of both organisms, phenol (0.5 g) was added to loo-ml broth cultures which were incubated for 2 h at 37°C and stored at 4°C for 24 h. The organisms were centrifuged and washed twice with lOOml volumes of 0.5% phenol saline and then resuspended in this medium to a concentration of 10’ organisms per milliliter using Wellcome opacity tubes (Burroughs Wellcome and Co., United Kingdom) and stored at 4°C.
Gel Filtration
of Serum,
Lymph,
Methods
and
Bile
Immunoglobulins in the serum and lymph of rats were separated by gel filtration through a 90 x 2.5-cm column of Sephadex G-200 (Pharmacia, Uppsala, Sweden) eluting with 0.15 M phosphate-buffered saline, pH 7.2. The immunoglobulins in bile were separated on a column of Ultrogel AcAzz as previously described (1). Assessment of separation was made by immunoelectrophoresis using the antisera described below.
Antisera The preparations of rabbit antirat IgM and goat antirat IgA have been described previously (19). Rabbit antirat IgG was purchased from Miles-Yeda Limited [Miles Laboratories, Inc., Elkhart, Ind).
of Antibody-Producing
Cells
The staphylococcal protein A assay of Gronowicz et al. (26) was used to detect cells secreting immunoglobulins of each class. The antirat sera were those described above and used at concentrations that gave optimal numbers of plaques. Horse erythrocytes [Commonwealth Serum Laboratories (CSL), Melbourne, Australia] were sensitized with staphylococcal protein A (Pharmacia). Suspensions of lymphoid cells from blood, spleen, and liver were prepared as follows. Blood (5.0 ml) was drawn from the abdominal aorta into a heparinized syringe. After centrifugation, the white cell layer was removed, washed once in Hanks’ balanced salt solution (C.S.L.) and resuspended in 1 ml of the same medium. Individual spleens were teased through a fine wire mesh in 10 ml of Hanks’ solution. The cells were washed once, resuspended in 5 ml Hanks’ solution, and filtered through a plug of glass wool. In order to remove the majority of blood, livers were perfused with 10 ml of phosphate-buffered saline via the portal vein. They were then teased through the wire mesh and the cell suspension was centrifuged. The upper layer of white cells was removed, washed once in Hanks’ solution, resuspended in 1 ml of the same medium, and filtered through glass wool. The numbers of mononuclear cells in each suspension were counted in a hemocytometer. Viability of cells was assessed at >95%. The times of incubation were 1.5 h for the initial plating of cells, 1 h with antiserum, and 1 h with guinea pig complement (CSL).
Results Natural Normal
Antibodies Rats
in the Body
Fluids
of
The concentrations of IgA, IgG, and IgM were examined in the serum, bile, and TDL of 8 normal rats. The results, given in Table 1, indicate that bile and lymph are enriched in IgA as compared with serum. In contrast, the concentrations of IgG and IgM
July
Table
ORIGINS
1984
1. Concentrations
of lmmunoelobulins Ig
&A
‘I Represents
Duct Lvmoh of Normal
0.23 ? 0.01 12.5 ? 1.1 0.7 + 0.08 mean
value
Origin of Serum and Biliary Immunoglobulins The contribution made by the intestinal lymphoid tissues to serum and biliary immunoglobulin
Bile/Serum
Lvmuh 1.6 5 0.3 4.1 2 0.6 0.22 2 0.02
from 8 rats + standard
are far higher in serum than in either bile or TDL. Two further points are noted: first, the IgG to IgM ratios in serum and lymph are similar; and second, although the results are not shown here, the concentrations of IgA, IgG, and IgM in serum and bile changed by
175
Rats
Ratio
Serum
1.12 -t 0.1 0.21 -t 0.02 40.03
Ig = immunoglobulin.
A. G. and M in Bile. Serum. and Thoracic
IMMlJNOGLOBLJLINS
concentration” (mgiml)
Bile
IziG IN
OF BILIARY
LvmphiSerum
4.8 t 0.5 0.02 2 0.003
6.9 + 1.1 0.33 f 0.04 0.32 k 0.04
deviation
levels was examined by determining the concentrations of IgA, IgG, and IgM in serum or bile, or both, of normal rats after thoracic duct cannulation or ligation. First, the thoracic ducts of 6 normal rats were cannulated and the change in serum immunoglobulin levels was followed for 24 h (Figure 1). The concentrations of all three immunoglobulin classes fell rapidly over the first 6 h, the fall in IgA levels being most marked. After this time, however, the change in IgA and IgG levels was not significant. In a second experiment the common bile duct of 5 normal rats was cannulated to obtain a sample of normal bile. After 1 h of bile collection the thoracic duct was also cannulated and the immunoglobulin content of bile was examined for 6 h thereafter. The results, given in Figure 2, show that the concentration of IgA in bile declined by 75% over 5 h of continuous thoracic duct drainage. Thereafter the IgA level remained constant at 25% of its initial level. In contrast, the IgG concentration in bile altered by
Table
2. Natural Antibodies of Normal Rats Agglutination
in Serum, Lymph, and Bile
titer” against (log,)
Lactobacillus Serum Lymph Bile
8.5 r 0.1 8.3 zk 0.1 9.6 ? 0.2
Distributionb of antiS. aureus activity in Ig classes (%)
S. aureus
IgA
IgG
IgM
7.4 t 0.7 10.8 + 0.6 8.0 C 0.4
ND 94 99
60 3 Trace
30 3 -
Ig = immunoglobulin. ND = not determined as no fraction free from IgG was obtained. a Represents mean value from 8 rats + standard deviation. b Percent distribution calculated from titers of pooled fractions from gel filtration, in which a single class of immunoglobulin was shown to be present by immunoelectrophoresis.
176
MANNING
Table
GASTROENTEROLOGY
ET AL.
3. Natural
in the Serum and Bile of Rats After
Antibodies
Agglutination
titer against
No.
1
512 512 512 512 1024
2 3 4 5
t=o
128 128 128 128 256
Average change
Agglutination
1024 512 1024 1024 128
t=6h
160 320 160 640 320
iactobacillus
t=o
40 80 40 160 80
-
75% fall
1
Serum
t=o
t=6h
1024 512 512 1024 128
titer against
Bile
Serum t=6h
t=o
87, No.
Duct Cannulation
S. aweus
Bile Animal
Thoracic
Vol.
t=6h
320 320 320 640 320
320 320 320 640 320
75% fall
in titer
as that found in thoracic duct-cannulated animals (Figure 3). Also, it was found that after thoracic duct occlusion the IgA content of the portal circulation was significantly higher than that of the systemic circulation, whereas the IgG concentration was found to be similar in the two circulations (Table 4). In the normal animal and in those animals having thoracic duct cannulation, the IgA content of the portal and systemic circulations was found to be similar. Antibody
Synthesis
synthesis occurs in the liver and might contribute to bile. By use of the hemolytic plaque assay of Gronowicz et al. (26) the antibody-forming cells (AFC) in the spleen, liver, and blood of 6 normal rats were enumerated. The results presented in Table 5 demonstrate the occurrence of AFC in these tissues and, in particular, show the presence of a proportionally larger number of cells of all classes in the liver than in the other sites.
Discussion
in the Liver
The amount of IgA in bile not originating from TDL is still appreciable, being -50% above the normal level in serum. This, and the observation described above, that the concentration of biliary IgG did not decrease during the period of thoracic duct cannulation despite a marked reduction in serum levels, prompted the notion that immunoglobulin
Immunogenic stimulation by antigens in the environment, particularly within the intestinal tract, is considered to be responsible for the appearance and maintenance of natural antibodies present in body fluids. In most discussions, however, it is the serum IgM and IgG natural antibodies that are primarily considered, notwithstanding the fact that initial stimulation is at a mucosal surface and thus IgA antibodies might be expected to predominate. It is not surprising therefore that analysis of the natural
-.-v 4
\
6
12
18
f \
‘g*
i
-Y I
I
24
b*
I
after
Figure 1. Relative
thoracic
duct cannulation
concentrations rats after thoracic duct ulin.
I
I
4
2 Time
W
P-
I
6
(hours)
of IgM, IgG, and IgA in serum of cannulation. Ig = immunoglob-
Time Figure
after
thoracic
duct
cannulation
(hours)
2. Relative concentrations of IgG and IgA in the bile of rats after thoracic duct cannulation. Ig = immunoglobulin.
July 1984
ORIGINS
OF BILIARY
IMMlJNOGLOBlJLINS
Table 5. Antibody-Forming Cells in Blood, Spleen, and Liver of Normal Rats
lgG
No.” of AFCIlO”
W
lymphoid
cells
Tissue
IgM
IgG
IgA
Blood” Spleen Liver
299 t 214 50 * 36 2043 ? 655
243 + 180 172 + 132 3881 + 1556
253 t 175 188 + 124 4244 + 1981
AFC = antibody-forming cells. Ig = immunoglobulin. sents mean value from 6 rats t standard deviation. prepared from 5 ml of blood.
25
I
I
I
2
Time after
thoracic
I
I
I
4
bacterial activity in lymph and bile indicates the availability of the vehicle to transport antigen absorbed through mucosal surfaces, as has been proposed (28-30). The present studies show that when TDL is prevented from entering the blood, the level of total IgA in bile is reduced selectively, by 75%, in accord with the results of Dahlgren et al. (18), thus providing further evidence that the bulk of biliary IgA is derived from the GALT. This is in keeping with previous findings that, although lymph is rich in IgA, serum levels are relatively low (12) due to selective removal of IgA within the liver (2,3,7), and that, in the rat, intravenously injected IgA from lymph is rapidly removed from the circulation by biliary secretion (13). Alternatively, it has been suggested (5) that after synthesis, IgA directly enters the venous drainage of the gut and reaches the liver directly via the portal vein without entering the systemic circulation. Direct testing of this by thoracic duct ligation, thus preventing lymphatic drainage of the gut and favoring the venous pathway, showed that despite a significantly increased concentration of IgA in the portal circulation, this pathway was not sufficient to halt the decline in biliary IgA levels. Hence we propose that the venous pathway together with lymphovascular anastomosis, linking the mesenteric lymph duct and portal vein, are of minor importance in delivering intestinal IgA to the liver. From the concentrations of IgA in serum and TDL it can be calculated (Table 6) that -95% of TDL IgA is newly
(hours)
Relative concentrations of IgG and IgA in the bile of rats after thoracic duct ligation. Ig = immunoglobulin.
antibodies against two common bacterial species showed high amounts of IgA present in lymph and bile whereas activity in serum was predominantly associated with IgM and IgG fractions. The demonstration of the loss of specific antibacterial antibody activity from bile after thoracic duct cannulation indicates the origin of the majority of biliary IgA to be the GALT. Supporting evidence for this was recently found when it was shown that bile duct occlusion of normal rats caused an increase in levels of IgA natural antibodies against Vibrio cholerae (27) and Salmonella spp. (unpublished). Dahlgren et al. (18) have found that when rats were immunized in their Peyer’s patches, some did not show a decline in specific antibody in bile after thoracic duct cannulation. These rats, freshly immunized with a new antigen, had high numbers of IgA AFC in nonmucosal sites, suggesting that relocalization back to the submucosa of the gut had not been fully established as would be the situation of AFC directed against the ever-present normal flora. Nonetheless, one must still account for a significant amount of natural IgA antibacterial antibody in bile, other than that derived from lymph. With regard to function, the identification of antiTable 4. Immunoglobulin Levels in Portal and Systemic
Blood of Rats
IgA (mgiml)” Treatment
IgC (mg,iml)”
Portal
Systemic
P/S
Systemic
Nil (8)” (4) Bile duct cannulated’ + thoracic Bile duct cannulated’ duct cannulated (5)
0.347 0.243 0.119
2 0.09 + 0.03 + 0.03
0.343 0.231 0.109
t 0.08 2 0.02 t 0.01
0.99 + 0.03 0.95 f 0.03 0.92 r 0.02
11.20 2 2.03 9.54 t 3.54 6.57 2 1.63
+ thoracic
0.182
t 0.02
0.250
+ 0.04
1.37 r 0.03
6.47 + 0.78
Bile duct cannulated’ duct ligated (4)
” Repre“Cells
1
6
duct occlusion
177
Ig = immunoglobulin. S = systemic. P = portal. “ Represents of rats in group. ‘. Samples taken 6-7 h after cannulation.
mean
value
2 standard
deviation.
” Figures
Portal
PIS
t 1.55 f 3.57 + 1.24
0.94 t 0.16 1.12 t 0.30 0.97 r 0.13
6.Y2 f 0.63
1.07 * 0.07
10.45 10.64 6.39
in parentheses
indicate
number
178
Table
GASTROENTEROLOGY
MANNING ET AL.
6.
Relative Contributions to Jmmunoglobulins A and G of Thoracic Duct Lymph of Synthesis Gut-Associated Lymphoid Tissues and Transfer From Blood
Expected Expected % derived O/nderived
lymph lymph from from
to serum ratio” concentrationb (mgiml) blood (expected/observedb) GALT
in
&A
I&
0.386 0.09 6 94
0.4 5 100
Ig = immunoglobulin. GALT = gut-associated lymphoid tissues. a Empirical results from Vaerman et al. (31) assuming IgA to be entirely dimeric. b Based on values from Table 1.
synthesized, agreeing with previous findings (31). Further, based on the values of the flow rates of bile and lymph, the concentrations of IgA in these fluids, and the result that 75% of biliary IgA is derived from of lymph IgA arrivTDL, it is calculated that -50% ing in blood is subsequently removed by the liver. This finding is consistent with previous observations on IgA transport (2,13). The remainder not transported to bile could be accounted for by either of three possibilities or a combination of them. First, it may contribute to the serum pool of IgA. Second, it could be removed from the circulation at various mucosal sites (32,33), thereby augmenting the system of relocation of IgA plasma cells from one site to another (reviewed in Reference 34). Finally, the possibility that IgA is degraded within hepatocytes must be considered although the bulk of the evidence supports a nondegradative pathway (reviewed in Reference 35). A brief comment is needed upon the decline in immunoglobulin levels in the blood of rats after thoracic duct cannulation (Figure 1). An equilibrium is rapidly reached (4-6 h), which suggests that the immunoglobulins leaving the circulation are being replaced from other sources at approximately the same rate. The decline in IgA is greatest, as would be expected, because it is selectively removed in the liver. Further, with regard to IgM, the similar IgG to IgM ratios for serum and lymph confirm the results of others and indicate a contribution of locally synthesized IgM to the blood pool (11,36). The final matter to be addressed is immunoglobulin synthesis in the liver. We have demonstrated the presence of IgA-, IgG-, and IgM-secreting cells in the liver at levels that are distinguishable from those in blood and spleen, suggesting a predilection for these cells in this site. This is in keeping with previous reports (37,38) which have indicated the presence of AFC in the liver after immunization and the recent finding by Reichert et al. (39) that functional immunoblasts from Peyer’s patches, when reinjected, localize preferentially in the liver. Although synthesis has been shown to occur, this
Vol. 87, No. 1
does not necessarily corroborate the second part of the hypothesis that secretion to bile follows, although there would be little difficulty in accepting that dimeric IgA produced in the liver could be translocated. The trace amount of IgM in bile could well leak from the plasma, although we have recently reported that the transient increase in biliary IgM that occurs after immunization can be explained by the localization of IgM cells (19). IgG occurs in bile at about one-sixtieth the level in serum, yet after thoracic duct cannulation the fall in serum concentration was not paralleled in bile. One explanation is that the serum level is still sufficiently high to allow passive exudation at the same rate. Alternatively, the large number of IgG cells present in the liver could maintain the local concentration and so preserve the degree of output in bile. The origin of cells detected in the current study, their location, and whether or not they are sufficient to account for the levels of immunoglobulin in bile not originating from other sources remain to be resolved. Nevertheless, inasmuch as local synthesis of at least IgA has been claimed to be significant in humans (16), we suggest that further studies of cellular activity in the rat liver may prove useful.
References 1.
2.
3. 4.
5.
6.
7.
8.
9.
10.
Lemaitre-Coelho I, Jackson GDF, Vaerman J-P. Rat bile as a convenient source of secretory IgA and free secretory component. Eur J Immunol 1977;7:588-90. Jackson GDF, Lemaitre-Coelho I, Vaerman J-P, Bazin H, Beckers A. Rapid disappearance from serum of intravenously injected rat myeloma IgA and its secretion into bile. Eur J Immunol 1978;8:123-6. Orlans E, Peppard J, Reynolds J, Hall J. Rapid active transport of IgA from blood to bile. J Exp Med 1978;147:588-92. Vaerman J-P, Lemaitre-Coelho I, Jackson GDF. Role of the liver in the rat intestinal sIgA system. Adv Exp Med Biol 1978;107:233-40, Renston RH, Jones AL, Christiansen WD, Hradek GT, Underdown BJ. Evidence for a vesicular transport mechanism in hepatocytes for biliary secretion of immunoglobulin A. Science 1980;208:1276-8. Birbeck MSC, Cartwright P, Hall JG, Orlans E, Peppard J. The transport by hepatocytes of immunoglobulin A from blood to bile visualized by autoradiography and electron microscopy. Immunology 1979;37:477-84. Fisher MM, Nagy B, Bazin H, Underdown BJ. Biliary transport of IgA: role of secretory component. Proc Nat1 Acad Sci USA 1979;76:2008-12. Orlans E, Peppard J, Fry JF, Hinton RH, Mullock BM. Secretory component as the receptor for polymeric IgA on rat hepatocytes. J Exp Med 1979;150:1577-81. Crabbe PA, Carbonara AD, Heremans JF. The normal intestinal mucosa as a major source of plasma cells containing gamma-A-immunoglobulin. Lab Invest 1965;14:235-48. Crabbe PA, Bazin H, Eyssen H, Heremans JF. The normal microbial flora as a major stimulus for proliferation of plasma cells synthesizing IgA in the gut. Int Arch Allergy Appl Immunol 1968;34:362-75.
July 1984
11. Vaerman J-P, Heremans JF. Origin and molecular size of IgA in mesenteric lymph of the dog. Immunology 1970;18:27-38. 12. Kaartinen M, lmir T, Klockars M, Sandholm M, Makela 0. IgA in blood and thoracic duct lymph. Concentration and degree of polymerization. Stand J lmmunol 1978;7:229-32. 13. Reynolds J, Gyure L, Andrew E, Hall JG. Studies of the transport of polyclonal IgA antibody from blood to bile in rats. Immunology 1980;39:463-7. 14. Dive Ch, Heremans JF. Nature and origin of proteins of bile. I. A comparative analysis of serum and bile in man. Eur J Clin Invest 1974;4:235-9. 15. Dive Ch, Nadalini RA, Vaerman J-P, Heremans JF. Origin and nature of the proteins of bile. II. A comparative analysis of serum, hepatic lymph and bile proteins in the dog. Eur J Clin Invest 1974;4:241-6. 16. Delacroix DL, Hodgson HJF, McPherson A, Dive C, Vaerman J-P. Selective transport of polymeric immunoglobulin A in bile. J Clin Invest 1982;70:230-41. 17. Spencer J, Gyure LA, Hall JG. IgA antibodies in the bile of rats. III. The role of intrathoracic lymph nodes and the migration pattern of their blast cells. Immunology 1983;48:687-93. 18. Dahlgren U, Ahlstedt S, Andersson T, Hedman L, Hanson LA. IgA antibodies in rat bile are not solely derived from thoracic duct lymph. Stand J Immunol 1983;17:569-74. 19. Jackson GDF, Walker PG. The transient appearance of IgM antibodies in the bile of rats injected with Salmonella enteritidis. lmmunol Lett 1983;7:41-5. 20. Lambert R. Surgery of the digestive system in the rat. Springfield, Ill.: Charles C Thomas, 1965. 21. Bollman JL, Cain JC, Grindlay JH. Techniques for the collection of lymph from the liver, small intestine and thoracic duct of the rat. J Lab Clin Med 1948;33:1349-52. 22. Gowans JL. The recirculation of lymphocytes from blood to lymph in the rat. J Physiol (Lond) 1959;146:54-69. 23. Holdeman LV, Moore WEC. Anaerobe laboratory manual. 3rd ed. Blacksburg: Virginia Polytechnic Institute and State University Anaerobe Laboratory, 1975. 24. Mancini G, Carbonara AD, Heremans JF. lmmunochemical quantitation of antigens by single radial immunodiffusion. Immunochemistry 1965;2:235-54. 25. Kenrick KG, Cooper GN. Antibodies in the intestinal secretions of rats. Primary and secondary responses to polymeric flagellin. Aust J Exp Biol Med Sci 1978;56:441-53.
ORIGINS OF BILIARY IMMUNOGLOBULINS
179
26. Gronowicz E, Coutinho A, Melchers F. A plaque assay for all cells secreting Ig of a given type or class. Eur J Immunol 1976;6:588-90. 27. Jackson GDF, Cooper GN. Immune responses of rats to live Vibrio cholerae: Secretions of antibodies in bile. Parasite lmmunol 1981;3:127-35. 28. Russell MW, Brown TA, Mestecky J. Role of serum IgA. Hepatobiliary transport of circulating antigen. J Exp Med 1981;153:968-76. 29. Peppard J, Orlans E, Payne AWR, Andrew E. The elimination of circulating complexes containing polymeric IgA by excretion in the bile. Immunology 1981;127:83-9. 30. Socken DJ, Simms ES, Nagy BR, Fisher MM, Underdown BJ. Secretory component dependent hepatic transport of IgA antibody-antigen complexes. J Immunol 1981;127:316-9. 31. Vaerman J-P, Andre C, Bazin H, Heremans JF. Mesenteric lymph as a major source of serum IgA in guinea pigs and rats. Eur J lmmunol 1973;3:580-4. 32. Jackson GDF, Lemaitre-Coelho I, Vaerman J-P. Transfer of MOPC-315 IgA to secretions in MOPC-315 tumour bearing and normal Balb/c mice. Protides Biol Fluids 1977;25:919-22. 33. Virella G, Montgomery P, Lemaitre-Coelho I. Transport of oligomeric IgA of systemic origin into external secretions. Adv Exp Med Biol 1978;107:241-51. 34. Tomasi TB, Larson DO, Challacombe S, McNabb P. Mucosal immunity: the origin and migration patterns of cells in the secretory system. J Allergy Clin Immunol 1980;65:12-9. 35. Jones AL, Renston RH, Burwen SJ. Uptake and intracellular disposition of plasma-derived proteins and apoproteins by hepatocytes. Prog Liver Dis 1982;7:51-69. 36. Kaartinen M, Kosunen TV, Makela 0. Complement and immunoglobulin levels in the serum and thoracic duct lymph of the rat. Eur J Immunol 1973;3:556-9. 37. Garvey JS, Linker-Israeli M. Anti-SRBC plaque-forming cells in liver, spleen and hepatocyte cell suspensions of rabbits and mice. Immunol Commun 1972;1:507-21. 38. Hiramoto RN, Habeeb AFSA, Hamlin M. Antibody-forming cells in five organs of hyperimmunized guinea pigs. Proc Sot Exp Biol Med 1970;134:975-9. 39. Reichert RA, Gallatin WM, Weissman IL, Butcher EC. Germinal centre cells lack homing receptors necessary for normal lymphocyte recirculation. J Exp Med 1983:157:813-27.
1984:87:173- 9
Studies on the Origins of Biliary Immunoglobulins in Rats RONALD J. MANNING, PATRICIA G. WALKER, LISA CARTER, PATRICIA J. BARRINGTON, and GRAHAM D. F. JACKSON School of Microbiology, Australia
The University
of New South
This study was designed to investigate the origins of biliary immunoglobuiins in rats. Natural antibodies to species of Lactobacillus and Staphylococcus were detected in blood, bile, and lymph, with predominant activity being associated with immunoglobulin A in lymph and bile and with immunoglobuiins M and G in blood. Six hours after thoracic duct cannuIation, serum immunoglobulins G, M, and A had fallen to SO%, 65% and 40% respectively, of the preoperative levels, whereas the level of total and specific immunoglobulin A in bile was reduced to 25% with no significant decline of immunoglobulin G. Thoracic duct occlusion caused a similar selective reduction in biliary immunoglobulin A. It is calculated that 50% of thoracic duct lymph immunoglobulin A entering’ the blood is secreted in bile. Synthesis of immunoglobulins M, G, and A in the liver was established by the detection of antibody-secreting cells. It is proposed that the major part of biliary immunoglobulin A is derived from intestinal lymphoid tissues and that a portion of the remainder, as well as part of biliary immunoglobulins M and G, results from local synthesis. Although rat bile is rich in secretory immunoglobulin (Ig) A (l),its exact origin is not known. The bulk of evidence, however, suggests that the immediate source of biliary secretory IgA is polymeric IgA in the circulation which is actively and selectively transported from blood to bile by the liver (2-4). This process, mediated by membrane-bound vesicles (5,6).occurs after polymeric IgA complexes with
Wales,
Kensington,
New South
secretory component expressed on the surface of hepatocytes (7,8). Not surprisingly, the origins of biliary IgG and IgM have not been considered insofar as the low levels (1)could be presumed to arise by passive transfer from serum. IgA in the circulation appears to be derived from &A-secreting cells in the intestinal tract (9) stimulated by intestinal antigens including the resident microbial flora (10).It reaches the blood via the mesenteric and thoracic duct lymph (TDL) which is enriched in IgA (e.g., ll,l2). Furthermore, it has been shown that 50°b of TDL IgA injected intravenously into rats is rapidly transported to bile (13).
In addition, sources of biliary immunoglobulins other than TDL have been proposed. First, synthesis in the liver itself has been suggested (14,15) and in humans up to 50% of the IgA may be of local origin (16).Second, recent studies on the appearance of specific antibodies in rat bile after injection of antigen have revealed the involvement of other lymphoid tissues in the production of IgA (17,18)and a role for the spleen in the transient appearance of biliary IgM (19). The present study has examined aspects of the sources of natural biliary immunoglobulins in the rat, confirming TDL as the major source of IgA. The detection of antibody-forming cells in the liver suggests local synthesis may contribute to biliary immunoglobulins. Materials
and Methods
Experimental Received November 7, 1983. Accepted February 13, 1984. Address requests for reprints to: Dr. G. D. F. Jackson, School of Microbiology, University of New South Wales, P.O. Box 1, Kensington, N.S.W. 2033, Australia. This work was supported by a grant from the National Health and Medical Research Council of Australia. 0 1984 by the American Gastroenterological Association 0016-5085/84/$3.00
Wales,
Animals
Male and female rats of the outbred Melbourne University Albino strain weighing between 200 and 250 g were used. Abbreviations used in this paper: GALT, gut-associated thoracic duct lymph.
lymphoid
AFC. antibody-forming tissue: Ig, immunoglobulin:
cells; TDL,
174
MANNING
GASTROENTEROLOGY Vol. 87, No. 1
ET AL.
Collection
of Body
Serological
Fluids
Serum. One- to two-milliliter volumes of blood were obtained by puncture of a tail vein. Terminal samples were obtained from the systemic circulation of anesthetized rats by severing the axillary artery, and from the portal circulation by puncturing the portal vein near the hilum of the liver. Sera were prepared by clotting and centrifugation and then stored at -zo”C. Bile. Bile duct cannulations were performed by Lambert’s method (20). Bile was collected into tubes containing 1 mg of sodium azide and stored at -20°C. Lymph. The thoracic duct was cannulated in the abdomen above the cisterna chyli by the method of Bollman et al. (21) as modified by Gowans (22). Lymph was collected and stored in the same manner as bile.
Thoracic Ligations
Duct
Ligation
were performed
as described
Determination of immunoglobulin concentrations in serum, lymph, and bile. The IgG, IgA, and IgM content of samples was determined by the method of Mancini et al. (24) using the antisera described above and the standards as described by Kenrick and Cooper (25). Agglutination tests for S. aureus and Lactobacillus antibodies. Serial twofold dilutions of serum, bile, lymph, and separated fractions of each were made in O.l-ml volumes in Dreyer tubes. The bacterial suspension (0.1 ml) of S. aureus or Lacotbacillus was then added to each dilution. After 2 h at 37°C the tubes were left at room temperature overnight. The titers recorded refer to the reciprocal of the highest dilution in which agglutination was visible with the aid of a hand lens. Fractions containing both IgA and IgM were pretreated for 1 h at 37°C with an equal volume of 0.2 M 2-mercaptoethanol before titration for antibody by the above method.
by Lambert
(20).
Detection
Bacterial
Suspensions
Staphylococcus aureus (University of New South Wales type culture collection No. 0485) was recovered from the lyophilized state and maintained on nutrient agar (Oxoid, London). A species of Lactobacillus (University of New South Wales type culture collection No. RSl3), which had been isolated from the small intestine of rats used in this study, was kindly provided by Associate Professor A. Lee. Cultures were prepared by inoculating supplemented brain heart infusion broth (23). For agglutinating suspensions of both organisms, phenol (0.5 g) was added to loo-ml broth cultures which were incubated for 2 h at 37°C and stored at 4°C for 24 h. The organisms were centrifuged and washed twice with lOOml volumes of 0.5% phenol saline and then resuspended in this medium to a concentration of 10’ organisms per milliliter using Wellcome opacity tubes (Burroughs Wellcome and Co., United Kingdom) and stored at 4°C.
Gel Filtration
of Serum,
Lymph,
Methods
and
Bile
Immunoglobulins in the serum and lymph of rats were separated by gel filtration through a 90 x 2.5-cm column of Sephadex G-200 (Pharmacia, Uppsala, Sweden) eluting with 0.15 M phosphate-buffered saline, pH 7.2. The immunoglobulins in bile were separated on a column of Ultrogel AcAzz as previously described (1). Assessment of separation was made by immunoelectrophoresis using the antisera described below.
Antisera The preparations of rabbit antirat IgM and goat antirat IgA have been described previously (19). Rabbit antirat IgG was purchased from Miles-Yeda Limited [Miles Laboratories, Inc., Elkhart, Ind).
of Antibody-Producing
Cells
The staphylococcal protein A assay of Gronowicz et al. (26) was used to detect cells secreting immunoglobulins of each class. The antirat sera were those described above and used at concentrations that gave optimal numbers of plaques. Horse erythrocytes [Commonwealth Serum Laboratories (CSL), Melbourne, Australia] were sensitized with staphylococcal protein A (Pharmacia). Suspensions of lymphoid cells from blood, spleen, and liver were prepared as follows. Blood (5.0 ml) was drawn from the abdominal aorta into a heparinized syringe. After centrifugation, the white cell layer was removed, washed once in Hanks’ balanced salt solution (C.S.L.) and resuspended in 1 ml of the same medium. Individual spleens were teased through a fine wire mesh in 10 ml of Hanks’ solution. The cells were washed once, resuspended in 5 ml Hanks’ solution, and filtered through a plug of glass wool. In order to remove the majority of blood, livers were perfused with 10 ml of phosphate-buffered saline via the portal vein. They were then teased through the wire mesh and the cell suspension was centrifuged. The upper layer of white cells was removed, washed once in Hanks’ solution, resuspended in 1 ml of the same medium, and filtered through glass wool. The numbers of mononuclear cells in each suspension were counted in a hemocytometer. Viability of cells was assessed at >95%. The times of incubation were 1.5 h for the initial plating of cells, 1 h with antiserum, and 1 h with guinea pig complement (CSL).
Results Natural Normal
Antibodies Rats
in the Body
Fluids
of
The concentrations of IgA, IgG, and IgM were examined in the serum, bile, and TDL of 8 normal rats. The results, given in Table 1, indicate that bile and lymph are enriched in IgA as compared with serum. In contrast, the concentrations of IgG and IgM
July
Table
ORIGINS
1984
1. Concentrations
of lmmunoelobulins Ig
&A
‘I Represents
Duct Lvmoh of Normal
0.23 ? 0.01 12.5 ? 1.1 0.7 + 0.08 mean
value
Origin of Serum and Biliary Immunoglobulins The contribution made by the intestinal lymphoid tissues to serum and biliary immunoglobulin
Bile/Serum
Lvmuh 1.6 5 0.3 4.1 2 0.6 0.22 2 0.02
from 8 rats + standard
are far higher in serum than in either bile or TDL. Two further points are noted: first, the IgG to IgM ratios in serum and lymph are similar; and second, although the results are not shown here, the concentrations of IgA, IgG, and IgM in serum and bile changed by
175
Rats
Ratio
Serum
1.12 -t 0.1 0.21 -t 0.02 40.03
Ig = immunoglobulin.
A. G. and M in Bile. Serum. and Thoracic
IMMlJNOGLOBLJLINS
concentration” (mgiml)
Bile
IziG IN
OF BILIARY
LvmphiSerum
4.8 t 0.5 0.02 2 0.003
6.9 + 1.1 0.33 f 0.04 0.32 k 0.04
deviation
levels was examined by determining the concentrations of IgA, IgG, and IgM in serum or bile, or both, of normal rats after thoracic duct cannulation or ligation. First, the thoracic ducts of 6 normal rats were cannulated and the change in serum immunoglobulin levels was followed for 24 h (Figure 1). The concentrations of all three immunoglobulin classes fell rapidly over the first 6 h, the fall in IgA levels being most marked. After this time, however, the change in IgA and IgG levels was not significant. In a second experiment the common bile duct of 5 normal rats was cannulated to obtain a sample of normal bile. After 1 h of bile collection the thoracic duct was also cannulated and the immunoglobulin content of bile was examined for 6 h thereafter. The results, given in Figure 2, show that the concentration of IgA in bile declined by 75% over 5 h of continuous thoracic duct drainage. Thereafter the IgA level remained constant at 25% of its initial level. In contrast, the IgG concentration in bile altered by
Table
2. Natural Antibodies of Normal Rats Agglutination
in Serum, Lymph, and Bile
titer” against (log,)
Lactobacillus Serum Lymph Bile
8.5 r 0.1 8.3 zk 0.1 9.6 ? 0.2
Distributionb of antiS. aureus activity in Ig classes (%)
S. aureus
IgA
IgG
IgM
7.4 t 0.7 10.8 + 0.6 8.0 C 0.4
ND 94 99
60 3 Trace
30 3 -
Ig = immunoglobulin. ND = not determined as no fraction free from IgG was obtained. a Represents mean value from 8 rats + standard deviation. b Percent distribution calculated from titers of pooled fractions from gel filtration, in which a single class of immunoglobulin was shown to be present by immunoelectrophoresis.
176
MANNING
Table
GASTROENTEROLOGY
ET AL.
3. Natural
in the Serum and Bile of Rats After
Antibodies
Agglutination
titer against
No.
1
512 512 512 512 1024
2 3 4 5
t=o
128 128 128 128 256
Average change
Agglutination
1024 512 1024 1024 128
t=6h
160 320 160 640 320
iactobacillus
t=o
40 80 40 160 80
-
75% fall
1
Serum
t=o
t=6h
1024 512 512 1024 128
titer against
Bile
Serum t=6h
t=o
87, No.
Duct Cannulation
S. aweus
Bile Animal
Thoracic
Vol.
t=6h
320 320 320 640 320
320 320 320 640 320
75% fall
in titer
as that found in thoracic duct-cannulated animals (Figure 3). Also, it was found that after thoracic duct occlusion the IgA content of the portal circulation was significantly higher than that of the systemic circulation, whereas the IgG concentration was found to be similar in the two circulations (Table 4). In the normal animal and in those animals having thoracic duct cannulation, the IgA content of the portal and systemic circulations was found to be similar. Antibody
Synthesis
synthesis occurs in the liver and might contribute to bile. By use of the hemolytic plaque assay of Gronowicz et al. (26) the antibody-forming cells (AFC) in the spleen, liver, and blood of 6 normal rats were enumerated. The results presented in Table 5 demonstrate the occurrence of AFC in these tissues and, in particular, show the presence of a proportionally larger number of cells of all classes in the liver than in the other sites.
Discussion
in the Liver
The amount of IgA in bile not originating from TDL is still appreciable, being -50% above the normal level in serum. This, and the observation described above, that the concentration of biliary IgG did not decrease during the period of thoracic duct cannulation despite a marked reduction in serum levels, prompted the notion that immunoglobulin
Immunogenic stimulation by antigens in the environment, particularly within the intestinal tract, is considered to be responsible for the appearance and maintenance of natural antibodies present in body fluids. In most discussions, however, it is the serum IgM and IgG natural antibodies that are primarily considered, notwithstanding the fact that initial stimulation is at a mucosal surface and thus IgA antibodies might be expected to predominate. It is not surprising therefore that analysis of the natural
-.-v 4
\
6
12
18
f \
‘g*
i
-Y I
I
24
b*
I
after
Figure 1. Relative
thoracic
duct cannulation
concentrations rats after thoracic duct ulin.
I
I
4
2 Time
W
P-
I
6
(hours)
of IgM, IgG, and IgA in serum of cannulation. Ig = immunoglob-
Time Figure
after
thoracic
duct
cannulation
(hours)
2. Relative concentrations of IgG and IgA in the bile of rats after thoracic duct cannulation. Ig = immunoglobulin.
July 1984
ORIGINS
OF BILIARY
IMMlJNOGLOBlJLINS
Table 5. Antibody-Forming Cells in Blood, Spleen, and Liver of Normal Rats
lgG
No.” of AFCIlO”
W
lymphoid
cells
Tissue
IgM
IgG
IgA
Blood” Spleen Liver
299 t 214 50 * 36 2043 ? 655
243 + 180 172 + 132 3881 + 1556
253 t 175 188 + 124 4244 + 1981
AFC = antibody-forming cells. Ig = immunoglobulin. sents mean value from 6 rats t standard deviation. prepared from 5 ml of blood.
25
I
I
I
2
Time after
thoracic
I
I
I
4
bacterial activity in lymph and bile indicates the availability of the vehicle to transport antigen absorbed through mucosal surfaces, as has been proposed (28-30). The present studies show that when TDL is prevented from entering the blood, the level of total IgA in bile is reduced selectively, by 75%, in accord with the results of Dahlgren et al. (18), thus providing further evidence that the bulk of biliary IgA is derived from the GALT. This is in keeping with previous findings that, although lymph is rich in IgA, serum levels are relatively low (12) due to selective removal of IgA within the liver (2,3,7), and that, in the rat, intravenously injected IgA from lymph is rapidly removed from the circulation by biliary secretion (13). Alternatively, it has been suggested (5) that after synthesis, IgA directly enters the venous drainage of the gut and reaches the liver directly via the portal vein without entering the systemic circulation. Direct testing of this by thoracic duct ligation, thus preventing lymphatic drainage of the gut and favoring the venous pathway, showed that despite a significantly increased concentration of IgA in the portal circulation, this pathway was not sufficient to halt the decline in biliary IgA levels. Hence we propose that the venous pathway together with lymphovascular anastomosis, linking the mesenteric lymph duct and portal vein, are of minor importance in delivering intestinal IgA to the liver. From the concentrations of IgA in serum and TDL it can be calculated (Table 6) that -95% of TDL IgA is newly
(hours)
Relative concentrations of IgG and IgA in the bile of rats after thoracic duct ligation. Ig = immunoglobulin.
antibodies against two common bacterial species showed high amounts of IgA present in lymph and bile whereas activity in serum was predominantly associated with IgM and IgG fractions. The demonstration of the loss of specific antibacterial antibody activity from bile after thoracic duct cannulation indicates the origin of the majority of biliary IgA to be the GALT. Supporting evidence for this was recently found when it was shown that bile duct occlusion of normal rats caused an increase in levels of IgA natural antibodies against Vibrio cholerae (27) and Salmonella spp. (unpublished). Dahlgren et al. (18) have found that when rats were immunized in their Peyer’s patches, some did not show a decline in specific antibody in bile after thoracic duct cannulation. These rats, freshly immunized with a new antigen, had high numbers of IgA AFC in nonmucosal sites, suggesting that relocalization back to the submucosa of the gut had not been fully established as would be the situation of AFC directed against the ever-present normal flora. Nonetheless, one must still account for a significant amount of natural IgA antibacterial antibody in bile, other than that derived from lymph. With regard to function, the identification of antiTable 4. Immunoglobulin Levels in Portal and Systemic
Blood of Rats
IgA (mgiml)” Treatment
IgC (mg,iml)”
Portal
Systemic
P/S
Systemic
Nil (8)” (4) Bile duct cannulated’ + thoracic Bile duct cannulated’ duct cannulated (5)
0.347 0.243 0.119
2 0.09 + 0.03 + 0.03
0.343 0.231 0.109
t 0.08 2 0.02 t 0.01
0.99 + 0.03 0.95 f 0.03 0.92 r 0.02
11.20 2 2.03 9.54 t 3.54 6.57 2 1.63
+ thoracic
0.182
t 0.02
0.250
+ 0.04
1.37 r 0.03
6.47 + 0.78
Bile duct cannulated’ duct ligated (4)
” Repre“Cells
1
6
duct occlusion
177
Ig = immunoglobulin. S = systemic. P = portal. “ Represents of rats in group. ‘. Samples taken 6-7 h after cannulation.
mean
value
2 standard
deviation.
” Figures
Portal
PIS
t 1.55 f 3.57 + 1.24
0.94 t 0.16 1.12 t 0.30 0.97 r 0.13
6.Y2 f 0.63
1.07 * 0.07
10.45 10.64 6.39
in parentheses
indicate
number
178
Table
GASTROENTEROLOGY
MANNING ET AL.
6.
Relative Contributions to Jmmunoglobulins A and G of Thoracic Duct Lymph of Synthesis Gut-Associated Lymphoid Tissues and Transfer From Blood
Expected Expected % derived O/nderived
lymph lymph from from
to serum ratio” concentrationb (mgiml) blood (expected/observedb) GALT
in
&A
I&
0.386 0.09 6 94
0.4 5 100
Ig = immunoglobulin. GALT = gut-associated lymphoid tissues. a Empirical results from Vaerman et al. (31) assuming IgA to be entirely dimeric. b Based on values from Table 1.
synthesized, agreeing with previous findings (31). Further, based on the values of the flow rates of bile and lymph, the concentrations of IgA in these fluids, and the result that 75% of biliary IgA is derived from of lymph IgA arrivTDL, it is calculated that -50% ing in blood is subsequently removed by the liver. This finding is consistent with previous observations on IgA transport (2,13). The remainder not transported to bile could be accounted for by either of three possibilities or a combination of them. First, it may contribute to the serum pool of IgA. Second, it could be removed from the circulation at various mucosal sites (32,33), thereby augmenting the system of relocation of IgA plasma cells from one site to another (reviewed in Reference 34). Finally, the possibility that IgA is degraded within hepatocytes must be considered although the bulk of the evidence supports a nondegradative pathway (reviewed in Reference 35). A brief comment is needed upon the decline in immunoglobulin levels in the blood of rats after thoracic duct cannulation (Figure 1). An equilibrium is rapidly reached (4-6 h), which suggests that the immunoglobulins leaving the circulation are being replaced from other sources at approximately the same rate. The decline in IgA is greatest, as would be expected, because it is selectively removed in the liver. Further, with regard to IgM, the similar IgG to IgM ratios for serum and lymph confirm the results of others and indicate a contribution of locally synthesized IgM to the blood pool (11,36). The final matter to be addressed is immunoglobulin synthesis in the liver. We have demonstrated the presence of IgA-, IgG-, and IgM-secreting cells in the liver at levels that are distinguishable from those in blood and spleen, suggesting a predilection for these cells in this site. This is in keeping with previous reports (37,38) which have indicated the presence of AFC in the liver after immunization and the recent finding by Reichert et al. (39) that functional immunoblasts from Peyer’s patches, when reinjected, localize preferentially in the liver. Although synthesis has been shown to occur, this
Vol. 87, No. 1
does not necessarily corroborate the second part of the hypothesis that secretion to bile follows, although there would be little difficulty in accepting that dimeric IgA produced in the liver could be translocated. The trace amount of IgM in bile could well leak from the plasma, although we have recently reported that the transient increase in biliary IgM that occurs after immunization can be explained by the localization of IgM cells (19). IgG occurs in bile at about one-sixtieth the level in serum, yet after thoracic duct cannulation the fall in serum concentration was not paralleled in bile. One explanation is that the serum level is still sufficiently high to allow passive exudation at the same rate. Alternatively, the large number of IgG cells present in the liver could maintain the local concentration and so preserve the degree of output in bile. The origin of cells detected in the current study, their location, and whether or not they are sufficient to account for the levels of immunoglobulin in bile not originating from other sources remain to be resolved. Nevertheless, inasmuch as local synthesis of at least IgA has been claimed to be significant in humans (16), we suggest that further studies of cellular activity in the rat liver may prove useful.
References 1.
2.
3. 4.
5.
6.
7.
8.
9.
10.
Lemaitre-Coelho I, Jackson GDF, Vaerman J-P. Rat bile as a convenient source of secretory IgA and free secretory component. Eur J Immunol 1977;7:588-90. Jackson GDF, Lemaitre-Coelho I, Vaerman J-P, Bazin H, Beckers A. Rapid disappearance from serum of intravenously injected rat myeloma IgA and its secretion into bile. Eur J Immunol 1978;8:123-6. Orlans E, Peppard J, Reynolds J, Hall J. Rapid active transport of IgA from blood to bile. J Exp Med 1978;147:588-92. Vaerman J-P, Lemaitre-Coelho I, Jackson GDF. Role of the liver in the rat intestinal sIgA system. Adv Exp Med Biol 1978;107:233-40, Renston RH, Jones AL, Christiansen WD, Hradek GT, Underdown BJ. Evidence for a vesicular transport mechanism in hepatocytes for biliary secretion of immunoglobulin A. Science 1980;208:1276-8. Birbeck MSC, Cartwright P, Hall JG, Orlans E, Peppard J. The transport by hepatocytes of immunoglobulin A from blood to bile visualized by autoradiography and electron microscopy. Immunology 1979;37:477-84. Fisher MM, Nagy B, Bazin H, Underdown BJ. Biliary transport of IgA: role of secretory component. Proc Nat1 Acad Sci USA 1979;76:2008-12. Orlans E, Peppard J, Fry JF, Hinton RH, Mullock BM. Secretory component as the receptor for polymeric IgA on rat hepatocytes. J Exp Med 1979;150:1577-81. Crabbe PA, Carbonara AD, Heremans JF. The normal intestinal mucosa as a major source of plasma cells containing gamma-A-immunoglobulin. Lab Invest 1965;14:235-48. Crabbe PA, Bazin H, Eyssen H, Heremans JF. The normal microbial flora as a major stimulus for proliferation of plasma cells synthesizing IgA in the gut. Int Arch Allergy Appl Immunol 1968;34:362-75.
July 1984
11. Vaerman J-P, Heremans JF. Origin and molecular size of IgA in mesenteric lymph of the dog. Immunology 1970;18:27-38. 12. Kaartinen M, lmir T, Klockars M, Sandholm M, Makela 0. IgA in blood and thoracic duct lymph. Concentration and degree of polymerization. Stand J lmmunol 1978;7:229-32. 13. Reynolds J, Gyure L, Andrew E, Hall JG. Studies of the transport of polyclonal IgA antibody from blood to bile in rats. Immunology 1980;39:463-7. 14. Dive Ch, Heremans JF. Nature and origin of proteins of bile. I. A comparative analysis of serum and bile in man. Eur J Clin Invest 1974;4:235-9. 15. Dive Ch, Nadalini RA, Vaerman J-P, Heremans JF. Origin and nature of the proteins of bile. II. A comparative analysis of serum, hepatic lymph and bile proteins in the dog. Eur J Clin Invest 1974;4:241-6. 16. Delacroix DL, Hodgson HJF, McPherson A, Dive C, Vaerman J-P. Selective transport of polymeric immunoglobulin A in bile. J Clin Invest 1982;70:230-41. 17. Spencer J, Gyure LA, Hall JG. IgA antibodies in the bile of rats. III. The role of intrathoracic lymph nodes and the migration pattern of their blast cells. Immunology 1983;48:687-93. 18. Dahlgren U, Ahlstedt S, Andersson T, Hedman L, Hanson LA. IgA antibodies in rat bile are not solely derived from thoracic duct lymph. Stand J Immunol 1983;17:569-74. 19. Jackson GDF, Walker PG. The transient appearance of IgM antibodies in the bile of rats injected with Salmonella enteritidis. lmmunol Lett 1983;7:41-5. 20. Lambert R. Surgery of the digestive system in the rat. Springfield, Ill.: Charles C Thomas, 1965. 21. Bollman JL, Cain JC, Grindlay JH. Techniques for the collection of lymph from the liver, small intestine and thoracic duct of the rat. J Lab Clin Med 1948;33:1349-52. 22. Gowans JL. The recirculation of lymphocytes from blood to lymph in the rat. J Physiol (Lond) 1959;146:54-69. 23. Holdeman LV, Moore WEC. Anaerobe laboratory manual. 3rd ed. Blacksburg: Virginia Polytechnic Institute and State University Anaerobe Laboratory, 1975. 24. Mancini G, Carbonara AD, Heremans JF. lmmunochemical quantitation of antigens by single radial immunodiffusion. Immunochemistry 1965;2:235-54. 25. Kenrick KG, Cooper GN. Antibodies in the intestinal secretions of rats. Primary and secondary responses to polymeric flagellin. Aust J Exp Biol Med Sci 1978;56:441-53.
ORIGINS OF BILIARY IMMUNOGLOBULINS
179
26. Gronowicz E, Coutinho A, Melchers F. A plaque assay for all cells secreting Ig of a given type or class. Eur J Immunol 1976;6:588-90. 27. Jackson GDF, Cooper GN. Immune responses of rats to live Vibrio cholerae: Secretions of antibodies in bile. Parasite lmmunol 1981;3:127-35. 28. Russell MW, Brown TA, Mestecky J. Role of serum IgA. Hepatobiliary transport of circulating antigen. J Exp Med 1981;153:968-76. 29. Peppard J, Orlans E, Payne AWR, Andrew E. The elimination of circulating complexes containing polymeric IgA by excretion in the bile. Immunology 1981;127:83-9. 30. Socken DJ, Simms ES, Nagy BR, Fisher MM, Underdown BJ. Secretory component dependent hepatic transport of IgA antibody-antigen complexes. J Immunol 1981;127:316-9. 31. Vaerman J-P, Andre C, Bazin H, Heremans JF. Mesenteric lymph as a major source of serum IgA in guinea pigs and rats. Eur J lmmunol 1973;3:580-4. 32. Jackson GDF, Lemaitre-Coelho I, Vaerman J-P. Transfer of MOPC-315 IgA to secretions in MOPC-315 tumour bearing and normal Balb/c mice. Protides Biol Fluids 1977;25:919-22. 33. Virella G, Montgomery P, Lemaitre-Coelho I. Transport of oligomeric IgA of systemic origin into external secretions. Adv Exp Med Biol 1978;107:241-51. 34. Tomasi TB, Larson DO, Challacombe S, McNabb P. Mucosal immunity: the origin and migration patterns of cells in the secretory system. J Allergy Clin Immunol 1980;65:12-9. 35. Jones AL, Renston RH, Burwen SJ. Uptake and intracellular disposition of plasma-derived proteins and apoproteins by hepatocytes. Prog Liver Dis 1982;7:51-69. 36. Kaartinen M, Kosunen TV, Makela 0. Complement and immunoglobulin levels in the serum and thoracic duct lymph of the rat. Eur J Immunol 1973;3:556-9. 37. Garvey JS, Linker-Israeli M. Anti-SRBC plaque-forming cells in liver, spleen and hepatocyte cell suspensions of rabbits and mice. Immunol Commun 1972;1:507-21. 38. Hiramoto RN, Habeeb AFSA, Hamlin M. Antibody-forming cells in five organs of hyperimmunized guinea pigs. Proc Sot Exp Biol Med 1970;134:975-9. 39. Reichert RA, Gallatin WM, Weissman IL, Butcher EC. Germinal centre cells lack homing receptors necessary for normal lymphocyte recirculation. J Exp Med 1983:157:813-27.