- Email: [email protected]
Inhibitory effects of bilirubin on cellular immune responses in man
690
May 1975 The Journal o f P E D I A T R I C S
Inhibitory effects of bilirubin on cellular immune responses in man A significant depression in cell-mediated immunity as measured by lymphoproliferative responses to phytohemagglutinin and responsiveness to mixed lymphocyte culture was observed when adult lymphocytes or cord blood lymphocytes were incubated with increasing concentrations o f bilirubin. The inhibitory effect o f bilirubin could only be demonstrated with suboptimal concentrations o f PHA (0.01 and 0.005%) and was more marked in premature infants than in term neonates or adults. This effect was partially reversible after short preincubation with bilirubin, but was more protracted with preincubations o f 24 hours or more. Inhibition o f M L C responsiveness o f 80.1 +_ 5.1% was also demonstrated at a bilirubin concentration o f 20 mg/dl. Specific cytotoxicity to rubella virus-infected cells, measured by a 51Cr-release microassay, was not found to be depressed. Bilirubin thus appears to have an inhibitory effect on immune responsiveness which is greater on the afferent limb than on the efferent limb o f immunity.
Marek Rola-Pleszezynski, M . D . , Monroe M. Vincent, B.S.,
W a s h i n g t o n , D . C., Sally A. Hensen, B.S.,
Bethesda, Md.,
and Joseph A. Bellanti, M . D . , *
W a s h i n g t o n , D . C.
DESPITE THE FACT that unconjugated bilirubin has long been known to exert toxic effects on various tissues, 1-3its influence on immune responsiveness has not been fully elucidated. Ansaldi and associates 4found slightly lower serum levels of IgM and IgA in neonates with bilirubin levels above 16 mg/dl when compared to normal neonates. In 12 infants with neonatal hyperbilirubinemia (>16 mg/dl) Nejedla 5 found antibody responses to diphtheria-pertussis-tetanus immunization to be significantly lower than those of normal control subjects. Diaz de Castillo 6made similar observations in infants following measles immunization. In 1968, De Sanctis and associates 7studied three infants with serum bilirubin levels of 13, 18, and 20 mg/dl, respectively, From the Departments o f Pediatrics and Microbiology, Georgetown University School o f Medicine, and MicrobiologicalAssociates Inc. Supported in part by a Canadian Medical Research Council Fellowship grant and by training grant No. HDOO261from the National Institutes o f Health. *Reprintaddress:Departmentof Pediatrics, Georgetown University, Washington,D. C. 20007.
Vol. 86, No. 5, pp. 690-696
and found that their sera inhibited the morphologic responses of autologous and heterologous lymphocytes to phytohemagglutinin. The present report describes the effects of varying concentrations of bilirubin on the afferent limb of immunity, as measured by lymphoproliferative responses to phytohemagglutinin and in mixed lymphocyte cultures, as well as on the efferent limb of immunity, employing a 51Cr microassay of cell-mediated immunity to viruses recently developed in our laboratory. 8 Abbreviations used CMI: cell-mediatedimmunity PHA-P: phytohemagglutinin MLC: mixedlymphocyte culture cpm: countsper minute M A T E R I A L S AND M E T H O D S Approximately 25 ml of heparinized blood were obtained from ten adult healthy donors, each of whom had positive rubella hemagglutination-inhibition titers. For studies of cord blood, 10 to 15 ml of heparinized
Volume 86 Number 5
Bilirubin effects on cellular immune responses
691
Table I. Effects of varying concentrations of bilirubin on lymphoproliferative responses to varying concentrations of PHA-P and in MLC reactions expressed in cpm __+SD Bilirubin concentration (mg/dl)
0 5 10 15 20
Responses at varying concentrations of PHA-P 0%
700 • 112 644• 82* 802• 136, 726• 70* 692• 100,
0.005%
13,629 • 9,391 • 8,287• 7,987• 7,292•
1,556 1,220" 1,094t 952t 1,239t
0.01%
35,491• 27,329• 3,279* 23,779• 22,608• 22,040•
0.1%
63,248 + 7 , 9 7 2 62,045_ 7,033* 66,067 • 10,841, 64,418___ 8,992* 62,581 + 7,201,
MLC responses
8,374___2,128 4,506 • 712, 2,948 • 461" 1,968 • 366t 1,666 • 241t
*Values significantlydifferent from control values (p(0.05). 1"Valuessignificantlydifferent from control values (p(0.01). ~Values not significantlydifferent from control values (p>0.05).
blood were obtained by venipuncture of the umbilical vein immediately after removal of the placenta. Purified suspensions of mononuclear leukocytes containing 95% lymphocytes were prepared by centrifugation on a Hypaque-Ficoll gradient and were washed and resuspended in RPMI 1640 medium; the m e d i u m contained 10% heat-inactivated fetal bovine serum (final albumin concentration: 0.3 gm/dl), 100 units of penicillin, and 100 /xg of streptomycin/ml, as described previously. 9 Five different concentrations of unconjugated bilirubin (Pfansteehl Labs, Inc., Waukegan, Ill.) were prepared in complete RPMI 1640 medium; the final concentrations were 2, 5, 10, 15, and 20 mg/dl, respectively. All cell cultures were performed in Falcon Plastics Microtest II tissue culture plates at 37~ in a moisture-laden atmosphere of 5% CO 2. The phytohemagglutinin (PHA-P) stimulation studies were performed with test cultures containing 2 x 105 lymphocytes in 0.2 ml of m e d i u m and varying concentrations of PHA-P; control samples were identical except for the omission of PHA-P. Mixed lymphocyte cultures were prepared with 2 x 105 lymphocytes from each of two unrelated donors in a two-way reaction mixed lymphocyte culture. All test cultures were performed in triplicate at concentrations of bilirubin ranging from 2 to 20 mg/dl; control cultures were identical except for the omission of bilirubin. P h y t o h e m a g g l u t i n i n c u l t u r e s were i n c u b a t e d for three days; mixed lymphocyte cultures for six days. During these incubation periods peak responses were n o t e d in a t i m e c o u r s e s t u d y with all P H A a n d / o r bilirubin concentrations used. Six hours prior t o harvesting, 1/xCi of 3H-thymidine was added to each culture. A special apparatus was employed for the separation of the lymphocytes on glass-fiber filter paper, for washing t h e m free of m e d i u m with isotonic saline, and
for their recovery and quantitation of radioactive uptake. 8 The filter paper was air dried and transferred into vials containing 5 ml of toluene and left for six hours at room temperature until all the bilirubin was dissolved. The filter paper was then transferred into vials containing 10 ml of Bray's solution for counting in a Packard Tricarb Liquid Scintillation Spectrometer. The results were expressed as average counts per minute of triplicate samples. In order to measure the reversibility of the effects of bilirubin, lymphocytes were incubated with bilirubin for v a r i a b l e l e n g t h s of t i m e a f t e r w h i c h t h e y were washed by centrifugation on a Hypaque-Ficoll gradient as described above, and then washed twice in complete RPMI medium. The lymphocytes were left to recover for varying periods of time before P H A - P was added to them as described above. A 51Cr microassay procedure for m e a s u r e m e n t of cell-mediated immunity to rubella virus was performed as described previously. 8 Test target cells consisted of baby hamster-kidney (BHK-21) cell lines chronically infected with rubella virus. The same virus-free cell line served as control target cells. A concentration of 2 • 106 infected or control target cells were labeled with 100 /xCi of sodium 51Cr-chromate ( N a 2 51CRO4; CJS-1P; Amersham/Searle, Arlington Heights, Ill.), incubated at 37~ for one hour, washed three times, and resuspended in complete RPMI medium. Test cultures consisted of 5 x l0 s lymphocytes and 5 x 103 target cells in 0.2 ml of medium, giving an attacker-to-target cell ratio of 100:1. Additional control cultures consisted of target cells without lymphocytes. Experiments were performed in triplicate. The cultures were incubated at 37~ on a rocker platform and harvested after 18 hours by an apparatus which separated the medium containing the released 51Cr from the reacting cells.8
692
Rola-Pleszczynski et aL
The Journal of Pediatrics May 1975
60
0 r
~,
20
ZlO~
r O
0
40
20~
ce-
O ~ v ,m
0 o
60 l,,e-,,
c=-o PHA 0.1% PHA 0.01% PHA 0.005%
cO.
80 I
5
I000
I
I0
I
15
,I
Fig. 1. Effects of addition of bilirubin to lymphocyte cultures stimulated with optimal (0.1%) and suboptimal (0.01%, 0.005%) concentrations of PHA-P. Lymphocytes from ten adults and ten neonates. The percent release for each cell line was calculated as follows:
Percent release -
(hrs) Fig. 2. Effects of addition of bilirubin (20 mg/dl) to lymphocyte cultures at 0, 6, 12, and 24 hours after the start of incubation with PHA-P (0.005%). Lymphocytes from ten adults and ten neonates.
20
Concentration of bilirubin in cultures rng / 100 rnl
cpm 51Cr released cpm 51Cr released from target cells _spontaneously from and lymphocytes target cells alone during incubation during incubation total 51Cr 5tCr released at 0 releasable - time incubation
6 12 24 Time of bilirubin addition
x 100
The specific i m m u n e release was obtained by subtracting the percentage release for control cells from that for rubella-infected cells. Viability of lymphocyte and target cells was assessed at the time of harvesting by the trypan blue exclusion method; it was found to be above 90% after 24 hours and above 70% after six days, and was comparable to control samples. F i n a l results, e x p r e s s e d as p e r c e n t i n h i b i t i o n o f response, were calculated by subtracting from 100% the percentage ratio of cpm with bilirubin over cpm without bilirubin. RESULTS The effects of varying concentrations of bilirubin on lymphoproliferative responses to varying concentrations of P H A - P are presented in Table I and Fig. 1. A significant dose-response inhibitory effect of bilirubin on lymphocyte proliferation was found at concentra-
tions of 0.01% and 0.005% of P H A - p (p<0.05), but none at a concentration of 0.1%. Maximal inhibition of 46.5% was detected at bilirubin concentrations of 20 mg/dl (Fig. 1). In order to determine at what phase of lymphocyte transformation the inhibition was most active, a separate set o f e x p e r i m e n t s was p e r f o r m e d in w h i c h bilirubin was added at 0, 6, 12, and 24 hours after the start of the incubation with 0.005% P H A - P (Fig. 2). Maximal inhibition was observed when bilirubin was added at the start of the reaction and was less marked when bilirubin was added 6 or 12 hours !ater; no inhibition was observed when bilirubin was added after 24 hours. The reversibility of the action of bilirubin was measured by incubating lymphocytes with bilirubin at a concentration of 20 mg/dl for 15 or 30 minutes, washing t h e m free of bilirubin, and then adding 0.005% P H A - P after varying periods of recovery (Fig. 3). No significant inhibition was noted after 15 minutes of initial incubation throughout 240 minutes of recovery; in contrast, lymphocytes preincubated with bilirubin for 30 minutes were significantly inhibited for 180 minutes, but complete reversibility was observed w h e n the lymp h o c y t e s were allowed to re'cover for 240 m i n u t e s . Following more prolonged incubation times of 24 and 72 hours, the inhibitory effects of bilirubin were protracted and remained irreversible ~ven after a recovery period of 72 hours (Table II). The effects of varying concentrations of bilirubin on m i x e d l y m p h o c y t e c u l t u r e r e s p o n s i v e n e s s are presented in Table I and Fig. 4. A significant dose-response effect of bilirubin was again seen; inhibiton, however, was more marked than with mitogenic stimulation with
Volume 86 Number 5
Bilirubin effects on cellular immune responses
x
I
N
15' initial incubation
o-o 30' initial incubation
/
I
0
693
I
15 30
I
60
120
I
I
180
240
'
Recovery time (min)
Fig. 3. Effects of varying periods of incubation with bilirubin (20 mg/dl) and recovery time on lymphoproliferative responses to PHA-P (0.005%). Lymphocytes from ten adults and ten neonates. Table II. Effects of varying periods of incubation with bilirubin (20 mg/dl) and recovery time on lymphoproliferative responses to P H A - P (0.005%), expressed in cpm + SD Reco very period (hr) Incubation with bilirubin (hr)
0 24 72
6
11,210 _--_1,215 5,412 + 640 NT
24
48
13,760 -e 1,611 6,697 • 7,233 2,955 • 342
12,015 • 1,466 5,140 • 697 2,161 • 288
72
14,172 + 1,558 7,072 • 774 2,818-• 375
NT = not tested. PHA-P. Inhibition was detected at bilirubin concentrations as low as 2 mg/dl, the lowest concentration tested, and a maximal inhibition of 80.1% was observed at 20 mg/dl (Fig. 4). A separate set of experiments was performed to compare the inhibitory effects of bi!irubin on lymphocytes from adults and newborn infants (Fig. 5). A n inhibitory effect was observed with iymphocytes from adults and from term and premature infants with concentrations of bilirubin as low as 5 mg/dl. A significant increase in inhibition was seen at higher concentrations of bilirubin. No significant differences were observed in lymphoproliferative responses between lymphocytes from adults and term neonates; however, at concentrations of bilirubin of 5 mg/dl, lymphocytes from three premature infants (<32 weeks of gestation) showed a more marked inhibition (p<0.05) than those from term infants or adults. Higher concentrations of bilirubin failed to reveal any significant difference between the groups studied. The effects of b i l i r u b i n on l y m p h o c y t o t o x i c responses tO rubella virus were studied employing a 51Cr-
release assay and are presented in Table III. No reduction in specific c e l l - m e d i a t e d c y t o t o x i c i t y to r u b e l l a virus could be detected by addition of bilirubin to the test system. DISCUSSION The results of the present studies indicate that unconjugated bilirubin has an inhibitory effect on the lymphoproliferative responses to P H A - P and on responsiveness to mixed lymphocyte culture, no effect, however, was demonstrated on the killer function of lymphocytes as measured by specific lymphocytotoxicity. These results can be explained on the basis of either a direct toxicity of bilirubin or on an inhibitory effect on some intermediary metabolic step. The direct toxicity of unconjugated bilirubin is well known 1~ and cellular viability is rapidly affected upon exposure to it. Toxicity is nevertheless prevented, if bilirubin is bound to albumin; it is diminished, if albumin is added a short time after bilirubin. ~~ Since cellular viability was not affected in our assays and since specific lymphocytotoxicity was not impaired by bilirubin, it seems quite
694
Rola-Pleszczynski et al.
The Journal of Pediatrics May1975
20 c" O
~
40
g+l
60 ::~
m
80-
I00
q)
--
I 5
~ 10
t 15
I 20
Concentration of bilirubin in cultures mg / 100 m l Fig. 4. Effects of varying concentrations of bilirubin on responsiveness to mixed lymphocyte culture. Lymphocytes from ten adults and ten neonates. Table III. Effect of addition of bilirubin to c u l t u r ~ of lymphocytes and target cells
Donors
Neonates (cord blood) Adults
Concentration of Specific immune bilirubin (mg/dl) release(%)*
0 0 5 10 15 20
0 24.2 • 4.1]" 21.5 • 3.8 28.6 _+ 6.1 25.5 • 4,4 21.0 • 4.1
*Mean • SD. tNo statistically significant difference with varying concentrations of bilirubin.
unlikely that our results can be explained by direct toxicity. Furthermore, the albumin content of fetal bovine serum that was used in preparing the culture m e d i u m is sufficient to bind approximately 96% of the bilirubin ~3 and hence prevent its direct toxicity. It is estimated that in the plasma of an infant with unconjugated hyperbilirubinemia, 98-99% of the bilirubin is normally bound to albumin, up to bilirubin levels of approximately 20 mg/dl. 13'14 Bilirubin may exert its effect through some interm ediary metabolic step involved in the lymphoproliferative response. The strong affinity of bilirubin for lipids
m a y m o d i f y t h e s u r f a c e s t r u c t u r e of l y m p h o c y t e m e m b r a n e s and thus conceivably inhibit their function a n y w h e r e in t h e s e q u e n c e of e v e n t s f r o m initial stimulation to final blastic transformation. In the present studies bilirubin curtailed the stimulatory effects of specific antigens, such as the surface ones o f all~geneic l y m p h o c y t e s , as well as n o n s p e c i f i c mitogens such as PHA-P. Since a significant degree of inhibition was observed at 6 and 12 hours of incubation, some alteration of subsequent metabolic steps may be involved. The inhibitory effect of bilirubin may also be the result of respiratory inhibition of mitochondria, as suggested by the studies of Noir and associates who r e p o r t e d that b i l i r u b i n is a m u l t i s i t e i n h i b i t o r o f mitochondrial respiratory metabolism. 10 The inhibitor effects of bilirubin could also be a reflection of some interaction with lysosomat m e m b r a n e s since these are k n o w n to be a c t i v a t e d in t h e l y m p h o p r o l i f e r a t i v e response. 15 The present studies extend the observations of De Sanctis and associates 7 who showed a morphologic in, hibition of blastogenesis in three hyperbilirubinemic infants. In these latter studies the measurement of inhibited blastogenesis in hyperbilirubinemia may have been biased by the use of whole serum as the source of bilirubin, thus introducing the possibility of interference by other s e r u m factors. The increased sensitivity of lymphocytes from premature infants to bilirubin, at least with low concentrations of this substance, may have clinical implications upon subsequent immunologic development. The studies of Steele and associates, 16 for example, have shown that even after one year of life, premature infants have a delay in the development o f specific cell-mediated immunity to rubella virus after immunization. The results of the present studies thus suggest the importance of prevention and prompt correction of hyperbilirubinemia in the newborn infant, and especially in the premature infant, whose levels of hyperbilirubinemia are c o m m o n l y h e i g h t e n e d and m o r e p r o t r a c t e d . If t h e results of the present in vitro studies can be extrapolated to the infant, even so-called "physiologic" icterus of up to 12 mg/dl of bilirubin may present a hazard by decreasing significantly the ability of lymphocytes to respond to extrinsic antigenic stimulation. Several studies have shown a long-term deficit in antibody production following immunization in infants with neonatal hyperb i l i r u b i n e m i a . 46 - The i r r e v e r s i b i l i t y of t h e action of bilirubin after an exposure of 24 hours or more is an ind i c a t i o n of t h e effects t h a t h y p e r b i l i r u b i n e m i a of r e l a t i v e l y s h o r t d u r a t i o n m a y h a v e on t h e i m m u n e system.
Volume 86 Number 5
Bilirubin effects on cellular immune responses
• |
[ ] Adults 9 Full-terminfants
20--~'~
[ ] Premature infants
695
tO
A 4 0 - -
"6 c" O o_
+I
60-"-i (a'3
80--
I00 0.
5
i0
15
20
Concentration of bilirubin mg/ 100 ml Fig. 5. Effects of varying concentrations of bilirubin on lympho-proliferative responses to 0.005% PHA-P of lymphocytes from adults and neonates.
The initial step in the cytotoxicity assay involves activation of lymphocytes by infected target ceils and could be considered a part of the afferent limb of the i m m u n e response, with the efferent "killer" function closely linked. The lack of interference by bilirubin with lymphocyte cytotoxicity indicates that not every part of the afferent limb is affected by bilirubin and is also another example of the functional dissociation of the afferent and efferent limbs of immunity, as has been reported by Rocklin and associates) 7-18and Epstein and associates. 19 Nevertheless, any substance which interferes with the initial steps of i m m u n e recognition may well c o m p r o m i s e t he w h o l e m e c h a n i s m of i m m u nologic responsiveness. The present studies thus lend further support for the clinical importance of monitoring and early treatment of the hyperbilirubinemic infant. We are indebted to Dr. Chantal Beaudet-Pleszczynska and Dr. Yee Hing Thong for their invaluable encouragement and counsel during this study, and to Mrs. Mary Jean Navaretta for her secretarial assistance. REFERENCES
1. Bowen WR, and Waters WJ: Bilirubin encephalopathy. Studies related to the site of inhibitory action of bilirubin on brain metabolism, Am J Dis Child 93:21, 1957. 2. Odell GB: Studies in kernicterus. The protein binding of bilirubin, J Clin Invest 38:823, 1959. 3. Bernstein J, and Landing BH: Extraneural lesions associated with neonatal hyperbilirubin and kernicterus, Am J Pathol 40:371, 1962.
4. Ansaldi N, Fiandino G, and Ciriotti G: Changes of the immunoglobulin level in hyperbilirubinemic premature infants, Minerva Pediatr 20:1982, 1968. 5. Nejedla Z: The development of immunological factors in infants with hyperbilirubinemia, Pediatrics 45:102, 1970. 6. Diaz de Castillo E: Neonatal hyperbilirubinemia. Its relation to the immunological response capacity of the newborn infant, Gac Med Mex 105:185, 1973. 7. De Sanctis C, Malandra C, Zanetti P, Fabris C, and Ponzone A: Neonatal hyperbilirubinemia and response of lyrnphocytes to phytohemagglutinin, Minerva Pediatr 20:2010, 1968. 8. Steele RW, Hensen SA, Vincent MM, Fucillo DA, and Bellanti JA: A 51Cr microassay technique for cell-mediated immunity to viruses, J Immunol 110:1502, 1973. 9. Thorsby E: Cell specific and common antigens on human granulocytes and lymphocytes demonstrated with cytotoxic heteroantibodies, Vox Sang 13:194, 1967. 10. Noir BA, Boveris A, Garaza-Pereira AM, and Stoppani ADM: Bilirubin: a multi-site inhibitor of mitochondrial respiration, FEBS Lett. 27:270, 1972. 11. Cowger ML: Mechanisms of bilirubin toxicity on tissue culture cells: factors that affect toxicity, reversibility by albumin and comparison with other respiratory poisons and surfactants, Biochem Med 5:1, 1971. 12. Tuilie M, and Lardinois R: The binding of unconjugated bilirubin by human sera and purified albumins, Biol Neonate 21:447, 1972. 13. Clarenburg R, and Barnhart JL: Interaction of serum albumin and bilirubin at low concentrations, Am J Physiol 225:493, 1973. 14. Schmid R: Bilirubin metabolism in man, N Engl J Med 287:703, 1972. 15. Hirshchorn K and Hirshchorn R: Role of lysosomes in the lymphocyte response, Lancet 1:1046, 1965.
696
Rola-Pleszczynski et al.
16. Steele RW, Hensen SA, Vincent MM, Fucillo DA, and Bellanti JA: The development of specific cellular and humoral immune responses in children immunized with live rubella virus vaccine, J Infect Dis 130:499, 1974. 17. Rocklin RE, Reardon G, Sheffer A, Churchill WH, and David JR: Dissociation between two in vitro correlates of delayed hypersensitivity: absence of migration inhibition factor (MIF) in the presence of antigen-induced incorporation of 3 H-Thymidine, in Harris, JE, editor: Proceed-
The Journal of Pediatrics May 1975
ings of the Fifth Leukocyte Culture Conference New York, 1970, Academic Press, Inc., p 639-648. 18. Rocklin RE: Production of migration inhibitory factor by non-dividing lymphocytes, J Immunol 110:674, 1973. 19. Epstein LB, Stevens DA, and Merigan TC: Selective increase in lymphocyte interferon response to vaccinia antigen after revaccination. Proc Nat Acad Sci USA 69:2632, 1972.
May 1975 The Journal o f P E D I A T R I C S
Inhibitory effects of bilirubin on cellular immune responses in man A significant depression in cell-mediated immunity as measured by lymphoproliferative responses to phytohemagglutinin and responsiveness to mixed lymphocyte culture was observed when adult lymphocytes or cord blood lymphocytes were incubated with increasing concentrations o f bilirubin. The inhibitory effect o f bilirubin could only be demonstrated with suboptimal concentrations o f PHA (0.01 and 0.005%) and was more marked in premature infants than in term neonates or adults. This effect was partially reversible after short preincubation with bilirubin, but was more protracted with preincubations o f 24 hours or more. Inhibition o f M L C responsiveness o f 80.1 +_ 5.1% was also demonstrated at a bilirubin concentration o f 20 mg/dl. Specific cytotoxicity to rubella virus-infected cells, measured by a 51Cr-release microassay, was not found to be depressed. Bilirubin thus appears to have an inhibitory effect on immune responsiveness which is greater on the afferent limb than on the efferent limb o f immunity.
Marek Rola-Pleszezynski, M . D . , Monroe M. Vincent, B.S.,
W a s h i n g t o n , D . C., Sally A. Hensen, B.S.,
Bethesda, Md.,
and Joseph A. Bellanti, M . D . , *
W a s h i n g t o n , D . C.
DESPITE THE FACT that unconjugated bilirubin has long been known to exert toxic effects on various tissues, 1-3its influence on immune responsiveness has not been fully elucidated. Ansaldi and associates 4found slightly lower serum levels of IgM and IgA in neonates with bilirubin levels above 16 mg/dl when compared to normal neonates. In 12 infants with neonatal hyperbilirubinemia (>16 mg/dl) Nejedla 5 found antibody responses to diphtheria-pertussis-tetanus immunization to be significantly lower than those of normal control subjects. Diaz de Castillo 6made similar observations in infants following measles immunization. In 1968, De Sanctis and associates 7studied three infants with serum bilirubin levels of 13, 18, and 20 mg/dl, respectively, From the Departments o f Pediatrics and Microbiology, Georgetown University School o f Medicine, and MicrobiologicalAssociates Inc. Supported in part by a Canadian Medical Research Council Fellowship grant and by training grant No. HDOO261from the National Institutes o f Health. *Reprintaddress:Departmentof Pediatrics, Georgetown University, Washington,D. C. 20007.
Vol. 86, No. 5, pp. 690-696
and found that their sera inhibited the morphologic responses of autologous and heterologous lymphocytes to phytohemagglutinin. The present report describes the effects of varying concentrations of bilirubin on the afferent limb of immunity, as measured by lymphoproliferative responses to phytohemagglutinin and in mixed lymphocyte cultures, as well as on the efferent limb of immunity, employing a 51Cr microassay of cell-mediated immunity to viruses recently developed in our laboratory. 8 Abbreviations used CMI: cell-mediatedimmunity PHA-P: phytohemagglutinin MLC: mixedlymphocyte culture cpm: countsper minute M A T E R I A L S AND M E T H O D S Approximately 25 ml of heparinized blood were obtained from ten adult healthy donors, each of whom had positive rubella hemagglutination-inhibition titers. For studies of cord blood, 10 to 15 ml of heparinized
Volume 86 Number 5
Bilirubin effects on cellular immune responses
691
Table I. Effects of varying concentrations of bilirubin on lymphoproliferative responses to varying concentrations of PHA-P and in MLC reactions expressed in cpm __+SD Bilirubin concentration (mg/dl)
0 5 10 15 20
Responses at varying concentrations of PHA-P 0%
700 • 112 644• 82* 802• 136, 726• 70* 692• 100,
0.005%
13,629 • 9,391 • 8,287• 7,987• 7,292•
1,556 1,220" 1,094t 952t 1,239t
0.01%
35,491• 27,329• 3,279* 23,779• 22,608• 22,040•
0.1%
63,248 + 7 , 9 7 2 62,045_ 7,033* 66,067 • 10,841, 64,418___ 8,992* 62,581 + 7,201,
MLC responses
8,374___2,128 4,506 • 712, 2,948 • 461" 1,968 • 366t 1,666 • 241t
*Values significantlydifferent from control values (p(0.05). 1"Valuessignificantlydifferent from control values (p(0.01). ~Values not significantlydifferent from control values (p>0.05).
blood were obtained by venipuncture of the umbilical vein immediately after removal of the placenta. Purified suspensions of mononuclear leukocytes containing 95% lymphocytes were prepared by centrifugation on a Hypaque-Ficoll gradient and were washed and resuspended in RPMI 1640 medium; the m e d i u m contained 10% heat-inactivated fetal bovine serum (final albumin concentration: 0.3 gm/dl), 100 units of penicillin, and 100 /xg of streptomycin/ml, as described previously. 9 Five different concentrations of unconjugated bilirubin (Pfansteehl Labs, Inc., Waukegan, Ill.) were prepared in complete RPMI 1640 medium; the final concentrations were 2, 5, 10, 15, and 20 mg/dl, respectively. All cell cultures were performed in Falcon Plastics Microtest II tissue culture plates at 37~ in a moisture-laden atmosphere of 5% CO 2. The phytohemagglutinin (PHA-P) stimulation studies were performed with test cultures containing 2 x 105 lymphocytes in 0.2 ml of m e d i u m and varying concentrations of PHA-P; control samples were identical except for the omission of PHA-P. Mixed lymphocyte cultures were prepared with 2 x 105 lymphocytes from each of two unrelated donors in a two-way reaction mixed lymphocyte culture. All test cultures were performed in triplicate at concentrations of bilirubin ranging from 2 to 20 mg/dl; control cultures were identical except for the omission of bilirubin. P h y t o h e m a g g l u t i n i n c u l t u r e s were i n c u b a t e d for three days; mixed lymphocyte cultures for six days. During these incubation periods peak responses were n o t e d in a t i m e c o u r s e s t u d y with all P H A a n d / o r bilirubin concentrations used. Six hours prior t o harvesting, 1/xCi of 3H-thymidine was added to each culture. A special apparatus was employed for the separation of the lymphocytes on glass-fiber filter paper, for washing t h e m free of m e d i u m with isotonic saline, and
for their recovery and quantitation of radioactive uptake. 8 The filter paper was air dried and transferred into vials containing 5 ml of toluene and left for six hours at room temperature until all the bilirubin was dissolved. The filter paper was then transferred into vials containing 10 ml of Bray's solution for counting in a Packard Tricarb Liquid Scintillation Spectrometer. The results were expressed as average counts per minute of triplicate samples. In order to measure the reversibility of the effects of bilirubin, lymphocytes were incubated with bilirubin for v a r i a b l e l e n g t h s of t i m e a f t e r w h i c h t h e y were washed by centrifugation on a Hypaque-Ficoll gradient as described above, and then washed twice in complete RPMI medium. The lymphocytes were left to recover for varying periods of time before P H A - P was added to them as described above. A 51Cr microassay procedure for m e a s u r e m e n t of cell-mediated immunity to rubella virus was performed as described previously. 8 Test target cells consisted of baby hamster-kidney (BHK-21) cell lines chronically infected with rubella virus. The same virus-free cell line served as control target cells. A concentration of 2 • 106 infected or control target cells were labeled with 100 /xCi of sodium 51Cr-chromate ( N a 2 51CRO4; CJS-1P; Amersham/Searle, Arlington Heights, Ill.), incubated at 37~ for one hour, washed three times, and resuspended in complete RPMI medium. Test cultures consisted of 5 x l0 s lymphocytes and 5 x 103 target cells in 0.2 ml of medium, giving an attacker-to-target cell ratio of 100:1. Additional control cultures consisted of target cells without lymphocytes. Experiments were performed in triplicate. The cultures were incubated at 37~ on a rocker platform and harvested after 18 hours by an apparatus which separated the medium containing the released 51Cr from the reacting cells.8
692
Rola-Pleszczynski et aL
The Journal of Pediatrics May 1975
60
0 r
~,
20
ZlO~
r O
0
40
20~
ce-
O ~ v ,m
0 o
60 l,,e-,,
c=-o PHA 0.1% PHA 0.01% PHA 0.005%
cO.
80 I
5
I000
I
I0
I
15
,I
Fig. 1. Effects of addition of bilirubin to lymphocyte cultures stimulated with optimal (0.1%) and suboptimal (0.01%, 0.005%) concentrations of PHA-P. Lymphocytes from ten adults and ten neonates. The percent release for each cell line was calculated as follows:
Percent release -
(hrs) Fig. 2. Effects of addition of bilirubin (20 mg/dl) to lymphocyte cultures at 0, 6, 12, and 24 hours after the start of incubation with PHA-P (0.005%). Lymphocytes from ten adults and ten neonates.
20
Concentration of bilirubin in cultures rng / 100 rnl
cpm 51Cr released cpm 51Cr released from target cells _spontaneously from and lymphocytes target cells alone during incubation during incubation total 51Cr 5tCr released at 0 releasable - time incubation
6 12 24 Time of bilirubin addition
x 100
The specific i m m u n e release was obtained by subtracting the percentage release for control cells from that for rubella-infected cells. Viability of lymphocyte and target cells was assessed at the time of harvesting by the trypan blue exclusion method; it was found to be above 90% after 24 hours and above 70% after six days, and was comparable to control samples. F i n a l results, e x p r e s s e d as p e r c e n t i n h i b i t i o n o f response, were calculated by subtracting from 100% the percentage ratio of cpm with bilirubin over cpm without bilirubin. RESULTS The effects of varying concentrations of bilirubin on lymphoproliferative responses to varying concentrations of P H A - P are presented in Table I and Fig. 1. A significant dose-response inhibitory effect of bilirubin on lymphocyte proliferation was found at concentra-
tions of 0.01% and 0.005% of P H A - p (p<0.05), but none at a concentration of 0.1%. Maximal inhibition of 46.5% was detected at bilirubin concentrations of 20 mg/dl (Fig. 1). In order to determine at what phase of lymphocyte transformation the inhibition was most active, a separate set o f e x p e r i m e n t s was p e r f o r m e d in w h i c h bilirubin was added at 0, 6, 12, and 24 hours after the start of the incubation with 0.005% P H A - P (Fig. 2). Maximal inhibition was observed when bilirubin was added at the start of the reaction and was less marked when bilirubin was added 6 or 12 hours !ater; no inhibition was observed when bilirubin was added after 24 hours. The reversibility of the action of bilirubin was measured by incubating lymphocytes with bilirubin at a concentration of 20 mg/dl for 15 or 30 minutes, washing t h e m free of bilirubin, and then adding 0.005% P H A - P after varying periods of recovery (Fig. 3). No significant inhibition was noted after 15 minutes of initial incubation throughout 240 minutes of recovery; in contrast, lymphocytes preincubated with bilirubin for 30 minutes were significantly inhibited for 180 minutes, but complete reversibility was observed w h e n the lymp h o c y t e s were allowed to re'cover for 240 m i n u t e s . Following more prolonged incubation times of 24 and 72 hours, the inhibitory effects of bilirubin were protracted and remained irreversible ~ven after a recovery period of 72 hours (Table II). The effects of varying concentrations of bilirubin on m i x e d l y m p h o c y t e c u l t u r e r e s p o n s i v e n e s s are presented in Table I and Fig. 4. A significant dose-response effect of bilirubin was again seen; inhibiton, however, was more marked than with mitogenic stimulation with
Volume 86 Number 5
Bilirubin effects on cellular immune responses
x
I
N
15' initial incubation
o-o 30' initial incubation
/
I
0
693
I
15 30
I
60
120
I
I
180
240
'
Recovery time (min)
Fig. 3. Effects of varying periods of incubation with bilirubin (20 mg/dl) and recovery time on lymphoproliferative responses to PHA-P (0.005%). Lymphocytes from ten adults and ten neonates. Table II. Effects of varying periods of incubation with bilirubin (20 mg/dl) and recovery time on lymphoproliferative responses to P H A - P (0.005%), expressed in cpm + SD Reco very period (hr) Incubation with bilirubin (hr)
0 24 72
6
11,210 _--_1,215 5,412 + 640 NT
24
48
13,760 -e 1,611 6,697 • 7,233 2,955 • 342
12,015 • 1,466 5,140 • 697 2,161 • 288
72
14,172 + 1,558 7,072 • 774 2,818-• 375
NT = not tested. PHA-P. Inhibition was detected at bilirubin concentrations as low as 2 mg/dl, the lowest concentration tested, and a maximal inhibition of 80.1% was observed at 20 mg/dl (Fig. 4). A separate set of experiments was performed to compare the inhibitory effects of bi!irubin on lymphocytes from adults and newborn infants (Fig. 5). A n inhibitory effect was observed with iymphocytes from adults and from term and premature infants with concentrations of bilirubin as low as 5 mg/dl. A significant increase in inhibition was seen at higher concentrations of bilirubin. No significant differences were observed in lymphoproliferative responses between lymphocytes from adults and term neonates; however, at concentrations of bilirubin of 5 mg/dl, lymphocytes from three premature infants (<32 weeks of gestation) showed a more marked inhibition (p<0.05) than those from term infants or adults. Higher concentrations of bilirubin failed to reveal any significant difference between the groups studied. The effects of b i l i r u b i n on l y m p h o c y t o t o x i c responses tO rubella virus were studied employing a 51Cr-
release assay and are presented in Table III. No reduction in specific c e l l - m e d i a t e d c y t o t o x i c i t y to r u b e l l a virus could be detected by addition of bilirubin to the test system. DISCUSSION The results of the present studies indicate that unconjugated bilirubin has an inhibitory effect on the lymphoproliferative responses to P H A - P and on responsiveness to mixed lymphocyte culture, no effect, however, was demonstrated on the killer function of lymphocytes as measured by specific lymphocytotoxicity. These results can be explained on the basis of either a direct toxicity of bilirubin or on an inhibitory effect on some intermediary metabolic step. The direct toxicity of unconjugated bilirubin is well known 1~ and cellular viability is rapidly affected upon exposure to it. Toxicity is nevertheless prevented, if bilirubin is bound to albumin; it is diminished, if albumin is added a short time after bilirubin. ~~ Since cellular viability was not affected in our assays and since specific lymphocytotoxicity was not impaired by bilirubin, it seems quite
694
Rola-Pleszczynski et al.
The Journal of Pediatrics May1975
20 c" O
~
40
g+l
60 ::~
m
80-
I00
q)
--
I 5
~ 10
t 15
I 20
Concentration of bilirubin in cultures mg / 100 m l Fig. 4. Effects of varying concentrations of bilirubin on responsiveness to mixed lymphocyte culture. Lymphocytes from ten adults and ten neonates. Table III. Effect of addition of bilirubin to c u l t u r ~ of lymphocytes and target cells
Donors
Neonates (cord blood) Adults
Concentration of Specific immune bilirubin (mg/dl) release(%)*
0 0 5 10 15 20
0 24.2 • 4.1]" 21.5 • 3.8 28.6 _+ 6.1 25.5 • 4,4 21.0 • 4.1
*Mean • SD. tNo statistically significant difference with varying concentrations of bilirubin.
unlikely that our results can be explained by direct toxicity. Furthermore, the albumin content of fetal bovine serum that was used in preparing the culture m e d i u m is sufficient to bind approximately 96% of the bilirubin ~3 and hence prevent its direct toxicity. It is estimated that in the plasma of an infant with unconjugated hyperbilirubinemia, 98-99% of the bilirubin is normally bound to albumin, up to bilirubin levels of approximately 20 mg/dl. 13'14 Bilirubin may exert its effect through some interm ediary metabolic step involved in the lymphoproliferative response. The strong affinity of bilirubin for lipids
m a y m o d i f y t h e s u r f a c e s t r u c t u r e of l y m p h o c y t e m e m b r a n e s and thus conceivably inhibit their function a n y w h e r e in t h e s e q u e n c e of e v e n t s f r o m initial stimulation to final blastic transformation. In the present studies bilirubin curtailed the stimulatory effects of specific antigens, such as the surface ones o f all~geneic l y m p h o c y t e s , as well as n o n s p e c i f i c mitogens such as PHA-P. Since a significant degree of inhibition was observed at 6 and 12 hours of incubation, some alteration of subsequent metabolic steps may be involved. The inhibitory effect of bilirubin may also be the result of respiratory inhibition of mitochondria, as suggested by the studies of Noir and associates who r e p o r t e d that b i l i r u b i n is a m u l t i s i t e i n h i b i t o r o f mitochondrial respiratory metabolism. 10 The inhibitor effects of bilirubin could also be a reflection of some interaction with lysosomat m e m b r a n e s since these are k n o w n to be a c t i v a t e d in t h e l y m p h o p r o l i f e r a t i v e response. 15 The present studies extend the observations of De Sanctis and associates 7 who showed a morphologic in, hibition of blastogenesis in three hyperbilirubinemic infants. In these latter studies the measurement of inhibited blastogenesis in hyperbilirubinemia may have been biased by the use of whole serum as the source of bilirubin, thus introducing the possibility of interference by other s e r u m factors. The increased sensitivity of lymphocytes from premature infants to bilirubin, at least with low concentrations of this substance, may have clinical implications upon subsequent immunologic development. The studies of Steele and associates, 16 for example, have shown that even after one year of life, premature infants have a delay in the development o f specific cell-mediated immunity to rubella virus after immunization. The results of the present studies thus suggest the importance of prevention and prompt correction of hyperbilirubinemia in the newborn infant, and especially in the premature infant, whose levels of hyperbilirubinemia are c o m m o n l y h e i g h t e n e d and m o r e p r o t r a c t e d . If t h e results of the present in vitro studies can be extrapolated to the infant, even so-called "physiologic" icterus of up to 12 mg/dl of bilirubin may present a hazard by decreasing significantly the ability of lymphocytes to respond to extrinsic antigenic stimulation. Several studies have shown a long-term deficit in antibody production following immunization in infants with neonatal hyperb i l i r u b i n e m i a . 46 - The i r r e v e r s i b i l i t y of t h e action of bilirubin after an exposure of 24 hours or more is an ind i c a t i o n of t h e effects t h a t h y p e r b i l i r u b i n e m i a of r e l a t i v e l y s h o r t d u r a t i o n m a y h a v e on t h e i m m u n e system.
Volume 86 Number 5
Bilirubin effects on cellular immune responses
• |
[ ] Adults 9 Full-terminfants
20--~'~
[ ] Premature infants
695
tO
A 4 0 - -
"6 c" O o_
+I
60-"-i (a'3
80--
I00 0.
5
i0
15
20
Concentration of bilirubin mg/ 100 ml Fig. 5. Effects of varying concentrations of bilirubin on lympho-proliferative responses to 0.005% PHA-P of lymphocytes from adults and neonates.
The initial step in the cytotoxicity assay involves activation of lymphocytes by infected target ceils and could be considered a part of the afferent limb of the i m m u n e response, with the efferent "killer" function closely linked. The lack of interference by bilirubin with lymphocyte cytotoxicity indicates that not every part of the afferent limb is affected by bilirubin and is also another example of the functional dissociation of the afferent and efferent limbs of immunity, as has been reported by Rocklin and associates) 7-18and Epstein and associates. 19 Nevertheless, any substance which interferes with the initial steps of i m m u n e recognition may well c o m p r o m i s e t he w h o l e m e c h a n i s m of i m m u nologic responsiveness. The present studies thus lend further support for the clinical importance of monitoring and early treatment of the hyperbilirubinemic infant. We are indebted to Dr. Chantal Beaudet-Pleszczynska and Dr. Yee Hing Thong for their invaluable encouragement and counsel during this study, and to Mrs. Mary Jean Navaretta for her secretarial assistance. REFERENCES
1. Bowen WR, and Waters WJ: Bilirubin encephalopathy. Studies related to the site of inhibitory action of bilirubin on brain metabolism, Am J Dis Child 93:21, 1957. 2. Odell GB: Studies in kernicterus. The protein binding of bilirubin, J Clin Invest 38:823, 1959. 3. Bernstein J, and Landing BH: Extraneural lesions associated with neonatal hyperbilirubin and kernicterus, Am J Pathol 40:371, 1962.
4. Ansaldi N, Fiandino G, and Ciriotti G: Changes of the immunoglobulin level in hyperbilirubinemic premature infants, Minerva Pediatr 20:1982, 1968. 5. Nejedla Z: The development of immunological factors in infants with hyperbilirubinemia, Pediatrics 45:102, 1970. 6. Diaz de Castillo E: Neonatal hyperbilirubinemia. Its relation to the immunological response capacity of the newborn infant, Gac Med Mex 105:185, 1973. 7. De Sanctis C, Malandra C, Zanetti P, Fabris C, and Ponzone A: Neonatal hyperbilirubinemia and response of lyrnphocytes to phytohemagglutinin, Minerva Pediatr 20:2010, 1968. 8. Steele RW, Hensen SA, Vincent MM, Fucillo DA, and Bellanti JA: A 51Cr microassay technique for cell-mediated immunity to viruses, J Immunol 110:1502, 1973. 9. Thorsby E: Cell specific and common antigens on human granulocytes and lymphocytes demonstrated with cytotoxic heteroantibodies, Vox Sang 13:194, 1967. 10. Noir BA, Boveris A, Garaza-Pereira AM, and Stoppani ADM: Bilirubin: a multi-site inhibitor of mitochondrial respiration, FEBS Lett. 27:270, 1972. 11. Cowger ML: Mechanisms of bilirubin toxicity on tissue culture cells: factors that affect toxicity, reversibility by albumin and comparison with other respiratory poisons and surfactants, Biochem Med 5:1, 1971. 12. Tuilie M, and Lardinois R: The binding of unconjugated bilirubin by human sera and purified albumins, Biol Neonate 21:447, 1972. 13. Clarenburg R, and Barnhart JL: Interaction of serum albumin and bilirubin at low concentrations, Am J Physiol 225:493, 1973. 14. Schmid R: Bilirubin metabolism in man, N Engl J Med 287:703, 1972. 15. Hirshchorn K and Hirshchorn R: Role of lysosomes in the lymphocyte response, Lancet 1:1046, 1965.
696
Rola-Pleszczynski et al.
16. Steele RW, Hensen SA, Vincent MM, Fucillo DA, and Bellanti JA: The development of specific cellular and humoral immune responses in children immunized with live rubella virus vaccine, J Infect Dis 130:499, 1974. 17. Rocklin RE, Reardon G, Sheffer A, Churchill WH, and David JR: Dissociation between two in vitro correlates of delayed hypersensitivity: absence of migration inhibition factor (MIF) in the presence of antigen-induced incorporation of 3 H-Thymidine, in Harris, JE, editor: Proceed-
The Journal of Pediatrics May 1975
ings of the Fifth Leukocyte Culture Conference New York, 1970, Academic Press, Inc., p 639-648. 18. Rocklin RE: Production of migration inhibitory factor by non-dividing lymphocytes, J Immunol 110:674, 1973. 19. Epstein LB, Stevens DA, and Merigan TC: Selective increase in lymphocyte interferon response to vaccinia antigen after revaccination. Proc Nat Acad Sci USA 69:2632, 1972.