Vous êtes sur la page 1sur 4


, 2000
© 2000 by Am. Coll. of Gastroenterology ISSN 0002-9270/00/$20.00
Published by Elsevier Science Inc. PII S0002-9270(99)00813-8

Probiotics and Immune Response

Susanna Cunningham-Rundles, Ph.D., Siv Ahrné, Ph.D., Stig Bengmark, M.D., Ph.D,
Rosemary Johann-Liang, M.D., Florence Marshall, M.D., Linda Metakis, B.S., Claudia Califano, B.A.,
Ann-Margaret Dunn, R.N., C.P.N.P., Claudia Grassey, R.N., C.P.N.P., Gilberto Hinds, R.N., and
Joseph Cervia, M.D.
Cornell Medical Center, New York, New York

ABSTRACT evaluate and to explore the possible significance of probiotic

Current evidence supports the concept that oral administra- bacteria and how some of these organisms may be used
tion of probiotic lactobacilli may be therapeutic in prevent- therapeutically (6, 7).
ing antibiotic-associated diarrhea in children and in reestab- The present study was undertaken to determine whether
lishing normal flora in the gastrointestinal tract. Children dietary supplementation with a well-characterized probiotic,
with human immunodeficiency virus (HIV) infections may Lactobacillus plantarum 299v (DSM 9843) (8), might be
have episodes of diarrhea and frequently experience malab- able to colonize the GI tract and improve growth in children
sorption associated with possible bacterial overgrowth; to- congenitally exposed to human immunodeficiency virus
gether these may interact to produce the growth abnormal- (HIV) infection who display clinical evidence of growth
ities characteristic of this group. The overall objective of failure. Although growth failure is likely to be multifactorial
this investigation has been to determine whether oral ad- in HIV disease, malabsorption and altered GI milieu are
ministration of the probiotic Lactobacillus plantarum 299v important contributing factors (9, 10). A recent study has
could improve nutrient status and promote growth in chil- shown that lactobacillus administration confers intestinal
dren congenitally exposed to HIV. In addition, the possible resistance to the opportunistic pathogen Cryptosporidium
beneficial effect of Lactobacillus plantarum 299v in mod- parvum in immunosuppressed mice (11). Preliminary data
ulating immune response was evaluated. In preliminary are presented here on immune response to L. plantarum
results described here, we report on the ability of Lactoba- 299v after oral supplementation, and we address potential
cillus plantarum 299v to colonize children with HIV and to safety issues in these profoundly immune-deficient children.
elicit specific systemic immune response after oral
supplementation. (Am J Gastroenterol 2000;95(Suppl.):
S22–S25. © 2000 by Am. Coll. of Gastroenterology)
Children were enrolled into the study after informed consent
INTRODUCTION of the parent, in accordance with the protocol (approved by
the Cornell University Medical College Internal Review
The development of an immune response toward foreign
Board). The first child to be studied received a juice formula
antigens of microbial and viral origin provides an essentially
systematic way through which potentially pathogenic organ- (12) prepared by the Skånemejerier Dairy in Lund Sweden,
isms are recognized and controlled. Although the gastroin- whereas subsequent children received packets containing
testinal (GI) tract contains many prokaryotic organisms, either L. plantarum 299v (8, 13) or an identical placebo
which, based on cell number alone far outnumber all of the prepared as a lyophilized powder in an oatmeal base. The
cells of the body, healthy blood is sterile, and deep tissues lyophilized preparations were given by the parent to the
operate in the absence of bacteria. This environment is child either orally by mixing the powder into a beverage, in
achieved and maintained by the physical structure of the GI formula once per day or, in one case, into the percutaneous
tract, as well as the activity of the immune system in enteral feeding tube. Colonization was not evaluated in the
preventing bacterial overgrowth and translocation (1, 2). first case, whereas in the second phase of study colonization
Whereas some microbes tend to produce toxins or com- was directly tested by culture of rectal swabs onto Rugosa
pete with the host for common substrates, others produce agar plates, a medium selective for lactobacilli. Lactobacilli
nutrients, absorbable peptides, and vitamins essential for the that grew out were subsequently genetically typed for pres-
host (3, 4). Furthermore, colonization of the GI tract with ence of L. plantarum 299v. Antigenic recognition of this
beneficial microbes may prevent overgrowth from more lactobacilli was evaluated as proliferative response to heat-
pathogenic ones if these bacteria have a selective advantage inactivated whole organisms of L. plantarum 299v in a
(5). In recent years the science of probiotics has emerged to standardized microtiter plate system (14, 15).
AJG – January, Suppl., 2000 Probiotics and Immune Response S23

Figure 1. Weight change in response to L. plantarum 299v in congenital acquired immune deficiency.

RESULTS appeared after about 1 month or less after the end of sup-
The first congenitally HIV-infected child to be given L. Mononuclear cells isolated from peripheral blood from 17
plantarum 299v was 11 yr of age at the time of study and children congenitally exposed to HIV were evaluated for
received juice containing L. plantarum 299v every day for 1 immune response in vitro to L. plantarum 299v before
month. At the beginning of the study, he was evaluated as
supplementation. Positive response was defined as mean
having failure-to-thrive, observed to have persistent mouth
(counts per minute of triplicate wells) of maximum response
ulcers associated with candidal esophagitis, and had chronic
to any concentration of the L. plantarum 299v antigen ⱖ3.5
diarrhea. This child was a fussy eater, unable to finish meals,
times the background response of unstimulated cells cul-
and was quoted by his grandmother as saying that he “would
tured in parallel. Ten children were evaluated as having
like to kill myself if I were not afraid to die.” He had virtual
positive response and seven children were considered non-
absence of CD4⫹ T cells in peripheral blood and was
responsive. The mean difference in response between re-
receiving azidothymidine (AZT). After 1 month of treat-
sponders and nonresponders was significantly different (Fig.
ment with L. plantarum 299v, his growth improved (Fig. 1),
2) and validated the use of this differentiation of response
and his mouth ulcers, candidiasis, and diarrhea resolved.
The child’s appetite improved markedly; he ate three full from background proliferation for this particular antigen in
meals each day, and he was able to enjoy playing again. The the setting of HIV infection. Responders to L. plantarum
improvement in weight was statistically significant if the 299v did not have a statistically different higher response to
two pretreatment values were compared either with the first phytohemagglutinin (PHA, a standard T-cell mitogen used as
three posttreatment values (p ⬍ 0.03) over 34 days or over a positive lymphocyte activator) than did those who did not
142 days (p ⬍ 0.001). This dramatic response in one child respond (data not shown). This result suggests that immune
led to the second phase of study, which was undertaken as deficiency alone was not the reason for lack of response.
a double-blind, placebo-controlled study of L. plantarum Included among the group of nonresponsive children
299v supplementation in congenital HIV infection. were two children who also did not respond to PHA. One of
Of 17 children who were enrolled into the study, no child these, a child with marked growth failure born to an HIV-
was colonized with L. plantarum 299v before receiving the positive mother, was subsequently found to be HIV-nega-
supplement orally, and no child who did not receive active tive by polymerase chain reaction (PCR) and eventually
product became colonized with L. plantarum 299v. Approx- showed seroreversion of the previously positive HIV anti-
imately 50% of swab cultures from these children grew out body reactivity. This child did show overall improvement in
a few lactobacilli representing species other than L. planta- height and weight while receiving L. plantarum 299v, but
rum 299v. this could not be directly attributed to L. plantarum 299v
All of the children tolerated the supplement, and no child because of inadequate follow-up visits and limited base-
was withdrawn from the study because of side effects or line data close to the time of administration. Interestingly,
problems with taste. Although flatulence has sometimes a marked augmentation in immune response was ob-
been reported with use of L. plantarum 299v in adults, this served (Fig. 3). This difference reflected a change from gen-
was not observed in the present study. Colonization ap- eralized anergy to normal immune response within 1 month
peared to take place within 2 wk. Although it was not and occurred in the absence of any other intervention or treat-
evaluated extensively, L. plantarum 299v colonization dis- ment.
S24 Cunningham-Rundles et al. AJG – Vol. 95, No. 1, Suppl., 2000

Figure 2. Natural immune response to L. plantarum in congenital HIV exposure.

DISCUSSION effects mediated through modulation of immune response

after supplementation with L. plantarum 299v. This finding
Therefore, preliminary data from this study indicate that is indirectly supported by other studies in volunteers show-
approximately 69% of HIV-positive children who have ing that oral supplementation with different lactobacilli did
never received L. plantarum 299v do have significant pre- act to modulate and increase phagocytic activity in periph-
existing immune response to this bacteria, probably due to eral blood (16). Although there was no effect on lymphocyte
cross-reactivity with other endogenous strains of L. planta- response to mitogens in this study, this may not be surpris-
rum or with strains of different lactobacilli species. The ing as immune response was already in the normal range in
observation that immune response to L. plantarum 299v was these healthy adults. These findings may indicate a potential
not directly related to response to PHA is important for two for therapeutic use of lactobacilli in HIV infected children.
reasons. First, absence of relationship suggests that lack of
response was not a simple reflection of immune deficiency.
Second, because antigen response is often lost early in HIV SUMMARY AND CONCLUSIONS
infection, the studies indicate that response to L. plantarum
299v may be specifically retained in HIV-positive children. This study examined the effect of L. plantarum 299v sup-
As noted, all children tolerated this supplementation, there plementation in children congenitally exposed to HIV. Re-
were no side effects, and children were not permanently sults show that 69% of HIV-positive children had preexist-
colonized. ing cross-reactive immune response to this antigen as
These preliminary results suggest L. plantarum 299v may measured by proliferative response to L. plantarum 299v
have the potential to improve growth and development. before oral administration. No child was colonized with L.
Furthermore, the data in one child indicate the possibility of plantarum 299v at baseline, as shown by genetic typing of

Figure 3. Immune response after L. plantarum supplementation in HIV-exposed child with failure to thrive.
AJG – January, Suppl., 2000 Probiotics and Immune Response S25

lactobacilli grown from rectal swabs. Mononuclear cells 6. Bengmark S, Jeppsson B. Gastrointestinal surface protection
from 10 confirmed HIV-positive children responded to L. and mucosa reconditioning. J Parenter Enteral Nutr 1995;19:
410 –5.
plantarum 299v in vitro. Seven children did not react. Re-
7. Majamaa H, Isolauri E. Probiotics: A novel approach in the
sponders had a significantly higher response to L. plantarum management of food allergy. J Allergy Clin Immunol 1997;
299v than did their nonreactive counterparts, but they did 99:179 – 85.
not have a significantly stronger response to PHA. These 8. Johansson ML, Molin G, Jeppsson B, et al. Administration of
results demonstrate both that natural immune response to L. different Lactobacillus strains in fermented oatmeal soup: In
plantarum 299v occurs frequently and that this response is vivo colonization of human intestinal mucosa and effect on the
indigenous flora. J App Environ Microbiol 1993;59:15–20.
highly conserved in the HIV-positive child. The data sug-
9. Crocker KS. Gastrointestinal manifestations of the acquired
gest that L. plantarum 299v may be given safely to the immunodeficiency syndrome. Nurs Clin North Am 1989;24:
immunocompromised host and may indeed have a positive 395– 406.
effect on immune response. 10. Kotler DP, Tierney AR, Culpepper-Morgan JA, et al. Effect of
home total parenteral nutrition on body composition in pa-
tients with acquired immunodeficiency syndrome. J Parenter
Reprint requests and correspondence: Susanna Cunningham-
Enteral Nutr 1990;14:454 – 8.
Rundles, Ph.D., Cornell Medical Center, 525 E. 68th Street, New
11. Alak JI, Wolf BW, Mdurvwa EG, et al. Effect of Lactobacillus
York, NY 10021.
reuteri on intestinal resistance to Cryptosporidium parvum
Received Mar. 17, 1999; accepted Sep. 7, 1999.
infection in a murine model of acquired immunodeficiency
syndrome. J Infect Dis 1997;175:218 –21.
12. Johansson ML, Berggren A, Nobaek S, et al. Survival of
REFERENCES Lactobacillus plantarum DSM 9843, and effect on the short
1. Prindull G, Ahmad M. The ontogeny of the gut mucosal chain fatty acid content in feces after ingestion of a rose hip
immune system and the susceptibility to infections in infants drink with fermented oats. Int J Food Microbiol 1999 (in
of developing countries. Eur J Ped 1993;152:786 –92. press).
2. Cunningham-Rundles S. Malnutrition and gut immune func- 13. Johansson ML, Quednau M, Ahrne S, et al. Classification of
tion. Curr Op Gastroenterol 1994;10:664 –70. Lactobacillus plantarum by restriction endonuclease analysis
3. Murry AC, Lewis RD, Amos HE. The effect of microbial of total chromosomal DNA, using conventional agarose gel
phytase in a pearl millet-soybean meal diet on apparent di- electrophoresis. Int J Syst Bacteriol 1995;45:670 –5.
gestibility and retention of nutrients, serum mineral concen- 14. Cunningham-Rundles S, Yeger-Arbitman R, Edelson P, et al.
tration, and bone mineral density of nursery pigs. J Anim Sci Experimental approach to the study of immune function in
1997;75:1284 –91. children with possible human immunodeficiency virus, HIV,
4. Torrallardona D, Harris CI, Coates ME, et al. Microbial amino infection. J Clin Lab Analysis 1990;4:399 – 404.
acid synthesis and utilization in rats: Incorporation of 15N 15. Paxton H, Cunningham-Rundles S, O’Gorman MRG. Labo-
from 15NH4Cl into lysine in the tissues of germ-free and ratory evaluation of the cellular immune system. In: Henry JB,
conventional rats. Br J Nutr 1996;76:689 –700. ed. Clinical diagnosis and management by laboratory methods.
5. Drago L, Gismondo MR, Lombardi A, et al. Inhibition of in Philadelphia: WB Saunders and Co, 1995:887–912.
vitro growth of enteropathogens by new Lactobacillus isolates 16. Schiffrin EJ, Rochat F, Link-Amster H, et al. Immunomodu-
of human intestinal origin. FEMS Microbiol Lett 1997;153: lation of human blood cells following the ingestion of lactic
455– 63. acid bacteria. J Dairy Sci 1995;78:491–7.