Does sphingosine 1-phosphate play a protective role
in the course of pulmonary tuberculosis?
Sanjay K. Garga,1,2, Marilina B. Santuccia,2, Miriam Panittia, Leo Pucillob,
Marialuisa Bocchinoc,3, Fumikazu Okajimad, Prakash S. Bisene,
Cesare Saltinic,4, Maurizio Frazianoa,*
aDepartment of Biology, University of Rome “Tor Vergata” Via della Ricerca Scientifica-00133, Rome, Italy
bLaboratory of Clinical Pathology, National Institute for Infectious Diseases (INMI) “L. Spallanzani-IRCCS”, Rome, Italy
cDivision of Respiratory Medicine of the University of Rome “Tor Vergata” at the INMI “L. Spallanzani-IRCCS”, Rome, Italy
dInstitute for Molecular and Cellular Regulation, Gunma University, Maebashi, Japan
eInstitute of Biotechnology and Allied Sciences, Seedling Academy of Design, Technology and Management, Korebariyan,
Jagatpura, Jaipur, India
Received 22 May 2006; accepted with revision 7 September 2006
Available online 16 October 2006
terial activity in vitro and in a mouse model of in vivo Mycobacterium tuberculosis infection.
However, its role in the course of pulmonary tuberculosis in humans is still not known. This study
shows that S1P levels in airway surface fluid of tuberculosis (TB) patients are significantly less
than those observed in non-TB control patients. Moreover, the in vitro stimulation of
bronchoalveolar lavage cells coming from TB patients with S1P significantly reduces intracellular
growth of endogenous mycobacterial isolates. These results show that, in the course of
pulmonary TB, airway epithelial fluid-associated S1P may play a protective role in the
containment of intracellular mycobacterial growth and that its decrease may represent a novel
pathogenic mechanism through which M. tuberculosis favors its replication.
© 2006 Elsevier Inc. All rights reserved.
Sphingosine 1-phosphate (S1P) has recently been reported to induce antimycobac-
Tuberculosis is a global disease causing more than 2 millions
deaths annually . Approximately one-third of the world
population is infected with the causative agent, Mycobac-
terium tuberculosis, and thus is at risk of developing active
disease during lifetime. Primary infection occurs when
aerosol-droplet nuclei containing a small number of bacilli
are deposited in the alveoli of the lung and subsequently
phagocytosed by alveolar macrophages. During these early
* Corresponding author. Fax: +39 06 72594224.
E-mail address: email@example.com (M. Fraziano).
1Present address: Department of Biochemistry, University of
Nebraksa, Lincoln, USA.
2Both authors contributed equally to the manuscript.
3Present address: Department of Clinical and Experimental Medi-
cine, School of Biotechnology, University of Naples “Federico II”,
4Present address: Department of Internal Medicine, University of
1521-6616/$ – see front matter © 2006 Elsevier Inc. All rights reserved.
available at www.sciencedirect.com
Clinical Immunology (2006) 121, 260–264
stages of host pathogen interaction, the possibility to get a
latent infection or active disease infection crucially depends
by the intensity of early innate immune response and by its
interplay with adaptive immune response . Pulmonary
innate immune response involves several cell types (mostly
macrophages, neutrophils and epithelial cells) and several
host antimicrobial factors, present in airway epithelial lining
fluid, which are able to kill directly invading microbes .
These factors include antimicrobial proteins, such as lyso-
1 (HBD-1), HBD-2, LL-37 (a cathelicidin), human neutrophil
In this context, recently reported results highlighted the
possible contribution of some bioactive lipids in the enhance-
ment of antimycobacterial activity of macrophages .
Sphingosine 1-phosphate (S1P) is an important lipid
mediator that has been implicated in many biological
processes and has been suggested to be a unique signaling
molecule due to its both intracellular and extracellular mode
of action . Extracellularly, S1P is prevalently associated
with serum HDL  and can signal to the target cells through
endothelial differentiation gene (EDG) receptors . After
binding with cognate receptors, S1P elicits a wide range of
cell migration in different cell types . We have recently
reported that S1P enhances antimycobacterial activity in
which in turn promotes phagolysosome maturation .
Moreover, the treatment of M. tuberculosis-infected mice
significantly reduced granulomatous response in the lung,
with a prevalent macrophage infiltrate without evidences of
necrosis, suggesting that S1P can reduce tissue damage from
the T-cell-mediated immune response while increasing the
local innate antimicrobial activity . Interestingly, the
liver, testis, brain, intestine, heart, muscle and spleen,
showed that the lung contained the highest amount of S1P
(29.11±1.44 nmol/g organ) , suggesting its possible
involvement in the induction of innate antimicrobial immune
response at the interface with external airway environment.
On these grounds, we asked whether S1P content in the
pulmonary epithelial lining fluid is modulated during the
course of pulmonary tuberculosis. We also asked if, as in the
experimental animal , lung cells from tuberculosis
patients are susceptible to S1P-mediated reduction of
intracellular mycobacterial growth. Data reported herein
show that S1P level on the alveolar surface of the lung in
tuberculosis patients is significantly lower than that
observed in patients with lung disorders other than tubercu-
losis. Further, S1P stimulation of lung cells obtained ex vivo
from tuberculosis patients is capable of inhibiting intracel-
lular growth of “endogenous” mycobacteria.
Materials and methods
Nine patients with culture-confirmed pulmonary tuberculosis
(age 31.8±8.1, 3 males, 6 Caucasians) and eleven non-
tuberculosis-related lung disease control patients (age 63.1
±12.9, 5 males, 11 Caucasians) whose final diagnosis were
interstitial lung disease (1 case), lung cancer (3 cases),
community-acquired pneumonia (2 cases), acute bronchitis
(1 case), interstitial pneumonia (1 case), lung abscess (1
case) and 2 cases with isolated hemoptysis (with no evidence
of both lung and bronchial involvement) were enrolled in the
study. Bronchoalveolar lavage was carried out as previously
described , before the initiation of anti-mycobacterial
treatment. All patients gave informed consent under a study
protocol approved by the institutional Ethics Committee.
Sphingosine 1-phosphate quantification in BAL fluid
Sphingosine 1-phosphate quantification analysis was carried
out, as previously described  with small modifications, on
BAL fluid (BALF) coming from TB as well as control patients.
BALF was filtered through 100 μm sterile gauze, centrifuged
at 1300 RPM for 10 min and supernatant collected and stored
at −80°C until use. In order to avoid interferences between
transfected CHO cells and incubated on ice for 30 min. Cells
were then washed twice with binding buffer consisting of
20 mM Tris–HCl (pH 7.5), 100 mM NaCl, 15 mM NaF, 0,4% BSA,
incubated for 30 min on ice with 5 nM of [3H]S1P (American
Radiolabeled Chemicals, Inc., St. Louis, MO) and washed
again with binding buffer to remove unbound ligand. Finally,
cells were solubilized with a solution composed by 0.1% SDS,
0.4% NaOH and 2% Na2CO3and radioactivity was counted by a
liquid scintillation counter (Microbeta Trilux, Wallac). Com-
parative analysis of S1P content in BAL samples coming from
two different groups has been preliminarily estimated by
calculating the percentage of binding inhibition of S1P
present in BAL samples with 5 nM radiolabeled S1P through
the following formula: cpm obtained in absence of competi-
tion−cpm obtained in the presence of sample/cpm obtained
in absence of competition−cpm obtained in the presence of
1000 nM S1P. Then, S1P concentration was estimated by
from 1 to 103nM) in place of BALF sample. Exponential
regression coefficient of standard replacement curve was
in BALF was recalculated as S1P epithelial lining fluid
concentration by the urea method , by multiplying the
value obtained by extrapolation from the standard displace-
ment curve with the dilution factor calculated as the ratio
Determination of urea in BALF and in plasma has been carried
confirmed by the absence of competition detected by using
sphingosine in place of sphingosine 1-phosphate at the same
range of concentrations (see additional data).
Quantification of intracellular endogenous
mycobacteria in bronchoalveolar lavage
cells from tuberculosis patients
The assay was carried out on bronchoalveolar lavage (BAL)
cells coming from further four subjects with culture-
confirmed pulmonary tuberculosis enrolled before the
initiation of anti-mycobacterial treatment (age 30.7±12.2,
2 males, 2 Caucasians). Briefly, BAL cells were suspended in
complete medium consisting of RPMI 1640 supplemented
with 10% FBS, 2 mM L-Glu, 1 mM non-essential amino acids,
1 mM sodium piruvate, 5 μg/ml ampicillin and 2 μg/ml
fluconazole (all by Invitrogen, Milan, Italy). Finally, 3×106
cells per well were incubated for 48 h in 6-well plates in the
presence or absence of 5 μM D-erythro sphingosine 1-
phosphate (Calbiochem, San Diego, CA) and analyzed at
the time indicated by colony forming unit assays, as
previously described . Any modification, in terms of cell
viability, was not detected in the course of in vitro S1P
stimulation of BAL cells (data not shown) as well as of human
primary macrophages .
Statistical analysis was carried out by Graphpad Prism 3.0
software package. Comparison between groups was done
using Student's t test as appropriate for normally distributed
data. Two tailed Mann–Whitney test was performed for data
that were not normally distributed. p<0.05 was considered
In order to analyze the role of S1P in the pulmonary
antimycobacterial immune response, we determined the
amount of sphingosine 1-phosphate in pulmonary epithelial
lining fluid of tuberculosis and non-tuberculosis disease
control patients. Firstly, the percentage of binding inhibition
of [3H]S1P with EDG1 transfected CHO cells by bronchoal-
veolar lavage fluid samples was calculated. Strikingly,
bronchoalveolar lavage fluid samples from patients with
disorders other than tuberculosis competed with radiola-
beled S1P with higher efficiency (mean±SE: 74.3±7.9% of
binding inhibition) than that observed by bronchoalveolar
lavage fluid samples from tuberculosis patients (mean±SE:
48.2±7.2% of binding inhibition, p=0.01 by two tailed Mann–
Whitney test) (Fig. 1). Then, S1P content was estimated by
extrapolation from a standard displacement curve. Since
urea content was under the threshold of detectability in
eight bronchoalveolar lavage samples, this analysis was
performed only on six non-tuberculosis controls and six
tuberculosis patients. Remarkably, the concentration of S1P
in the epithelial lining fluid from tuberculosis patients was
also significantly decreased in comparison with non-tuber-
culosis disease patients (tuberculosis patients, mean±SE:
47.5± 36.2 nM; non-tuberculosis disease control patients,
mean± SE: 2623±1576 nM; p=0.02 by two tailed Mann–
Whitney test). As the average age of the non-tuberculosis
patient population was higher than that of the tuberculosis
patient population, the percentage of binding inhibition
between [3H]S1P and EDG1 transfected CHO cells by
bronchoalveolar lavage samples from non-tuberculosis
patients was plotted against of age in order to exclude an
augmenting effect of age upon lung S1P levels. In fact, no
significant correlation between S1P levels and age was
found, thus ruling out the possibility that the difference in
S1P levels between tuberculosis and non-tuberculosis
patients could be due to older age of the non-tuberculosis
patients (r2=0.089, p=0,37; see additional data).
We then examined the possibility that S1P might
effectively increase in vivo mycobacterial killing by alveolar
macrophages in tuberculosis patients. To this end, bronch-
oalveolar lavage cells from further four tuberculosis patients
were monitored for intracellular growth of in vivo macro-
phage infecting mycobacteria after 48 h of in vitro S1P
stimulation. As shown in Fig. 2, a significant reduction of
intracellular colony forming units after S1P stimulation was
[3H]sphingosine 1-phosphate (S1P) with EDG1 transfected CHO
cells after previous incubation with bronchoalveolar lavage fluid
coming from non-tuberculosis control (open circles) and
tuberculosis (closed circles) patients. The percentage of binding
inhibition was calculated by using the following formula: cpm
obtained in absence of competition−cpm obtained in the
presence of bronchoalveolar lavage sample/cpm obtained in
absence of competition−cpm obtained in the presence of 1 μM
S1P. p=0.01 by two tailed Mann–Whitney test.
Percentage of binding inhibition of 5 nM radiolabeled
ity by sphingosine 1-phosphate (S1P). Bronchoalveolar lavage
(BAL) cells from four pulmonary tuberculosis patients were
stimulated in vitro with 5 μM S1P. Colony forming unit assay was
performed before the addition and at 48 h after S1P stimulation
as indicated in Materials and methods. Data are expressed as
means±SD of the CFU performed in triplicates. Differences
between S1P-treated (open circles) and non-treated (closed
circles) were compared by Student's t test and p values indicated
in each quadrant.
Enhancement of pulmonary antimycobacterial activ-
262 Rapid Communication
detected in bronchoalveolar lavage cells from all patients
Microbe elimination from the alveolar surface of the lung is
achieved in man by unique mechanism of defense char-
acterized by a multifaceted innate immune system, which
has evolved to protect the lung while minimizing acquired
immune response and local inflammation . In this
context, a number of proteins, peptides and other small
molecules have been identified to play a role in antimicro-
bial activity . Recently published results reported the
efficacy of some selected lipids, enclosing S1P, in the
enhancement of anti-mycobacterial innate immune
defenses [5,9]. In this context, in vivo evidences showed
that S1P treatment of M. tuberculosis-infected mice
significantly reduced mycobacterial burden . Moreover,
S1P has also been described to induce anti-inflammatory
activity that was exerted by sequestering lymphocytes in
lymph nodes  and by shifting adaptive immune response
towards a T helper 2 phenotype . Interestingly, the lung
is one of the described tissues containing the highest amount
of S1P  where it may play a protective role against
inflammatory lung injury [16,17]. Altogether, these evi-
dences suggest that S1P can represent a component of
pulmonary mucosa deputed to the maintenance of alveolar
sterility and to the regulation of tissue damaging inflamma-
tory response. The results reported herein show that S1P
content in airway epithelial lining fluid of tuberculosis
patients is significantly lower than that observed in patients
with lung diseases other than tuberculosis. The data have
been obtained by a radioreceptor binding assay, previously
published to be a specific, easy and versatile method for S1P
quantification in biological samples . Although a number
of other biochemical methods have been reported for S1P
quantification, such as mass spectrometry based [18,19] or
thin layer chromatography-based [20,21] systems, the
bioassay reported herein is particularly feasible for the
management of infected samples and, in particular, for the
mass scale analysis. Moreover, as the radioreceptor binding
assay does not crucially depends from sample pretreatment
or S1P purification, it allowed us to get information
regarding the content of biologically active and available
S1P is prevalently released by platelets, mast cells and
macrophages . Although the precise mechanism for S1P
release in alveolar epithelial fluid is not known, alveolar
macrophages may represent both an important source of S1P
as well as its possible target . Thus, it is plausible to
hypothesize that the reduced pulmonary content reported
herein can be due, at least in part, to the inhibitory activity,
exerted by M. tuberculosis, on macrophage sphingosine
kinase-1, as previously described following in vitro infection
with live, but not dead, M. tuberculosis [24,25]. However,
the possibility that sphingolipids represent a “functional”
constituent of food  and the evidence that malnutrition
may represent both a cause and effect of tuberculosis ,
suggests that the nutritional status of tuberculosis patient
may also play an important additional role in the observed
reduction of S1P content.
The possibility that S1P may play a protective role
during pulmonary anti-mycobacterial response has been
explored by evaluating the capability of S1P to enhance
antimycobacterial activity in bronchoalveolar lavage cells
coming from tuberculosis patients. The results reported
herein show that the incubation of bronchoalveolar lavage
cells with 5 μM S1P reduces intracellular growth of
endogenous mycobacterial isolates. Bronchoalveolar lavage
cells are a mixed cell population representative of that
present in deep lung and comprise prevalently alveolar
macrophages, T lymphocytes and neutrophils. Thus, the ex
vivo S1P-induced reduction of intracellular growth of
mycobacteria may possibly reflect what can occur in vivo
in the course of active disease in humans. On these
grounds, it is plausible to hypothesize a therapeutic value
for S1P in order to enhance pulmonary antimycobacterial
activity by simultaneously minimizing T-cell-mediated
tissue damaging immune response. It is noteworthy that
immunotherapeutic disruption of granulomas has recently
been suggested as a possible adjuvant immunotherapy to
make bacilli more accessible to pharmacological therapy
. In this context, the replacement of S1P in the lung of
tuberculosis patients may represent a therapeutic strategy
that can be associated with conventional antibiotics aimed
to the reduction of time therapy and to the treatment of
The present study was financially supported by the Italian
Ministry of Health (Grant 4AD/F9) and by the Italian Ministry
of University (COFIN 2004, FIRB 2001). G.S.K. was financially
supported by a UNESCO-ROSTE study fellowship.
Appendix A. Supplementary data
in the online version, at doi:10.1016/j.clim.2006.09.002.
 C. Dye, S. Scheele, P. Dolin, V. Pathania, M.C. Raviglione,
Consensus statement. Global burden of tuberculosis: estimated
incidence, prevalence, and mortality by country. W.H.O.
Global Surveillance and Monitoring Project, JAMA 282 (1999)
 S.H. Kaufmann, How can Immunology contribute to the control
of tuberculosis? Nat. Rev., Immunol. 1 (2001) 20–30.
 P. Zhang, W.R. Summer, G.J. Bagby, S. Nelson, Innate
immunity and pulmonary host defense, Immunol. Rev. 173
 S.M. Travis, P.K. Singh, M.J. Welsh, Antimicrobial peptides and
proteins in the innate defense of the airway surface, Curr. Opin.
Immunol. 13 (2001) 89–95.
 E. Anes, M.P. Kuhnel, E. Bos, J. Moniz-Pereira, A. Habermann,
G. Griffiths, Selected lipids activate phagosome actin assembly
and maturation resulting in killing of pathogenic mycobacteria,
Nat. Cell Biol. 5 (2003) 793–802.
 S. Spiegel, S. Milstien, Sphingosine 1-phosphate, a key cell
signaling molecule, J. Biol. Chem. 277 (2002) 25851–25854.
 N. Murata, K. Sato, J. Kon, H. Tomura, M. Yanagita, A.
Kuwabara, M. Ui, F. Okajima, Interaction of sphingosine 1-
phosphate with plasma components, including lipoproteins, Download full-text
regulates the lipid receptor-mediated actions, Biochem. J. 3
 S. Spiegel, S. Milstien, Sphingosine-1-phosphate: an enigmatic
signalling lipid, Nat. Rev., Mol. Cell Biol. 4 (2003) 397–407.
 S.K. Garg, E. Volpe, G. Palmieri, M. Mattei, D. Galati, A.
Martino, M.S. Piccioni, E. Valente, E. Bonanno, P. De Vito, P.M.
Baldini, L.G. Spagnoli, V. Colizzi, M. Fraziano, Sphingosine 1-
phosphate induces antimicrobial activity both in vitro and in
vivo, J. Infect. Dis. 189 (2004) 2129–2138.
 N. Murata, K. Sato, J. Kon, H. Tomura, F. Okajima, Quantitative
measurement of sphingosine 1-phosphate by radioreceptor-
binding assay, Anal. Biochem. 282 (2000) 115–120.
 M.B. Santucci, M. Bocchino, V. Colizzi, C. Saltini, M. Fraziano,
Expansion of CCR5+CD4+T lymphocytes in the course of active
pulmonary tuberculosis, Eur. Respir. J. 24 (2004) 638–643.
 S.I. Rennard, G. Basset, D. Lecossier, K.M. O'Donnell, P.
Pinkston, P.G. Martin, R.G. Crystal, Estimation of volume of
epithelial lining fluid recovered by lavage using urea as marker
of dilution, J. Appl. Physiol. 60 (1986) 532–538.
 A.M. Le Vine, J.A. Whitsett, Pulmonary collectins and innate
immune defense of the lung, Microbe Infect. 3 (2001) 161–166.
 S. Mandala, R. Hajdu, J. Bergstrom, E. Quackenbush, J. Xie, J.
Milligan, R. Thornton, G.J. Shei, D. Card, C. Keohane, M.
Rosenbach, J. Hale, C.L. Lynch, K. Rupprecht, W. Parsons, H.
Rosen, Alteration of lymphocyte trafficking by sphingosine-1-
phosphate receptor agonists, Science 296 (2002) 346–349.
 M. Idzko, E. Panther, S. Corinti, A. Morelli, D. Ferrari, Y. Herouy,
S. Dichmann, M. Mockenhaupt, P. Gebicke-Haerter, F. Di
Virgilio, G. Girolomoni, J. Norgauer, Sphingosine 1-phosphate
induces chemotaxis of immature and modulates cytokine-
release in mature human dendritic cells for emergence of Th2
immune responses, FASEB J. 16 (2002) 625–627.
 B.J. McVerry, X. Peng, P.M. Hassoun, S. Sammani, B.A. Simon, J.
G. Garcia, Sphingosine 1-phosphate reduces vascular leak in
murine and canine models of lung injury, Am. J. Respir. Crit.
Care Med. 170 (2004) 987–993.
 X. Peng, P.M. Hassoun, S. Sammani, B.J. McVerry, M.J. Burne, H.
Rabb, D. Pearse, R.M. Tuder, J.G. Garcia, Protective effects of
Sphingosine 1-phosphate in murine endotoxin-induced inflam-
matory lung injury, Am. J. Respir. Crit. Care Med. 189 (2004)
 P.P. Van Veldhoven, P. De Ceuster, R. Rozenberg, G.P.
Mannaerts, E. de Hoffmann, On the presence of phosphorylated
sphingoid bases in rat tissues. A mass-spectrometric approach,
FEBS Lett. 350 (1994) 91–95.
 N. Mano, Y. Oda, K. Yamada, N. Asakawa, K. Katayama,
Simultaneous quantitative determination method for sphingo-
lipid metabolites by liquid chromatography/ionspray ionization
 Y. Yatomi, F. Ruan, J. Ohta, R.J. Welch, S. Hakomori, Y.Igarashi,
Quantitative measurement of sphingosine 1-phosphate in
biological samples by acylation with radioactive acetic anhy-
dride, Anal. Biochem. 230 (1995) 315–320.
 L.C. Edsall, S. Spiegel, Enzymatic measurement of sphingosine
1-phosphate, Anal. Biochem. 272 (1999) 80–86.
 H. Rosen, E.J. Goetzl, Sphingosine 1-phosphate and its
receptors: an autocrine and paracrine network, Nat. Rev.,
Immunol. 5 (2005) 560–570.
 C. Hornuβ, R. Hammermann, M. Fuhrmann, U.R. Juergens, K.
Rackè, Human and rat alveolar macrophages express multiple
EDG receptors, Eur. J. Pharmacol. 429 (2001) 303–308.
 Z.A. Malik, C.R. Thompson, S. Hashimi, B. Porter, S.S. Iyer, D.J.
Kusner, Mycobacterium tuberculosis blocks Ca2+ signaling and
phagosome maturation in human macrophages via specific
inhibition of sphingosine kinase, J. Immunol. 170 (2003)
 C.R. Thompson, S.S. Iyer, N. Melrose,R. VanOosten, K. Johnson,
S.M. Pitson, L.M. Obeid, D.J. Kusner, Sphingosine kinase 1 (SK1)
is recruited to nascent phagosomes in human macrophages:
inhibition of SK1 translocation by Mycobacterium tuberculosis,
J. Immunol. 174 (2005) 3551–3561.
 H. Vesper, E.M. Schmelz, M.N. Nikolova-Karakashian, D.L.
Dillehay, D.L. Lynch, A.H. Merryll Jr., Sphingolipids in food
and the emerging importance of sphingolipids to nutrition, J.
Nutr. 129 (1999) 1239–1250.
 D.C. Macallan, Malnutrition in tuberculosis, Diagn. Microbiol.
Infect. Dis. 34 (1999) 153–157.
 R.S. Wallis, Reconsidering adjuvant therapy for tuberculosis,
Clin. Infect. Dis. 41 (2005) 201–208.