|
|
||||||||






* Cancer Biology Program, Hematology/Oncology Division, Beth Israel-Deaconess Medical Center, and Harvard Medical School, Boston, Massachusetts;
Harvard School of Public Health, Boston, Massachusetts;
Microbiology/Immunology, Wright State University, Dayton, Ohio;
|| Core Research in Evolution, Science, and Technology (CREST), Chiba University, Chiba, Japan; and
Kirin Brewery Corporation, Ltd., Japan
Correspondence: Dr. Mark A. Exley, Cancer Biology, HIM 1047, Hematology/Oncology, Beth Israel-Deaconess Medical Center, 330 Brookline Avenue, Boston, MA 02215. E-mail: mexley{at}caregroup.harvard.edu
| ABSTRACT |
|---|
|
|
|---|
14J
281
TCR-
chain (V
invt T cells) and produce Th2 and Th1
cytokines rapidly in response to CD1d, but their physiological
function(s) remain unclear. We have found that CD1d-reactive T cells
mediate to resistance against the acute, cytopathic virus diabetogenic
encephalomyocarditis virus (EMCV-D) in relatively Th1-biased,
C57BL/6-based backgrounds. We show now that these results generalize to
Th2-biased, hypersensitive BALB/c mice. CD1d-KO BALB/c mice were more
susceptible to EMCV-D. Furthermore,
-galactosylceramide
(
-GalCer), a CD1d-presented lipid antigen that specifically
activates V
invt T cells, protected wild-type (WT) mice
against EMCV-D-induced encephalitis, myocarditis, and diabetes. In
contrast, neither CD1d-KO nor J
281-KO mice were protected by
-GalCer. Finally, disease in J
281-KO mice was comparable to WT,
indicating for the first time equivalent roles for CD1d-reactive
V
invt and noninvariant T cells in resistance to acute
viral infection. A model for how CD1d-reactive T cells can initiate
immune responses, which synthesizes current results, is
presented.
Key Words: diabetes encephalitis IL-12 BALB/c mice V
invt knockout mice
| INTRODUCTION |
|---|
|
|
|---|
-chain and a limited Vß repertoire (V
invt T
cells) [1
2
3
4
5
6
]. In addition to T cells using the
invariant TCR
-chain, there are further polyclonal, CD1d-reactive T
cells that may have functionally similar properties
[7
8
9
10
11
]. CD1d-reactive T cells produce large amounts of
interleukin (IL)-4 and/or interferon (IFN)-
early in immune
responses [1
2
3
4
5
6
7
8
9
10
11
12
] and also have NK-like or CD1d-specific
cytotoxic activities [13
, 14
].
Many immunoregulatory functions have been proposed for CD1d-reactive T
cells. A role for V
invt T cells in particular and
CD1d-reactive T cells in general in driving humoral Th2-type responses
has been proposed. Indeed, multiple doses of a specific, activating
ligand for CD1d-reactive V
invt T cells,
-galactosylceramide (
-GalCer) [15
], produce
Th2-like V
invt T cells [16
] and Th2
antigen-specific responses [17
]. In further support of
this hypothesis, V
invt T cells contribute to
anti-immunoglobulin (Ig)D-mediated IgE production [18
].
Quantitative and qualitative defects in V
invt T cells
are associated with progression in human and model type-1 diabetes
[19
, 20
]. However, CD1d-deficient mice,
which lack both populations of these T cells, are able to generate
model Th2 responses [21
22
23
]. An alternative function
for CD1d-reactive T cells, also consistent with a protective role in
autoimmune disease, is in antigen-specific suppression induced by
immunization at immune, privileged sites [24
], an effect
known as immune deviation [25
].
V
invt T cells are also shown to augment model
Th1-like, responses such as IL-12-mediated IFN-
production and
anti-tumor responses [26
27
28
]. However, evidence of
involvement of CD1d-reactive T cells in physiological, Th1-like
protective-immune responses is more contradictory. NK1.1+ T cells
appear to augment anti-Toxoplasma gondii responses
in MHC class II-knockout (KO) mice [29
] and model
antimycobacterial responses [30
, 31
].
CD1d-reactive T cells are not, however, essential for protection
against Mycobacteria tuberculosis
[32
]. Furthermore, blocking CD1d recognition by antibody
is protective against listeriosis [33
]. In the first
published data linking CD1d-reactive T cells to resistance against a
virus, therapeutic activation by
-GalCer was shown to inhibit
replication of hepatitis B virus (HBV) in a transgenic model system
[34
]. However,
-GalCer can produce therapeutic
effects, which do not reflect physiological roles, through systemic,
cytokine induction. For example, a further proposed function for
CD1d-reactive T cells to promote IgG responses to endogenous and
microbial glycophosphatidylinositol-linked molecules
[35
], although supported by the finding of CD1d-binding
to such molecules [36
], has not been confirmed by
further results [37
]. Nonetheless,
-GalCer is
therapeutically active in this system [38
]. Taken
together, these data implicate V
invt T cells in
particular and CD1d-reactive T cells in general in regulation of the
immune response, but the critical functions of CD1d-reactive T cells in
physiological, immune responses remain to be identified.
The picornavirus, diabetogenic encephalomyocarditis virus (EMCV-D),
causes lethal, acute disease in young male mice of appropriate strains,
and female and older male mice are resistant [39
40
41
42
43
].
The severity of EMCV-D-induced disease varies in different genetic
backgrounds, BALB/c being most sensitive [40
] and
C57/Bl6 being most resistant [41
, 42
].
Resistance is dependent on a functional IFN-
receptor and is
associated with increased IL-12 and consequent IFN-
production
[43
44
45
46
]. Contributions of NK cells, macrophages, and T
cells to responses against various EMCV strains also vary in different
strains [42
, 47
, 48
], as with
other viruses [49
]. Resistance is also conferred in part
by an NK-like spleen cell [44
]. The findings that
CD1d-KO mice have increased susceptibility to EMCV-D and that
-GalCer could protect certain strains of wild-type (WT) mice against
EMCV-D-induced paralysis and diabetes provided evidence for a role of
CD1d-reactive T cells, including V
invt T cells in
physiological, Th1-like, anti-viral responses (Exley et al.,
unpublished results). The results described here show that both
populations of CD1d-reactive T cells contribute to protective, immune
responses against paralysis, diabetes, and myocarditis induced by this
acute, viral infection. A model for the role of CD1d-reactive T cells
integrating current results is presented.
| MATERIALS AND METHODS |
|---|
|
|
|---|
281-KO mice (F10)
have been described [21
, 26
]. A single dose
of
-GalCer [15
] or
-mannosylceramide (
-ManCer)
treatment (100 µg/Kg in 0.5% Tween-20) was given intraperitoneally
(i.p.) 5 h prior to infection. WT and KO (129xC57/Bl6)F2 or BALB/c
mice were infected i.p. with 800 plaque-forming units (PFU) EMCV-D
[39
]. Paralysis score: No paralysis = 1; 2 =
limp or partial use of one paw; 3 = one completely paralyzed hind
paw; 4 = loss of two hind limbs; 5 = paralysis of three or
four paws. Glucose tolerance testing was performed by i.p. injection of
2 g/Kg glucose with blood collected into glucosidase inhibitor-treated
tubes at 1 h [39
]. Organs were fixed for histology
at this time. | RESULTS |
|---|
|
|
|---|
invt T cells
[21
]. Studies with CD1d-KO (129xC57BL/6)F2 and C57/BL6 mice demonstrated a role for CD1d-reactive T cells in general in resistance to EMCV-D (Exley et al., unpublished results). Because genetic background plays an important role in resistance to pathogens, the generality of the effect of loss of CD1d-reactive T cells on EMCV-D infection was examined in another strain. Therefore, similar experiments were then performed in CD1d-KO F8 BALB/c mice, which have a relatively Th2-biased response and are known to be hypersensitive to EMCV-D [40 ]. EMCV-D-induced disease can be measured by hind-limb paralysis (a manifestation of encephalitis) and glucose-tolerance testing (diabetes), reflecting acute, cytopathic effects of the virus on neuronal cells and islet cells, respectively [39 ]. Paralysis was rated on a scale of 1 (no paralysis) to 5 (three or more limbs paralyzed). Although relatively high levels of paralysis were seen with even 8-week-old WT males of this hypersensitive strain [40 ], a greater incidence and severity of disease were seen with CD1d-KO mice (Fig. 1 ). This confirms and extends results in two other genetic backgrounds (Exley et al., unpublished results).
|
invt T cells: EMCV-D infection of mice
selectively deficient in the V
invt T-cell subset
invt T cells play a critical role
in resistance to EMCV-D infection, stimulation of these cells might be
protective against viral infection. The glycolipid
-GalCer is a
potent and highly specific, CD1d-presented, activating ligand for
V
invt T cells [15
]. In preliminary
experiments, it was shown that
-GalCer suppressed paralysis and
diabetes caused by EMCV-D infection of WT mice of several relatively
resistant strains at ages 58 weeks (Exley et al., unpublished
results). Because 8-week-old, male, BALB/c mice had a relatively high
susceptibility to EMCV-D (Fig. 1)
, older, 10-week-old, BALB/c, WT males
(still relatively severe) and females (more resistant in this and other
backgrounds) were treated with a single dose of
-GalCer (100
µg/Kg) or
-ManCer, a control lipid that does not activate these
cells, 5 h prior to infection. Similar to the results with other
WT mice (Exley et al., unpublished results),
-GalCer protected
almost completely WT-BALB/c mice against paralysis and diabetes induced
by infection with EMCV-D (Fig. 2a
and b
). In contrast, the majority of control-treated
mice manifested abnormal glucose tolerance and severe paralysis (Fig. 2a
and 2b)
. Further control experiments showed that
-GalCer did not
protect CD1d-KO mice, which lack V
invt T cells and other
CD1d-reactive, T-cell populations (unpublished results). Therefore,
specific activation of CD1d-reactive V
invt T cells by a
single dose of
-GalCer on the day of infection could protect mice
against paralysis and diabetes induced by EMCV-D.
|
invt T cells, which use diverse TCR
[7
8
9
10
11
]. As a further demonstration of the specificity
of the
-GalCer treatment, mice specifically lacking only
V
14J
281 V
invt T cells [26
] were
challenged with EMCV-D. Ten-week-old, male, BALB/c, WT mice were
relatively sensitive to viral challenge, providing the greatest
opportunity to observe any protective effect (Fig. 2a
and 2b)
.
Ten-week-old, female, WT and BALB/c J
281-KO mice were more resistant
than male mice (Fig. 2c
and 2d)
, as in other backgrounds described
above. Notably, although WT-BALB/c mice were protected by
-GalCer,
there was no protection of highly sensitive male or relatively
resistant female J
281-KO mice by
-GalCer (Fig. 2a
2b
2c
2d)
. These
results confirmed the selective specificity of this treatment for
V
invt T cells among CD1d-reactive, T-cell populations,
as also suggested by the results with CD1d-KO mice.
The results with CD1d-KO mice (Exley et al., unpublished results) (Fig. 1)
showed that lack of all CD1d-restricted T cells results in
exacerbated, EMCV-D-induced disease incidence and severity. To see
whether the noninvariant, CD1d-reactive T cells can function similarly
to V
invt T cells in EMCV-D infection, BALB/c J
281-KO
mice were compared with WT of the same background. Similar results to
those seen with age-matched WT mice were found with sensitive male and
even resistant female, BALB/c J
281-KO mice, with which exacerbation
of disease might be most easily detected (Fig. 2a 2b
2c
2d)
. Therefore,
although systemic activation of V
invt T cells can
provide protection against EMCV-D infection, other CD1d-reactive T
cells can also be protective in the physiological response to EMCV-D.
EMCV-D-induced myocarditis in
-GalCer lipid, antigen-treated
mice
A further manifestation of EMCV-D infection is myocarditis. EMCV
and other virus-induced myocarditis involve limited, direct, viral
damage in conjunction with extensive mononuclear-cell infiltration and
a substantial auto-aggressive, T-cell response [54
].
IL-10 treatment inhibits myocarditis in the EMCV model
[55
], supporting the proposal that much of the
myocarditis and cardiac damage is a result of an excessive immune
response in the heart in mice, which successfully overcomes infection
in other sites. Consistent with this proposal, extensive multi-focal
areas of predominantly mononuclear infiltrates and substantial necrosis
were observed in the hearts of most WT mice at 7 days post-infection
(Fig. 3A
). Significantly, however, these infiltrates were observed in the
majority of WT mice, which did not develop paralysis or diabetes. In
contrast, the hearts from the minor population of paralyzed and
hyperglycemic WT mice contained few infiltrates and did not sustain
substantial myocardial damage following infection (unpublished
results). These observations indicated that much of the myocardial
damage was indeed because of an excessive local immune response.
Similar results were obtained in the CD1d-KO mice (unpublished
results). Because myocarditis was associated with protective immune
responses in WT and CD1d-KO mice, myocarditis was also assessed in
-GalCer and control, lipid-treated animals. Control lipid
-ManCer
treatment did not influence the course of the myocarditis seen in
EMCV-D-infected, WT mice (Fig. 3)
. However, in marked contrast to
-ManCer treatment, myocarditis was reduced greatly in the
EMCV-D-infected WT animals protected by
-GalCer treatment
(Fig. 3B)
. Therefore, therapeutic stimulation of CD1d-reactive, T-cell
activity was protective against myocarditis as well as encephalitis and
diabetes.
|
| DISCUSSION |
|---|
|
|
|---|
-GalCer in a total of 4/4 strains of WT mice
tested.
-GalCer activates CD1d-reactive, V
invt T
cells specifically [15
]. A single dose of
-GalCer was
fully protective even when administered on the day of EMCV-D infection,
where disease is manifested within 5 days. Furthermore,
-GalCer was
acting through CD1d-reactive, V
invt T cells because
there was no protection of CD1d-KO or J
281-KO mice. However,
J
281-KO mice, which selectively lack V
invt T cells
but retain other CD1d-reactive T cells and CD1d itself, were able to
mount protective responses comparable with WT controls. Therefore,
noninvariant, CD1d-reactive T cells can apparently, also mediate
protection against this acute viral infection.
Although these cells can mediate cytotoxicity directly, our previous
results implicate loss of IL-12 production by antigen-presenting cells
(APC) and subsequent NK-cell activation as major defects in CD1d-KO
mice (Exley et al., unpublished results). Furthermore,
-GalCer
stimulation of V
invt T cells results in rapid, NK-cell
activation [56
]. These in vivo results are
also consistent with in vitro studies in which IL-12
production by APC can be regulated by local V
invt,
T-cell, IFN-
production and by direct V
invt,
T-cell:APC contact through CD154:CD40 interaction [57
,
58
]. Thus, initial, local, V
invt, T-cell,
IFN-
production could augment APC-IL-12 release (Fig. 4
). This IL-12 could then induce more general IFN-
production. As
IFN-
is a relatively weak anti-viral, the effector mechanism likely
involves CD1-reactive T-cell IFN-
-activated macrophages
[47
], NK cells [42
, 57
; Exley
et al., unpublished results], as well as conventional T cells
[48
].
|
responses implies that these cells must respond to early, innate,
immune signals. Candidates include IFN-
, IL-12, and IL-18, produced
by APC in response to viral infections including EMCV
[59
]. Co-stimulator expression or presentation of an
endogenous lipid antigen in response to cytopathic viral infection
could contribute to activation of CD1d-reactive T cells. We propose
that a physiological role for CD1d-reactive T cells is to integrate
rapidly CD1d-dependent and cytokine signals from APC to optimize the
innate, immune response and depending on the nature of the antigen
challenge, to influence the Th1 versus Th2 decision (Fig. 4)
. A role
for CD1d-reactive T cells in directing immune responses suggests that
appropriate therapeutic activation could ameliorate infectious disease,
autoimmunity, allergic, and anti-tumor responses.
Our results in antiviral, Th1-like, immune responses may also bare on a
role for CD1d-reactive T cells in influencing Th2 responses, as has
been widely supposed. A further role for CD1d-reactive T cells has been
shown in immune deviation induced by antigen at an immune-privileged
site [24
]. To reconcile these diverse, potential roles,
we propose that the default pathway for activated V
invt
T cells is to suppress Th1 responses and that one or more signals from
the innate, immune system indicative of an intracellular infection are
required to shift CD1d-reactive T cells toward a Th1-promoting response
(Fig. 4)
. The physiological role for CD1d-reactive T cells would then
be to integrate rapidly CD1d-dependent, cytokine, and innate, immune
signals and to influence the Th1 versus Th2 decision based on the
nature of the antigen challenge. In conclusion, therapeutic activation
of CD1d-reactive T cells has the potential to optimize natural and
vaccine-induced, anti-microbial or anti-tumor, immune responses.
| ACKNOWLEDGEMENTS |
|---|
Received October 23, 2000; revised December 14, 2000; accepted December 15, 2000.
| REFERENCES |
|---|
|
|
|---|
24+ CD4-CD8-T cells J. Exp. Med. 186,109-120
14 NKT cells by glycosylceramides Science 278,1626-1629
-galactosylceramide polarizes CD1-reactive NK T cells towards Th2 cytokine synthesis Eur. J. Immunol. 29,2014-2025[Medline]
-galactosylceramide directs conventional T cells to the acquisition of a Th2 phenotype J. Immunol. 163,2373-2377
24J
Q T cells in type 1 diabetes Nature 391,177-180[Medline]
14 NKT cells in IL-12-mediated rejection of tumors Science 278,1623-1626
producing CD4+ NK T cells by M. bovis bacillus Calmette Guerin Eur. J. Immunol. 29,650-659[Medline]
-galactosylceramide demonstrates its immunopotentiating effect by inducing IL-12 production by dendritic cells and IL-12 receptor expression on NK T cells J. Exp. Med. 189,1121-1125
14+ CD4+ NKT cells: stimulation of IL-12 production by antigen presenting cells in the innate immune system J. Immunol. 163,93-101Related Articles
This article has been cited by other articles:
![]() |
G. J. Renukaradhya, M. A. Khan, D. Shaji, and R. R. Brutkiewicz Vesicular Stomatitis Virus Matrix Protein Impairs CD1d-Mediated Antigen Presentation through Activation of the p38 MAPK Pathway J. Virol., December 15, 2008; 82(24): 12535 - 12542. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Grubor-Bauk, J. L. Arthur, and G. Mayrhofer Importance of NKT Cells in Resistance to Herpes Simplex Virus, Fate of Virus-Infected Neurons, and Level of Latency in Mice J. Virol., November 15, 2008; 82(22): 11073 - 11083. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Tsunoda, T. Tanaka, and R. S. Fujinami Regulatory Role of CD1d in Neurotropic Virus Infection J. Virol., October 15, 2008; 82(20): 10279 - 10289. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. J. Raftery, M. Hitzler, F. Winau, T. Giese, B. Plachter, S. H. E. Kaufmann, and G. Schonrich Inhibition of CD1 Antigen Presentation by Human Cytomegalovirus J. Virol., May 1, 2008; 82(9): 4308 - 4319. [Abstract] [Full Text] [PDF] |
||||
![]() |
K.-H. Sonoda, T. Nakamura, H. A. Young, D. Hart, P. Carmeliet, and J. Stein-Streilein NKT Cell-Derived Urokinase-Type Plasminogen Activator Promotes Peripheral Tolerance Associated with Eye J. Immunol., August 15, 2007; 179(4): 2215 - 2222. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Raghuraman, Y. Geng, and C.-R. Wang IFN-beta-Mediated Up-Regulation of CD1d in Bacteria-Infected APCs J. Immunol., December 1, 2006; 177(11): 7841 - 7848. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. J. Raftery, F. Winau, S. H. E. Kaufmann, U. E. Schaible, and G. Schonrich CD1 Antigen Presentation by Human Dendritic Cells as a Target for Herpes Simplex Virus Immune Evasion J. Immunol., November 1, 2006; 177(9): 6207 - 6214. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. O. Ilyinskii, R. Wang, S. P. Balk, and M. A. Exley CD1d Mediates T-Cell-Dependent Resistance to Secondary Infection with Encephalomyocarditis Virus (EMCV) In Vitro and Immune Response to EMCV Infection In Vivo J. Virol., July 15, 2006; 80(14): 7146 - 7158. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. R. Brutkiewicz CD1d Ligands: The Good, the Bad, and the Ugly J. Immunol., July 15, 2006; 177(2): 769 - 775. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. E. Faunce, J. L. Palmer, K. K. Paskowicz, P. L. Witte, and E. J. Kovacs CD1d-Restricted NKT Cells Contribute to the Age-Associated Decline of T Cell Immunity J. Immunol., September 1, 2005; 175(5): 3102 - 3109. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Bilenki, S. Wang, J. Yang, Y. Fan, A. G. Joyee, and X. Yang NK T Cell Activation Promotes Chlamydia trachomatis Infection In Vivo J. Immunol., September 1, 2005; 175(5): 3197 - 3206. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. A. Huber and D. Sartini Roles of Tumor Necrosis Factor Alpha (TNF-{alpha}) and the p55 TNF Receptor in CD1d Induction and Coxsackievirus B3-Induced Myocarditis J. Virol., March 1, 2005; 79(5): 2659 - 2665. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Lin, T. J. Roberts, P. M. Spence, and R. R. Brutkiewicz Reduction in CD1d expression on dendritic cells and macrophages by an acute virus infection J. Leukoc. Biol., February 1, 2005; 77(2): 151 - 158. [Abstract] [Full Text] [PDF] |
||||
![]() |
V. V. Parekh, A. K. Singh, M. T. Wilson, D. Olivares-Villagomez, J. S. Bezbradica, H. Inazawa, H. Ehara, T. Sakai, I. Serizawa, L. Wu, et al. Quantitative and Qualitative Differences in the In Vivo Response of NKT Cells to Distinct {alpha}- and {beta}-Anomeric Glycolipids J. Immunol., September 15, 2004; 173(6): 3693 - 3706. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Durante-Mangoni, R. Wang, A. Shaulov, Q. He, I. Nasser, N. Afdhal, M. J. Koziel, and M. A. Exley Hepatic CD1d Expression in Hepatitis C Virus Infection and Recognition by Resident Proinflammatory CD1d-Reactive T Cells J. Immunol., August 1, 2004; 173(3): 2159 - 2166. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. J. Roberts, Y. Lin, P. M. Spence, L. Van Kaer, and R. R. Brutkiewicz CD1d1-Dependent Control of the Magnitude of an Acute Antiviral Immune Response J. Immunol., March 15, 2004; 172(6): 3454 - 3461. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. E. Faunce, A. Terajewicz, and J. Stein-Streilein Cutting Edge: In Vitro-Generated Tolerogenic APC Induce CD8+ T Regulatory Cells That Can Suppress Ongoing Experimental Autoimmune Encephalomyelitis J. Immunol., February 15, 2004; 172(4): 1991 - 1995. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. A. Campos, M. Szczepanik, A. Itakura, M. Akahira-Azuma, S. Sidobre, M. Kronenberg, and P. W. Askenase Cutaneous Immunization Rapidly Activates Liver Invariant V{alpha}14 NKT Cells Stimulating B-1 B Cells to Initiate T Cell Recruitment for Elicitation of Contact Sensitivity J. Exp. Med., December 15, 2003; 198(12): 1785 - 1796. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Rauch, J. Gumperz, C. Robinson, M. Skold, C. Roy, D. C. Young, M. Lafleur, D. B. Moody, M. B. Brenner, C. E. Costello, et al. Structural Features of the Acyl Chain Determine Self-phospholipid Antigen Recognition by a CD1d-restricted Invariant NKT (iNKT) Cell J. Biol. Chem., November 28, 2003; 278(48): 47508 - 47515. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Skold and S. M. Behar Role of CD1d-Restricted NKT Cells in Microbial Immunity Infect. Immun., October 1, 2003; 71(10): 5447 - 5455. [Full Text] [PDF] |
||||