Content uploaded by A. Catharine Ross
Author content
All content in this area was uploaded by A. Catharine Ross
Content may be subject to copyright.
Vitamin A and retinoids in antiviral responses
A. CATHARINE ROSS” AND CHARLES B. STEPHENSENt
*Depailment of Nutrition, Pennsylvania State University, University Park, Pennsylvania 16802, USA; and
tDepatlment of International Health, School of Public Health, University of Alabama at Birmingham, Birmingham,
Alabama 35294, USA
ABSTRACT Vitamin A deficiency results in multi-
ple derangements that impair the response to infec-
tion. This review focuses on experimental models of
specific virus infections and on cytokines and cells
with cytolytic activity important to antiviral defenses.
Altered specific antibody responses and greater epi-
thelial damage in vitamin A-deficient hosts are con-
sistent findings. The cytolytic activity of natural killer
cells and various cytokine responses are altered. The
inflammatory response to infection may also result
in derangements in the transport and metabolism of
retinol. We speculate that interaction of several
factors may combine to explain the greater severity
of infection seen in vitamin A-deficient animals and
children. In addition to a preexisting lack of tissue
vitamin A, these factors may include reduced mobi-
lization and increased excretion of retinol during the
acute phase response to infection, poor innate and
specific immune response to virus, and delayed re-
pair of damaged epithelia. Foci of vitanun A-deficient
epithelia may be sites of penetration of bacteria and
other agents, leading to secondary infections and
contributing to an increased severity of infections
and poor outcome in vitamin A-deficient animals and
humans.-Ross, A. C., Stephensen, C. B. Vitamin A
and retinoids in antiviral responses. FASEB J. 10,
979-985 (1996)
Key Words: Newcastle disease virus Herpes simplex virus ‘natu-
ral killer cells’ cyzokines .retinol binding protein .vitamin A
slat us
AN ASSOCIATION BETWEEN VITAMIN Aand the immune re-
sponse to infection was postulated in the late 1920s. The
impetus for renewed interest in this subject comes from
research in several disciplines. First, it is now recognized
from clinical and cellular studies of hematopoiesis that
retinol, retinoic acid, and related retinoids affect lympho-
cyte development and maturation (reviewed in ref 1).
Second, cell culture and animal nutritional studies have
shown that vitamin A and related retinoids are capable of
modulating lymphocyte responses that lead to antibody
production, T cell activation, and cytokine production
(see ref 2 for a general review). Third, the results of nu-
merous epidemiologic studies have shown convincingly
that mortality, and sometimes morbidity, are greater in vi-
tamin A-deficient children even when xerophthalmia is
not present (3). Diarrhea, measles, and, in some studies,
respiratory infections are common in vitamin A-deficient
children (4). Diarrhea and respiratory infections are
caused by many different etiologic agents, including vi-
ruses and bacteria, whereas measles is caused by a single
agent, the measles virus. For many vitamin A-deficient
children, the cause of death is likely to be an overwhelm-
ing infection of viral or bacterial origin. A meta-analysis
of eight field-based studies showed a highly significant
effect of vitamin A supplementation in preschool children
in poor regions of Asia, Malaysia, and Africa, with an
overall reduction in mortality of 23% (5). Although most
data suggest that vitamin A status has little effect on the
incidence of infectious disease, the severity of some in-
fections, especially those associated with diarrhea, may
be reduced when vitamin A status is improved (6). In
contrast, there seems to be no benefit of vitamin A in re-
ducing the severity of childhood pneumonia in developing
countries (7).
In this review, we first consider experimental studies
designed to understand the requirements for and func-
tions of vitamin A in the in vivo response to infection
with specific viruses. Second, we review the role of vita-
min A in maintaining cells and cytokines with antiviral
properties, focusing on natural killer (NK)2 cells and re-
lated cytokines. Third, we discuss briefly how infection
may alter the normal transport and metabolism of retinol.
From these data, we postulate an interplay between vita-
min A and infection in which vitamin A deficiency re-
sults in derangements of the immune response, while the
inflammatory response initiated by infection reduces the
mobilization of retinol and its transport to tissues, thereby
contributing to tissue depletion of retinoids needed for
the repair of infection-damaged epithelia.
To whom correspondence and reprint requests should be addressed,
at: Department of Nutrition, 126-S Henderson Bldg., University Park, PA
16802, USA.
2Abbreviations: NK, natural killer; NDV, Newcastle disease virus; IL,
interleukin; Th, T helper; HSV, Herpes simplex virus; HIV, human
immunotrophic virus; RBP, retinol binding protein.
Severity of infection Increased severity of clinical signs; probable
increase in mortality for some infections
VITAMIN A DEFICIENCY AND VIRUS-HOST
INTERACTIONS
General features
Vitamin Adeficiency can affect several aspects of a vi-
rus’ interaction with its host. Infection begins when a vi-
rus binds to its cellular receptor and is internalized. The
virus must then “uncoat,” replicate its genome, produce
structural and nonstructural viral proteins, and assemble
into infectious progeny virions that then (typically) exit
the cell in order to find another susceptible cell to infect
and, eventually, to find another host. The host’s initial re-
sponse to viral infection is through antigen-independent
(innate) mechanisms, particularly the production of cy-
tokines of the interferon family at the site of initial infec-
tion and the activity of cytotoxic NK cells, which are
themselves activated by interferon. After this initial re-
sponse, the host develops a specific immune response to
viral proteins that typically includes both antibody pro-
duction (leading to viral neutralization) and cell-mediated
effector mechanisms, particularly the generation of virus-
specific cytotoxic T lymphocytes that kill virus-infected
cells. After clearance of the virus, the host must then re-
pair damage to cells and tissues caused by the virus itself
as well as the host’s inflammatory response to infection.
Several aspects of the host response have been examined
in experimental animal models, as summarized in Ta-
ble 1.
Newcastle disease virus infection in chickens
The best-characterized model for examining the interac-
tion of vitamin A status and viral infection is Newcastle
disease virus (NDV) infection of chickens. NDV is a
member of the. Paramyxovirus genus of the family
Paramyxoviridae; members of this genus that infect hu-
mans include parainfluenza virus types 1 through 4,
which cause respiratory infections. Viruses from other
genera in this family include the important human patho-
gens measles virus and respiratory syncytial virus. Like
other paramyxoviruses, NDV is transmitted by the respi-
ratory route. Tissue tropism and age of susceptibility of
the host vary for NDV strains, but the virus isolates used
in the studies described here primarily infect epithelial
cells lining the respiratory tract. The age dependence of
lentogenic strains (those causing disease in young chicks
but not adults) can be altered by vitamin A deficiency in
that the high level of virus replication seen in 1-wk-old
chicks is maintained through 3 wk of age by feeding
chicks a vitamin A-deficient diet (8). Vitamin A metabo-
lites promote the differentiation of epithelial cells in the
respiratory tract. These authors speculated that delayed
epithelial maturation in vitamin A-deficient animals,
rather than a direct effect of vitamin A on the virus or on
the immune response, accounted for the increased shed-
ding of virus at 3 wk of age. This age dependence may be
due to expression of a trypsin-like enzyme in the respira-
tory epithelium, which is necessary to cleave the F0 pro-
tein of the NDV virion and render it infectious (9).
The effect of vitamin A deficiency on the replication of
mesogenic NDV strains (those that produce disease in in-
fected adult chickens) is not certain. Such NDV strains
do produce more severe clinical signs of infection (respi-
ratory signs, general weakness) in vitamin A-deficient
animals than they do in control animals (10, 11). In-
creased virus replication could account for this increased
virulence, but other factors, such as a decreased immune
response or impaired epithelial recovery, could also be
important. Mortality is also increased in adult chickens
fed vitamin A-deficient diets, at least when the deficiency
is advanced enough to impair weight gain in uninfected
animals (10).
The immune response to NDV infection is also im-
paired by vitamin A deficiency. The total number of both
B and T lymphocytes is decreased in the peripheral blood
as well as in both primary and secondary lymphoid or-
TABLE1. Summary of effects of vitamin A deficiency on host response to viral infections in animal models
Aspect of virus infection or host response
Virus replication
Interferon-a and -f response at site of initial infection
NK cell activity
Serum lgc and 1gMresponse
Secretory IgA response
Cell-mediated immune response
Regeneration of virus-damaged epithelia
Principal effect of vitamin A deficiency
Not affected? (little data)
Unknown
Decreased (decreased number of cells)
Often not affected; possible increase?
Decreased
Decreased? (little data)
Severely impaired
gans (12, 13). Lymphopenia normally occurs during the
acute phase of NDV infection, followed by a rebound to
greater than normal levels. This pattern-both the de-
crease and subsequent increase of lymphocytes-is
blunted by vitamin A deficiency. That the NDV-specific
cytotoxic T lymphocyte response is also impaired by vita-
min A deficiency is particularly important. The activity of
cytotoxic T lymphocytes against NDV-infected cells is
decreased both during the primary infection and during a
secondary challenge infection by approximately 40%
(14). Such a decrease measured in vitro suggests that vi-
rus clearance in vivo may also be affected, although this
was not directly demonstrated.
In contrast, the serum antibody response to both a pri-
mary and secondary NDV infection is not affected by vi-
tamin A deficiency, although addition of supplemental
retinoic acid to the diet can increase the secondary re-
sponse above control levels (12, 15, 16). Similarly, the
serum antibody responses of chickens to another viral
respiratory pathogen (infectious bronchitis virus) and to
an enteric pathogen (reovirus) are not affected by vitamin
A deficiency (17).
The NDV-specific secretory IgA response has not been
examined; however, the concentration of total IgA in bile
is decreased in vitamin A-deficient animals while levels
in the respiratory tract (tracheal homogenates) are not af-
fected (15, 16).
Recovery from NDV infection involves regeneration of
respiratory epithelium, which has been damaged by virus
replication and the immune response. This process is
dramatically impaired by vitamin A deficiency. A hall-
mark of vitamin A deficiency is the replacement of the
normal ciliated cuboidal or columnar epithelium and mu-
cus-secreting goblet cells by a keratinized squamous epi-
thelium (18, 19). Such metaplastic changes may be
important in the response to infection because these
metaplastic sites in the airways are more susceptible to
colonization by some bacterial respiratory pathogens. This
colonization could increase the risk of secondary bacte-
rial infection after a primary viral infection, as is often
the case after an episode of measles (20) or other respira-
tory viral infections (21).
Rotavirus infection of mice
Rotavirus is an important human pathogen that causes
diarrheal disease in children generally under 2 years of
age. Human rotavirus isolates have only recently been
adapted to limited growth in mice, but a naturally occur-
ring mouse rotavirus has been psed as a model system
(22). As with NDV infection of the respiratory tract, the
most dramatic effect of vitamin A deficiency on the host’s
response to rotavirus infection is the inability of deficient
animals to repair virus-induced damage to the intestinal
epithelium. The cytopathic effects of rotavirus infection
are much more severe in the majority of vitamin A-defi-
cient animals than in vitamin A-sufficient controls, with
the villous tips in some areas of the small intestine being
completely destroyed, leaving the lamina propria exposed
to the gut lumen. This level of pathology was not seen in
uninfected, deficient animals or in infected animals on
control diets (22).
It was also noted that the serum antibody response was
decreased in vitamin A-deficient animals; however, the
secretory IgA response, which is important in protection
against reinfection by enteric pathogens, was not meas-
ured (23). Cell-mediated immunity, measured by dermal
sensitization of animals with picryl chloride 1 day after
rotavirus infection, and the subsequent delayed-type hy-
persensitivity response to rechallenge with picryl chloride
were decreased by approximately 30% (24). Although
such delayed-type hypersensitivity responses are not a
key component of an antiviral response (but are involved
in the clearance of other intracellular pathogens, such as
tuberculosis bacilli), they have been traditionally used to
measure the effect of malnutrition on cell-mediated im-
munity and suggest the impairment of cell-mediated im-
munity during vitamin A-deficiency.
Influenza A virus infection of mice
Influenza A virus causes mild to severe respiratory infec-
tions in humans. Some strains of influenza A have been
adapted to mice by serial passage. Infection of mice in
the upper respiratory tract can cause a mild infection
whereas inoculation of the lungs can cause severe viral
pneumonia. Virus replication and the rate of virus clear-
ance from the respiratory tract are not affected by vitamin
A deficiency in this model (25). Nor does the extent of
inflammatory lesions during viral pneumonia differ be-
tween deficient and control mice. As with NDV infection,
however, regeneration of the normal epithelium is im-
paired and florid metaplastic lesions are seen in areas
where virus replication resulted in inflammation and cel-
lular damage. Adjacent areas are histologically normal. In
spite of this, animals recover and there is no difference in
mortality or in the clinical severity of infection, at least in
mildly vitamin A-deficient animals (i.e., before they have
reached a weight plateau as a result of vitamin A defi-
ciency).
The influenza-specific serum IgG response, which
helps protect the lungs against infection, is either not af-
fected or may be increased by twofold in vitamin A-defi-
cient animals (26). This increase in mice is largely due
to an increase in IgG2a, an isotype that predominates in
the response to viral infections. The murine IgG2a re-
sponse is driven by interferon-’y-producing T helper (Th)
cells designated as type 1 (Thi) cells (27). In contrast,
the influenza A-specific IgA response, which helps pro-
tects the upper respiratory tract against infection, is dra-
matically diminished in vitamin A-deficient mice, with
salivary IgA levels being less than 10% of those in ani-
mals fed control diet. The expression in mice of the poly-
meric total IgA (as compared to influenza A-specific IgA)
concentrations in saliva and bile are higher in vitamin A-
deficient mice than controls. This is apparently due to in-
00’) in k.I., lone. Tk.. csccli orcc ajr cTcDt-lc,.jccM
creased expression of the polymeric immunoglobulin re-
ceptor, which transports IgA across mucosal surfaces and
into the bile (C. B. Stephensen, unpublished observa-
tions). This increase differs from the reduction in biliary
and intestinal IgA levels found in vitamin A-deficient
chickens (15, 16) and rats (28-30). The decreased influ-
enza-specific IgA appears to be due to a reduction in the
number of antibody-secreting plasma cells (C. B.
Stephensen, unpublished observations), as has been re-
ported in other systems for serum antibody response
against purified protein antigens (24). The IgA response
is driven mainly by T cells of the Th2 type, characterized
by a high level of production of interleukin-4 (IL-4) and
IL-5 (27). The decreased influenza-specific IgA response
is consistent with work from other laboratories indicating
that antibody responses driven by Th2 cells are often di-
minished by vitamin A deficiency in inbred mice,
whereas interferon-y production by Thi cells from vita-
min A-deficient mice may be increased in vitro (31). This
increased production of interferon-gg, if it also occurs in
vivo, might explain the higher IgG2a response of vitamin
A-deficient mice.
Herpes virus infection of rats
Herpes simplex virus (HSV) infects humans and other
species, commonly causing ulcerative lesions on mucosal
surfaces (herpes labialis and urogenital infections) but
also encephalitis and keratitis. HSV causes both acute
and latent infections, the latter being reactivated by nu-
merous stimuli. Nauss et al. (32) investigated the effects
of vitamin A deficiency on ocular infections in HSV-in-
fected rats. HSV infection of the cornea induced an in-
flammatory cell infiltrate, first producing corneal opacity,
followed by epithelial ulceration and stromal degenera-
lion with eventual perforation; the severity of lesions is
determined by the quantity of virus inoculum.
After large inocula, ocular pathology developed more
rapidly in vitamin A-deficient animals than in controls,
but eventually the extent of damage was similar. With
smaller inocula, a higher percentage of vitamin A-defi-
cient animals developed lesions and the severity of these
lesions was greater, as judged by slit-lamp scores and the
extent of microscopic pathologic changes. The greater pa-
thology in the deficient rats did not appear to be due to a
decreased cell-mediated immune response to HSV be-
cause the HSV-specific proliferative response of cells in
draining lymph nodes and the spleen of deficient animals
was equal to or substantially greater than in the control
animals at all postinfection time points (33). This more
intense immune response could have been triggered by a
higher level of virus replication in the deficient rats, but
virus replication was not measured in these experiments.
Splenic NK cell cytolytic activity (against heterologous
target cells) was decreased by approximately 30% in vi-
tamin A-deficient, HSV-infected animals. Diminished NK
activity could have impaired the initial host response to
HSV infection, and this may have contributed to the
greater severity of infection found in these vitamin A-de-
ficient animals. The serum antibody response to HSV was
not affected by vitamin A status. Parenteral administra-
lion of vitamin A to nondeficient rabbits decreased the
severity of HSV keratitis (34, 35). This beneficial effect
of supplemental vitamin A could be due either to im-
proved healing of HSV-induced epithelial lesions or to
some enhancement of the immune response.
Retroviral infections
No animal model studies have examined the effect of vi-
tamin A on infection with human immunodeficiency virus
(HIV) or its simian counterpart, SlY. However, one study
with a murine retrovirus reported increased survival of
mice fed diets supplemented with very high levels of vita-
min A, and the investigators suggested that enhancement
of the immune response was important in improving sur-
vival (36). With respect to HIV infection, the long-termi-
nal repeat region of the HIV virus contains a functional
retinoic acid response element (37). Retinoic acid treat-
ment in vitro can either increase or decrease HIV repli-
cation, depending on the cell types used (38). The
implications of such regulation for viral pathogenesis are
not yet clear, but these data suggest that the replication
of HIV as well as other viruses with DNA genomes or in-
termediates (39) can be directly modulated via the activ-
ity of retinoid receptors.
ANTIVIRAL CELL-MEDIATED CYTOLYSIS AND
CYTOKINE PRODUCTION
As indicated earlier, the initial response to viral infec-
tions is largely mediated by nonspecific mechanisms, in-
cluding the cytolytic action of NK cells against
virus-infected host cells. NK cells are a quantitatively
minor population of lymphocytes found predominantly in
blood, spleen, and liver that have abundant cytoplasmic
granules that contain the pore-forming protein perform
and a number of proteases. The cytolytic activity of NK
cells is increased substantially upon exposure to IL-2 or
interferon (mainly aand ). NK cell are capable of a
rapid response, without prior sensitization, to virus-in-
fected cells (or heterologous cells used as targets in vitro)
through the release of their lytic mediators. Activated NK
cells are also major producers of interferon-’y and tumor
necrosis factor-a and thus have a potentially major role
in regulating other immune responses, including antibody
production. In general, but in a manner that may differ
for specific infectious agents, mnterferon-y (e.g., from NK
or Thi cells) facilitates cell-mediated responses and may
inhibit the antibody responses facilitated by cytokines
produced by Th2 cells.
Experiments by Nauss and Newberne (33) showed re-
duced NK cell cytotoxicity in splenic lymphocytes from
vitamin A-deficient, HSV-infected rats. In uninfected
rats, NK cell activity was also reduced by 30 to >50% in
vitamin A-deficient rats (40, 41), and restored to normal
levels by dietary treatment with retinol or retinoic acid
(42). However, it was not clear whether reduced NK cell
activity resulted from low cytolytic activity per cell or a
reduction in the number of the NK cell phenotype. Using
a monoclonal antibody that recognizes rat NK cells, Zhao
et al. (41) showed that the majority of the reduced NK
cell cytotoxic activity in peripheral blood and nearly half
of the reduction in spleen could be explained by a iow
number of NK cells. Nonetheless, NK cells from both vi-
tamin A-deficient and control rats were activated essen-
tially equally by interferon-aJ or IL-2, which suggests
that although cell numbers were reduced, possibly due to
impaired lymphopoiesis and NK cell maturation, the
functional activity of NK cells reaching maturity was not
impaired.
There is little information concerning NK cells in chil-
dren with vitamin A deficiency. Griffin et al. (43) noted
reduced NK cell activity in the peripheral blood of Peru-
vian children with measles, which remained low for at
least 3 wk after the onset of rash. The ability of NK cells
to respond to IL-2 was retained. Although the vitamin A
status of these children was not determined, it is tempt-
ing to speculate that it may have been marginal. In vita-
min A-deficient rats treated with dietary retinoic acid or
N-(4-hydroxyphenyl) retinamide, the number of NK cells
returned to normal within a week (42). Retinoids may
also stimulate the cytotoxicity of NK cells from normal
animals, increasing NK cell lytic efficiency (cytolytic ac-
tivity per NK cell) (42). A possible mechanism for the re-
duced cell-mediated responses to viruses and other
infectious agents in vitamin A-deficient animals and chil-
dren may be related to impaired production or maturation
of NK cells, reducing their immediate impact via cy-
tolysis and possibly resulting in a reduction in the secre-
tion of cytokines that participate in the regulation of
antigen-specific antibody responses.
Evidence from in vivo studies indicates that the pro-
duction of cytokines may also be altered during vitamin A
deficiency. Cells from vitamin A-deficient mice produced
less IL-4 and IL-5, consistent with reduced antibody re-
sponses. Conversely, IL-12 and interferon-I were ex-
pressed constitutively in deficient mice; these effects
were proposed to shape the immune response toward a
Thi type and away from a Th2 type of response, which
would favor antibody production (31, 44). However, the
delayed-type hypersensitivity response to picryl chloride,
considered a Thi response, in mice and the cytotoxic T
lymphocyte response of NDV-infected chicks are both
low. Conversely, specific antibody responses may not be
decreased (as in the case of the elevated IgG2a response
of vitamin A-deficient mice infected with influenza virus).
Evidence from in vitro experimentation with normal
human lymphocytes supports a possible role for retinoic
acid in the regulation of IL-2-mediated responses that are
central to T cell-mediated immunity, antibody responses,
and the activation of NK cells. The addition of retinoic
acid to cultured human B cell lines or thymocytes led to
an apparent inactivation of an inhibitor of transcription of
the IL-2 receptor asubunit and an increase in the level
of its mRNA (45, 46). On the basis of cell culture stud-
ies, it has also been proposed that retinoic acid down-
regulates the transcription of the IL-2 gene (47). If
retinoic acid functions to enhance expression of high-af-
finity IL-2 receptors while down-regulating IL-2 produc-
tion, the result could be a more focused response limited
to those cells expressing the high-affinity form of the IL-2
receptor, and therefore responsive to lower amounts of IL-
2.
EFFECTS OF INFECTION AND
INFLAMMATION ON RETINOL TRANSPORT
AND METABOLISM
Vitamin A is transported from liver, its major site of stor-
age, to target tissues in a complex consisting of one mole-
cule of retinol bound noncovalently in the binding cavity
of retinol binding protein (RBP), which in turn is associ-
ated noncovalently with transthyretin (48). The liver is
the major site of synthesis and release of these relatively
short-lived, homeostatically regulated transport proteins.
Most, if not all, target organs receive retinol via this com-
plex; in addition, plasma also contains retinoic acid at a
lower concentration. Although the mechanism of retinol
uptake by tissues is uncertain, there is reason to believe,
based on studies in cultured cells, that the Km for uptake
(49) is similar to the normal plasma concentration
(-1.5-3 l.tmol/l), such that delivery of the vitamin to tis-
sues may be reduced even during moderate hyporet-
inemia. However, it is not yet known whether reductions
in the concentration, of retinol or RBP during inflamma-
tion (below) impair the delivery of retinol to target or-
gans. The kidney is especially important with respect to
the degradation of RBP and the recycling of retinol,
which is known to recirculate between plasma and tissues
several times before being degraded (48).
Evidence is increasing that several aspects of the
transport and metabolism of retinol are altered during in-
fection or experimental inflammation. During infections
in children, plasma retinol concentrations may be low (as,
for example, in respiratory syncytial virus infection; 50).
In adults with pneumonia and sepsis, the urinary excre-
tion of retinol and RBP, which normally is low, was ele-
vated significantly, especially during fever (51);
similarly, excretion was elevated in children with acute
diarrhea (52). Thus, loss of retinol may be a mechanism
contributing to the development of poor vitamin A status
in young children or of clinical vitamin A deficiency if
their vitamin A status already is poor. The inflammatory
response to infection may also have a significant effect on
the hepatic synthesis and secretion of RBP and, hence,
on the delivery of retinol to target tissues. In normal, vita-
min A-sufficient rats, the induction of inflammation by
lipopolysaccharide caused a significant (-‘50%) reduc-
tion the plasma concentrations of retinol and RBP; these
changes were preceded by a reduction in RBP mRNA in
liver (53). These changes, which corresponded to the
Iacute
Iphase
+response
Immunesystem
responses
1) altered lymphocyte
number and/or subsets
2) reduced innate
responses
3) altered cytoklne
productIon
4) altered specIfic
antibody responses
Reduced plasma
vitamin A
1) decreased vitaili A
inala andlor
2) reduced hepatic
mobilization
3) increased retinol
excretion
Epithelial damage
requiring repair!
regeneration
1) less retlnold for
differentIatIon
2) Increased secondary
Infections
INCREASED
.SEVERITY OF
DISEASE
nbA SI..,] In I..L. IflOC TI..-. CA CUD OCSCC * kin CtflLJCkiCKi
INFECTION
Figure 1. Model of the possible interactions between vitamin A status, infection and increased severity of
infectious disease.
acute-phase response to infection, resulted in an acute
hyporetinemia even though liver vitamin A reserves were
more than adequate. The results of these studies imply
that infection may induce changes in the transport and
catabolism of vitamin A by reducing both the delivery of
retinol from liver to tissues via RBP and the renal reten-
tion and recycling of retinol back to plasma.
POSSIBLE INTERACTIONS BETWEEN
RETINOID METABOLISM AND THE SEVERITY
OF INFECTION IN ANIMALS AND HUMANS
A conclusion drawn from the results of field-based
epidemiologic studies is that the incidence or rate of in-
fection does not differ between vitamin A-supplemented
children and untreated controls. On the other hand, the
data support a greater severity of diarrheal disease and
measles in vitamin A-deficient vs. vitamin A-supple-
mented children. The lack of an effect of vitamin A on
the incidence of infection has been interpreted as sug-
gesting that preexisting differences in epithelial integrity
are unlikely to be a main cause of the increased morbid-
ity and mortality in vitamin A-deficient children. How-
ever, the results of animal studies discussed above, most
notably with NDV and rotavirus, suggest that epithelial
damage resulting from the infection is greater, and of
longer duration, when vitamin A deficiency is also pre-
sent. These results suggest that, although infection results
in damage to the epithelia regardless of vitamin A status,
a preexisting deficiency of vitamin A leads to a more pro-
found and consequential impairment. At the same time,
inflammation may reduce retinol transport both in indi-
viduals with normal vitamin A status and, we speculate,
in those with impending vitamin A deficiency. In the for-
mer case, such an effect may be transient, hut in the lat-
ter case the response to infection may precipitate a
critical depletion of tissue retinoids. As noted previously,
the foci of vitamin A-deficient epithelia may be sites of
penetration of bacteria and other agents leading to secon-
dary infections. Thus, as proposed in Fig. 1, vitamin A
deficiency may result in a weak or aberrant immune re-
sponse, whereas infection itself could play a twofold role
by inducing epithelial damage and initiating an inflam-
matory response that further impairs retinol transport
and/or increases catabolism. Derangements in immunity,
retinoid metabolism, and tissue repair responses may in-
teract, or even synergize, to increase the severity of infec-
lions in vitamin A-deficient animals and humans.
We acknowledge with gratitude the many contributions of our co-
workers and the support of National Institutes of Health research grants
DK-41479 (A.C.R.) and HD-30293 (C.B.S.)
REFERENCES
1. Gottgens, B., and Green, A. R. (1995) Retinoids and the differentiation of
lymphohaemopoietic stem cells. BioEssays 17. 187-189
2. Ross, A. C., and Hammerling, U. (1994) Retinoids and the immune system.
In The Retinoids. Biology, Chemistry and Medicine, 2nd Ed (Spom, M. B.,
Roberts, A. B., Goodman, D. S., eds) pp. 521-544, Raven Press, New York
3. Sommer, A. (1992) Vitamin A deficiency and childhood mortality. Lancet
339,864
4. Glasziou, P. P., and Mackerras, D. E. (1993) Vitamin A supplementation in
infectious diseases: a mets-analysis. Br. Med. J. 306, 366-370
5. Beaton, C. H., Martorell, R., Aronson, K. A., Edmonston, B., McCabe, C.,
Ross, A. C., and Harvey, B. (1994) Vitamin A supplementation and child
morbidity and mortality in developing countries. Food Nwr. Bull. 15,
282-289
6. Kirkwood, B. R., Arthur, P.,Ross, D. A., Morris, S. S., Gyapong, J.0..
Tomkins, A. M. (1993) Vitamin A supplementation in northern Ghana:
effects on clinic attendances, hospital admissions, and child mortality.
Lancet 342, 7-12
7. Son,mer, A., Rahmathuilah, L., Underwood, B., Milton, R. Reddy, V., West,
K.,Daulaire, N., Stukel, T., Herrera, C. Stansfield, S., Ross, D., Kirkwood,
B. R., Arthur, P., Morris, S., Kjolhede, C., Dibley, M., Barreto, M., Bhan,
M. K.,and Cove, 5. (1995) Potential interventions for the prevention of
childhood pneumonia in developing countries. A mets-analysis of data from
field trials to assess the impact of vitamin A supplementation on pneumonia
morbidity and mortality. Bull. World Health Org. 73, 609-619
8. Bang, F. B., and Foard, M. A. (1971) The effect of acute vitamin A deficiency
on the susceptibility of chicks to Newcastle disease and influenza viruses.
Johns Hopkins Med.). 129, 100-109
9. Alexander, D. J. (1991) Newcastle disease and other paramyxovirus infec-
tions. In Diseases of Poultry, 9th ed (Calnek, B. W., ed) Chapt. 19, pp.
496-519, Iowa State University Press, Ames, Iowa
10. Squibb, R. L.. and Veros, H. (1961) Avian disease virus and nutrition
relationships. 1. Effect of vitamin A on growth, symptoms mortality and
vitamin A reserves of white leghorn chicks infected with Newcastle disease
virus. Poultry Sci. 40,425-453
11. Sijtsma,S. R., West, C. E., Rombout,J. H.,and Van derZijpp, A. i. (1989)
The interaction between vitamin A status and Newcastle disease virus
infection in chickens.). Nazi-. 119,932-939
12. Sijtsma, S. R., Rombout, J. H., Kiepurski, A., West, C. E,, and van derZijpp,
A. J. (1991) Changes in lymphoid organs and blood lymphocytes induced by
vitamin A deficiency and Newcastle disease virus infection in chickens. Dev.
Comp.immunol. 15,349-356
13. Rombout, J. H., van Rens, B. 1., Sijtsma, S. R., van der Weide, M. C., and
West, C. E. (1992) Effects of vitamin A deficiency and Newcastle disease
virus infection on lymphocyte subpopulations in chicken blood. Vet. Immu-
nol. Immunopathol. 31,155-166
14. Sijisma, S. R., Rombout, .1. H., West, C. E., and van der Zijpp, A. J. (1990)
Vitamin A deficiency impairs cytotoxic I lymphocyte activity in Newcastle
disease virus-infected chickens. Vet. Immunol. Irnmunopathol. 26,191-201
15. Davis, C. Y., and Sell, J. L. (1989) Immunoglobulin concentrations in serum
and tissues of vitamin A-deficient broiler chicks after Newcastle disease virus
vaccination. Poultry Sci. 68, 136-144
16. Rombout, J. H.,Sijtsma,S. R.,West,C. E., Karabinis, Y., Sijtsma, 0. K.,
Van der Zijpp, A. J., and Koch, G. (1992) Effect of vitamin A deficiency and
Newcastle disease virus infection on IgA and 1gM secretion in chickens. Br.
J.Mar. 68, 753-763
17. West, C. E., Sijtsma, S. R., Kouwenhoven, B., Rombout, J. H., and van der
Zijpp, A. j. (1992) Epithelia-damaging virus infections affect vitamin A
status in chickens.). Nutr. 122, 333-339
18. Bang, F. B., Bang. B. G., and Foard, M. (1975) Acute Newcastle viral
infection of the upper respiratory tract of the chicken. II. The effect of diets
deficient in vitamin A on the pathogenesis of the infection. Am. J. Pathol.
78,417-426
19. Bang, B. G., and Bang, F. B. (1969) Replacement of virus-destroyed epithe..
hum by keratinized squamous cells in vitamin A-deprived chickens. Proc.
Soc. Exp. Biol. Med. 132,50-54
20. Hussey G. D., and Klein M. (1990) A randomized, controlled trial of vitamin
A in children with severe measles. New Engl. I. Med. 323, 160-164
21. Leigh M. W., Carson J. L., and Denny F. W. (1991) Pathogenesis of
respiratory infections due to influenza virus: implications for developing
countries. Rev. Infect. Dis. 13, S501-S508
22. Ahmed, F.,Jones, D. B., and Jackson, A. A. (1990) The interaction of vitamin
A deficiency and rotavirus infection in the mouse. Br. I. Nutr. 63, 363-373
23. Ahmed, F., Jones, D. B., and Jackson, A. A. (1991) Effect of vitamin A
deficiency on the immune response to epizootic diarthoea of infant mice
(EDIM) rotavirus infection in mice. Br.). Nutr. 65,475-485
24. Smith, S. M., and Hayes. C. E. (1987) Contrasting impairments in 1gM and
IgC responses of vitamin A-deficient mice. Proc. Nod. Acad. Sci. USA 84,
5878-5882
25. Stephensen, C. B., Blount, S. R., Schoeb. 1. R.,and Park, J.Y. (1993)
Vitamin A deficiency impairs some aspects of the host response to influenza
A virus infection in BALB/c mice.). Nuir. 123,823-833
26. Stephensen, C. B., Moldoveanu, Z., and Cangopadhyay, N. N. (1996) Vitamin
A deficiency diminishes the salivary immunoglobulin A response and en-
hances the serum immunoglobulin G response to influenza A virus infection
in BALB/c mice.). Nuir. 126,94-102
27. Mossmann, T. R., and Sad, S. (1996) The expanding universe of 1-cell
subsets: Thl, Th2 and more. Immunol. Today 17, 138-146
28. Sirisinha, S., Darip, M. D., Moongkamdi, P., Ongsakul, M., and Lamb, A. J.
(1980) Impaired local immune response in vitamin A-deficient rats. Clin.
Exp.Immunol. 40, 127-135
29. Puengtomwatanakul, S., and Sirisinha, S. (1986) Impaired biliary secretion
of immunoglobulin A in vitamin A-deficient rats. Proc. Soc. Exp. Bid. Med.
182,437-442
30. Wiedermann, U., Hanson, L. A., Holmgren, J., Kahu, H., and Dahlgren. U.
I. (1993) Impaired mucosal antibody response to cholera toxin in vitamin
A-deficient rats immunized with oral cholera vaccine [published erratum
appears in Infect. Immunol. (1993 61, (12) 5431]. Infect. Immunol. 6,
3952-3957
31. Cantoma, M. 1., Nashold, F. E., and Hayes, C. E. (1994) In vitamin A
deficiency multiple mechanisms establish a regulatory I helper cell imbal.
ance with excess Th1 and insufficient ‘I’h2 function. ). Immwiol. 152,
1515-152
32. Nauss, K. M., Anderson, C. A., Conner, M. W., and Newberne, P. M. (1985)
Ocular infection with herpes simplex virus (HSV-1) in vitamin A-deficient
and control rats.). Nuzr. 115,1300-1315
33. Nauss, K. M., and Newbeme, P. M. (1985) Local and regional immune
function of vitamin A-deficient rats with ocular herpes simplex virus (HSV)
infections. I. Nuzr. 115,1316-1324
34. Smolin,C.,Okumoto, M., Fnedlaender, M., and Kwok, 5. (1979) Herpes
simplex keratitis treatment with vitamin A. Arch. Ophthai. 97,2181-2183
35. Starr, M. B., Dawson, C. R., Briones, 0., and Oh, J. (1981) Vitamin A in
experimental herpetic keratitis. Arch.Op!uhdmol. 99,322-326
36. Watson, R. R., Yahya, M. D., Darban, H. R., and Prahhala, R. H. (1988)
Enhanced survival by vitamin A supplementation during a retrovirus infec-
tion causing murine AIDS. L(e Sci. 43. 13-18
37. Towers, C., Harris, J., Lang, C., Collins, M. K., and Latchman, D. S. (1995)
Retinoic acid inhibits both the basal activity and phorbol ester-mediated
activation of the HIV long terminal repeat promoter. AIDS 9, 129-136
38. Poli, G., Kinter, A. L., Justement, J. S., Brassier, P., Kehrl, J. H., and Fauci,
A. S. (1992) Retinoic acid mimics transforming growth factor beta in the
regulation of human immunodeficiency virus expression in monocytic cells.
Proc. Nazi. Acad. Sci. USA 89,2689-2693
39. Huan, B., and Siddiqui, A. (1992) Retinoid X receptor RXR alphabindsto
and trans-activates the hepatitis B virus enhancer. Proc. Nod. Acad. Sci. USA
89,9059-9063
40. Bowman, l.A.,Goonewardene, I. M., Pasatiempo. A. M. C., Ross, A. C., and
Taylor, C. E. (1990) Vitamin A deficiency decreases natural killer cell
activity and interferon production in rats.). Nusr. 120, 1264-1273
41. Zhao, Z. Murasko, D. M., and Ross, A. C (1994) The role of vitamin A in
natural killercell cytotoxicity, number and activation in the rat. Nat. immun.
Cell Growth Regal. 13,29-41
42. Zhao, Z., Matsuura,I., Popoff, K., end Ross, A. C. (1994) Effects of
N-(4-hydroxyphenyl) retinamide on the number and cytotoxicity of natural
killer cells in vitamin A-sufficient and -deficient rats. Nat. Immun. Cell
Growth Regal. 13,280-288
43. Griffin, D. E., Ward, B. J., .Jauregui, E.. Johnson, R. I., and Vaisberg, A.
(1990) Natural killer cell activity during measles. Clin. Exp. Immunol. 81,
218-224
44. Cantoma, M. I., Nashold, F. E., and Hayes, C. E. (1995) Vitamin A
deficiency results in a priming environment conducive forTh! cell develop-
ment. Eur. J. Immunol. 25, 1673-1679
45. Bhatti, L, and Sidell, N. (1994) Transcriptional regulation by retinoic acid of
interleukin-2a (alpha) receptors in human B cells. Immunology 81,273-279
46. Sidell, N., Chang, B., and Bhatti, L. (1993) Up-regulation by retinoic acid of
interleukin-2-receptor mRNA in human 1 lymphocytes. Cell.Immunol. 146,
28-37
47. Felli, M. P., Vacca, A., Meco, D., Screpanti, I., Farina, A. R.. Maroder, M.,
Martinotti, S., Petrangeli, E., Frati, L., and Gulino, A. (1991) Retinoic
acid-induced down-regulation of the interleukin-1 promoterviacia-regula-
torysequences containing an octamer motif. Mol. Cell. Biol. 11,4771-4778
48. Blomboff, R., Green, M. H., Green, J. B., Berg, T., and Norum, K. R. (1991)
Vitamin A metabolism: new perspectives on absorption, transport and stor-
age. Physiol. Rev. 71,951-990
49. Shingleton, J. L, Skinner, M. K., and Ong, 0. E. (1989) Characteristics of
retinol accumulation from serum retinol-binding protein by cultured Sertoli
cells. Biochemistry 28,9641-9647
50. Neuzil, K. M., Gruber, W. C., Chytil, F. Stahlman, M, I.,Engelhardt,B..and
Graham, B. S. (1994) Serum vitamin A levelsin respiratory syncytial virus
infection. ). Pediatr. 124,433-436
51. Siephensen, C. B., Alvarez, J. 0.,Kohatsu,J.,Hardmeier,R.,Kennedy,J.I.,
Jr., and Gammon, R. B., Jr. (1994) Vitamin A is excreted in the urine during
acute infection. Am.). Clin. Nutr. 60,388-392
52. Alvarez, J.0.,Salazar-Lindo, E., Kohatsu,J.Miranda,P.,and Stephensen,
C.B.(1995) Urinary excretion of retinol in children with acute diarrhea. Am.
J. Clin. Nuir. 61, 1273-1276
53. Rosales, F. J.,Ritter,S.J.,Zolfaghari,R.,Smith,i.E.,and Ross, A. C. (1996)
Effects of acute inflammation on plasma retinol, retinol-binding protein, and
its mRNA in the liver and kidneys of vitamin A-sufficient rats.). Lipid Res.
37,962-971
