| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Biology Division, California Institute of Technology, Pasadena, CA 91125
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-galactosidase has no discernable
effect on inflammation. In addition, comparison of the CFA responses in
LIF knockout vs wild-type skin reveals that LIF is an important
regulator of IL-1, IL-6, IL-7, IL-2R
, and IFN-
in cutaneous
inflammation. These and our previous data indicate that both endogenous
and exogenous LIF are anti-inflammatory in the CFA model and that
LIF is a key regulator of the cytokine cascade. The results also
indicate that adenoviral gene delivery can be an effective therapeutic
approach in this paradigm. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and TGF- also being implicated. The cytokines that play key
regulatory roles in this cascade are of clinical interest as potential
targets of therapeutic intervention; e.g., subcutaneous delivery of
IL-10 or -11 to patients with psoriasis can ameliorate this condition
(5, 6). Leukemia inhibitory factor (LIF)6 is also important in this context, as this cytokine can induce proinflammatory as well as anti-inflammatory effects, depending on the tissue and the form of perturbation (reviewed in Ref. 7). Experiments with LIF knockout (KO) mice have shown that endogenous LIF is required for normal neuronal and glial responses to injury in the central and peripheral nervous systems (7, 8). Following sciatic nerve injury, LIF is required for normal inflammatory cell infiltration, and LIF is directly and indirectly chemotactic for macrophages (8, 9). In the periphery, injection of high concentrations of LIF into skin or joints can induce swelling and leukocyte invasion (10, 11). In striking contrast, we found that endogenous LIF is anti-inflammatory in the CFA model of cutaneous inflammation and that injection of LIF strongly reduces pain in this model (12). To clarify further the role of LIF in the cytokine cascade of cutaneous inflammation and to determine the efficacy of viral delivery in treating this condition, we have investigated CFA-induced inflammation in normal and LIF KO mice and used a LIF adenoviral construct to deliver the cytokine. We also report the first findings on the cytokine cascade induced by CFA in the skin, some of which are surprising.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The strain of LIF KO-/- mice used in this work was that of Stewart et al. (13), which has been intermittently backcrossed with the C57BL/6 strain to maintain fertility and viability. Genotypes were determined by the PCR (12), and all comparisons were made only among mice of the same age. All animal procedures conformed to the requirements of the Caltech Research Animal Care and Use Committee.
Adenovirus construction
The adenoviral vector was derived from adenovirus serotype 5 deleted of its E1 and E3 regions. Recombinant adenovirus containing lacZ (AdlacZ), LIF (AdLIF), or no transgene (AdPacI) were constructed with the transgene under the control of the CMV promoter. The 293 cells expressing the E1A and E1B proteins support the replication of E1A-defective mutants (obtained from N. Davidson’s laboratory at Caltech) and were maintained in DMEM (Life Technologies, Rockville, MD) containing 2% FCS. Following infection, cells were harvested, resuspended in culture medium, and subjected to three rounds of freezing and thawing. Cell debris were eliminated by centrifugation at 3000 x g at 4°C for 15 min. Crude viral supernatants were layered on a discontinuous CsCl gradient and centrifuged for 2 h at 36,000 rpm. Isolated viral particles were then subjected to a second round of CsCl centrifugation. Viral particles were collected, dialyzed against buffer containing 10 mM HEPES (pH 7.5), 140 mM NaCl, 1 mM MgCl, and 10% glycerol, aliquoted, and frozen at -80°C. Viral titers were assayed by plaque assay on 293 cells and expressed as PFUs/ml.
Assay of LIF expression by adenovirus
Confluent rat skin fibroblasts were prepared from dorsal skin of adult rats by standard methods (14). Small pieces of dermis (1–3 mm3) were explanted to 60-mm dishes and incubated in DMEM for 7–10 days. Fibroblasts growing out on the dish were harvested with 0.05% trypsin and subcultured in 24-well plates for up to 10 passages. Cells were infected with virus at 37°C for 2 h and then washed with DMEM. Wells were incubated with 250 µl of DMEM for 1–3 days. This conditioned medium (100 µl) was incubated on M1 cells (5 x 102 cells/well) in 96-well microtiter plates for 3–7 days. Cell number was assayed using the MTT method (15)
In vivo gene transfer
Mice were anesthetized with isoflurane. Viral stocks of AdPacI, AdlacZ, and AdLIF were thawed and diluted in PBS (1 mM Ca2+ +1 mM Mg2+), and a final volume of 10 µl (3 x 109 particles) was injected intradermally with a Hamilton microsyringe (80301) and a Hamilton needle (90033; Hamilton, Reno, Nevada) into the skin of the footpad.
CFA-induced inflammation
An acute inflammatory lesion was produced by injection into the plantar surface of the hind paw, under isoflurane anesthesia, of 20 µl of CFA (1 mg/ml Mycobacterium tuberculosis (H37Ra, ATCC 25177) in 0.85 ml of paraffin oil and 0.15 ml of mannide monooleate (Sigma, St. Louis, MO). Edema was assayed by measuring paw diameter using calipers applied across the dorsoventral midplane of the hind paw.
Histochemical staining for -galactosidase
Footpads were removed from mice following perfusion with PBS
buffer and then 2% paraformaldehyde, postfixed with 2%
paraformaldehyde for 30 min, and then immersed in 30% sucrose in PBS
overnight before OCT embedding. Frozen sections of 10 µm were taken
at multiple levels through the block. Following immersion in
5-bromo-4-chloro-3-indolyl -D-galactoside (X-Gal)
reagent (5 mM K3Fe(CN)6, 5
mM K4Fe(CN)6, 25 mM
MgCl2, and 2 mg/ml X-Gal; Roche Molecular
Biochemicals, Indianapolis, IN) for 4 h at 37°C, sections were
counterstained with eosin and examined by light microscopy.
Double staining for LIF protein and -galactosidase
Sections stained with X-Gal reagent were blocked with 2% swine serum for 2 h at room temperature (RT) and then immunostained overnight at 4°C with polyclonal goat anti-LIF Abs (R&D Systems, Minneapolis, MN) at a 1:2000 dilution in 2% swine serum in PBS. Biotinylated swine anti-goat IgG (Roche) was then applied for 2 h at RT. Slides were mounted with Permount and examined by light microscopy.
Analysis of cellular infiltration
Macrophages were identified by immunostaining with the F4/80 Ab, while the Mac-1 Ab was used to stain macrophages and neutrophils (BioSource International, Camarillo, CA). Frozen sections were rinsed three times in PBS at RT and endogenous peroxidase was blocked with 3% H2O2 in PBS for 10 min at RT. Nonspecific Ab binding was blocked with 2% normal rabbit serum (Sigma) for 2 h. The primary Abs were diluted with 2% normal rabbit serum in PBS (1:2500 for Mac-1 and 1:250 for F4/80) and applied to sections overnight at 4°C. Sections were incubated with biotinylated rabbit anti-rat IgG at a 1:200 dilution for 1 h, followed by the developing solutions from the avidin-biotin complex kit and substrate very intense purple kits (Vecastain Elite avidin-biotin complex kit; Vector Laboratories, Burlingame, CA). Immunostaining was quantified by optically scanning photographs using frames placed in the AdLIF or AdlacZ injection sites. Four frames were used for each section and three sections were used per footpad. Data from four footpads of two mice of each group were used to calculate the mean and SEM. Background OD outside of the skin tissue was subtracted from each frame before quantification.
RNase protection assay
Footpad skin was removed under deep terminal pentobarbital
anesthesia, and total RNA was extracted using the single-step method
(16). Total RNA from two feet of the same mouse was
treated as one sample. Three mice were used for each group. Multiprobe
template mCK-3b and mCK-L (custom template sets; PharMingen, San Diego,
CA) contained DNA templates for the following cytokines: TNF- and
-, IL-6, IFN-, TGF-1, -2, and -3, IL-12p32 and p40,
IL-7, IL-1, IL-1, IL-1R antagonist, LIF, IL-2R, and the
housekeeping genes L32 (ribosomal RNA) and GAPDH. The assay was
performed according to the manufacturer’s protocol. Briefly, DNA
templates were used to synthesize probes, which were labeled with
[-32P]UTP (3000 Ci/mmol, 10 µCi/µl;
DuPont NEN Research Products, Boston, MA) in the presence of a GACU
pool using a T7 polymerase. Each probe was hybridized overnight at
56°C with 15–20 µg of total RNA, followed by digestion with RNase
A and T1. The samples were then treated with proteinase K-SDS mixture,
extracted with phenol-chloroform, and precipitated in the presence of
ammonium acetate. Samples were loaded on an acrylamide-urea sequencing
gel, with undigested labeled probes, and run at 50 W with 0.5x TBE
(208 g Tris base, 55 g Boric acid, 40 ml 0.5 M Tris EDTA (pH 8.0),
autoclaved 20 min.) for 2.5 h. The gel was dried on filter paper
under vacuum. For quantitation, the dried gel was placed in a
PhosphorImager (Molecular Dynamics, Sunnyvale, CA) cassette and exposed
to a Kodak Storage phosphor screen (Kodak, Rochester, NY) at RT for
1–2 wk. This screen was scanned with a QuantiScan sensitometer, and
analysis was done with ImageQuant (Molecular Dynamics) software.
Background OD in the same lane (obtained from a frame of the same size)
was subtracted from the cytokine band and the resulting value was
expressed as a ratio to L32 and GAPDH.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The AdLIF and AdlacZ viruses were produced as described above and tested for LIF activity using the M1 cell assay. Recombinant LIF at 10 ng/ml inhibits growth 68%, while supernatants from cells infected with AdLIF and AdlacZ inhibit growth 44 and 4%, respectively. In a similar experiment, we found that an anti-LIF Ab is able to block this activity.
In addition, anti-LIF Abs stain dermal cells in the skin of the
footpad 4 days following infection with AdLIF and AdlacZ
(Fig. 1
C) but not following
infection with AdlacZ alone (Fig. 1D). To confirm
infection by the latter virus, sections were stained for
-galactosidase and are positive (Fig. 1A). We were also
able to independently localize viral infection in the case of AdLIF by
coinfecting with AdlacZ and double-staining sections for LIF
and -galactosidase (Fig. 1C). As expected, both viruses
infect the same localized area, and in a minority of cases, single
cells stain for both proteins.
|
To determine the optimal time frame in which to study cutaneous inflammation in this CFA model, edema as well as Mac-1 and F4/80 staining of macrophages was quantified in wild-type (WT) mice at 6 h and 1, 2, 3, and 6 days after CFA injection. The peak of edema and F4/80 staining is clearly at 2 days, whereas peak Mac-1 staining is nearly equivalent at 1 and 2 days (data not shown). Therefore, the 2-day time point was chosen for subsequent experiments.
Injection of CFA increases footpad thickness 55% in WT and 92% in the
LIF KO mice (Fig. 2). This result is very
similar to the 2-fold difference seen in our previous work
(12), indicating that endogenous LIF suppresses the edema
induced by CFA. In this experiment, the controls were injected with
AdlacZ virus as well as CFA to control for effects of viral
infection in the AdLIF experiments. Additional experiments showed,
however, that injection of AdlacZ causes no additional
swelling, with or without CFA injection (data not shown). Infection
with AdLIF at the same time as CFA results in suppression of the
swelling response to CFA, both in the WT and KO. In the WT, the
exogenous LIF suppresses swelling by 
45% and in the KO, exogenous
LIF suppresses swelling by about 35%. Both of these differences are
significant (Student’s t test, p < .001;
n = 4).
|
To assess inflammatory cell infiltration following CFA injection,
sections were stained with Mac-1 and F4/80. Injection of CFA increases
Mac-1 staining of macrophages and neutrophils 5-fold in the WT and
7-fold in the LIF KO (Figs. 3 and 4). This difference between the genotypes
in their response to CFA was observed previously, where we found
4.8-fold more neutrophils and 2-fold more mast cells in the inflamed
dermis of KO vs WT mice, both in the absence of AdlacZ
infection (12). The fact that Mac-1 staining does not rise
even higher in the LIF KO could reflect that action of other
anti-inflammatory cytokines such as IL-10 or IL-11, as considered
in Discussion. Injection of AdLIF suppresses the CFA-induced
increase in Mac-1 staining in both genotypes; in the WT, the increase
with CFA in the presence of AdLIF is only about one-seventh of normal
and about one-third in the KO with AdLIF (p <
0.05). As a control for the viral infection in the AdLIF experiments,
the controls were also injected with AdlacZ. No difference
in Mac-1 or F4/80 staining was observed with AdlacZ
infection when compared with paired footpads, with or without CFA (data
not shown).
|
|
4-fold in the WT and over 4-fold in
the KO (Figs. 5 and 6). These differences are significant
(p < 0.05). Thus, exogenous LIF, delivered
simultaneously with CFA, effectively suppresses inflammatory cell
infiltration nearly to background levels.
|
|
To further assess the role of LIF in this paradigm, we assayed the
mRNA levels for a variety of cytokines and their receptors. This was
done in both WT and KO footpads, with and without CFA injection. The
results are expressed as the fold induction of each cytokine caused by
CFA. Although this type of data has not been previously published for
this paradigm to our knowledge, we predicted that CFA would increase
the levels of proinflammatory cytokines. Indeed, in the WT footpads,
CFA strongly induces IL-6 (p < 0.03), and
induces other proinflammatory cytokines such as TNF-
(p < 0.08), IL-1 (p
< 0.01), and the receptor for IL-2
(IL-2R)(p < 0.03) to a lesser degree (Fig. 7). Interestingly, CFA lowers the level
of IFN- (p < 0.05), a proinflammatory
cytokine, and raises the levels of the anti-inflammatory cytokine
TGF-1 (p < 0.03). It may be that at this
2-day time point, the anti-inflammatory response has begun to
restore the normal state of the tissue. The other apparent effects of
CFA on cytokine expression in the WT (TNF-, TGF-, TGF-3, IL-7,
and IL-1,) are not statistically significant. The p
values given here represent comparisons of cytokine mRNA levels in the
presence and absence of CFA to those in WT, where n = 3
for each.
|
response in the LIF KO than in the WT
(p < 0.05), which reproduces our earlier
finding on IL-1 protein (12). Moreover, in the absence
of LIF, the increases in IL-2R (p < 0.02)
and IL-7 (p < 0.04) are greater than in the
WT, and the CFA-induced suppression of IFN- is turned into an
increase in the absence of LIF (p < 0.001).
These results are consistent with an anti-inflammatory role of
endogenous LIF. Surprisingly, however, the CFA-induced increase in IL-6
is much less in the LIF KO than in the WT (p <
0.03) (Fig. 7
). This strong effect is considered further in
Discussion. Clearly, in the CFA model, the endogenous LIF
response to CFA regulates a number of important cytokines and their
receptors. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The present results with the LIF KO mouse confirm our previous
experiments showing that CFA-induced edema and inflammatory cell
infiltration are more severe in the absence of endogenous LIF. The
prior work also found that injection of recombinant LIF protein reduced
the CFA-mediated induction of IL-1 and nerve growth factor, but this
injection was unable to suppress edema (12). The present
approach of raising LIF levels via an adenoviral vector is more
efficacious, reducing swelling in the WT by almost half as well as very
strongly suppressing inflammatory cell infiltration and proinflammatory
cytokine levels. The suppression of macrophage infiltration was 3- to
7-fold, depending on the genotype of the mouse and the macrophage
marker used. These strong effects may be surprising in view of the fact
that only a small minority of skin cells were obviously infected by the
virus, and production and release of LIF would be delayed in comparison
to the injection of LIF protein itself. The relative effectiveness of
the viral approach may be due to its longer lasting effects relative to
an injected protein. We have found strong anti-LIF staining for at
least 2 wk following delivery of AdLIF to the brain (K. Oishi, S.
C. Lee, and P. H. Patterson, unpublished observations). The use of
an adenoviral vector in this context is not without potential problems,
however, as there are reports of a similar virus causing an
inflammatory reaction (cf Ref. 17). Under the conditions
used here, we did not observe any evidence of such a reaction to
either theAdlacZ or the AdPac1, and the AdLIF construct
suppressed inflammation rather than inducing it. This lack of obvious
response to the control vectors could be due to the relatively low
levels of virus used in these experiments. It would be worthwhile
investigating the effect of a second challenge with virus, as this has
been reported to elicit a strong inflammatory reaction at the original
site of infection, at least in the brain (18).
Our findings that both endogenous and exogenous LIF are anti-inflammatory in the skin are inconsistent with results reported by others. Injection of high doses of LIF into the ear pinnae of mice increased ear thickness (11), and systemic injection of high levels of LIF induced a prolonged hypersensitivity to mechanical stimulation (19). We previously found that injection of low amounts of LIF into the footpad is analgesic (12). It may be that exogenous LIF displays a biphasic dose dependence in peripheral inflammation, with low doses, which likely correspond to endogenous levels induced by injury, being anti-inflammatory (7). It is also worth noting that within the peripheral and CNS, LIF is strongly proinflammatory, inducing inflammatory cell infiltration, microglial and astrocyte activation, and neuronal injury responses (7, 8). It may be that the skin contains a cell type not present within the nervous system, and this cell may respond to LIF by producing a strong anti-inflammatory agent.
We compared the responses of a variety of cytokines and their receptors
to CFA in WT vs LIF KO skin. These data show that LIF is an important
regulator of IL-1, IL-2R, IL-6, IL-7, and IFN-. Although there
were no previous data available on cytokine expression in the CFA skin
model, considerable work has been done on cytokine changes in other
forms of cutaneous inflammation, particularly psoriasis (1, 2, 3, 4, 20). A number of type 1 cytokines are up-regulated, including
IL-2, IL-6, IL-8, IL-12, IL-18, IFN-, and TNF-. In contrast,
relatively low levels of anti-inflammatory cytokines such as IL-1R
antagonist and IL-10 have been found in this chronic condition.
Perturbation of endogenous cytokines or application of exogenous
cytokines has been used to assess their relative roles. In a study of
the role of endogenous IL-10 in cutaneous wound healing, it was shown
that local injection of an anti-IL-10 Ab caused increased
inflammatory cell infiltration and increased levels of chemokines and
proinflammatory cytokines (21). Transgenic mice
overexpressing IL-1 in keratinocytes exhibit spontaneous skin
disease and focal inflammatory lesions (22). Injection of
IL-12 can overcome UV-induced immune suppression, and this cytokine can
overcome or reverse superantigen-induced suppression (23, 24). Local injection of IL-1, but not IL-1 or TNF-,
mimicked cytokine changes induced by allergen, and injection of
anti-IL-1 Abs prevented sensitization to allergen
(25). Notably, preliminary clinical trials have reported
amelioration of psoriatic lesions and suppression of proinflammatory
cytokines upon local application of IL-10 (5) or IL-11
(6). The latter result can be viewed as very consistent
with our results, as IL-11 and LIF belong to the same cytokine family
and induce the same signaling cascades. It has not yet been shown,
however, that the receptors for LIF and IL-11 are on the same cells in
human or mouse skin. The targets and sources of these cytokines in
cutaneous inflammation is an area that deserves more attention.
The roles of IL-1 and TNF have also been investigated in their
respective receptor KO mice. The skin of such mice does not display
serious abnormalities in response to organic solvent, detergent, or
mechanical disruption (26), despite the fact that levels
of these cytokines are normally increased by such treatments. In
contrast, IL-7 is a keratinocyte-derived growth factor important for
the development of 
T cells, including mouse dendritic epidermal T
cells (DETC) (27). Expression of this cytokine is
increased in psoriasis (28), and injection of an
anti-IL-7 Ab inhibits T cell accumulation in lymph nodes
following skin sensitization (29, 30). Moreover,
transgenic mice overexpressing IL-7 in keratinocytes develop a
lymphoproliferative skin disease (31, 32, 33, 34). Our finding
that IL-7 is up-regulated by CFA and that LIF suppresses this is
therefore of added interest.
In view of our findings that IL-1, IL-7, IFN-, and IL-2R are
all increased in the absence of LIF, it is surprising that the response
of IL-6 to CFA is reduced in the LIF KO. This is not consistent with
the view of IL-6 being proinflammatory and LIF exerting
anti-inflammatory effects in this model. We found previously that
endogenous IL-6 and LIF are each proinflammatory in brain injury, but
that the double KO did not exhibit a more pronounced phenotype than the
single KOs (8). We suggested that rather than acting in
redundant, parallel pathways, these two cytokines may be acting in
series, in the same pathway, one controlling the expression of the
other. Our present result in cutaneous inflammation can be viewed as
consistent with this notion, suggesting an anti-inflammatory role
for IL-6. Supporting this hypothesis is the result from a model of skin
damage with UVB irradiation, where it was found that IL-6 KO mice
displayed a more severe reaction, and IL-10 levels were suppressed
compared with WT (35). This is also consistent with the
IL-11 results discussed above, as LIF, IL-6, and IL-11 belong to the
same family and induce the same signal cascades. Each of these
cytokines is increased in lesional psoriatic skin (36) and
is thus a potential therapeutic agent.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Department of Pathology, University of California, Los Angeles School of Medicine, 675 Young Drive South, Los Angeles, CA 90095.
3 Current address: Department of Pharmacology, Meiji Pharmaceutical University, 2-522-1 Noshio, Kiyose, Tokyo 204-8588, Japan.
4 Current address: Protein Engineering Division, Korea Research Institute of Bioscience and Biotechnology, P.O.B. 115 Yusong, Taejon 305-600, Korea.
5 Address correspondence and reprint requests to Dr. Paul H. Patterson, 216-76, California Institute of Technology, Pasadena, CA 91125.
6 Abbreviations used in this paper: LIF, leukemia inhibitory factor; AdlacZ, adenovirus coding lacZ; AdLIF, adenovirus coding LIF; AdPacI, adenovirus with no transgene; X-Gal, 5-bromo-4-chloro-3-indolyl -D-galactoside; RT, room temperature; KO, knockout; WT, wild type.
Received for publication July 20, 2000. Accepted for publication November 8, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in the initiation of primary immune-responses in skin. J. Immunol. 150:3698.[Abstract]
T cells. Proc. Natl. Acad. Sci. USA 93:7172. cells involved in contact sensitivity preferentially rearrange the V3 region and require interleukin-7. Eur. J. Immunol. 27:206.[Medline]
This article has been cited by other articles:
|
|
 |
|
  R. A. Linker, N. Kruse, S. Israel, T. Wei, S. Seubert, A. Hombach, B. Holtmann, F. Luhder, R. M. Ransohoff, M. Sendtner, et al. Leukemia Inhibitory Factor Deficiency Modulates the Immune Response and Limits Autoimmune Demyelination: A New Role for Neurotrophic Cytokines in Neuroinflammation J. Immunol., February 15, 2008; 180(4): 2204 - 2213. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Bauer and P. H. Patterson Leukemia Inhibitory Factor Promotes Neural Stem Cell Self-Renewal in the Adult Brain. J. Neurosci., November 15, 2006; 26(46): 12089 - 12099. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. F. Berry, T. J. Pirolli, V. Jayasankar, K. J. Morine, M. A. Moise, O. Fisher, T. J. Gardner, P. H. Patterson, and Y. J. Woo Targeted overexpression of leukemia inhibitory factor to preserve myocardium in a rat model of postinfarction heart failure J. Thorac. Cardiovasc. Surg., December 1, 2004; 128(6): 866 - 875. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Bauer, S. Rasika, J. Han, C. Mauduit, M. Raccurt, G. Morel, F. Jourdan, M. Benahmed, E. Moyse, and P. H. Patterson Leukemia Inhibitory Factor Is a Key Signal for Injury-Induced Neurogenesis in the Adult Mouse Olfactory Epithelium J. Neurosci., March 1, 2003; 23(5): 1792 - 1803. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. K. Tofaris, P. H. Patterson, K. R. Jessen, and R. Mirsky Denervated Schwann Cells Attract Macrophages by Secretion of Leukemia Inhibitory Factor (LIF) and Monocyte Chemoattractant Protein-1 in a Process Regulated by Interleukin-6 and LIF J. Neurosci., August 1, 2002; 22(15): 6696 - 6703. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
