,
* Institutes of Microbiology & Immunology and
Neuroscience, School of Life Science, and
Center for Neuroscience, National Yang-Ming University, Taipei, Taiwan, Republic of China; and
Biomedical Group, Takara Shuzo Co., Ltd., Otsu, Shiga, Japan
Correspondence: Nan-Chi A. Chang, Ph.D., Institute of Microbiology & Immunology, School of Life Science, National Yang-Ming University, Taipei, Taiwan 112, R.O.C. E-mail: ncchang{at}ym.edu.tw
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
barrel structure with a carbohydrate-binding cleft similar to that of triose-phosphate isomerases. To further delineate the physiological significance of Ym1, we examined its expression patterns during mouse embryonic development and inflammation states elicited by agents other than parasitic infections in the peritoneal cavity and brain. This is the first report revealing prominent expression of Ym1 in early myeloid precursor cells of hematopoietic tissues—initially in the yolk sac and subsequently in fetal liver, spleen, and bone marrow. In nonhematopoietic systems, Ym1 was not detected in most of the tissues examined, with the exception of lung. Although no expression was detected up to gestation day 16.5 (E16.5), an increasing level of Ym1 expression in lung was detected from E18.5 on and persisted through adulthood. While most resident macrophages in various tissues examined are Ym1-negative, transient expression of Ym1 may be induced in their activated counterparts during inflammation in response to different stimuli in vivo, ranging from various chemical agents to brain injuries. The temporal and spatial expression in myeloid precursors and its transient induction in activated macrophages support the notion that Ym1 may be involved in hematopoiesis and inflammation. In addition, its putative functional association with heparin/heparan sulfate is discussed.
Key Words: murine development • myeloid precursor • activated macrophage
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
We have previously reported the identification, purification, and molecular characterization of a novel mammalian lectin Ym1, transiently expressed and secreted by activated peritoneal macrophages in response to nematode infections [8
]. The induced expression of Ym1 and the profound cellular changes paralleling its appearance in the murine peritoneal cavity 15 days postinfection suggest that Ym1 may bear functional significance to the development of host defense against the nematode. As a lectin, Ym1 has a binding specificity toward saccharides with a free amine group such as glucosamine (GlcN), galactosamine, oligomers of GlcN, and heparin. We have also resolved by X-ray crystallography the three-dimensional structure of Ym1 at 2.5
resolution, which revealed its saccharide binding site inside the triose phosphate-isomerase domain [9
]. Our initial data suggest that the physiological ligand of Ym1 may be heparin/heparan sulfate (HS) on cell surface or extracellular matrix (ECM) in vivo [8
]. Heparin/HS are the major structural constituents in proteoglycans at the cell surface and ECM. By selective binding to various mammalian lectins, cell-cell recognition and cell-ECM interactions are modulated in ways that are pivotal to processes such as matrix organization, cell adhesion, growth factor, or cytokine interaction with cell and ECM during embryonic development, hematopoiesis, inflammation, and tumor metastasis [1
, 10
].
During the pursuit of characterizing this novel macrophage mediator, we have noticed that Ym1 is also expressed constitutively in adult bone marrow, spleen, and lung [8 ]. However, the onset and the time course of Ym1 expression in these tissues during development require detailed studies. The apparent different modes of expression, i.e., inducible in the peritoneal residents and constitutive in lung and the bone marrow progenitors, strongly suggest that Ym1 expression is regulated differentially and that it may be endowed with distinct, functional significance in different tissues and states of macrophage development.
In the present study, we have therefore explored further the biological significance of Ym1, in particular, the temporal course and tissue loci of its expression during hematopoiesis and in experimental models of inflammation. The expression patterns of Ym1 during embryonic development were examined by immunohistochemistry and multitissue Western blot and/or Northern blot analyses. Cells expressing Ym1 were verified by double immunofluorescence using markers defining macrophages at a particular stage of differentiation, that is, ER-MP54 for early myeloid precursors [11 ] and F4/80 for mature macrophages [12 ]. Experimental paradigms to elicit an inflammatory state of peritoneal cavity and brain were also used to recruit activated macrophages and/or microglia at respective loci [13 ]. Expression of Ym1 in different tissues under distinct inducing conditions was examined. We also discuss the possible association of Ym1 with HS-mediating functions.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Induction of Ym1 expression and culture of peritoneal exudate cells (PEC)
Ym1 expression was examined by induction paradigms using intraperitoneal (i.p.) injection in each mouse, 1 ml sterilized agents; i.e., 4% thioglycollate (Difco, Detroit, MI), 1% Sephadex G-50 (Pharmacia, Upsala, Sweden), or 0.5 mg/ml Corynebacterium parvum (Immunochem Research, Inc., Hamilton, MT). Ym1 expression was also induced by Trichinella spiralis infection as described [8
] to serve as positive control. PEC elicited 4 days after injection of artificial irritants, 7 days after injection of C. parvum, or 15 days after T. spiralis infection were harvested, respectively. Total cell number in the lavaged fluid was determined by using a hemocytometer. The cells were cultured for 1 day in serum-free Dulbecco’s modified Eagle’s medium (Life Technologies, Grand Island, NY), supplemented with 10 µg/ml gentamycin at 37°C in 5% CO2. The culture supernatant of PEC (PECCS) was collected by centrifugation (800 g for 10 min) to separate from cells. PEC collected were seeded routinely onto microcoverslips (Matsunami, Japan) and allowed to adhere for 2 h in serum-free medium. The attached cells were then washed twice with phosphate-buffered saline (PBS; 20 mM phosphate buffer, pH 7.4, 0.85% NaCl), fixed, permeabilized in acetone (-20°C) for 10 min, and processed for double immunofluorescence as described below.
Induction of Ym1 expression in brain
A stab wound in the brain was generated as follows: mice in groups of three were first anesthetized by sodium pentobarbital (i.p., 50 mg/kg body weight). Using a 30-gauge needle, a unilateral lesion was introduced approximately 5 mm in depth into the right parieto-temporal cortex. Skin over the craniotomies was sutured, and the mice were usually recovered in approximately 30 min. At selected time points, i.e., 1, 3, and 6 days postoperation, mice were anesthetized, and their brains were removed after perfusion for vibratome sectioning. Alternating coronal sections were collected and processed for immunohistochemical examination of Ym1 and glial fibrillary acidic protein (GFAP).
Experimental seizure was induced in mice by pentylenetetrazole [PTZ; Research Biochemicals International (RBI), Natick, MA] or kainic acid (KA; RBI). Mice were injected (i.p.) with a single dose of KA (15 mg/kg body weight) or PTZ (50 mg/kg) three times at 2-day intervals. Mice that developed behavioral seizures were selected for analysis of Ym1 expression in brain. Three hours after the third injection of PTZ and 1, 3, or 6 days after the treatment of KA, mice were killed and perfusion-fixed, and vibratome sections of brains were collected and processed for immunocytochemistry. Brain sections obtained from mice received sterile saline injections were used as control.
Northern blot analysis
Total RNA was isolated by the guanidinium isothiocyanate method. Total RNA of various tissues (40 µg/lane) or PEC (2 µg/lane) was separated on 1% formaldehyde-agarose gels and transferred onto immobilon-N membranes (Millipore, Bedford, MA). Northern blot hybridization was carried out according to Sambrook et al. [14
] and probed with [-32P]dCTP (Amersham, Little Chalfont, UK)-labeled Ym1 cDNA fragment P1–P8 (nt 73–366 of the open reading frame) generated by polymerase chain reaction as described [8
]. The membranes were exposed to Kodak X-OmatTM film (Eastman Kodak, Rochester, NY) with an intensifying screen for autoradiography or were subjected to analysis in a Phosphoimager (Molecular Dynamics, Sunnyvale, CA).
Antibodies
Specific polyclonal antibodies elicited against purified Ym1 were produced in this laboratory as described previously [8
]. Preimmune sera obtained from respective rabbits were used as negative controls. Other primary antibodies used included anti--tubulin [N356, mouse monoclonal antibody (mAb), Amersham], ER-MP54 (rat mAb, Biogenesis, Bournemouth, UK), F4/80 (CI: A3-1, rat mAb, Serotec, Oxford, UK), scavenger receptor (SR; 2F8, rat mAb, Serotec), and GFAP (G-A-5, mouse mAb, Sigma Chemical Co., St. Louis, MO). Secondary antibodies, including fluorescein isothiocyanate (FITC)-conjugated donkey anti-rabbit immunoglobulin G (IgG), rhodamine-conjugated donkey anti-rat IgG, horseradish peroxidase-conjugated rabbit anti-mouse IgG, goat anti-rabbit IgG, and rabbit peroxidase-anti-peroxidase (PAP) antibodies were from Jackson ImmunoResearch Laboratories (Bar Harbor, ME).
Western blot analysis
Concentration of proteins extracted from selected tissues was determined by Coomassie blue protein assay reagent (Pierce, Rockford, IL). Proteins obtained from various tissues (150 µg/lane) or that released from PEC elicited by different agents (5 µg/lane) were separated in 10% sodium dodecyl sulfate-polyacrylamide gel according to Laemmli [15
]. Protein obtained from the culture supernatant of PEC elicited by T. spiralis infection (1 µg) was used as positive control. Subsequently, proteins resolved on gel were transferred onto 0.45 µm Immobilon-P (Millipore) using a semi-dry electroblotter (Pharmacia) at constant current (0.8 mA/cm2 membrane) for 2 h. Membranes were blocked, incubated with specific antibodies against Ym1 or preimmune sera (1:4000 dilution) overnight at 4°C, and processed for immunodetection [16
]. The same blots were subsequently subjected to -tubulin detection as loading controls. The positive bands were revealed with the chemiluminescent substrate (enhanced chemiluminescence, Amersham) on X-ray films (Eastman Kodak).
Immunohistochemistry and double immunofluorescence
Mice were each anesthetized by i.p. injection of sodium pentobarbital (50 mg/kg body weight) and heparin (100 IU/mouse). Tissues were snap-frozen and embedded in OCT Tissue-Tek® cryoprotectant. Tissues were sectioned on a cryostat (Leica, Knowlhill, UK; 15 µm), then air-dried and post-fixed in cold acetone at 4°C for 20 min. The sections were subjected to immunostaining according to the PAP method [17
] or double immunofluorescence method [18
].
For immunohistochemistry, sections were bleached with 0.3% H2O2 in PBS for 30 min to remove endogenous peroxidase activity, rinsed in PBS, and blocked from nonspecific sites with 3% bovine serum albumin containing 2% normal goat serum for 30 min. The sections were incubated with properly diluted primary antibodies at 4°C overnight and subsequently subjected to the sequential incubations of secondary antibodies and tertiary antibodies. Sections were counter-stained with 0.1% toluidine blue or cresyl violet and mounted with Entellan media (Merck, Rahway, NJ) for light microscopic photography (Olympus IMT2 with Differential Interference Contrast attachment).
To colocalize macrophage markers with Ym1 in various tissues, samples stained by the double immunofluorescence method were analyzed by confocal laser-scanning microscopy [18 ]. Tissue sections were incubated with two types of properly diluted primary antibodies, rabbit polyclonal antibody against Ym1 and rat mAb against ER-MP54 or F4/80, and were followed by FITC or rhodamine-labeled secondary antibodies. The coverslips were mounted using 50% glycerol in PBS and examined under a confocal microscope (Leica, TCS NT). Control sections were processed the same way as described, except that respective primary antibody was omitted.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
 View larger version (42K): [in a new window] |
Figure 1. Expression of Ym1 protein and transcript in various tissues of adult mice. (A) Western blot analysis of Ym1 protein expression was used using specific rabbit anti-Ym1 antibodies. PECCS was included as positive control. Preimmune serum was used as negative control. (B) Northern blot analysis of Ym1 transcript expression was conducted using Ym1 cDNA (P1–P8, 294bp) as probe. PEC was included as positive control. Ym1 transcripts were visualized by autoradiography. Molecular weight (MW) markers are as marked.
|
 View larger version (52K): [in a new window] |
Figure 2. Expression of Ym1 in yolk sac. (A) On E10.5, Ym1-positive cells were detected in the extra-embryonic yolk sac by immunocytochemistry. (B) Control section was stained using preimmune serum. (C–K) Double-immunofluorescence analysis of cells expressing Ym1 in the yolk sac on E10.5. (C) Ym1-positive cells were labeled by FITC, and (D) the same section was stained using antibody against ER-MP54 (rhodamine-labeled). (E) Merged image clearly indicated that Ym1+ cells are coexpressing ER-MP54, a marker for myeloid precursor cells. Ym1+ cells (F), however, were not able to express F4/80, a marker for mature macrophage. Results are shown in G and H. (I–K) The negative controls for each antibody were shown, respectively. Original bars, 50 µm.
|
Temporal course of Ym1 expression in liver
Northern blot data indicated that Ym1 transcripts are expressed transiently in fetal liver from E16.5 to P7.5. The transcript is no longer detectable from P14.5 and on (Fig. 3A
). Parallel expression of Ym1 protein was detected in fetal liver from E16.5 through P7.5 (Fig. 3B)
. Peak level of expression was reached by E18.5 to neonate (P0.5). The level then declined to "undetectable" from P14.5 on through adulthood (Fig. 3B)
. Immunohistochemical staining confirmed Ym1 expression in fetal liver cells at E16.5. Time-course studies indicated that Ym1+ cells increase in numbers from E18.5 to P0.5 and are rounded in shape (Fig. 3C
and 3D)
. From P2.5 to P10.5, the number of Ym1+ cells decreased markedly, and the cells were distributed unevenly in small clusters. No Ym1+ cells were detectable in the liver after P14.5. The dynamic change of Ym1 expression in liver during development as revealed by immunostaining results correlates well with that of Ym1 mRNA and protein. Double immunofluorescence staining and confocal image analyses further revealed that, in fetal liver, Ym1 is coexpressed with ER-MP54 (Fig. 3E
3F
3G)
but not F4/80 (Fig. 3H 3I
3J)
.
 View larger version (55K): [in a new window] |
Figure 3. Expression of Ym1 in liver during development. (A) Liver mRNA was prepared at selected stages of development and examined for the expression of Ym1 transcript in Northern blot using Ym1 cDNA (P1–P8, 294 bp in length) as probe. PEC was included as a positive control, and 18S ribosomal RNA (rRNA) of all samples was used as loading controls. (B) Protein extracts of liver obtained from mice at selected stages of development were analyzed by Western blot for Ym1 protein expression. PECCS was included as a positive control. Subsequently, the same membrane was probed for -tubulin expression as internal controls. A MW marker of 47.5 kDa is marked. (C, D) Immunocytochemical detection of Ym1 in fetal liver on E18.5. Ym1-positive cells were clearly detectable in the fetal liver (C). Sections stained using preimmune serum (D) were used as negative control. Original bars, 50 µm. (E–J) Double-immunofluorescence analysis of cells expressing Ym1 in the fetal liver on E18.5. Ym1-positive cells labeled by FITC (E) was also labeled by rhodamine (F), thereby coexpressing ER-MP54 (G; merge) in fetal liver. However, Ym1-positive cells (H) do not express mature macrophage marker, F4/80 (I, J). Original bars, 10 µm.
|
 View larger version (69K): [in a new window] |
Figure 4. Expression of Ym1 in spleen during development. (A) Spleen mRNA was prepared at selected stages of development and examined for the expression of Ym1 transcript in Northern blot using Ym1 cDNA (P1–P8, 294 bp in length) as probe. PEC was included as positive control, and 18S rRNA of all samples was used as loading controls. (B) Protein extracts of spleen obtained from mice at selected stages of development were analyzed by Western blot for Ym1 protein expression. PECCS was included as positive control. Subsequently, the same membrane was probed for -tubulin expression as loading controls. A MW marker of 47.5 kDa is marked. (C, D) Immunocytochemical detection of Ym1 in fetal spleen on E18.5. Ym1-positive cells were detected in the fetal spleen (C). Sections stained using preimmune serum were used as negative control (D). Original bars, 50 µm. (E–J) Double-immunofluorescence analysis of cells expressing Ym1 in the adult spleen. Ym1+ cells were labeled with FITC (E); ER-MP54+ cells were labeled with rhodamine (F). Merged image (G) indicated the colocalization of Ym1 and ER-MP54 in the same cells. Sections of adult spleen were double-stained with antibody against Ym1 and F4/80. (H) Results as shown indicate that Ym1+ cells were labeled with FITC, and F4/80+ cells were labeled with rhodamine (I); however, the merged confocal image (J) indicates that Ym1+ cells do not coexpress the mature macrophage marker F4/80. Original bars, 50 µm.
|
 View larger version (78K): [in a new window] |
Figure 5. Expression of Ym1 in bone marrow. (A) Immunocytochemical detection of Ym1 in bone marrow on E18.5. (B) Section stained with preimmune serum was used as negative control. (C–K) Double-immunofluorescence analysis of cells expressing Ym1 in the bone marrow on E18.5. Sections of bone marrow were stained with antibody against Ym1 and ER-MP54. Ym1+ cells were labeled by FITC (C, F); ER-MP54+ cells were labeled with rhodamine (D, G). Merged images (E, H) indicate that Ym1 and ER-MP54 are coexpressed in bone marrow cells on E18.5. In contrast, marrow cells coexpressing Ym1 (I) and mature macrophage marker, F4/80 (J), were scant (K; merge). (A–E and I–K) Original bars, 50 µm; (F–H) original bars, 10 µm.
|
 View larger version (56K): [in a new window] |
Figure 6. Expression of Ym1 in lung during development. (A) Lung mRNA was prepared at selected stages of development and examined for Ym1 expression in Northern blot using Ym1 cDNA (P1–P8, 294 bp) as probe. (B) Protein extracts of lung at selected stages of development were analyzed by Western blot using Ym1 antiserum. The same membrane was probed with -tubulin as loading controls. A MW marker of 47.5 kDa is marked. (C, D) Immunocytochemical detection of Ym1 in adult lung. Ym1 immunoreactivity is present in cells and ECM of adult lung (C). Section stained with preimmune serum (D) was used as negative control. Original bars, 50 µm. (E–J) Double immunofluorescence analysis of cells expressing Ym1 in adult lung. Ym1+ cells were labeled with FITC (E, H), and coexpression of Ym1 and F4/80 (F) or SR (I) was evidenced by the merged images (G, J), respectively. Original bars, 10 µm.
|
|
View this table: [in a new window] |
Table 1. Ym1 Expression in Murine Tissues During Developmenta
|
Induced expression of Ym1 in activated macrophages of peritoneal cavity
PEC were elicited by i.p. injection of different agents, i.e., thioglycollate, Sephadex G-50, and C. parvum. The total cell number in the lavage fluid obtained from each mouse was, on the average, as follows: unstimulated control, 8 x 106, thioglycollate-elicited: 1.5 x 107, Sephadex G-50-elicited: 2.6 x 107, C. parvum-elicited: 2.7 x 107, and T. spiralis infection-elicited: 4.2 x 107 cells. Results obtained from Western blot reveal that Ym1 expression was not detectable in the culture supernatant of PEC obtained from unstimulated mice. However, a different level of Ym1 was detected in the culture supernatant of PEC elicited by thioglycollate, Sephadex G-50, or C. parvum, although none was comparable to the extent elicited by T. spiralis infection (Fig. 7A
). Double immunofluorescence staining further revealed that those resident macrophages marked by F4/80 and harvested from control peritoneal cavity are not expressing Ym1 (Fig. 7B
7C
7D)
, and those harvested from peritoneal cavity in the state of inflammation may be induced to express Ym1 (Fig. 7E
7F
7G)
.
 View larger version (35K): [in a new window] |
Figure 7. Induction of Ym1 expression in peritoneal exudate macrophages. (A) Western blot analysis of Ym1 protein in culture supernatants of PEC elicited by different agents. MW markers are as marked. (B–G) Double-immunofluorescence analysis of cell expressing Ym1 in PEC. PEC were retrieved from control mice (B–D) or mice infected with T. spiralis for 15 days (E–G). Ym1+ cells were labeled with FITC (B, E), and cells expressing F4/80 were labeled with rhodamine (C, F). The merged images reveal that Ym1 was undetectable in the peritoneal macrophages of normal mice (D). However, significant expression of Ym1 was evident in PEC of infected mice (E). Merged image (G) indicated that Ym1+ cells coexpress F4/80. Original bars, 10 µm.
|
 View larger version (108K): [in a new window] |
Figure 8. Induction of Ym1 expression in brain. (A–D) Ym1 expression in injured brain after a stab-wound. Although none was detected in the contralateral control area (A), significant expression of Ym1 was detected in the lesioned area in the right parieto-temporal cortex 3 days after the stab-wound (B). At higher magnification, the infiltrated Ym1+ cells were found along the needle tract at the site of lesion (C). GFAP+ astrocytes were also noted in the lesioned site, however in an area surrounding the infiltrating Ym1+ cells (D). (E–H) Induction of Ym1 expression in activated microglia elicited by drug-induced seizure. (E) Ym1 was not detected in the brain of mice receiving normal saline. However, significant expression was detected in microglia in mice treated with PTZ or KA. Ym1-expressing, activated parenchymal microglia (F and G) and perivascular microglia are shown (H). (A–D) Original bars, 100 µm; (E and F) original bars, 50 µm; (G and H) original bars, 25 µm.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Hematopoiesis is initiated within the extra-embryonic yolk sac and/or the intraembryonic aorta-gonad-mesonephros region at around E8.5. Subsequently, the hematopoietic centers will move to fetal liver and spleen along with the migration of stem cells and will finally colonize in the bone marrow throughout adulthood [22 , 23 ]. It was rather striking to find that Ym1 expression correlates well with the ontogeny of hematopoiesis. Significant expression of Ym1 was noted first in the extra-embryonic yolk sac on E10.5 (Fig. 2) . Concomitant expression of Ym1 in myeloid precursors was found in the yolk sac, which is at its peak activity of myelopoiesis [24 , 25 ]. As fetal liver starts to become the major hematopoietic organ for erythropoiesis from E12.5 on, the yolk sac loses its hematopoietic capacity with time [23 , 26 ]. The level of Ym1 transcript and protein increased in the fetal liver from E16.5 to P0.5, a stage with most active hepatic myelopoiesis (Fig. 3A and 3B) [24 ]. The Ym1+ cells were characterized later as myeloid precursors by cell marker studies (Fig. 3E 3F 3G 3H 3I 3J) . By P7.5 and on, the level of Ym1 declined gradually as the hematopoietic stem cells moved away from liver. Therefore, the time window of Ym1 expression in myeloid precursors of fetal liver parallels closely to that of the perinatal myelopoiesis.
In embryos of late gestation, spleen and bone marrow assume the role as sites for hematopoiesis from E16.5 and on [22 , 23 ]. In our study, the expression of Ym1 protein in spleen peaked during E18.5–P0.5 and subsided by P7.5 (Fig. 4B) . The time course of Ym1 expression again correlates with the hematopoietic activity of spleen, which exhibits predominant myelopoiesis from E16.5 to P4.5 and maintains a basal level from then on [22 , 23 ]. Moreover, as shown in Figure 4E 4F 4G , Ym1 was consistently found in subpopulations of myeloid precursors in the red pulps, the hematopoietic compartments for myelopoiesis [27 , 28 ]. As hematopoiesis commences at E16.5 in the fetal bone marrow, cells with persistent Ym1 expression were detected through adulthood. The presence of Ym1-expressing myeloid precursor cells as revealed by immunostaining is in accordance with the detection of Ym1 protein and transcript in the adult marrow (Figs. 1 and 5) . Together, these findings strongly suggest that the temporal order of Ym1 expression in myeloid-lineage precursors closely correlates to the sequential switch of myelopoiesis in hematopoietic sites during murine development.
During hematopoiesis, it is known that common myeloid progenitors could give rise to cells of all myeloid lineages [29 30 31 ]. In our study, Ym1 expression in myeloid precursors is consistent with a previous study demonstrating that Ym1 was detected in a committed myeloid progenitor cell line (EPRO), but not in a stem cell line (EML) [32 ]. To investigate the roles of Ym1 in hematopoiesis, the effects of exogenous addition of purified Ym1 into cultures of bone marrow cells were examined. Preliminary results indicated that Ym1 has no ability to stimulate stem cell colony formation in the absence or in the presence of conditioned medium of L929-containing potent, colony-stimulating activity (data not shown). However, structurally specific heparan sulfate, namely O-sulfated-HS glycosaminoglycans, together with cytokine (e.g., interleukin-3) and matrix protein (e.g., thrombospondin), have been shown to regulate the growth and differentiation of human hematopoietic progenitor cells in stem cell "niches" in the marrow microenvironment [33 34 35 36 ]. Considering that Ym1 is a secreted lectin capable of binding HS, HS is abundant in the marrow microenvironment and has been implicated as an important supportive component for hematopoiesis, and significant expression of Ym1 was found in all sites of hematopoiesis during ontogeny and in adult bone marrow (Figs. 2 3 4 5) , it is tempting to hypothesize that Ym1 may be a novel HS-binding protein, secreted at the "niches" in the marrow. By binding to HS rich in GlcN, myeloid precursor cells may be anchored for subsequent proliferation and/or differentiation during hematopoiesis. Or else Ym1 may also represent an activity that negates the excessive proliferation of progenitor cells via similar adhesive interactions [36 ]. A definitive role of Ym1 during hematopoiesis warrants further studies.
In the nonhematopoietic system, Ym1 was undetectable in all tissues examined except in lung. Ym1 expression in lung as assessed at the transcriptional and the translational levels was increased after birth and persisted through adulthood (Fig. 6A and 6B) . Subcellular localization of Ym1 in lung of normal mice indicated its presence in putative macrophages and at ECM of respiratory epithelium, despite the absence of inflammation. Data obtained from confocal microscopy indicate the coexpression of F4/80 or SR with Ym1, thereby confirmimg the cell type expressing Ym1 as pulmonary resident macrophages (Fig. 6C 6D 6E 6F 6G 6H 6I 6J) .
The temporal increase of Ym1 in pulmonary tissue during development may be related to the function of resident alveolar macrophages in response to gas exchange—the primary function of lung. As respiratory epithelium is exposed repeatedly to inhaled stimuli, mediators of innate immunity have been identified in clearing foreign materials and protecting host from infection [37 , 38 ]. Among these mediators, nitric oxide (NO), produced by NO synthase (NOS) and its degradation product HNO2, is not only produced constantly by pulmonary epithelial cells and macrophages, but also found to cleave HS at the N-unsubstituted GlcN residues; the same moiety of HS, Ym1 exhibits specific binding affinity toward [8 , 39 , 40 ]. HS of lung contains a relatively high level (approximately 5%) of N-unsubstituted GlcN [41 ]. The constitutive expression and secretion of Ym1 into the ECM of lung may represent a protective mechanism shielding the pulmonary tissues from excessive HS degradation by NO. Ym1 thereby may be crucial in maintaining the integrity of HS and/or homeostasis of lung.
In genetic mutant mice; i.e., motheaten (mev/mev)-deficient in SHP-1 (protein tyrosine phosphatase) or CD-40-L-deficient mice known to have dysregulated hematopoiesis and immune response or inherent immunodeficiency, respectively, hyperactive macrophages over-expressing Ym1 and resulting in eosinophilic crystal formations in the lung have been shown [42 ]. It is plausible that the endogenous turnover mechanism for Ym1 may not be sufficient to handle its overproduction. The intrinsic tendency of Ym1 to form crystals [9 ] may have facilitated the progressive and fatal lung injuries observed [42 ]. High incidences of similar Ym1-hyalinosis in respiratory epithelium and Ym1 crystals in alveolar macrophage pneumonia, a major cause of death in aging 129S4/SvJ mice, have also been shown [43 ]. Albeit it is difficult at the present stage to ascribe a particular mechanism underlying the hyperactivation of alveolar macrophages in these mutants or aging mice, the importance of Ym1 in maintaining the homeostasis of lung is further implicated.
With the exception of lung, Ym1 was not detected in F4/80-marked resident macrophages of most tissues at a noninflammatory state. In Figure 7B 7C 7D , the F4/80+-PEC obtained from normal control mice are Ym1-negative. However, F4/80+-PEC obtained from mice subjected to T. spiralis infections revealed that all have been induced to express Ym1 (Fig. 7E 7F 7G) . Because of the systemic migration of infective larvae in the host, T. spiralis infection represents an ideal in vivo model system in which the extent of inflammation and tissue damage was sufficient to reveal the production of Ym1 in the peritoneal cavity. The same response can be elicited by another nematode, Ascaris suum, in which systemic migration of larvae is also required for the maturation processes of the adult worms. The time course of peritoneal accumulation of exudate cells and the extent of Ym1 production are comparable to that of the T. spiralis model [8 ]. In the present study, we have provided further evidence, as shown in Figure 7A , that direct injection of agents such as thioglycollate, Sephadex G-50, and C. parvum into the peritoneal cavity of mice can also elicit Ym1 expression, yet to a lesser extent as compared with that by T. spiralis infection. The data suggest that the induction of Ym1 expression in macrophages is not restricted to inflammation caused by parasitic infections; therefore, Ym1 may assume a broader role in cellular immunity. The differential levels of expression may reflect the varied extents of inflammation caused by distinct types of assault to the immune system.
In addition to the peritoneal cavity, we have explored the ability of other tissue macrophages to express Ym1 during inflammation. Brain is unique, for it is immune-privileged, and two types of glia, astrocyte and microglia, are principal immune cells within. Microglia constituting as many as 12% of the cells in the central nervous system (CNS) with characteristic morphology are considered to have major roles in host defense against invading microbes and neoplastic cells, and thereby is counterpart to peripheral macrophages [13 ]. Two paradigms were used in an attempt to activate microglia such that Ym1 expression may be evaluated. In the stab wound model, as shown in Figure 8B and 8C) , cellular infiltration is evident, and essentially all cells adjacent to the wound are Ym1-positive by day 3. The Ym1-expressing cells assume the typical morphology of phagocytic microglia. These Ym1+ cells are very likely those recruited monocytes from peripheral, which accumulated and differentiated into microglia after entering the parenchyma at the lesioned parietal cortex [44 ]. However, the present study was neither designed for nor able to provide definitive evidence to conclude the cellular origin of these Ym1+ microglia. In addition, scar-forming GFAP+ astrocytes were found at the periphery of lesions occupied predominantly by microglia expressing Ym1 (Fig. 8C and 8D) . It was intriguing to find that as Ym1 may be induced transiently to express in activated, peritoneal macrophages in a 3-day interval, i.e., 15–17 days post-infection during the inflammatory course of parasitic infection [8 ], a similar pattern of expression was observed in the brain microglia elicited by the wound, because by day 6, Ym1 expression subsided in the microglia. In contrast, the astrocytes surrounding the lesion remained GFAP+ (data not shown). The simultaneous presence, yet complementary locations, of these two types of glial cells at the inflammatory locus further support the notion that each are endowed with distinct functions, subsequent wound healing, and tissue remodeling, but may require the interactions of both cell types [45 , 46 ]. In the drug-induced seizure models, irrespective of the drug used, during a course of 1 week after behavioral seizure was observed, no cellular infiltration was found at hippocampus, the primary site known to be affected by KA and PTZ. At day 3, however, Ym1 expression in the parenchymal and perivascular microglia was evident (Fig. 8E 8F 8G 8H) . Those solitary Ym1+ parenchymal cells with ramified morphology may represent microglia derived directly from the resident population [47 ], and the Ym1+ perivascular microglia without ramified process may represent the recruitment of macrophages derived from circulating monocytes [48 , 49 ]. Ym1-expressing microglia were rarely seen by day 6 in the seizured brain.
Our data obtained from the "stab wound" and seizure induction models conclude that assaults in the CNS by physical or chemical means were sufficient to activate microglia, parenchymal, and perivascular alike in order to express Ym1 transiently, which suggests the mode of expression is tightly regulated. As mentioned, we are unable to verify whether Ym1-expressing cells were from resident parenchymal microglia or recruited peripheral macrophages. However, transplantation of bone marrow cells marked by gene transfer [50 ] provides a feasible approach to delineate the origin of Ym1+ microglia in these CNS models of inflammation. Adding Ym1 to the list of the secretory products of microglia may seem justified at this time; however, to ascribe Ym1’s function as beneficial or harmful to the brain requires further study. Its transient nature of expression locally and its binding to N-unsubstituted GlcN would imply that Ym1 may be there to antagonize the toxic effect of NO produced at the site of brain injury [13 ].
The models of inflammation in central and peripheral mentioned above demonstrated that remarkable Ym1 production by activated microglia and macrophages is not restricted to parasitic infection; its level of expression is dependent on the eliciting agents and the extent of inflammation. Judging from the structural features and biochemical characteristics of Ym1, the putative functions of Ym1 during inflammation may be associated with its heparin/HS binding ability at acidic pH [8 ]. Heparin sulfate of ECM modulates the local immune response to regain homeostasis by serving as an available reservoir of cytokines and chemokines, which control the proliferation, activation, and migration of leukocytes [51 52 53 ]. In addition, its structural polymorphism also provides the specificity for cell adhesion; e.g., L-selectin expressed in leukocytes binds selectively to N-unsubstituted GlcN residues of HS [54 , 55 ]. However, degradation of HS moieties by heparanase and NO has been implicated in the inflamed milieu. In particular, NO synthesized by inducible NOS in activated macrophages mediates the acidic pH-dependent HS cleavage [40 , 56 ]. We propose that Ym1 synthesized and released by activated macrophages may modulate the function of HS in the inflamed loci by protecting some moieties of HS from degradation, thereby maintaining the scaffold of ECM. In addition, upon binding to N-unsubstituted GlcN residues of HS, Ym1 may also compete for the adhesive sites of L-selectin such that adhesion and migration of leukocytes on the HS-encoded "path" may be regulated. The action mechanism of Ym1 in inflammation and tissue remodeling will be pursued.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received December 19, 2001; revised December 19, 2001; accepted January 31, 2002.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  Q. Liu, L. I. Cheng, L. Yi, N. Zhu, A. Wood, C. M. Changpriroa, J. M. Ward, and S. H. Jackson p47phox Deficiency Induces Macrophage Dysfunction Resulting in Progressive Crystalline Macrophage Pneumonia Am. J. Pathol., January 1, 2009; 174(1): 153 - 163. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Donnelly, C. M. Stack, S. M. O'Neill, A. A. Sayed, D. L. Williams, and J. P. Dalton Helminth 2-Cys peroxiredoxin drives Th2 responses through a mechanism involving alternatively activated macrophages FASEB J, November 1, 2008; 22(11): 4022 - 4032. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. D. Bird, K. H. Tan, P. F. Olsson, M. Zieba, S. J. Flecknoe, D. R. Liddicoat, R. Mollard, S. B. Hooper, and T. J. Cole Identification of glucocorticoid-regulated genes that control cell proliferation during murine respiratory development J. Physiol., November 15, 2007; 585(1): 187 - 201. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Greenlee, D. B. Corry, D. A. Engler, R. K. Matsunami, P. Tessier, R. G. Cook, Z. Werb, and F. Kheradmand Proteomic Identification of In Vivo Substrates for Matrix Metalloproteinases 2 and 9 Reveals a Mechanism for Resolution of Inflammation J. Immunol., November 15, 2006; 177(10): 7312 - 7321. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. Reece, M. C. Siracusa, and A. L. Scott Innate Immune Responses to Lung-Stage Helminth Infection Induce Alternatively Activated Alveolar Macrophages Infect. Immun., September 1, 2006; 74(9): 4970 - 4981. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Hoenerhoff, M. F. Starost, and J. M. Ward Eosinophilic Crystalline Pneumonia as a Major Cause of Death in 129S4/SvJae Mice. Vet. Pathol., September 1, 2006; 43(5): 682 - 688. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. G. Nair, I. J. Gallagher, M. D. Taylor, P. Loke, P. S. Coulson, R. A. Wilson, R. M. Maizels, and J. E. Allen Chitinase and Fizz Family Members Are a Generalized Feature of Nematode Infection with Selective Upregulation of Ym1 and Fizz1 by Antigen-Presenting Cells Infect. Immun., January 1, 2005; 73(1): 385 - 394. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
