Target ablation-induced regulation of macrophage recruitment into the olfactory epithelium of Mip-1{alpha}–/– mice and restoration of function by exogenous MIP-1{alpha}

Kevin Kwong1, Radhika A. Vaishnav2, Yushu Liu3, Nishikant Subhedar4, Arnold J. Stromberg3, Marilyn L. Getchell2,4 and Thomas V. Getchell1,4

1 Department of Physiology, University of Kentucky, Lexington, Kentucky
2 Department of Anatomy and Neurobiology, University of Kentucky, Lexington, Kentucky
3 Department of Statistics, University of Kentucky, Lexington, Kentucky
4 Sanders-Brown Center on Aging, University of Kentucky, Lexington, Kentucky


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
The chemokine macrophage inflammatory protein (MIP)-1{alpha} recruits macrophages to sites of epithelial remodeling. We showed previously that mRNA and protein levels of MIP-1{alpha} in the olfactory epithelium (OE) increased significantly at 3 days after bilateral olfactory bulbectomy (OBX). The first aim of this study was to investigate the effect of the absence of MIP-1{alpha} on macrophage recruitment to the OE 3 days after OBX in Mip-1{alpha}–/– mice compared with C57BL/6 mice and to test whether chemokine function could be restored by MIP-1{alpha} protein injection into Mip-1{alpha}–/– mice. OBX was performed on C57BL/6 and Mip-1{alpha}–/– mice. The mice received six subcutaneous injections at 12-h intervals of either 10 µg/ml MIP-1{alpha} protein in carrier or carrier only. Macrophage recruitment was evaluated with antibodies to CD68 for all macrophages and F4/80 for activated macrophages. Compared with C57BL/6 mice, at 3 days post-OBX the numbers of CD68+ and F4/80+ macrophages were significantly lower in carrier-injected Mip-1{alpha}–/– mice and were comparable in MIP-1{alpha} protein-injected Mip-1{alpha}–/– mice. The second aim was to determine the identity of genes regulated at 3 days post-OBX in the OE of carrier-injected Mip-1{alpha}–/– mice compared with carrier-injected C57BL/6 mice. Total RNA from the OE was hybridized to Affymetrix microarrays. A number of chemokine-, cytokine-, and growth factor-related genes were significantly regulated in the Mip-1{alpha}–/– mice and were restored in MIP-1{alpha} protein-injected Mip-1{alpha}–/– mice. The results illustrated that MIP-1{alpha} played a key role in recruitment of macrophages to the OE and provided insight into the genomic regulation involved in OE remodeling.

macrophage activation; olfactory bulbectomy; globose basal cell; proliferation; microarray


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
THROUGHOUT ADULT LIFE, the olfactory epithelium (OE) is a site of neurogenesis (26). In every section of OE, one finds olfactory sensory neurons (OSNs) in various states of development, ranging from nascent immature OSNs that have yet to make synaptic contact with their olfactory bulb (OB) targets to mature OSNs to OSNs undergoing apoptosis. Surgical ablation of the OB axotomizes the OSNs, whose somata reside in the OE, and removes their synaptic targets and trophic support. Bilateral olfactory bulbectomy (OBX) synchronizes the myriad events in the life cycle of the OSNs so that analysis of each stage of the degeneration-regeneration cycle can be more precisely resolved. Within hours following OBX, apoptosis of mature OSNs begins to occur (6, 34, 37) and continues for 3–5 days (37, 41). Within this time frame, resident macrophages secrete chemokines that recruit additional macrophages to the OE (37) to phagocytose apoptotic cellular debris (43). In addition, the recruited activated macrophages secrete a variety of bioactive molecules including chemokines such as macrophage inflammatory protein-1{alpha} (MIP-1{alpha}) and monocyte chemoattractant protein-1 (MCP-1), cytokines such as leukemia inhibitory factor (LIF), and growth factors (21, 22, 37). One target of LIF is the globose basal cell (GBC; 2, 37), which is an OSN progenitor cell that resides in the lower third of the OE (26, 41). LIF induces proliferation of GBCs, the progeny of which mature into OSNs that repopulate the OE.

Previously, we provided evidence implicating the ß-chemokine MIP-1{alpha} in the recruitment of macrophages to the OE and macrophage-derived factors in olfactory neurogenesis (22). We demonstrated that mRNA and protein levels of MIP-1{alpha} in the OE were maximally increased at 3 days after OBX. This increase in MIP-1{alpha} levels was coincident with the rise in the number of macrophages and activated macrophages in the OE and with increased numbers of proliferative GBCs. Although the evidence strongly suggested that the recruited macrophages play a key role in OE progenitor cell proliferation, the definitive reciprocal experiment showing the impact of a lack of recruited and activated macrophages on neurogenesis was not performed.

Therefore, we designed experiments, all performed at 3 days after OBX when MIP-1{alpha} mRNA and protein levels peaked in C57BL/6 mice, to test the hypothesis that the lack of MIP-1{alpha} would reduce both the number of macrophages recruited to the OE and, as a result, the number of proliferative GBCs. We used a transgenic mouse model in which the Mip-1{alpha} gene was disrupted (Mip-1{alpha}–/–). To confirm the involvement of MIP-1{alpha}, we injected MIP-1{alpha} protein into the Mip-1{alpha}–/– mice to determine whether macrophage recruitment and GBC proliferation in the OE were restored after OBX. Finally, we asked whether other chemokines, cytokines, or growth factors might play a role in compensating for the lack of MIP-1{alpha} in the OE. We used high-throughput microarray technology to determine OBX-induced differences in gene expression between Mip-1{alpha}–/– and C57BL/6 mice to gain insight into the variety of processes related to cytokine/growth factor expression, macrophage infiltration, and proliferation of progenitor cells in the OE at 3 days following OBX. We demonstrate that the lack of MIP-1{alpha} significantly reduced the recruitment of macrophages and the number of activated macrophages in the OE and significantly reduced the number of proliferative cells in the OE following OBX. These functions were restored in Mip-1{alpha}–/– mice that were injected with exogenous MIP-1{alpha}. We also report preliminary evidence for the regulation of a number of chemokines, cytokines, and growth factor genes that participate in OE remodeling at 3 days following OBX.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Animal and tissue preparation.
Forty-four 6-wk-old age-matched male mice were obtained from the Jackson Laboratory (Bar Harbor, ME); half of the mice were of the C57BL/6-Scya3 (Mip-1{alpha}–/–) strain, and half were of the wild-type C57BL/6 strain. All mice were maintained in the animal facility at the Department of Laboratory Animal Research on a 12:12-h light/dark cycle and were given food and water ad libitum. All protocols were implemented in accordance with the NIH guidelines and approved by the University of Kentucky Institutional Animal Care and Use Committee. To characterize the olfactory phenotype of the Mip-1{alpha}–/– mice and to investigate their response to bilateral OBX, 24 mice, including 12 Mip-1{alpha}–/– mice and 12 C57BL/6 mice, were utilized. Six of the C57BL/6 and six of the Mip-1{alpha}–/– mice underwent bilateral OBX; the remaining six mice of each strain underwent sham bulbectomies. Both procedures were performed according to this laboratory’s protocol as previously described (22, 37). These mice were used for histological and immunohistochemical studies. To determine the effect of exogenous MIP-1{alpha} on the response to OBX in both C57BL/6 and Mip-1{alpha}–/– mice, 20 mice, including 10 Mip-1{alpha}–/– mice and 10 C57BL/6 mice, underwent bilateral OBX. Five of the C57BL/6 mice were injected with 10 µg/ml MIP-1{alpha} protein (R&D Systems, Minneapolis, MN) in a carrier solution of 0.5% BSA in sterile saline; the other five were injected with carrier alone. Similarly, half of the Mip-1{alpha}–/– mice were injected with 10 µg/ml MIP-1{alpha} protein in the carrier solution, and the other half were injected with carrier alone. MIP-1{alpha} protein or carrier alone was administered in six subcutaneous injections of 0.5 ml/injection at 12-h intervals over the 3 days following OBX. Four Mip-1{alpha}–/– mice, half injected with protein and half with carrier alone, were used for histological and immunohistochemical studies; and six mice, half injected with protein and half with carrier alone, were used for microarray and real-time RT-PCR analyses. All mice whose tissues were used for histological and immunohistochemical studies were injected with 5-bromo-2'-deoxyuridine (BrdU, 100 mg/kg ip; Sigma Chemical, St. Louis, MO) in sterile saline 2 h prior to perfusion. On day 3 post-OBX, all mice were euthanized by CO2 asphyxiation. Mice whose tissues were used for histological and immunohistochemical studies were perfused transcardially with 0.1 M phosphate-buffered saline (PBS, pH 7.4) followed by 3% paraformaldehyde. Tissues were cryoprotected, embedded, frozen, and sectioned as previously described (22, 37). From the mice whose tissues were used for microarray and real-time RT-PCR studies, olfactory-nasal mucosa and liver were rapidly removed, weighed, frozen in liquid N2, and stored at –80°C.

RNA isolation.
Total RNA was extracted from each mouse olfactory mucosa under RNase-free conditions. Frozen tissue was pulverized in TRI Reagent (Sigma-Aldrich, St. Louis, MO) and processed through a QIAshredder column (Qiagen, Valencia, CA). The total RNA was further purified using the Qiagen RNeasy Mini-Kit according to the manufacturer’s protocol. Total RNA yield and purity was assessed with a spectrophotometer and with the model 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA); all samples had A260/A280 ratios of 1.9–2.1 and showed two sharp peaks corresponding to 18S and 28S RNA on the Bioanalyzer electropherograms.

High-density oligonucleotide arrays.
Affymetrix Murine Genome (MG) U74Av2 microarrays (Affymetrix, Santa Clara, CA) were processed at the University of Kentucky Microarray Core Facility. The microarray data set was deposited into the Gene Expression Omnibus (GEO) database; the series accession number is GSE1680. Eight GeneChips were used: two each for carrier-injected C57BL/6, carrier-injected Mip-1{alpha}–/–, MIP-1{alpha} protein-injected C57BL/6, and MIP-1{alpha} protein-injected Mip-1{alpha}–/–. For each condition, one of the GeneChips contained total RNA pooled from two mice; the other GeneChip contained total RNA from a single mouse. Each GeneChip contained 12,488 probe sets that included known genes, expressed sequence tags (ESTs), and internal housekeeping and control genes; each probe set consisted of 16 pairs of 25-bp oligonucleotides. Following target preparation, hybridization, and signal acquisition according to standard Affymetrix protocols, the normalized hybridization signals were analyzed with the Affymetrix Microarray Suite version 5; a threshold value of 1,500 was used to normalize the data across all the microarrays. The hybridization signal for each probe set was assigned an "absolute call" by Affymetrix Microarray Suite version 5 and designated as "present," "marginal," or "absent." To be included in the subsequent statistical analyses, a probe set was required to be identified as present on at least one of the microarrays. The mean hybridization signals were analyzed using a 2 x 2 ANOVA (overall P < 0.05), with strain (C57BL/6, Mip-1{alpha}–/–) as one factor and treatment (carrier injection, MIP-1{alpha} injection) as the other factor. Probe sets that had a significant interaction effect (P < 0.05) or had significant effects in both strain and treatment were considered for further analysis. The following comparisons were made using two sample t-tests: 1) carrier-injected C57BL/6 vs. carrier-injected Mip-1{alpha}–/–, 2) carrier-injected Mip-1{alpha}–/– vs. MIP-1{alpha} protein-injected Mip-1{alpha}–/–, and 3) carrier-injected C57BL/6 vs. MIP-1{alpha} protein-injected Mip-1{alpha}–/–. Data were analyzed with SAS (SAS Institute, Cary, NC), Excel (Microsoft, Redmond, WA), and SigmaPlot (SPSS, Chicago, IL). Functional categorization of each gene was derived from the Gene Ontology (GO) database (http://www.geneontology.org).

Real-time RT-PCR.
Real-time PCR was performed on aliquots of cDNA reverse-transcribed from the same pooled total RNA that was used for microarray hybridization. The reaction was performed using cDNA equivalent to 10 ng of RNA, 300 nM each of forward and reverse primers (Table 1), and QuantiTect SYBR Green PCR Master Mix (Qiagen) according to the manufacturer’s protocol in the ABI Prism 7000 Sequence Detection System (Applied Biosystems). Primers for all genes, except that encoding olfactory marker protein (Omp), which is an intronless gene, spanned at least one intron. Reaction parameters were as follows: 2 min at 50°C; 15 min at 95°C; 40 cycles of 15 s each at 94°C, 30 s at the appropriate annealing temperature (Table 1), and 1 min at 72°C. Control and experimental cDNA, cDNA for the derivation of standard curves, and no-template controls were run in triplicate. A dissociation curve was obtained at the end of each reaction to verify the presence of a single product with the appropriate melting point for each product. The presence of a single product from each reaction and its size were also confirmed on a 2% agarose gel stained with ethidium bromide. The comparative threshold cycle (CT) method was used to analyze relative changes in gene expression. The average CT values for each triplicate from each of the four control or experimental conditions were evaluated with an unpaired t-test.


View this table:
[in this window]
[in a new window]
 
Table 1. Real-time RT-PCR primers

 
Immunohistochemistry.
A standard indirect immunofluorescence technique was used as previously described (37) to visualize immunoreactivity for the mature OSN marker olfactory marker protein (OMP); CD68, a pan-macrophage marker; and F4/80, a marker for activated macrophages. The goat polyclonal antibody for OMP, a generous gift of Dr. F. L. Margolis, University of Maryland, Baltimore, MD, was used at a dilution of 1:1,000; immunoreactivity was visualized with FITC-conjugated donkey anti-goat IgG (Jackson ImmunoResearch, West Grove, PA) at a dilution of 1:100. Mouse monoclonal antibodies for CD68 (Serotec, Raleigh, NC) and F4/80 (Caltag Laboratories, Burlingame, CA) were used at 1:50 dilution. CD68 and F4/80 immunoreactivity were visualized with Rhodamine red X-conjugated anti-rat IgG (Jackson ImmunoResearch) at a dilution of 1:200. To determine the number of cells that had incorporated BrdU, tissue sections were sequentially pretreated with trypsin for 15 min at 37°C (Trypsin Histokit; Zymed Laboratories, South San Francisco, CA), rinsed in PBS, incubated with 2 N HCl for 45 min at 37°C, and rinsed in 0.1 M borate buffer followed by PBS. Sections were then incubated with a mouse monoclonal antibody to BrdU (1:500 dilution; Sigma-Aldrich) overnight at 4°C. The secondary antibody was FITC-conjugated donkey anti-mouse IgG (Jackson ImmunoResearch) at 1:100 dilution for 1 h. Methods used by our laboratory to determine the numbers of CD68- and F4/80-immunopositive macrophages and of BrdU-positive cells have been described in detail previously (37).

Peroxidase immunohistochemistry for ABL1, ACVR2B, SOCS3, and CCL22 was performed using the Vectastain Elite ABC Kit according to manufacturer’s directions (Vector Laboratories, Burlingame, CA). Briefly, endogenous peroxidase activity was quenched with 0.3% H2O2 in methanol for 30 min. Sections were incubated in diluted normal blocking serum for 20 min, in the primary antibody for 30 min, in the biotinylated secondary antibody for 30 min, in the Vectastain Elite ABC reagent 30 min, and finally in the peroxidase substrate aminoethyl carbazole (AEC; Zymed). Polyclonal antibodies against ABL1 (Cell Signaling Technology, Beverly, MA), and ACVR2B (Santa Cruz Biotechnology, Santa Cruz, CA) were used at 1:100 dilution. Polyclonal antibodies for CCL22 (R&D Systems) and SOCS3 (Santa Cruz Biotechnology) were used at 1:25 dilution.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Cellular phenotype of the OE does not differ in Mip-1{alpha}–/– and C57BL/6 mice.
Visual inspection of Mip-1{alpha}–/– mice revealed no obvious phenotypic differences compared with the C57BL/6 background mice. The Mip-1{alpha}–/– mice appeared healthy, and body weights between them and the C57BL/6 mice were not significantly different (Mip-1{alpha}–/–, 26.1 ± 1.2 g; C57BL/6, 27.8 ± 0.3 g; P > 0.05). The two strains did not differ statistically in the weight of the olfactory mucosa (Mip-1{alpha}–/–, 99.8 ± 23.9 mg; C57BL/6, 172.3 ± 18.6 mg; P > 0.05) or the thickness of the OE (Fig. 1, A and B, top, and E) when visualized with cresyl violet staining. At 3 days after OBX, there was a significant decrease in the thickness of the OE in both C57BL/6 mice (Fig. 1A, bottom, E; P < 0.001) and Mip-1{alpha}–/– mice (Fig. 1B, bottom, E; P < 0.001). The reduction in the thickness of the OE was a direct consequence of the loss of mature OSNs as assessed using immunohistochemistry for OMP (Fig. 1, C and D). These data indicate that the OE of Mip-1{alpha}–/– mice were histologically similar to that of the C57BL/6 mice. These data also confirmed that surgical ablation of the OBs resulted in a similar degree of OBX-induced neuronal degeneration in both strains of mice.



View larger version (46K):
[in this window]
[in a new window]
 
Fig. 1. Morphometric analysis of olfactory epithelial (OE) height in C57BL/6 and Mip-1{alpha}–/– mice 3 days after bilateral olfactory bulbectomy (OBX). A: when visualized using cresyl violet staining, the height of the OE at the level of the septum was reduced at 3 days post-OBX (top) compared with that of 3 days sham-OBX controls (bottom) in C57BL/6 mice. B: a similar reduction in the height of the OE at the same level of the septum was seen in Mip-1{alpha}–/– mice at 3 days post-OBX (top) compared with the sham controls (bottom). Olfactory marker protein (OMP) immunofluorescence revealed that the reduction in OE height at 3 days following OBX in Mip-1{alpha}–/– mice coincides with a decrease in the number of mature olfactory neurons (C) compared with sham-operated Mip-1{alpha}–/– mice (D). E: averaged data quantifying changes in the OE at 3 days post-OBX in C57BL/6 and Mip-1{alpha}–/– mice. Data obtained from 3 C57BL/6 mice (160 microscopic fields) and 3 Mip-1{alpha}–/– mice (96 microscopic fields). Arrowheads indicate the position of the OE basement membrane. Data are means ± SE. ***P < 0.001. Scale bars, 30 µm.

 
Lack of MIP-1{alpha} reduced OBX-induced recruitment and activation of macrophages in the OE.
Following OBX, the number of macrophages immunoreactive for CD68 in the OE of C57BL/6 mice was significantly greater than that in the sham-OBX controls (Fig. 2, A and B). In Mip-1{alpha}–/– mice injected with carrier following OBX, the number of CD68+ macrophages in the OE was significantly less than that in the C57BL/6 mice (Fig. 2A, bottom; B; C, top; and D). In Mip-1{alpha}–/– mice injected with MIP-1{alpha} protein following OBX, the number of CD68+ macrophages in the OE was not significantly different from that in the C57BL/6 mice (Fig. 2, B and D) and was significantly greater than that in the carrier-injected Mip-1{alpha}–/– mice.



View larger version (36K):
[in this window]
[in a new window]
 
Fig. 2. MIP-1{alpha} injection increases the recruitment of macrophages to the OE 3 days following OBX in Mip-1{alpha}–/– mice. Immunoreactivity to the CD68 cell surface antigen was used as a pan-macrophage marker. A and B: in C57BL/6 mice, the number of CD68+ cells increased in the OE at 3 days post-OBX compared with 3 days sham-operated controls. C and D: the number of CD68+ cells at 3 days post-OBX in carrier-injected Mip-1{alpha}–/– mice was decreased relative to that of C57BL/6 mice at 3 days post-OBX. Injection of MIP-1{alpha} protein in Mip-1{alpha}–/– mice resulted in an increase in the number of CD68+ macrophages at 3 days following OBX compared with that of carrier-injected Mip-1{alpha}–/– mice. Data in B were obtained from 3 C57BL/6 mice (21 microscopic fields); data in D were from 2 Mip-1{alpha}–/– mice (40 microscopic fields). Arrow heads indicate the position of the OE basement membrane. Data are means ± SE. ***P < 0.001. Scale bar, 30 µm.

 
Activated F4/80+ macrophages were distributed in a pattern similar to that of CD68+ macrophages in the OE, although there were significantly fewer of them (e.g., Fig. 3B compared with Fig. 2B). Following OBX, the number of activated macrophages in the OE of C57BL/6 mice was significantly greater than that of C57BL/6 mice that underwent sham OBX (Fig. 3, A and B). In the carrier-injected Mip-1{alpha}–/– mice following OBX, the number of activated macrophages in the OE was significantly less than that in the C57BL/6 mice (Fig. 3A, bottom; B; C, top; D). In contrast, in Mip-1{alpha}–/– mice injected with MIP-1{alpha} protein following OBX, the number of activated macrophages in the OE was significantly greater, similar to that in the C57BL/6 mice (Fig. 3, B and D), and was significantly greater than that in the carrier-injected Mip-1{alpha}–/– mice.



View larger version (44K):
[in this window]
[in a new window]
 
Fig. 3. MIP-1{alpha} injection increases the number of activated macrophages in the OE 3 days following OBX in Mip-1{alpha}–/– mice. Immunoreactivity to the F4/80 cell surface marker was used to identify activated macrophages. A and B: in C57BL/6 mice, the number of F4/80+ cells increased in the OE at 3 days post-OBX compared with 3 days sham-operated controls. C and D: the number of F4/80+ cells was increased in the OE of Mip-1{alpha}–/– mice that received MIP-1{alpha} protein injections compared with those that received carrier injections. Data in B were obtained from 3 C57BL/6 mice (36 microscopic fields); data in D were from 2 Mip-1{alpha}–/– mice (62 microscopic fields). Arrow heads indicate the position of the OE basement membrane. Data are means ± SE. ***P < 0.001. Scale bar, 30 µm.

 
Because MIP-1{alpha} was injected systemically, we investigated whether its effect on macrophage recruitment in the Mip-1{alpha}–/– mice following OBX was specific to the OE by comparing it to the number of macrophages in the liver. The numbers of both recruited and activated macrophages in the liver were not significantly different between MIP-1{alpha} protein- and carrier-injected Mip-1{alpha}–/– mice (Fig. 4). The number of macrophages per unit area was about sixfold greater in the liver than in the OE, and the proportion of activated macrophages was also greater in the liver (~54%) than in the OE (~12%; Figs. 2D, 3D, and 4).



View larger version (12K):
[in this window]
[in a new window]
 
Fig. 4. MIP-1{alpha} protein injection following OBX does not increase recruitment and activation of macrophages to the liver. A: MIP-1{alpha} protein injection in Mip-1{alpha}–/– mice did not alter the number of CD68+ macrophages recruited to the liver at 3 days post-OBX compared with Mip-1{alpha}–/– mice injected with carrier. B: the number of F4/80+ cells in animals injected with MIP-1{alpha} protein at 3 days post-OBX was no different than that in animals receiving carrier injections. Data are mean ± SE. ***P < 0.001.

 
These results demonstrated that at 3 days post-OBX, the absence of the Mip-1{alpha} gene resulted in a significant decrease in the number of macrophages that were recruited to the OE as well as a significant decrease in the number of activated macrophages in the OE. The injection of MIP-1{alpha} protein into Mip-1{alpha}–/– mice following OBX restored the numbers of both recruited and activated macrophages in the OE to levels approximating those in C57BL/6 mice.

Lack of MIP-1{alpha} decreased OBX-induced cellular proliferation in the OE.
The number of BrdU+ GBCs, identified on the basis of their characteristic shape and position in the OE, was greater at 3 days after OBX in C57BL/6 mice compared with the sham-OBX controls (Fig. 5, A and B). The number of BrdU+ GBCs in the sham-OBX Mip-1{alpha}–/– mice was higher than that in C57BL/6 controls (Fig. 5D). However, the number of BrdU+ cells in the Mip-1{alpha}–/– mice was not changed significantly following OBX (Fig. 5C, bottom; and D). In contrast, following OBX, Mip-1{alpha}–/– mice injected with MIP-1{alpha} protein displayed a significantly greater number of proliferating GBCs than that did the carrier-injected Mip-1{alpha}–/– mice (Fig. 5E). Injection of exogenous MIP-1{alpha} protein into Mip-1{alpha}–/– mice following OBX restored the number of proliferating GBCs to a level similar to that in the C57BL/6 mice following OBX.



View larger version (28K):
[in this window]
[in a new window]
 
Fig. 5. Enumeration of 5-bromo-2'-deoxyuridine (BrdU)-immunoreactive cells in the OE of Mip-1{alpha}–/– mice and the effect of MIP-1{alpha} protein injection. Immunoreactivity to BrdU was used as a marker for proliferating cells. A and B: In C57BL/6 mice, the number of cells that were BrdU+ was increased at 3 days post-OBX compared with the 3 days sham-operated controls. C and D: in Mip-1{alpha}–/– mice, no difference in the number of BrdU+ cells was detected between the 3 days OBX and 3 days sham OBX mice. E: the number of BrdU+ cells elicited 3 days post-OBX in Mip-1{alpha}–/– mice was increased in MIP-1{alpha} protein-injected mice compared with carrier-injected mice. Data in B were obtained from 3 C57BL/6 mice (266 microscopic fields); data in D and E were from 2 Mip-1{alpha}–/– mice (110 and 178 microscopic fields, respectively). Arrow heads indicate the position of the OE basement membrane. Data are means ± SE. ***P < 0.001. Scale bar, 30 µm.

 
Mip-1{alpha}–/– and C57BL/6 mice exhibit differential gene expression at 3 days after OBX.
The evidence presented thus far indicates that MIP-1{alpha} played a prominent role in the recruitment and subsequent activation of macrophages and neurogenesis of GBCs in the degeneration-regeneration cycle in the OE at 3 days following OBX. To identify genes that were differentially regulated in the OE in the absence of Mip-1{alpha} compared with C57BL/6 mice, as well as genes that were regulated upon injection of MIP-1{alpha} protein, we used the high-density oligonucleotide GeneChips from Affymetrix. Of the 12,488 probe sets contained on the Affymetrix MG U74Av2 GeneChip, several subsets were excluded from analysis: 66 quality control probe sets and 2,784 probe sets corresponding to ESTs and unknown genes (Fig. 6). Among the remaining 9,638 probe sets, 2,641 were identified as "always absent" (27.4%), which were not further analyzed, and 6,997 probe sets were identified as present (72.6%). A 2 x 2 ANOVA evaluation of the 6,997 probe sets revealed that 357 probe sets (5.1%) had a significant overall effect (P < 0.05). Roughly a third of the 357 probe sets (130 probe sets; 36.4%) had significant main effects of treatment or strain and were excluded from further analysis, because they were either unaffected by treatment (main effect of strain) or there were no differences between strains (main effect of treatment). However, probe sets in which there were significant main effects of both strain and treatment (39 probe sets; 10.9%) were further analyzed. Finally, 190 probe sets achieved statistical significance with respect to interaction effects (53.2%) and were also further analyzed.



View larger version (35K):
[in this window]
[in a new window]
 
Fig. 6. Schematic depicting global gene identification strategy. The flow chart illustrates the process by which genes were initially selected for analysis (shaded rectangles) as well as those omitted (open rectangles) from analysis.

 
There was very high correlation in the mean hybridization signals of the "present" probe sets between the two GeneChips that represent each experimental or control condition, indicating that the output signals obtained from the probe sets were highly correlated between the GeneChip pairs. In carrier-injected C57BL/6 mice, the correlation coefficient (r) was 0.92 (Fig. 7A). The correlation coefficient for MIP-1{alpha} protein-injected C57BL/6 mice was 0.97 (Fig. 7B). Correlation coefficients for carrier-injected and MIP-1{alpha} protein-injected Mip-1{alpha}–/– mice were 0.94 and 0.98, respectively (Fig. 7, C and D).



View larger version (36K):
[in this window]
[in a new window]
 
Fig. 7. A high correlation exists between the probe sets of the GeneChip pairs that comprise each experimental condition. The correlation between the mean hybridization signals of each pair of known and present probe sets was well-fitted for each experimental and control condition.

 
Results obtained thus far revealed a characteristic trend in the regulation of cellular processes: there was divergence between the Mip-1{alpha}–/– and the C57BL/6 mice in their response to OBX (i.e., macrophage recruitment, Fig. 2; macrophage activation, Fig. 3; GBC proliferation, Fig. 5). When exogenous MIP-1{alpha} protein was injected into the Mip-1{alpha}–/– mice, the response to OBX changed significantly to approximate that of the C57BL/6 mice expressing endogenous MIP-1{alpha} protein, suggesting that these three functions had been restored. We reasoned that the regulation of genes that were of potential interest would mirror this trend in the regulation of cellular processes. Therefore, we applied the pattern of cellular regulation as a model for the statistical pattern analysis of the 229 probe sets that had significant main effects of both strain and treatment or had an interactive effect. Specifically, we identified genes in mice that had undergone OBX that fulfilled three criteria: 1) a significant difference (P < 0.05) between the carrier-injected C57BL/6 mice and the carrier-injected Mip-1{alpha}–/– mice, 2) a significant difference (P < 0.05) between the carrier-injected Mip-1{alpha}–/– mice and the MIP-1{alpha} protein-injected Mip-1{alpha}–/– mice, and 3) no significant difference (P > 0.05) between the carrier-injected C57BL/6 mice and the MIP-1{alpha} protein-injected Mip-1{alpha}–/– mice. Among the 229 genes, 90 genes showed patterns in which an initial significant difference in levels of expression between carrier-injected C57BL/6 and Mip-1{alpha}–/– mice was restored to equivalent levels of expression in Mip-1{alpha}–/– mice injected with MIP-1{alpha} protein, and we undertook a detailed analysis of their patterns of expression (Tables 2 and 3). Of the 90 genes, 41 (45.6%) showed a significantly higher level of mRNA expression in the carrier-injected Mip-1{alpha}–/– mice compared with the carrier-injected C57BL/6 (e.g., Fig. 8D); differences in expression ranged from 1.1- to 5.1-fold (Table 2). Only a single chemokine-related gene followed this pattern of expression: chemokine (C-C motif) receptor 7 (Ccr7), a chemoattractant receptor for cells of both the myeloid and lymphoid lineages. Cytokine-related genes include syndecan-2 (Sdc2), a heparan sulfate proteoglycan, and interleukin-1 receptor 2 (Il1r2). Further in the progression of events subsequent to macrophage activation is proliferation and axonal growth in the OE. Genes that might play a role in these processes include suppressor of cytokine signaling 3 (Socs3), budding uninhibited by benzimidazoles 1 homolog (Bub1), and arginase 1 (Arg1). Among apoptosis genes, phospholipid scramblase 1 (Plscr1) displayed a pattern of expression identical to other genes in this group. The remaining 49 genes (54.4%) showed a significantly lower level of expression in the carrier-injected Mip-1{alpha}–/– mice compared with the carrier-injected C57BL/6 mice (e.g., Fig. 8, A-C); differences in expression ranged from 1.1- to 3.5-fold (Table 3). One chemokine identified that followed this pattern of expression was chemokine (C-C motif) ligand 22 (Ccl22). Among apoptosis-related genes, two pro-apoptotic genes were identified: death associated kinase 2 (Dapk2) and Tnfrsf12a. Anti-apoptosis genes were represented by Bcl2, bcl2-associated athanogene-2 (Bag2), and phospholipase D2 (Pld2). Genes involved in proliferation, such as activin A receptor 2B (Acvr2b), which is a member of the TGF-ß receptor superfamily, displayed an initial lower expression as did v-abl Abelson murine leukemia oncogene 1 (Abl1).


View this table:
[in this window]
[in a new window]
 
Table 2. Functional categorization of genes in which MIP-1{alpha} injection restored gene regulation in Mip-1{alpha}–/– mice at 3 days postolfactory bulbectomy: initially upregulated genes

 

View this table:
[in this window]
[in a new window]
 
Table 3. Functional categorization of genes in which MIP-1{alpha} injection restored gene regulation in Mip-1{alpha}–/– mice at 3 days postolfactory bulbectomy: initially downregulated genes

 


View larger version (27K):
[in this window]
[in a new window]
 
Fig. 8. Microarray data showing the regulation of 4 selected genes: Abl1 (A), Acvr2b (B), Ccl22 (C), and Socs3 (D). Mean hybridization signals were compared between carrier-injected and MIP-1{alpha} protein-injected C57BL/6 mice (open circles/dotted lines) and carrier-injected and MIP-1{alpha} protein-injected Mip-1{alpha}–/– mice (solid circles/dashed lines). The arrowheads indicate the direction of change in the mean hybridization signals in animals injected with MIP-1{alpha} protein compared with those injected with carrier only.

 
In addition to the 90 genes that displayed complete restoration of expression in the MIP-1{alpha} protein-injected Mip-1{alpha}–/– mice (when compared with the carrier-injected C57BL/6 mice), we identified an additional 17 genes that displayed a partial restoration of gene expression upon MIP-1{alpha}-protein injection (Table 2 and 3, asterisks). These are genes in which MIP-1{alpha} injection into the Mip-1{alpha}–/– mice did not result in a level of gene regulation identical to that of carrier-injected C57BL/6 mice, but regulation was in the direction of restoration. Statistically, these genes passed the first two criteria for pattern analysis and failed the third; however, gene regulation in the MIP-1{alpha}-injected Mip-1{alpha}–/– mice approached that of the carrier-injected C57BL/6 mice. In summary, the overall analysis revealed a number of chemokine, cytokine, and pro-proliferation genes that were regulated at 3 days post-OBX in the absence of endogenous MIP-1{alpha} protein. The patterns of expression of 46.7% (107/229) of these genes suggest that gene regulation was restored to some degree in the Mip-1{alpha}–/– mice injected with exogenous MIP-1{alpha} protein.

We selected four genes that were of particular interest to our laboratory in the analysis of OE remodeling. The level of Abl1 mRNA expression was 1.5-fold lower in the carrier-injected Mip-1{alpha}–/– mice compared with carrier-injected C57BL/6 mice (P < 0.05; Fig. 8A). Abl1 mRNA was significantly upregulated 1.5-fold in MIP-1{alpha} protein-injected Mip-1{alpha}–/– mice, to a level not significantly different from that of carrier-injected C57BL/6 mice (Fig. 8A). The level of Acvr2b mRNA expression was 1.5-fold lower in the carrier-injected Mip-1{alpha}–/– mice compared with the carrier-injected C57BL/6 mice (P < 0.001; Fig. 8B). Acvr2b mRNA was significantly upregulated 1.5-fold in MIP-1{alpha} protein-injected Mip-1{alpha}–/– mice, to a level not significantly different from that of the carrier-injected C57BL/6 mice (P > 0.05). Interestingly, Acvr2b expression was downregulated in MIP-1{alpha} protein-injected C57BL/6 mice compared with the carrier-injected C57BL/6 mice (P < 0.001; Fig. 8B), suggesting that very high levels of MIP-1{alpha} could feed back negatively on Acvr2b regulation in the OE. Ccl22 expression was 1.9-fold lower in the carrier-injected Mip-1{alpha}–/– mice compared with the carrier-injected C57BL/6 mice (P < 0.05; Fig. 8C). Expression of Ccl22 mRNA increased 2.1-fold in MIP-1{alpha} protein-injected Mip-1{alpha}–/– mice compared with the carrier-injected Mip-1{alpha}–/– mice (P < 0.01) to a level similar to that of carrier-injected C57BL/6 mice (P > 0.05; Fig. 8C). The level of Socs3 mRNA expression was, in contrast, 1.5-fold higher in the carrier-injected Mip-1{alpha}–/– mice compared with the carrier-injected C57BL/6 mice (P < 0.05; Fig. 8D). Socs3 mRNA significantly decreased 2.3-fold 3 days following OBX in the MIP-1{alpha} protein-injected Mip-1{alpha}–/– mice compared with the carrier-injected Mip-1{alpha}–/– mice, to a level similar to that of the carrier-injected C57BL/6 mice (P > 0.05; Fig. 8D).

Validation using real-time RT-PCR.
Regulation of the four selected genes in mice that had undergone OBX was validated using real-time RT-PCR (Fig. 9). Overall, the mRNA expression patterns (Fig. 10) correlated well with the microarray expression patterns (Fig. 8). The level of Abl1 mRNA expression was 1.3-fold lower in the carrier-injected Mip-1{alpha}–/– mouse compared with the carrier-injected C57BL/6 mouse (Figs. 9A and 10A). Abl1 mRNA expression increased 1.4-fold in the MIP-1{alpha} protein-injected Mip-1{alpha}–/– mouse compared with the carrier-injected Mip-1{alpha}–/– mouse, to a level similar to that of the carrier-injected C57BL/6 mouse (Fig. 10A), which was highly similar to pattern of expression detected in the microarray experiments (Fig. 8A). There was a 1.6-fold lower expression of the Acvr2b mRNA in the carrier-injected Mip-1{alpha}–/– mouse compared with the carrier-injected C57BL/6 mouse (Figs. 9B and 10B). Acvr2b mRNA expression increased 1.6-fold 3 days post-OBX in the MIP-1{alpha} protein-injected Mip-1{alpha}–/– mouse compared with the carrier-injected Mip-1{alpha}–/– mouse, to a level similar to that of the carrier-injected C57BL/6 mouse (Fig. 10B). Expression levels of Ccl22 mRNA were 2.1-fold lower in the carrier-injected Mip-1{alpha}–/– mouse compared with the carrier-injected C57BL/6 mouse (Figs. 9C and 10C). Ccl22 mRNA expression increased 1.7-fold in the MIP-1{alpha} protein-injected Mip-1{alpha}–/– mouse compared with the carrier-injected Mip-1{alpha}–/– mouse, to a level similar to that of the carrier-injected C57BL/6 mouse (Fig. 10C). However, real-time RT-PCR analysis also showed a 2.1-fold increase in Ccl22 mRNA expression in the MIP-1{alpha} protein-injected C57BL/6 mouse compared with the carrier-injected C57BL/6 mouse, which was less similar to the results obtained from the microarray analysis (Fig. 8C). The mRNA expression of Socs3 was similar for both the carrier-injected C57BL/6 and the Mip-1{alpha}–/– mice (Figs. 9D and 10D). Socs3 mRNA expression was 1.2-fold lower in the MIP-1{alpha} protein-injected Mip-1{alpha}–/– mouse compared with the carrier-injected Mip-1{alpha}–/– mouse; whereas Socs3 expression decreased 1.6-fold in the MIP-1{alpha} protein-injected C57BL/6 mouse compared with the carrier-injected C57BL/6 mouse (Fig. 8D).



View larger version (26K):
[in this window]
[in a new window]
 
Fig. 9. Raw data from real-time RT-PCR analysis of the 4 selected genes: Abl1 (A), Acvr2b (B), Ccl22 (C), and Socs3 (D). The average of the fluorescence values of 3 replicates minus background ({Delta}Rn) as a function of the cycle number. Insets: dissociation curves illustrating specificity of the primer pair selected for each gene. Bottom inset shows that the real-time RT-PCR amplicons are of the correct size. Solid symbols, MIP-1-{alpha} protein-injected mice; open symbols, carrier-injected mice; triangles, C57BL/6 mice; circles, Mip-1{alpha}–/– mice. RFU, relative fluorescence units; T, temperature. Horizontal lines indicate the threshold at which CT values were determined.

 


View larger version (19K):
[in this window]
[in a new window]
 
Fig. 10. Real-time RT-PCR validation of the 4 selected genes in the microarray analysis. Real-time RT-PCR analysis of the 4 selected genes, Abl1 (A), Acvr2b (B), Ccl22 (C), and Socs3 (D), show positive validation of microarray results. Open circles/dotted lines, C57BL/6 mice; solid circles/dashed lines, Mip-1{alpha}–/– mice. The arrowheads indicate the direction of change in the relative quantities of mRNA in animals injected with MIP-1{alpha} protein compared with those injected with carrier only.

 
At the end of each real-time RT-PCR run, a dissociation protocol was performed to establish that a unique RT-PCR product of the correct melting temperature was obtained. A single peak corresponding to an amplicon of the appropriate melting temperature was obtained for each of the four selected genes (Fig. 9, A-C, insets; 9D, left inset). The real-time RT-PCR products were of the expected sizes (Fig. 9D, right inset). The relative abundance of the four selected mRNAs was estimated from the relative CT values generated from a standard amount of RT-PCR product (Fig. 11). The least abundant message was Ccl22, followed by systematically increasing relative amounts of Socs3, Acvr2b, and Abl1. As a reference for internal calibration, the relative abundance of Omp was much greater than the other mRNAs (Fig. 11). Real-time RT-PCR techniques provided positive validation of the pattern of mRNA expression detected in the microarray analysis.



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 11. Relative abundance of selected interactive genes. The differences in CT values in the standard curves are an indication of the relative abundance among the selected genes. Each of the standard curves was well-fitted (R2 > 0.85). Omp ({triangledown}) was included as a reference gene. CT, threshold cycle; •, Abl1; {circ}, Acvr2b; {blacktriangledown}, Ccl22; and {blacksquare}, Socs3.

 
Immunohistochemical validation of the regulation of gene products.
We performed an immunohistochemical analysis of the OE from C57BL/6 and Mip-1{alpha}–/– mice that had undergone OBX followed by injection of either MIP-1{alpha} protein or carrier to determine whether the regulated genes identified in the microarray study resulted in a similar regulation of protein expression, and to localize the gene products. ABL1+ immunoreactivity was detected in structures corresponding to well-defined myelinated nerve bundles in the lamina propria (Fig. 12, A and B) that were not immunoreactive to OMP (not shown). No apparent differences in the levels of protein expression were evident between the carrier-injected and MIP-1{alpha} protein-injected Mip-1{alpha}–/– mice. ACVR2B protein was expressed in cells of the myeloid tissue. In carrier-injected Mip-1{alpha}–/– mouse, ACVR2B+ myeloid cells were confined within the myeloid tissue; no immunoreactivity was detected in the OE (Fig. 12C). In the MIP-1{alpha} protein-injected Mip-1{alpha}–/– mouse, ACVR2B+ cells occurred in the lamina propria and OE (Fig. 12D). Immunoreactivity to CCL22 was not detected in the olfactory mucosa. SOCS3 immunoreactivity occurred in small round cells that were distributed primarily in the in the lamina propria, but also in the OE to a lesser extent in the carrier-injected Mip-1{alpha}–/– mouse (Fig. 12E). In the MIP-1{alpha} protein-injected Mip-1{alpha}–/– mouse, SOCS3 immunoreactivity was not detected (Fig. 12F). The level of staining for ACVR2B and SOCS3 in the MIP-1{alpha} protein-injected Mip-1{alpha}–/– mouse compared with the carrier-injected Mip-1{alpha}–/– mouse mirrored the changes observed at the mRNA level in the microarray and real-time RT-PCR experiments.



View larger version (139K):
[in this window]
[in a new window]
 
Fig. 12. Immunohistochemical localization of selected genes. Immunohistochemistry was used to localize 3 of the selected genes in MIP-1{alpha} protein- or carrier-injected Mip-1{alpha}–/– mice at 3 days post-OBX. ABL1 was localized in the myelinated nerve bundles in the lamina propria in the region dorsal to the dorsal recess. No difference was detected between carrier- (A, arrow) and MIP-1{alpha} protein-injected (B, arrows) mice. ACVR2B was localized in myeloid cells that infiltrated into lamina propria in MIP-1{alpha} protein-injected mice (D, arrows), but not in carrier-injected mice (C). SOCS3 was localized in the OE and lamina propria in carrier-injected mice (E, arrows), but not in MIP-1{alpha} protein-injected mice (F). Scale bar, 30 µm.

 

    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
This study presented evidence illustrating the key role that MIP-1{alpha} plays in the response of the OE to OBX. The disruption of the Mip-1{alpha} gene resulted in a significant reduction in the recruitment of CD68+ and activated F4/80+ macrophages in the OE at 3 days following OBX. Injection of MIP-1{alpha} protein into the Mip-1{alpha}–/– mice facilitated the restoration in the number of CD68+ and F4/80+ macrophages in the OE. Disruption of the Mip-1{alpha} gene also abrogated the increase in the number of BrdU+ cells in the OE, and injection of MIP-1{alpha} protein facilitated the restoration of the number of proliferative cells elicited by OBX. Finally, this study used microarray technology to identify a number of genes that were differentially regulated at 3 days following OBX in the OE of Mip-1{alpha}–/– mice compared with C57BL/6 mice, as well as those in which MIP-1{alpha} protein had been administered systemically. The present work supports and extends to the molecular level prior research implicating the importance of MIP-1{alpha} in the recruitment of macrophages to the OE as a key component in OE remodeling (22, 37) and the participation of macrophage-derived pro-proliferative factors in OE neurogenesis and further defines its role in these processes.

Is MIP-1{alpha} required to recruit macrophages to the OE in response to OBX?
A significant reduction in the number of CD68+ macrophages (~50%) was detected in the OE in Mip-1{alpha}–/– mice. Furthermore, injection of MIP-1{alpha} protein into Mip-1{alpha}–/– mice fully restored the number of CD68+ macrophages in the OE in response to OBX. Together, these experiments convincingly demonstrate that MIP-1{alpha} plays a crucial role in the recruitment of macrophages to the OE following target ablation. These findings advance the hypothesis that after the removal of OSN targets, resident macrophages secrete chemokines that attract circulating monocytes into the OE from blood vessels in the lamina propria, leading to maximal macrophage recruitment in the OE by 3 days post-OBX. The reduction in macrophage recruitment in Mip-1{alpha}–/– mice was not due to a diminution in the total number of macrophages in the transgenic animals because Mip-1{alpha}–/– mice do not display abnormalities in peripheral blood hematopoietic parameters, including hematocrits, and total leukocyte and differential cell counts (11). Although this study did not assay the presence of CCR1, a cognate receptor for MIP-1{alpha}, we know from our previous experiments that the increase in CCR1 expression coincided with the increase in CD68+ macrophages, suggesting that MIP-1{alpha} acted in an autocrine fashion to increase the number of macrophages recruited to the OE after OBX (22).

Although the number of recruited macrophages was drastically reduced in Mip-1{alpha}–/– mice, it should be noted that disruption of the Mip-1{alpha} gene did not totally prevent recruitment of macrophages to the OE, which suggests that macrophage recruitment to the OE was not exclusively dependent on MIP-1{alpha}. This finding is also consistent with our previous study, in which we demonstrated that mRNA and protein expression of macrophage chemoattractant protein-1 (MCP-1) began to increase at 2 h post-OBX and peaked at 16 h (22). It has been shown that MCP-1 and MIP-1{alpha} are both highly efficient in monocyte chemotaxis (45). It is therefore likely that MCP-1 is active in recruiting macrophages to the OE in response to OBX, but at a time point earlier than the action of MIP-1{alpha}. Additionally, our study does not rule out the possibility of other chemotactic molecules functioning synergistically with MIP-1{alpha} or MCP-1 or independently at other times after OBX.

The systemic administration of MIP-1{alpha} protein resulted in increased recruitment of macrophages to the OE but not to the liver. The recruitment of leukocytes to peripheral tissues is mediated by ligands/receptors expressed by both blood-borne leukocytes and activated endothelial cells that affect the movement of leukocytes through blood vessels (31). Although MIP-1{alpha} itself can slow the passage of leukocytes and induce their "rolling" in retinal blood vessels (12) and may contribute to the reduction in the number of macrophages recruited to the target-ablated OE in the Mip-1{alpha}–/– mice, the recruitment of macrophages to the OE but not the liver following systemic administration of MIP-1{alpha} suggests that OBX regulates the expression of one or more factors in the OE in addition to MIP-1{alpha} that participate in macrophage adhesion to the vascular endothelium.

Is MIP-1{alpha} required to activate recruited macrophages in the OE in response to OBX?
Three days following target ablation, the number of F4/80+-activated macrophages was significantly reduced in the Mip-1{alpha}–/– mouse compared with the C57BL/6 mouse, but when MIP-1{alpha} protein was injected into the Mip-1{alpha}–/– mouse, the number of recruited and activated macrophages was restored to levels similar to that of the C57BL/6 mouse. These findings provide solid evidence for the pivotal role that MIP-1{alpha} plays in the recruitment of macrophages to the OE following target ablation. However, we investigated only one post-OBX time point in this study and were unable to determine the precise sequence of events leading to macrophage activation. There are at least three possibilities: 1) unactivated macrophages were first recruited to the OE by MIP-1{alpha} and then activated by MIP-1{alpha}, which has been shown to be able to directly activate macrophages (19); 2) unactivated macrophages were first recruited to the OE by MIP-1{alpha} and then activated by other local OBX-induced factors with or without the participation of MIP-1{alpha}; and 3) activated macrophages were recruited directly to the OE by MIP-1{alpha}. Further experiments are needed to elucidate the sequence of events.

Macrophage activation can occur through at least two pathways. Classic activation occurs through the combination of exposure to interferon-{gamma} (IFN-{gamma}) and tumor necrosis factor-{alpha} (TNF-{alpha}) or agents that elicit TNF-{alpha} (14, 20, 40). The alternative pathway of macrophage activation occurs through exposure to interleukin-4 (IL-4) or glucocorticoids. Whereas classically activated macrophages promote inflammation and are highly cytotoxic, alternatively activated macrophages are not; one of the hallmarks that differentiate alternatively activated macrophages from classically activated macrophages is their lack of production of the cytotoxic radicals NO or O2. Alternatively activated macrophages have been implicated in wound healing and cell proliferation (24, 35). Another hallmark of alternatively activated macrophages is the expression of Arg1, which was upregulated in the Mip-1{alpha}–/– mice. Arg1, which converts L-arginine to L-ornithine, results in the accumulation of polyamines (36); polyamines promote proliferation and wound repair (27). While the lack of indicators of an inflammatory response and cytotoxicity in the OE after OBX (37) suggests the predominance of the alternative pathway, our data present unequivocal evidence that MIP-1{alpha} played a key role in recruiting macrophages that subsequently exhibit activation.

What is the role of MIP-1{alpha} in the proliferation of OE cells?
Disruption of the Mip-1{alpha} gene reduced the number of proliferative cells following OBX. The dysfunction was abrogated following MIP-1{alpha} protein injection into the Mip-1{alpha}–/– mice. These data are consistent with the hypothesis that an increase in the secretion of cytokines and growth factors by activated macrophages following OBX resulted in the increased proliferation of GBCs, leading to neurogenesis. MIP-1{alpha}-induced proliferation likely depends on the recruitment of activated macrophages in the OE, because we did not detect CCR1 expression by GBCs (22). The activated macrophages were situated in the lower third of the OE, where secreted cytokines and growth factors could influence proliferation of GBCs in a paracrine fashion. For example, the macrophage-derived cytokine LIF increased proliferation of progenitor cells in vivo (2, 22, 37). Furthermore, we showed that the LIF receptor was transiently expressed on GBCs and that gp130, a LIF receptor transmembrane signaling partner, was upregulated coincident with the LIF receptor. The fact that both the number of activated macrophages and the number of proliferative GBCs in the OE after OBX was significantly reduced in Mip-1{alpha}–/– mice and that both were restored to near-normal levels by the injection of MIP-1{alpha} protein supports the key role of macrophage-derived mitogens in OE neurogenesis.

Does the lack of MIP-1{alpha} elicit regulation of other chemokines/cytokines/growth factors as a compensatory mechanism?
MIP-1{alpha} clearly played a major role in macrophage recruitment and activation as well as GBC proliferation in the OE following OBX. Had another chemokine fully compensated for the lack of MIP-1{alpha}, these parameters would have been much more similar between the two strains of mice. However, macrophage recruitment in the knockout mice was not totally absent, suggesting that other chemokines are also involved. One likely contributor was MCP-1, in part because its expression precedes that of MIP-1{alpha} (22). The only other chemokine that was regulated significantly in our microarray analysis was Ccl22. However, the observation that Ccl22 expression was actually lower in the Mip-1{alpha}–/– mice suggests that it did not function to compensate for the lack of MIP-1{alpha}. We did not detect a significant regulation in the expression of Ccr1, a gene encoding a cognate receptor for MIP-1{alpha} (4) that is expressed by macrophages in the OE following OBX (22). The only chemokine receptor that was found to be regulated was Ccr7, which was expressed at higher levels in the Mip-1{alpha}–/– mice. An increased expression of Ccr7 is typically identified with mature monocyte and lymphocyte derivatives, which migrate toward lymphoid tissue in response to CCL19 or CCL21 gradients (7). Further experiments are required to elucidate the role of these chemokines/receptors in the response to OBX.

What were other factors besides the chemokines themselves might have contributed to alterations in OE function in the knockout mouse? In addition to its cognate receptors, chemokines bind to low-affinity heparan sulfate proteoglycan molecules, which are transmembrane molecules (e.g., syndecans), or are linked to the plasma membrane via glycosyl phosphatidylinositol (GPI; e.g., glypicans; 3, 16). Heparan sulfate proteoglycans are thought to facilitate the action of chemokines in part by increasing the local concentration of chemokines, promoting oligomerization of the chemokine before its binding to the cognate receptor, and immobilizing chemokines on membrane surfaces to maintain a chemotactic gradient (1, 3, 29, 39). MIP-1{alpha} binds to two sulfate-rich S-domains on heparan sulfate (42). Our microarray data showed that Sdc2 was expressed at a higher level in the absence of MIP-1{alpha}. SDC2 has been shown to be highly expressed on endothelial cells and monocyte-derived macrophages, where it modulated chemokine activity and growth factor signaling (8, 25). It is likely that Sdc2 expression was increased in a negative feedback compensatory mechanism to boost the effective concentration or signaling of MIP-1{alpha}.

It is well documented that OBX elicits apoptosis of mature OSNs in the OE (6, 34), which peaks at 2 days post-OBX when measured by TUNEL (28). One of the functions of macrophages recruited to the OE by OBX is phagocytosis of dead OSNs (43). The upregulation of Plscr1 in Mip-1{alpha}–/– mice is particularly interesting because of its potential role in transforming an apoptotic cell for phagocytosis. PLSCR1 is a plasma membrane enzyme that actively shuttles phospholipids, phosphatidylserine in particular, from the inner leaflet of the plasma membrane to the outer leaflet, which is normally devoid of phosphatidylserine (46). The newly exposed phosphatidylserine marks the cell as apoptotic and signals for engulfment by macrophages (17). Other genes of particular interest included Bub1, whose increased expression is consistent with its role as a checkpoint gene that prevents cell entry into anaphase of the cell cycle (18). It is possible that this gene plays a role in the reduced proliferative response of GBCs to OBX in Mip-1{alpha}–/– mice. The expression of IL-1R2, which is a decoy receptor for IL-1 (9, 32), is increased by IL-4. IL-4 induces alternative activation of macrophages, which is consistent with the role of the induction of this pathway by OBX.

We validated and localized the expression of the proteins encoded by Ccl22, Abl1, Acvr2b, and Socs3. Since ABL1 is known to play a role in promoting progenitor cell activity, we expected to localize it in basal cells. We found immunoreactivity only in myelinated nerve bundles running through the olfactory mucosa. ACVR2B and SOCS3 protein expression validated the differential gene regulation indicated by the mRNA analysis. ACVR2B appeared to be localized in macrophages. Activins are growth and differentiation factors belonging to the TGF-ß family, which have been shown to regulate proliferation in the OE (38). The receptor encoded by this gene has been associated with oogonial proliferation in the ovary (33). The SOCS3 protein also appeared to be localized in macrophages. SOCS3 is a negative regulator of proliferation and blocks signaling of IL-6 (13) and LIF (44). We were unable to localize CCL22. Two factors may have contributed: 1) it is a soluble protein that may have been washed away during buffer washes, and 2) its low concentration in the extracellular space could be below the level of detection using peroxidase immunohistochemistry. The latter possibility is in agreement with finding that Ccl22 mRNA was the least abundantly expressed of the genes according to the low mean hybridization signal in the microarray analysis as well as a high CT value in the real-time RT-PCR. Godiska et al. (23), using Northern blot analysis, showed that cultured monocytes do not express detectable levels of CCL22 before 6 days postdifferentiation into macrophages. It is possible that the lack of CCL22 protein expression at 3 days after OBX might be due to the relative immaturity of recruited monocyte/macrophages.

Together, these findings indicate that MIP-1{alpha} played a major role in the recruitment and activation of macrophages to the OE 3 days following OBX. MIP-1{alpha}-recruited and -activated macrophages were key in promoting proliferation of GBCs in the OE, leading to neurogenesis and remodeling following OBX. Results from our microarray experiments provided initial strong evidence for the role of other factors, including cell growth, differentiation, cell cycle regulation, and cytoskeletal organization genes that might play a role in the response of the OE to OBX.


    GRANTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
This study was supported by National Institutes of Health Grants AG-16824 (to T. V. Getchell), T32-DC-00065 (to K. Kwong), and 1P20-RR-16481-01 (to A. J. Stromberg).


    ACKNOWLEDGMENTS
 
We thank Dr. Dharmen Shah and Mr. James Partin for technical expertise. We also thank Dr. Frank L. Margolis for the generous gift of the OMP antibody and Dr. Hal E. Broxmeyer for consultation on MIP-1{alpha} protein injection paradigm. Finally, we thank Dr. Subbarao Bondada and Dr. Lakshman Chelvarajan for guidance with the real-time RT-PCR experiments.


    FOOTNOTES
 
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).

Address for reprint requests and other correspondence: T. V. Getchell, 309 Sanders-Brown Center on Aging, Univ. of Kentucky, 800 S Limestone St., Lexington, KY 40536-0230 (E-mail: tgetche{at}uky.edu).

doi:10.1152/physiolgenomics.00187.2004.


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Ali S, Palmer AC, Banerjee B, Fritchley SJ, and Kirby JA. Examination of the function of RANTES, MIP-1{alpha}, and MIP-1ß following interaction with heparin-like glycosaminoglycans. J Biol Chem 275: 11721–11727, 2000.[Abstract/Free Full Text]
  2. Bauer S, Rasika S, Han J, Mauduit C, Raccurt M, Morel G, Jourdan F, Benahmed M, Moyse E, and Patterson PH. Leukemia inhibitory factor is a key signal for injury-induced neurogenesis in the adult mouse olfactory epithelium. J Neurosci 23: 1792–1803, 2003.[Abstract/Free Full Text]
  3. Bernfield M, Götte M, Park PW, Reizes O, Fitzgerald ML, Lincecum J, and Zako M. Functions of cell surface heparan sulfate proteoglycans. Annu Rev Biochem 68: 729–777, 1999.[CrossRef][ISI][Medline]
  4. Broxmeyer HE, Cooper S, Hangoc G, Gao JL, and Murphy PM. Dominant myelopoietic effector functions mediated by chemokine receptor CCR1. J Exp Med 189: 1987–1992, 1999.[Abstract/Free Full Text]
  5. Buiakova OI, Krishna NS, Getchell TV, and Margolis FL. Human and rodent OMP genes: conservation of structural and regulatory motifs and cellular localization. Genomics 20: 452–462, 1994.[CrossRef][ISI][Medline]
  6. Calof AL, Hagiwara N, Holcomb JD, Mumm JS, and Shou J. Neurogenesis and cell death in olfactory epithelium. J Neurobiol 30: 67–81, 1996.[CrossRef][ISI][Medline]
  7. Campbell DJ, Kim CH, Butcher EC. Chemokines in the systemic organization of immunity. Immunol Rev 195: 58–71, 2003.[CrossRef][ISI][Medline]
  8. Clasper S, Vekemans S, Fiore M, Plebanski M, Wordsworth P, David G, and Jackson DG. Inducible expression of the cell surface heparan sulfate proteoglycan syndecan-2 (fibroglycan) on human activated macrophages can regulate fibroblast growth factor action. J Biol Chem 274: 24113–24123, 1999.[Abstract/Free Full Text]
  9. Colotta F, Re F, Muzio M, Bertini R, Polentarutti N, Sironi M, Giri JG, Dower SK, Sims JE, and Mantovani A. Interleukin-1 type II receptor: a decoy target for IL-1 that is regulated by IL-4. Science 261: 472–475, 1993.[ISI][Medline]
  10. Columba-Cabezas S, Serafini B, Ambrosini E, Sanchez M, Penna G, Adorini L, and Aloisi F. Induction of macrophage-derived chemokine/CCL22 expression in experimental autoimmune encephalomyelitis and cultured microglia: implications for disease regulation. J Neuroimmunol 130: 10–21, 2002.[CrossRef][ISI][Medline]
  11. Cook DN, Beck MA, Coffman TM, Kirby SL, Sheridan JF, Pragnell IB, and Smithies O. Requirement of MIP-1{alpha} for an inflammatory response to viral infection. Science 269: 1583–1585, 1995.[ISI][Medline]
  12. Crane IJ, Xu H, Manivannan A, McKillop-Smith S, Lamont G, Wallace C, Liversidge J, Sharp PF, and Forrester JV. Effect of anti-macrophage inflammatory protein-1a on leukocyte trafficking and disease progression in experimental autoimmune uveoretinitis. Eur J Immunol 33: 402–410, 2003.[CrossRef][ISI][Medline]
  13. Croker BA, Krebs DL, Zhang JG, Wormald S, Willson TA, Stanley EG, Robb L, Greenhalgh CJ, Forster I, Clausen BE, Nicola NA, Metcalf D, Hilton DJ, Roberts AW, and Alexander WS. SOCS3 negatively regulates IL-6 signaling in vivo. Nat Immunol 4: 540–545, 2003.[CrossRef][ISI][Medline]
  14. Dalton DK, Pitts-Meek S, Keshav S, Figari IS, Bradley A, and Stewart TA. Multiple defects of immune cell function in mice with disrupted interferon-{gamma} genes. Science 259: 1739–1742, 1993.[ISI][Medline]
  15. Di Simone N, Hall HA, Welt C, and Schneyer AL. Activin regulates ßA-subunit and activin receptor messenger ribonucleic acid and cellular proliferation in activin-responsive testicular tumor cells. Endocrinology 139: 1147–1155, 1998.[Abstract/Free Full Text]
  16. David G, Lories V, Decock B, Marynen P, Cassiman JJ, and Van den Berghe H. Molecular cloning of a phosphatidylinositol-anchored membrane heparan sulfate proteoglycan from human lung fibroblasts. J Cell Biol 111: 3165–3176, 1990.[Abstract]
  17. Fadok VA, Voelker DR, Campbell PA, Cohen JJ, Bratton DL, and Henson PM. Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J Immunol 148: 2207–2216, 1992.[Abstract/Free Full Text]
  18. Fagin JA. Minireview: branded from the start-distinct oncogenic initiating events may determine tumor fate in the thyroid. Mol Endocrinol 16: 903–911, 2002.[Abstract/Free Full Text]
  19. Fahey TJ III, Tracey KJ, Tekamp-Olson P, Cousens LS, Jones WG, Shires GT, Cerami A, and Sherry B. Macrophage inflammatory protein 1 modulates macrophage function. J Immunol 148: 2764–2769, 1992.[Abstract/Free Full Text]
  20. Fujihara M, Muroi M, Tanamoto K, Suzuki T, Azuma H, and Ikeda H. Molecular mechanisms of macrophage activation and deactivation by lipopolysaccharide: roles of the receptor complex. Pharmacol Ther 100: 171–194, 2003.[CrossRef][ISI][Medline]
  21. Getchell TV, Subhedar NK, Shah DS, Hackley G, Partin JV, Sen G, and Getchell ML. Chemokine regulation of macrophage recruitment into the olfactory epithelium following target ablation: involvement of macrophage inflammatory protein-1{alpha} and monocyte chemoattractant protein-1. J Neurosci Res 70: 784–793, 2002.[CrossRef][ISI][Medline]
  22. Getchell TV, Shah DS, Partin JV, Subhedar NK, and Getchell ML. Leukemia inhibitory factor mRNA expression is upregulated in macrophages and olfactory receptor neurons after target ablation. J Neurosci Res 67: 246–254, 2002.[CrossRef][ISI][Medline]
  23. Godiska R, Chantry D, Raport CJ, Sozzani S, Allavena P, Leviten D, Mantovani A, and Gray PW. Human macrophage-derived chemokine (MDC), a novel chemoattractant for monocytes, monocyte-derived dendritic cells, and natural killer cells. J Exp Med 185: 1595–1604, 1997.[Abstract/Free Full Text]
  24. Gordon S. Alternative activation of macrophages. Nat Rev Immunol 3: 23–35, 2003.[CrossRef][ISI][Medline]
  25. Götte M. Syndecans in inflammation. FASEB J 17: 575–591, 2003.[Abstract/Free Full Text]
  26. Graziadei GA and Graziadei PP. Neurogenesis and neuron regeneration in the olfactory system of mammals. II. Degeneration and reconstitution of the olfactory sensory neurons after axotomy. J Neurocytol 8: 197–213, 1979.[ISI][Medline]
  27. Hesse M, Modolell M, La Flamme AC, Schito M, Fuentes JM, Cheever AW, Pearce EJ, and Wynn TA. Differential regulation of nitric oxide synthase-2 and arginase-1 by type 1/type 2 cytokines in vivo: granulomatous pathology is shaped by the pattern of L-arginine metabolism. J Immunol 167: 6533–6544, 2001.[Abstract/Free Full Text]
  28. Holcomb JD, Mumm JS, and Calof AL. Apoptosis in the neuronal lineage of the mouse olfactory epithelium: regulation in vivo and in vitro. Dev Biol 172: 307–323, 1995.[CrossRef][ISI][Medline]
  29. Hoogewerf AJ, Kuschert GS, Proudfoot AE, Borlat F, Clark-Lewis I, Power CA, and Wells TN. Glycosaminoglycans mediate cell surface oligomerization of chemokines. Biochemistry 36: 13570–13578, 1997.[CrossRef][ISI][Medline]
  30. Hosoi T, Okuma Y, and Nomura Y. Leptin regulates interleukin-1ß expression in the brain via the STAT3-independent mechanisms. Brain Res 949: 139–146, 2002.[CrossRef][ISI][Medline]
  31. Imhof BA and Aurrand-Lions M. Adhesion mechanisms regulating the migration of monocytes. Nat Rev Immunol 4: 432–444, 2004.[CrossRef][ISI][Medline]
  32. Mantovani A, Locati M, Vecchi A, Sozzani S, and Allavena P. Decoy receptors: a strategy to regulate inflammatory cytokines and chemokines. Trends Immunol 22: 328–336, 2001.[CrossRef][ISI][Medline]
  33. Martins da Silva SJ, Bayne RA, Cambray N, Hartley PS, McNeilly AS, and Anderson RA. Expression of activin subunits and receptors in the developing human ovary: activin A promotes germ cell survival and proliferation before primordial follicle formation. Dev Biol 266: 334–345, 2004.[CrossRef][ISI][Medline]
  34. Michel D, Moyse E, Brun G, and Jourdan F. Induction of apoptosis in mouse olfactory neuroepithelium by synaptic target ablation. Neuroreport 5: 1329–1332, 1994.[ISI][Medline]
  35. Mosser DM. The many faces of macrophage activation. J Leukoc Biol 73: 209–212, 2003.[Free Full Text]
  36. Munder M, Eichmann K, Morán JM, Centeno F, Soler G, and Modolell M. Th1/Th2-regulated expression of arginase isoforms in murine macrophages and dendritic cells. J Immunol 163: 3771–3777, 1999.[Abstract/Free Full Text]
  37. Nan B, Getchell ML, Partin JV, and Getchell TV. Leukemia inhibitory factor, interleukin-6, and their receptors are expressed transiently in the olfactory mucosa after target ablation. J Comp Neurol 435: 60–77, 2001.[CrossRef][ISI][Medline]
  38. Newman MP, Féron F, and Mackay-Sim A. Growth factor regulation of neurogenesis in adult olfactory epithelium. Neuroscience 99: 343–350, 2000.[CrossRef][ISI][Medline]
  39. Patel DD, Koopmann W, Imai T, Whichard LP, Yoshie O, and Krangel MS. Chemokines have diverse abilities to form solid phase gradients. Clin Immunol 99: 43–52, 2001.[CrossRef][ISI][Medline]
  40. Schroder K, Hertzog PJ, Ravasi T, and Hume DA. Interferon-{gamma}: an overview of signals, mechanisms and functions. J Leukoc Biol 75: 163–189, 2004.[Abstract/Free Full Text]
  41. Schwartz Levey M, Chikaraishi DM, and Kauer JS. Characterization of potential precursor populations in the mouse olfactory epithelium using immunocytochemistry and autoradiography. J Neurosci 11: 3556–3564, 1991.[Abstract]
  42. Stringer SE, Forster MJ, Mulloy B, Bishop CR, Graham GJ, and Gallagher JT. Characterization of the binding site on heparan sulfate for macrophage inflammatory protein 1{alpha}. Blood 100: 1543–1550, 2002.[Abstract/Free Full Text]
  43. Suzuki Y, Schafer J, and Farbman AI. Phagocytic cells in the rat olfactory epithelium after bulbectomy. Exp Neurol 136: 225–233, 1995.[CrossRef][ISI][Medline]
  44. Takahashi Y, Carpino N, Cross JC, Torres M, Parganas E, and Ihle JN. SOCS3: an essential regulator of LIF receptor signaling in trophoblast giant cell differentiation. EMBO J 22: 372–384, 2003.[Abstract/Free Full Text]
  45. Uguccioni M, D’Apuzzo M, Loetscher M, Dewald B, and Baggiolini M. Actions of the chemotactic cytokines MCP-1, MCP-2, MCP-3, RANTES, MIP-1{alpha} and MIP-1ß on human monocytes. Eur J Immunol 25: 64–68, 1995.[ISI][Medline]
  46. Zhao J, Zhou Q, Wiedmer T, and Sims PJ. Level of expression of phospholipid scramblase regulates induced movement of phosphatidylserine to the cell surface. J Biol Chem 273: 6603–6606, 1998.[Abstract/Free Full Text]




This Article
Free upon publication
Abstract
Full Text (PDF)
All Versions of this Article:
20/1/73    most recent
00187.2004v1
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Search for citing articles in:
ISI Web of Science (2)
Google Scholar
Articles by Kwong, K.
Articles by Getchell, T. V.
Articles citing this Article
PubMed
PubMed Citation
Articles by Kwong, K.
Articles by Getchell, T. V.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2004 by the American Physiological Society.