Introduction

Ribonucleases are prominent in the biology of all living systems, yet our understanding of the roles played by these proteins in mammalian biology remains incomplete. Of particular interest are the ribonucleases of the RNase A superfamily, the only enzyme family limited to vertebrate species.¹ The RNase superfamily includes multiple lineages of structurally related proteins that have diverged to promote specific ribonucleolytic functions within mammalian cells.^{2, 3, 4}

Much of the work in our laboratory has focused on the eosinophil ribonuclease lineages, which are RNase A ribonucleases expressed prominently (although not uniquely) in eosinophilic leukocytes.⁵ The human eosinophil ribonucleases, eosinophil-derived neurotoxin (EDN, RNase 2) and eosinophil cationic protein (ECP, RNase 3) are paralogous genes that have cytotoxic, antiviral, and ribonuclease activities, and are among the most rapidly diverging functional coding sequences known among primates.^{6, 7} The highly divergent rodent orthologs of EDN and ECP are found in species-limited gene clusters generated via an unusual evolutionary process defined as rapid-birth–death, a feature shared with the MHC, immunoglobulin, and T-cell receptor gene families.⁸ Larson et al⁹ isolated the first two of these 15 ribonucleases from Mus musculus, and coined the term mouse eosinophil-associated ribonucleases, or mEars.^{9, 10, 11, 12, 13} While all 15 are potential ribonucleases (ie, all 15 have appropriate structural and catalytic elements), ribonuclease and antiviral activity have thus far been documented only for mEar 2.^{14, 15} Most recently, Yang et al¹⁶ have shown that mEar 2 is also a chemoattractant for dendritic cells in vivo, a finding that promotes a new understanding of the multiple biologic role(s) of these proteins.

Among the more intriguing questions—despite the existence of 15 unique but related ribonucleases—only transcripts encoding mEars 1 and 2 are detected in significant amounts in peripheral tissues under homeostatic conditions (Table 1). We and others have suggested that the remaining mEars might be expressed in response to pathophysiologic stimuli, a hypothesis borne out by recent results demonstrating mEar 11 expression in response to allergic provocation in the lung.¹¹ In this study, we investigated the dynamics of mEar expression in Th2-dependent, eosinophil-enriched liver tissue characteristic of Schistosoma mansoni infection in mice.¹⁷ Most notable was the elevated level of expression of mEar 6, a relatively divergent member of the Mus musculus eosinophil ribonuclease gene cluster originally cloned by Cormier et al.¹² We have evaluated the evolutionary and enzymatic properties of mEar 6 in light of this novel expression pattern, and consider the significance of the activation-mediated modulation of this specific gene product.

Table 1 - Mouse eosinophil ribonuclease transcripts expressed in organs and tissues at homeostasis.

Full table

Results

Histopathology of S. mansoni infection in mice

Histologic sections of liver from mice infection with S. mansoni are shown in Figure 1a–d. At 8 weeks postexposure to cercariae, infected mice develop granulomata that are dense with inflammatory cells and fibroblasts (Figure 1c and d). Granulomata do not develop in uninfected mice (Figure 1a and b). Granulomata are characteristic of the Th2 phase of this disease, which also includes elevated levels of serum IL-5 and blood and tissue eosinophilia. Spleen tissue from S. mansoni-infected mice (8 weeks postexposure) is shown in Figure 1e and f.

Figure 1.

Histopathology of S. mansoni-infected mice. Liver (a–d) and spleen (e and f) sections were prepared from uninfected mice (a, b and e), mice at 8 weeks postexposure to cercariae of S. mansoni (c, d, and f). The sections were subjected to staining with hematoxylin and eosin. Images are photographed at original magnification of 10 (a and c), 40 (b, e, and f), or 100 (d).

Full figure and legend (165K)

Kinetics of ribonuclease protein expression and enzymatic activity

Although transcripts encoding mEar 2 can be detected in the normal liver (Table 1), no mEar protein was detected by Western blotting of liver homogenates prepared from uninfected mice (Figure 2a). Mouse Ear protein was first detected in homogenates from the infected mice at 6 weeks postexposure, reaching a plateau at 8–16 weeks. RNase activity was first detected over baseline level in infected wild-type mice at 4 weeks postexposure, and reached a plateau at t=8 weeks (Figure 2b). Immunodepletion experiments with the anti-mEars antibody demonstrated that the ribonuclease activity measured in the liver homogenates from infected mice was due to the activity of the mEars (Figure 2c).

Figure 2.

Detection of mouse eosinophil ribonuclease protein and activity in liver homogenates. (a) Total liver protein (50 g) subjected to gel electrophoresis and probed with polyclonal rabbit anti-mEars antibody. The mEars migrate as a single band with apparent molecular mass of 16 kDa under reducing conditions. (b) Ribonuclease activity in liver homogenates (200 g protein per sample) increases during the course of infection in wild-type mice. (c) Ribonuclease activity (40 g protein per sample) in infected wild-type mice liver homogenates is depleted by addition of 5 l of polyclonal rabbit anti-mEars antibody and precipitation with protein G Sepharose, but not by an irrelevant (anti-human major basic protein antibody, hMBP) control.

Full figure and legend (80K)

Expression of specific mEar transcripts in the liver and spleen in response to S. mansoni infection

Northern analysis demonstrated elevated levels of total mEar transcripts in the liver at 8 weeks postexposure to cercariae (Figure 3), which was consistent with levels of immunoreactive protein (Figure 2a) and enzymatic activity (Figure 2b). However, of this total, only transcripts encoding mEar 2 were detected in the livers of uninfected wild-type mice. In contrast, transcripts encoding mEars 1 and 6 in addition to mEar 2 were detected in liver tissue from S. mansoni-infected mice (Table 2). This is the first demonstration of significant mEar 6 expression in peripheral tissue. Interestingly, mEar 6 transcripts were not detected in liver tissue from IL-5-transgenic (IL-5-Tg) mice, despite constitutively elevated levels of serum IL-5 in vivo (Table 2). These results indicate that mEar 6 expression is not directly related to elevated levels of serum IL-5.

Figure 3.

Mouse eosinophil ribonuclease transcripts detected in the liver and spleen of uninfected and S. mansoni-infected wild-type mice. Total RNA (10 g) was probed with cDNA encoding mEar 2 (cross-reacts with all mEars) or -actin. Results shown are values of intensity of mEar 2/intensity of -actin. Values are means.e.m. from three individual mice. ^*P<0.05 vs corresponding uninfected mice.

Full figure and legend (36K)

Table 2 - Distribution of mouse eosinophil ribonuclease cDNAs isolated from liver RNA.

Full table

Despite the increased numbers of eosinophils in the blood, the livers of IL-5-Tg mice are not among the organs that are eosinophil-enriched.¹⁸ To explore the relationship between tissue eosinophilia and mEar 6 expression, spleen tissue, which is eosinophil-enriched in these IL-5-Tg mice, was evaluated for further experiments. Analogous to what was observed in the liver, total mEar mRNA increased in the spleens of wild-type mice in response to S. mansoni infection (Figure 3). Transcripts encoding mEars 1 and 2 were detected in spleen tissue from uninfected wild-type mice, and transcripts encoding mEar 6 were detected only in the spleens of mice infected with S. mansoni. No mEar 6 transcripts were detected in spleen tissue from uninfected, IL-5-Tg mice, despite constitutively elevated serum IL-5 and eosinophil infiltration¹⁸ (Table 3). These results suggest that neither IL-5 nor eosinophil infiltration per se is involved directly in promoting mEar 6 expression. Interestingly, transcripts encoding mEar 7 were detected in the spleens of IL-5-Tg mice (Table 3).

Table 3 - Distribution of mouse eosinophil ribonuclease cDNAs isolated from spleen RNA.

Full table

Characterization of recombinant mEar 6

Expression of mEar 6 in response to S. mansoni infection prompted us to provide a more thorough biochemical characterization of this protein (Figure 4). Mouse Ear 6 is among the most divergent members of the mEar cluster. The coding sequence of mEar 6 (mature protein, without signal peptide) shares 80% identity with the sequences of mEars 1 and 2. Mouse Ear 6 (calculated pI=8.4) is significantly less cationic than mEars 1 and 2 (calculated pIs of 9.2 and 9.7, respectively).

Figure 4.

Sequence divergence and kinetic characterization of mEar 6. (a) Amino-acid sequence comparisons among mEars 1, 2, and 6 (GenBank Accession No. U72032, U72031, AY015175 (AY316150), respectively). Indicated in bold are amino acids that are unique to mEar 6. Double dots above each cysteine residue indicate those that are disulfide bonded in the appropriately folded, secreted form of the protein. The single arrow indicates the position of the unpaired cysteine in the sequence of mEar 6. Calculated isoelectric points are as indicated. (http://www.embl-heidelberg.de/
cgi/pi-wrapper.pl). (b) Immunoblotting with anti-mEars polyclonal antiserum. In all, 50 g protein of liver lysate under reducing or nonreducing conditions was probed with polyclonal rabbit anti-mEars antibody. Bands corresponding to mEars (16 kDa) are indicated. The three bolder arrows above are additional bands detected under nonreducing conditions.

Full figure and legend (91K)

Mouse Ear 6 is the only member of this cluster that contains an unpaired cysteine, at position 113 (arrow in Figure 4a), just proximal to the catalytic histidine-122. In order to determine whether the unpaired cysteine of the mEar 6 protein has a free sulfhydryl group, we compared this protein directly to recombinant mEars 1 and 2, which do not have unpaired cysteines, and performed a quantitative measure of the free sulfhydryls using a Thiol Sulfide Quantitation Kit. The reagent used in this assay is reported to be 10 times more sensitive than the traditional Ellman reagent. Interestingly, all three proteins tested below the limits of detection (less than 0.1 nmol sulfhydryl groups in 0.6 nmol proteins, data not shown), suggesting that the unpaired cysteine in recombinant mEar 6 does not have a free sulfhydryl group. However, when natural mEar protein from liver tissue was evaluated in Western blots under both reducing and nonreducing conditions, additional bands present at 23, 28, and 30 kDa were present under nonreducing conditions only, all of which are of lower molecular weight than a putative mEar homodimer (32 kDa). These results suggest the possibility of heterologous covalent interactions between natural mEar 6 and other low-molecular-weight proteins (Figure 4b).

The catalytic constants defining ribonuclease activity of recombinant mEar 6 were determined. K_m and k_cat were calculated at 0.61 M and 0.53 s^-1, respectively, from a Lineweaver-Burk double-reciprocal plot. Although the activity of recombinant mEar 6 (k_cat/K_m=0.87 10⁶/M/s) is somewhat less than that determined for recombinant mEar 2,¹⁴ they are still within range of one another, and within range of several of the other RNase A family ribonucleases that have been evaluated among mammalian species². The ratio of nonsynonymous to synonymous substitutions (K_a/K_s) calculated for mEar 6 vs mEars 1 and 2 are +1.6 and +1.5, respectively (Table 4), suggesting that divergence is promoted by positive (Darwinian) selection (K_a/K_s>1.0).

Table 4 - Determination of synonymous (K_s) and nonsynonymous (K_a) substitution rates for divergent mEar pairs.

Full table

Discussion

In this study, we have characterized the expression of mEar 6, a divergent member of the eosinophil-associated ribonuclease gene cluster in mice, in response to infection with the helminthic parasite, S. mansoni.

Cormier et al¹¹ have shown that mEar 11, another of the divergent mEars, is induced in response to allergic provocation in mouse lung tissue and in macrophages stimulated by Th2 cytokines. We show here that mEar 6 expression in the liver and spleen is induced in response to another prominent Th2-skewed condition. Mouse Ear 6 was originally isolated from eosinophil granules,¹² and was cloned from mouse bone marrow cDNA.¹² However, our results demonstrate that neither elevated levels of serum IL-5 nor the presence of tissue eosinophilia, such as that observed in IL-5-Tg mice,^{18, 19} explain the magnitude of mEar 6 expression in this setting (Tables 2 and 3). Infection studies with IL-5-Tg mice may ultimately help to identify the factors necessary to elicit mEar 6 expression.

Accumulation of mEars in general, and mEar 6 specifically, may contribute to host defense against S. mansoni infection in several ways. Both mEar 2 and mEar 6 (and likely several, if not all, of the others) are strong ribonucleases, which may enhance the clearance of cellular debris and lead the way to tissue remodeling. Mouse Ear 2 also has chemotactic activity for dendritic cells in vitro and in vivo;¹⁶ mEar 6 may have a similar or related function.

The coding sequence of mEar 6 includes an unusual, unpaired cysteine, yet we failed to detect either dimerization or a free sulfhydryl group in the recombinant form of the protein. We are unable to explain this in a satisfactory manner, although the most likely explanation would be to suspect post-translational substitution unique to this recombinant form.

While there are 15 unique gene sequences in the Mus musculus mEar cluster, the physiologic constraints promoting gene duplication and divergence remain unclear. In earlier work, we proposed that the evolution of the eosinophil ribonucleases was in response to constraints promoting diversification of specific ribonuclease-dependent activities, and, toward this end, we have shown that the primate eosinophil ribonucleases, EDN and ECP, have ribonuclease-dependent activity against specific single-stranded RNA viruses in vitro.²⁰ At the same time, extensive gene duplication permits relaxation of functional constraints. This provides the possibility that one or more of the mEar cluster ribonucleases may be diverging from this central focus in order to promote ribonuclease-independent activities and functions.²¹ Future studies with these mEar cluster ribonucleases will ultimately elucidate a more complex picture of their independent and interdependent roles and relationships.

Materials and methods

Mice

Wild-type C57BL/6 mice were maintained in the NIAID 14BS Animal Facility. All procedures were reviewed and approved under protocols LAD-8E or LPD-16E.

Infection of mice with S. mansoni

Cercariae of a Puerto Rican strain of S. mansoni (NMRI) were obtained from infected Biomphalaria glabrata snails (Biomedical Research Institute, Rockville, MD, USA). C57BL/6 mice were exposed to cercariae of S. mansoni by immersion of their tails in water containing approximately 25–40 cercariae for 40 min, as previously described.²² All procedures described in this paper were reviewed and approved by NIAID ASP LPD-16E.

Preparation of liver homogenates

Liver tissues were collected from mice that had been killed by cervical dislocation at the appropriate time postinfection. In all, 500 mg (wet weight) liver tissue was homogenized with a homogenizer (Tekmar, Cincinnati, OH, USA) after addition of 1 ml PBS. The homogenate was centrifuged at 13 600 g for 20 min at 4°C to clear insoluble debris. The supernatant was stored at –80°C. Total protein was determined by a Bio-Rad Protein assay (Bio-Rad Laboratories, Hercules, CA, USA).

Production of recombinant mEar 6 protein

Recombinant mEar 6 was prepared in Escherichia coli in the pFlag-CTS expression vector (Sigma-Aldrich, St Louis, MO, USA) and the quantity of protein determined as described previously for other RNase A superfamily ribonucleases.²³

Ribonuclease assay

The ribonuclease assay with yeast tRNA substrate has also been described in detail elsewhere.²³ For these experiments, 40 g of yeast tRNA (Sigma Aldrich) was added to an 800 l of PBS, pH 7.4 containing 50 or 200 g protein of liver lysate, or 0.66 pmol of recombinant protein. The reaction was stopped at 5 min by the addition of a cold solution of 20 mM lanthanum nitrate with 3% perchloric acid to precipitate the undigested tRNA, which was removed by centrifugation at 13 600 g for 5 min at room temperature. The amount of solubilized RNA was determined by its absorbance at 260 nm with a t=0 control as a blank. Determination of pmol ribonucleotides from optical density measurements include the following approximations: the average molecular mass of a molecule of tRNA is 28 100 (75–90 ribonucleotides per molecule, MW=341 per ribonucleotide).

Immunoblotting

The liver homogenate (50 g) was mixed with a nonreducing or reducing Tris–glycine 2 SDS sample buffer (Invitrogen, Carlsbad, CA, USA) and subjected to SDS-polyacrylamide gel electrophoresis (14% acrylamide, 1 Tris–glycine system). After proteins in the gel were transferred to a nitrocellulose membrane (Invitrogen), nonspecific binding to the membrane was blocked with 5% nonfat dry milk in Tris-buffered saline (TBS, pH 7.5) prior to probing the membrane with a 1 : 600 dilution of polyclonal anti-mEars antiserum followed by horseradish peroxidase-conjugated goat anti-rabbit IgG (Amersham Biosciences, Piscataway, NJ, USA). The anti-mEars antiserum has been tested in our laboratory and specifically detects recombinant mEars 1, 2, 6, and 11 with equal intensity on Western blots (data not shown). Kodak Biomax film (Rochester, NY, USA) was exposed to the membrane immersed in the solution of ECL plus Western Blotting Detection System (Amersham Biosciences) and then developed.

Immunodepletion

A volume of 5 l each of either rabbit anti-mEars antiserum, rabbit anti-human eosinophil major basic protein antiserum (control), or PBS was added to each 50 g protein of three samples of the liver homogenate as described above. After equilibration at room temperature for 1 h, 5 l of protein G Agarose (Boehringer Manheim, Indianapolis, IN, USA) was added and samples were equilibrated as above for another 1 h. Immunoprecipitated proteins were removed by centrifugation at 13 600 g for 10 min at 4°C. RNase assay was performed on the resultant supernatant as described above.

RNA preparation

Livers from mice were immersed in RNAlater (Ambion, Austin, TX, USA) followed by blade homogenization in 7 ml of RNAbee reagent (Tel-Test, Friendswood, TX, USA). For bone marrow cells, 1 ml RNAzol B was added per 10⁶ cells and mixed. After chloroform was added (1 : 10 vol/vol), the specimen was mixed thoroughly and incubated on ice for 15 min. After a centrifugation at 13 600 g for 20 min at 4°C, the aqueous layer was transferred to fresh tubes. Equal volumes of ice-cold isopropanol were added, and RNA was precipitated at -20°C. Total RNA was pelleted by the centrifugation, washed twice in 80% ethanol, dried, and resuspended in diethyl pyrocarbonate-treated water. RNA was quantitated spectrophotometrically.

Northern blotting

RNA (10 g) was loaded in a 1% agarose gel containing 6.7% formaldehyde and 1 MOPS and electrophoresis was performed under 100 V constant for 1 h at room temperature. After electrophoresis, RNA in the gel was blotted with 20 SSPE. Hybridization with mEar 2 probe was performed as previously described.¹⁴ If brief, the membrane was prehybridized and hybridized in PerfectHyb Plus (Sigma-Aldrich) with 2 10⁶ cpm/ml random-primed radiolabeled probe at 55°C for 12 h. The blot was then washed three times in 0.1 SSPE containing 0.1% SDS for 10 min. After washing, the blot was exposed to Kodak Biomax film at –80°C and developed. After the detection of mEar mRNA, the probe was stripped and rehybridization with an endolabeled--actin probe was performed to check the amounts of RNA loaded. The blots were subject to densitometric analysis using an NIH Image software.

Identification of mEar mRNAs

The identification of mEar mRNA was performed as described elsewhere.^{15, 24} In brief, in order to remove any traces of genomic DNA contamination, 2 g total tissue RNA were treated with DNase I (Invitrogen) in a total volume of 20 l. After 15 min at room temperature, the enzyme was inactivated by the addition of 2 l 25 mM EDTA and by heating to 65°C for 10 min. DNA-free RNA, 0.7 g, was subjected to reverse transcription by the addition of 20 U AMV reverse transcriptase as described in the 1st Strand cDNA Synthesis Kit for RT-PCR (Roche, Indianapolis, IN, USA). PCR was performed using Pfx DNA polymerase (Invitrogen) and an aliquot of cDNA as a template. Primer sequences for PCR are as follows: 5'-ATGGGTCCGAAGCTGCTTGAGTC-3' and 5'-CTAAAATGTCCCATCCAAGTG-AAC-3', and are able to amplify the open reading frame of all known Ear genes. PCR products were gel purified with Geneclean Spin Kit (Qbiogene, Carlsbad, CA, USA) and subcloned into pCR-Blunt vectors (Invitrogen). Mouse Ear genes were identified by differential restriction digestion²⁴ and automated DNA sequencing.

Sequencing

Automated DNA sequencing was performed using an ABI PRISM 377 DNA sequencer (Applied Biosystems, Foster City, CA, USA) and a DYEnamic ET Terminator Cycle Sequencing Kit (Amersham Biosciences). Sequence analyses were performed with the assistance of the Sequencher programs (GeneCodes Corporation, Ann Arbor, MI, USA).

Detection of free sulfhydryl groups

Free sulfhydryl groups in recombinant mEar proteins were quantitated with a colorimetric assay, the Thiol and Sulfide Quantitation Kit (Molecular Probes, Eugene, OR, USA), according to the manufacturer's instructions. Briefly, 0.6 nmol of recombinant mEar proteins was incubated with 2 mM cystamine followed by 0.6 mg/ml Papain-SSCH3 for 1 h at room temperature. The reaction mixture was then incubated with 2.5 mM L-BAPNA for 1 h at room temperature, and absorbance was measured 410 nm. Thiol contents were calculated from a calibration curve using standardized cysteine solutions.

Statistical analysis

To compare the composition of mEar gene ratio in animals of each infection status, ²-analysis was performed. Otherwise, data are shown as the means.e.m. for the numbers of mice. The Student's t-test was performed for statistical analysis of the differences between the groups.

References

1. Lander ES, Linton LM, Birren B et al. Initial sequencing and analysis of the human genome. Nature 2001; 409: 860–921. | Article | PubMed | ISI | ChemPort |
2. Riordan J, D'Alessio G. Ribonucleases: Structures and Functions. Academic Press: San Diego, 1997.
3. Beintema JJ, Kleineidam RG. The ribonuclease A superfamily: general discussion. Cell Mol Life Sci 1998; 54: 825–832. | Article | PubMed |
4. Beintema JJ, Schuller C, Irie M, Carsana A. Molecular evolution of the ribonuclease superfamily. Prog Biophys Mol Biol 1988; 51: 165–192. | Article | PubMed | ISI | ChemPort |
5. Rosenberg HF. The eosinophil ribonucleases. Cell Mol Life Sci 1998; 54: 795–803. | Article | PubMed |
6. Rosenberg HF, Dyer KD, Tiffany HL, Gonzalez M. Rapid evolution of a unique family of primate ribonuclease genes. Nat Genet 1995; 10: 219–223. | Article | PubMed | ISI | ChemPort |
7. Zhang J, Rosenberg HF, Nei M. Positive Darwinian selection after gene duplication in primate ribonuclease genes. Proc Natl Acad Sci USA 1998; 95: 3708–3713. | Article | PubMed | ChemPort |
8. Zhang J, Dyer KD, Rosenberg HF. Evolution of the rodent eosinophil-associated RNase gene family by rapid gene sorting and positive selection. Proc Natl Acad Sci USA 2000; 97: 4701–4706. | Article | PubMed | ChemPort |
9. Larson KA, Olson EV, Madden BJ et al. Two highly homologous ribonuclease genes expressed in mouse eosinophils identify a larger subgroup of the mammalian ribonuclease superfamily. Proc Natl Acad Sci USA 1996; 93: 12370–12375. | Article | PubMed |
10. Singhania NA, Dyer KD, Zhang J et al. Rapid evolution of the ribonuclease A superfamily: adaptive expansion of independent gene clusters in rats and mice. J Mol Evol 1999; 49: 721–728. | PubMed | ISI | ChemPort |
11. Cormier SA, Yuan S, Crosby JR et al. T_H2-mediated pulmonary inflammation leads to the differential expression of ribonuclease genes by alveolar macrophages. Am J Respir Cell Mol Biol 2002; 27: 678–687. | PubMed |
12. Cormier SA, Larson KA, Yuan S et al. Mouse eosinophil-associated ribonucleases: a unique subfamily expressed during hematopoiesis. Mamm Genome 2001; 12: 352–361. | Article | PubMed | ISI | ChemPort |
13. Nittoh T, Hirakata M, Mue S, Ohuchi K. Identification of cDNA encoding rat eosinophil cationic protein/eosinophil-associated ribonuclease. Biochim Biophys Acta 1997; 1351: 42–46. | PubMed |
14. McDevitt AL, Deming MS, Rosenberg HF, Dyer KD. Gene structure and enzymatic activity of mouse eosinophil-associated ribonuclease 2. Gene 2001; 267: 23–30. | Article | PubMed |
15. Moreau JM, Dyer KD, Bonville CA et al. Diminished expression of an antiviral ribonuclease in response to pneumovirus infection in vivo. Antiviral Res 2003; 59: 181–191. | Article | PubMed |
16. Yang D, Rosenberg HF, Chen Q et al. Eosinophil-derived neurotoxin (EDN), an antimicrobial protein with chemotactic activities for dendritic cells. Blood 2003; 102: 3396–3403. | Article | PubMed |
17. Sabin EA, Kopf MA, Pearce EJ. Schistosoma mansoni egg-induced early IL-4 production is dependent upon IL-5 and eosinophils. J Exp Med 1996; 184: 1871–1878. | Article | PubMed | ISI | ChemPort |
18. Dent LA, Strath M, Mellor AL, Sanderson CJ. Eosinophilia in transgenic mice expressing interleukin 5. J Exp Med 1990; 172: 1425–1431. | Article | PubMed | ISI | ChemPort |
19. Tominaga A, Takaki S, Koyama N et al. Transgenic mice expressing a B cell growth and differentiation factor gene (interleukin 5) develop eosinophilia and autoantibody production. J Exp Med 1991; 173: 429–437. | Article | PubMed |
20. Rosenberg HF, Domachowske JB. Eosinophils, eosinophil ribonucleases, and their role in host defense against respiratory virus pathogens. J Leukoc Biol 2001; 70: 691–698. | PubMed | ISI | ChemPort |
21. Zhang J, Zhang YP, Rosenberg HF. Adaptive evolution of a duplicated pancreatic ribonuclease gene in a leaf-eating monkey. Nat Genet 2002; 30: 411–415. | Article | PubMed | ISI | ChemPort |
22. Hoffmann KF, McCarty TC, Segal DH et al. Disease fingerprinting with cDNA microarrays reveals distinct gene expression profiles in lethal type 1 and type 2 cytokine-mediated inflammatory reactions. FASEB J 2001; 15: 2545–2547. | PubMed | ChemPort |
23. Rosenberg HF, Domachowske JB. Eosinophil-derived neurotoxin. Methods Enzymol 2001; 341: 273–286. | PubMed |
24. Bystrom J, Wynn TA, Domachowske JB, Rosenberg HF. Gene microarray analysis reveals interleukin-5-dependent transcriptional targets in mouse bone marrow. Blood 2004; 103: 868–877. | Article | PubMed |

Acknowledgements

We are deeply indebted to Dr James Lee and Dr Nancy Lee, Mayo Scottsdale, for their generous gift of polyclonal anti-mEars antiserum and to Dr Klaus Matthaei, Australian National University, Canberra, Australia, for his gift of RNA samples from IL-5 transgenic mice. We are also grateful to the staff of the NIAID 14BS Animal Facility for the care of the mice used in these studies. This work was partly supported by a JSPS Research Fellowship for Japanese Biomedical and Behavioral Researchers at NIH (2004–2006).