• View in gallery
    Figure 1—

    Complete nucleotide sequence and translated amino acid sequence of the presumptive coding portion of cDNA of fMPO (A) and schematic representations of the hMPO CDS and partial representation of fMPO CDS organization based on comparison of the fMPO cDNA sequence with published sequences of fragments of the feline genome (B). In the nucleotide sequence, noncoding mRNA may have extended farther in the 5′ direction to this sequence but was not characterized. In the schematic representations, breaks indicate unpublished or nonaccessioned portions of the feline genome and thus were introns of unknown length. Exon 8 (161 bp) likewise existed in an unsequenced portion of the feline genome. Rectangles represent exons, and lines connecting rectangles are introns; values in rectangles indicate the number of nucleotides in each exon, and values between rectangles indicate the number of nucleotides in each intron. The CDS exons 1 to 3 lay within the feline genome fragment submitted under GenBank accession No. AANG01671927, CDS exons 4 to 7 lay within GenBank accession No. AANG01166014, and CDS exons 9 to 12 lay within GenBank accession No. AANG01671926. * = Stop codon. M = Methionine. K = Lysine. L = Leucine. A = Alanine. G = Glycine. I = Isoleucine. V = Valine. P = Proline. Q = Glutamine. S = Serine. E = Glutamic acid. T = Threonine. C = Cysteine. R = Arginine. D = Aspartic acid. Y = Tyrosine. F = Phenylalanine. H = Histidine. N = Asparagine. W = Tryptophan.

  • View in gallery
    Figure 2—

    Comparative amino acid sequences of feline, canine, human, and murine myeloperoxidase. Fully conserved amino acids are unshaded, residues with conservation of strong groups are shaded dark gray, residues with conservation of weak groups are in light gray, and residues with no consensus are shaded black. Numbers in the right column indicate the amino acid number of the last residue in the line from the methionine start codon. *Sequence of hMPO not displayed with 26 amino acids at the N-terminus because these amino acids lack homologous sequences in myeloperoxidase from other species.

  • View in gallery
    Figure 3—

    Photomicrographs of IFA staining of human (polyclonal rabbit anti-hMPO antibody; A) and feline (polyclonal rabbit anti-hMPO antibody; B) granulocytes, indicating an identical diffuse cytoplasmic fluorescence pattern. Immunofluorescence antibody staining of feline granulocytes with polyclonal rabbit anti-polyhistidine antibody, rabbit serum, or PBS solution alone resulted in minimal to no background fluorescence. Bar = 10 μm.

  • View in gallery
    Figure 4—

    Electron micrographs of a feline neutrophil with immunogold (polyclonal rabbit anti-hMPO antibody [1:100]) labeling of primary granules (arrowheads; A) and a human neutrophil with an identical labeling (polyclonal rabbit anti-hMPO antibody of primary granules [1:500]) pattern (B). The labeling confirmed that myeloperoxidase was localized to primary granules. *Nucleus. Bar =1 μm.

  • View in gallery
    Figure 5—

    Results of SDS-PAGE western immunoblotting of peroxidase-active feline granulocyte lysate (22.5 μL) indicating bands with similar molecular weights (left column; kDa) as those obtained with purified hMPO (22.5 ng). The myeloperoxidase heavy chain is visible at approximately 60 kDa, the light chain is visible at approximately 14 kDa, and autocatalytic fragments are visible at 40 and 20 kDa.

  • View in gallery
    Figure 6—

    Anti-hMPO ELISA reactivity (optical density, 405 nm) in hyperthyroid (n = 112) versus nonhyperthyroid (85) cats (A) and in hyperthyroid cats treated with surgery (1), methimazole (met; 39), radioactive iodine (4), a combination of treatments (combo; 7), or nothing (untreated; 14;B). No significant (P > 0.05) differences in reactivity were detected between or among groups. The mean + 2 SD optical density for the nonhyperthyroid cat group was 0.746.

  • 1.

    Klebanoff SJ. Myeloperoxidase: friend and foe. J Leukoc Biol 2005;77:598625.

  • 2.

    Hansson M, Olsson I, Nauseef WM. Biosynthesis, processing, and sorting of human myeloperoxidase. Arch Biochem Biophys 2006;445:214224.

  • 3.

    Yang JJ, Pendergraft WF, Alcorta DA, et al. Circumvention of normal constraints on granule protein gene expression in peripheral blood neutrophils and monocytes of patients with antineutrophil cytoplasmic autoantibody-associated glomerulonephritis. J Am Soc Nephrol 2004;15:21032114.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4.

    Brennan ML, Penn MS, Van Lente F, et al. Prognostic value of myeloperoxidase in patients with chest pain. N Engl J Med 2003;349:15951604.

  • 5.

    Bonomini F, Tengattini S, Fabiano A, et al. Atherosclerosis and oxidative stress. Histol Histopathol 2008;23:381390.

  • 6.

    Shao B, Oda MN, Oram JF, et al. Myeloperoxidase: an inflammatory enzyme for generating dysfunctional high density lipoprotein. Curr Opin Cardiol 2006;21:322328.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 7.

    Bonaci-Nikolic B, Nikolic MM, Andrejevic S, et al. Antineutrophil cytoplasmic antibody (ANCA)-associated autoimmune diseases induced by antithyroid drugs: comparison with idiopathic ANCA vasculitides. Arthritis Res Ther 2005;7:R1072R1081.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8.

    Gumá M, Salinas I, Reverter JL, et al. Frequency of antineutrophil cytoplasmic antibody in Graves' disease patients treated with methimazole. J Clin Endocrinol Metab 2003;88:21412146.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9.

    Harper L, Chin L, Daykin J, et al. Propylthiouracil and carbimazole associated-antineutrophil cytoplasmic antibodies (ANCA) in patients with Graves' disease. Clin Endocrinol (Oxf) 2004;60:671675.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 10.

    Lionaki S, Hogan SL, Falk RJ, et al. Association between thyroid disease and its treatment with ANCA small-vessel vasculitis: a case-control study. Nephrol Dial Transplant 2007;22:35083515.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11.

    Sato H, Hattori M, Fujieda M, et al. High prevalence of antineutrophil cytoplasmic antibody positivity in childhood onset Graves' disease treated with propylthiouracil. J Clin Endocrinol Metab 2000;85:42704273.

    • Search Google Scholar
    • Export Citation
  • 12.

    Nauseef WM. Lessons from MPO deficiency about functionally important structural features. Jpn J Infect Dis 2004;57:S4S5.

  • 13.

    Gresner P, Gromadzinska J, Wasowicz W. Polymorphism of selected enzymes involved in detoxification and biotransformation in relation to lung cancer. Lung Cancer 2007;57:125.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14.

    Xiao H, Heeringa P, Hu P, et al. Antineutrophil cytoplasmic autoantibodies specific for myeloperoxidase cause glomerulonephritis and vasculitis in mice. J Clin Invest 2002;110:955963.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15.

    Erdbrügger U, Hellmark T, Bunch DO, et al. Mapping of myeloperoxidase epitopes recognized by MPO-ANCA using humanmouse MPO chimers. Kidney Int 2006;69:17991805.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16.

    Heeringa P, Brouwer E, Tervaert JW, et al. Animal models of antineutrophil cytoplasmic antibody associated vasculitis. Kidney Int 1998;53:253263.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17.

    Patry YC, Nachman PH, Audrain MA, et al. Difference in antigenic determinant profiles between human and rat myeloperoxidase. Clin Exp Immunol 2003;132:505508.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18.

    Aucoin DP, Peterson ME, Hurvitz AI, et al. Propylthiouracil-induced immune-mediated disease in the cat. J Pharmacol Exp Ther 1985;234:1318.

    • Search Google Scholar
    • Export Citation
  • 19.

    Aucoin DP, Rubin RL, Peterson ME, et al. Dose-dependent induction of anti-native DNA antibodies in cats by propylthiouracil. Arthritis Rheum 1988;31:688692.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20.

    Peterson ME, Hurvitz AI, Leib MS, et al. Propylthiouracil-associated hemolytic anemia, thrombocytopenia, and antinuclear antibodies in cats with hyperthyroidism. J Am Vet Med Assoc 1984;184:806808.

    • Search Google Scholar
    • Export Citation
  • 21.

    Peterson ME, Kintzer PP, Hurvitz AI. Methimazole treatment of 262 cats with hyperthyroidism. J Vet Intern Med 1988;2:150157.

  • 22.

    Waldhauser L, Uetrecht J. Antibodies to myeloperoxidase in propylthiouracil-induced autoimmune disease in the cat. Toxicology 1996;114:155162.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 23.

    Fittschen C, Parmley RT, Bishop SP, et al. Morphometry of feline neutrophil granule genesis. Am J Anat 1988;181:195202.

  • 24.

    Bertram TA. Neutrophilic leukocyte structure and function in domestic animals. Adv Vet Sci Comp Med 1985;30:91129.

  • 25.

    Pontius JU, Mullikin JC, Smith DR, et al. Initial sequence and comparative analysis of the cat genome. Genome Res 2007;17:16751689.

  • 26.

    Apostolopoulos J, Ooi JD, Odobasic D, et al. The isolation and purification of biologically active recombinant and native autoantigens for the study of autoimmune disease. J Immunol Methods 2006;308:167178.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27.

    Cooray R, Petersson CG, Holmberg O. Isolation and purification of bovine myeloperoxidase from neutrophil granules. Vet Immunol Immunopathol 1993;38:261272.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 28.

    Bradley PP, Priebat DA, Christensen RD, et al. Measurement of cutaneous inflammation: estimation of neutrophil content with an enzyme marker. J Invest Dermatol 1982;78:206209.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 29.

    Ohashi YY, Kameoka Y, Persad AS, et al. Novel missense mutation found in a Japanese patient with myeloperoxidase deficiency. Gene 2004;327:195200.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30.

    Brown KE, Brunt EM, Heinecke JW. Immunohistochemical detection of myeloperoxidase and its oxidation products in Kupffer cells of human liver. Am J Pathol 2001;159:20812088.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 31.

    Taylor KL, Pohl J, Kinkade JM Jr. Unique autolytic cleavage of human myeloperoxidase. Implications for the involvement of active site MET409. J Biol Chem 1992;267:2528225288.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32.

    Green P. 2x genomes—does depth matter? Genome Res 2007;17:15471549.

  • 33.

    Bouali H, Nietert P, Nowling TM, et al. Association of the G-463A myeloperoxidase gene polymorphism with renal disease in African Americans with systemic lupus erythematosus. J Rheumatol 2007;34:20282034.

    • Search Google Scholar
    • Export Citation
  • 34.

    Chevrier I, Stücker I, Houllier AM, et al. Myeloperoxidase: new polymorphisms and relation with lung cancer risk. Pharmacogenetics 2003;13:729739.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 35.

    Zhang J, Zhu FY, Pu YP, et al. Analysis of multiple single nucleotide polymorphisms (SNPs) of myeloperoxidase (MPO) to screen for genetic markers associated with acute leukemia in Chinese Han population. J Toxicol Environ Health A 2007;70:901907.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 36.

    Pecoits-Filho R, Stenvinkel P, Marchlewska A, et al. A functional variant of the myeloperoxidase gene is associated with cardiovascular disease in end-stage renal disease patients. Kidney Int Suppl 2003;S172S176.

    • Search Google Scholar
    • Export Citation
  • 37.

    Borregaard N, Cowland JB. Granules of the human neutrophilic polymorphonuclear leukocyte. Blood 1997;89:35033521.

  • 38.

    Gullberg U, Andersson E, Garwicz D, et al. Biosynthesis, processing and sorting of neutrophil proteins: insight into neutrophil granule development. Eur J Haematol 1997;58:137153.

    • Search Google Scholar
    • Export Citation
  • 39.

    Villiers E, Baines S, Law AM, et al. Identification of acute myeloid leukemia in dogs using flow cytometry with myeloperoxidase, MAC387, and a canine neutrophil-specific antibody. Vet Clin Pathol 2006;35:5571.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 40.

    Liso V, Bennett J. Morphological and cytochemical characteristics of leukaemic promyelocytes. Best Pract Res Clin Haematol 2003;16:349355.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 41.

    Falk RJ, Jennette JC. Anti-neutrophil cytoplasmic autoantibodies with specificity for myeloperoxidase in patients with systemic vasculitis and idiopathic necrotizing and crescentic glomerulonephritis. N Engl J Med 1988;318:16511657.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 42.

    Savige J, Davies D, Falk RJ, et al. Antineutrophil cytoplasmic antibodies and associated diseases: a review of the clinical and laboratory features. Kidney Int 2000;57:846862.

    • Crossref
    • Search Google Scholar
    • Export Citation

Advertisement

Partial characterization of feline myeloperoxidase and investigation of its potential role as an autoantigen in hyperthyroid cats

Barrak M. PresslerDepartment of Veterinary Clinical Sciences, School of Veterinary Medicine, Purdue University, West Lafayette, IN 47907.

Search for other papers by Barrak M. Pressler in
Current site
Google Scholar
PubMed
Close
 DVM, PhD
,
Mark E. RobargeDepartment of Veterinary Clinical Sciences, School of Veterinary Medicine, Purdue University, West Lafayette, IN 47907.

Search for other papers by Mark E. Robarge in
Current site
Google Scholar
PubMed
Close
, and
Kathleen I. AndersonDepartment of Veterinary Clinical Sciences, School of Veterinary Medicine, Purdue University, West Lafayette, IN 47907.

Search for other papers by Kathleen I. Anderson in
Current site
Google Scholar
PubMed
Close
 MS

Abstract

Objective—To partially characterize the cDNA, amino acid sequence, and tertiary structure of feline myeloperoxidase, describe its cellular location in mature granulocytes, and determine whether hyperthyroid cats have anti-myeloperoxidase antibody.

Sample Population—Bone marrow RNA and whole blood from cats of various sources and feline serum samples submitted for measurement of total thyroxine concentration from September 2006 to July 2007.

Procedures—Feline myeloperoxidase cDNA was amplified from bone marrow RNA; presumptive splice sites were determined by comparison with human sequences. Intracellular localization of myeloperoxidase in granulocytes was determined by use of immunofluorescence and electron microscopy, and molecular weight and partial tertiary structure were determined by use of immunoblotting of granulocyte lysates. Anti-human myeloperoxidase (hMPO) antibody was detected via ELISA.

Results—A 2,493-bp sequence encompassing the 2,160-bp cDNA with presumably the same number and size of exons as hMPO was generated. Translation predicted 85% homology with hMPO. Feline myeloperoxidase was localized to neutrophil primary granules, and immunoblotting revealed heavy and light bands with molecular weights similar to those of hMPO. The prevalence of anti-hMPO antibody did not differ between nonhyperthyroid and hyperthyroid cats or among hyperthyroid cats subclassified by treatment modality.

Conclusions and Clinical Relevance—Moderate homology existed between feline myeloperoxidase and hMPO cDNA and protein. Although findings suggested a similar tertiary structure and function for the 2 proteins, they also suggested that inability to detect a high prevalence of anti-hMPO antibody in hyperthyroid cats may be attributable to antigenic differences between the human and feline proteins rather than a lack of autoantibody.

Abstract

Objective—To partially characterize the cDNA, amino acid sequence, and tertiary structure of feline myeloperoxidase, describe its cellular location in mature granulocytes, and determine whether hyperthyroid cats have anti-myeloperoxidase antibody.

Sample Population—Bone marrow RNA and whole blood from cats of various sources and feline serum samples submitted for measurement of total thyroxine concentration from September 2006 to July 2007.

Procedures—Feline myeloperoxidase cDNA was amplified from bone marrow RNA; presumptive splice sites were determined by comparison with human sequences. Intracellular localization of myeloperoxidase in granulocytes was determined by use of immunofluorescence and electron microscopy, and molecular weight and partial tertiary structure were determined by use of immunoblotting of granulocyte lysates. Anti-human myeloperoxidase (hMPO) antibody was detected via ELISA.

Results—A 2,493-bp sequence encompassing the 2,160-bp cDNA with presumably the same number and size of exons as hMPO was generated. Translation predicted 85% homology with hMPO. Feline myeloperoxidase was localized to neutrophil primary granules, and immunoblotting revealed heavy and light bands with molecular weights similar to those of hMPO. The prevalence of anti-hMPO antibody did not differ between nonhyperthyroid and hyperthyroid cats or among hyperthyroid cats subclassified by treatment modality.

Conclusions and Clinical Relevance—Moderate homology existed between feline myeloperoxidase and hMPO cDNA and protein. Although findings suggested a similar tertiary structure and function for the 2 proteins, they also suggested that inability to detect a high prevalence of anti-hMPO antibody in hyperthyroid cats may be attributable to antigenic differences between the human and feline proteins rather than a lack of autoantibody.

Myeloperoxidase is a heme-containing peroxidase that exists in primary (ie, azurophilic) mammalian neutrophil granules and in lesser amounts in monocyte granules. In humans, myeloperoxidase is transcribed and translated from a single 11-kb gene locus composed of 12 exons, and the product is cleaved after translation to form a heavy- and light-chain protomer.1,2 Two protomers then associate to form the mature myeloperoxidase protein, with disulfide bonds linking the heavy chains.1 In healthy humans, myeloperoxidase production takes place only in bone marrow neutrophil precursors, although de novo synthesis in circulating neutrophils takes place in humans with anti-myeloperoxidase antibody–associated vasculitis.2,3 Hydrogen peroxide and a halide (usually chloride) are converted to several toxic metabolites, including hydroxyl radicals, chlorine, singlet oxygen, and chloramines; therefore, much of the bactericidal activity of neutrophils that follows the respiratory burst is dependent on typical myeloperoxidase function.1

Myeloperoxidase has been implicated as a biomarker for or as being involved in the pathogenesis of several diseases in humans. For example, serum myeloperoxidase concentration is a predictor of myocardial infarction and associated adverse cardiac events in humans with chest pain.4 The reason for this association is unclear, although myeloperoxidase-induced oxidation and alteration of function of high-density lipoprotein have been suggested. Additionally, myeloperoxidase oxidation of arterial walls via production of reactive oxygen species and the commonness of neutrophil infiltration into atherosclerotic plaques also suggest a more direct role for myeloperoxidase in development of disease.5,6

Alternatively, in some forms of vasculitis, myeloperoxidase-specific ANCAs result in various combinations of rapidly progressive glomerular and pulmonary diseases. Anti-neutrophil cytoplasmic autoantibodies also develop in a subset of humans with hyperthyroidism. These hyperthyroidism-associated ANCAs most commonly have antimyeloperoxidase specificity, although associated clinical signs of vasculitis are not regularly manifested, and it is unclear whether ANCA development is associated with the disease or with antithyroidal drug treatment.7–11 Reduced or absent myeloperoxidase function is also evident in humans and results in an increase in susceptibility to infection attributable to defective neutrophil function.2,12 Altered myeloperoxidase activity may also modulate susceptibility to smoking-associated lung cancer, presumptively through reduced activation of inhaled chemicals to biochemically active carcinogens.13

Myeloperoxidase has been sequenced in a limited number of animal species. Murine myeloperoxidase has been sequenced and cloned, and myeloperoxidase-ANCA disease has been experimentally induced in mice.14 Experimental induction of disease was initially impeded by the fact that immunization of rodents with hMPO results in the production of anti-hMPO antibody, but the antibody does not cross-react with murine myeloperoxidase in vivo or is insufficient to cause disease. This phenomenon has been attributed to interspecies variation in protein sequence.15–17 Nevertheless, moderate sequence homology between the murine and hMPO genes, identical myeloperoxidase protein localization within neutrophils, and similarity in phenotypic expression of myeloperoxidase-ANCA disease likely imply that although myeloperoxidase in other mammalian species will likewise have variations in protein sequence and tertiary protein structure, its function will be similar. Myeloperoxidase gene sequences in higher mammals such as cats will likely have more similarities with gene sequences in the human protein; thus, these species may develop similar or identical myeloperoxidase-associated diseases.

Hyperthyroidism, one of the diseases associated with anti-myeloperoxidase ANCA in humans, is the most common endocrine disease of aging cats. Treatment modalities in both species are identical, including thyroidectomy, ablation of hypersecretory thyroid tissue with radioactive iodine, and suppression of thyroid gland function with antithyroidal medications. A considerable percentage of cats treated with antithyroidal medications develop hematologic or rheumatic idiosyncratic reactions, including anti-nuclear antibodies, fever, cytopenias, myasthenia gravislike disease, and intense pruritus leading to self-mutilation.18–21 When propylthiouracil is used for treatment, the frequency of these reactions is so high that this drug is no longer recommended. One study22 of clinically normal laboratory cats given propylthiouracil at higher than therapeutic doses revealed that 2 of 5 cats developed typical signs of propylthiouracil hypersensitivity, and circulating anti-hMPO antibody was detected in both cats by use of ELISA. Although the in vivo feline target autoantigen of these antibodies was not investigated, feline neutrophil granules are known to have peroxidase activity.23,24 Taken together, these findings suggest that anti-myeloperoxidase antibody may also develop in hyperthyroid cats treated with methimazole or propylthiouracil and could be a biomarker for development of treatment-associated adverse effects. If so, methimazole-treated hyperthyroid cats may serve as a natural example of drug-induced ANCA development. The purpose of the study reported here was to sequence the CDS for fMPO, predict the amino acid sequence of the translated protein, partially characterize circulating neutrophil fMPO, and determine whether treated or untreated hyperthyroid cats have anti-hMPO antibody.

Materials and Methods

Animals and specimens—A sample of femoral bone marrow was collected from a feral cat with severe respiratory disease immediately after euthanasia and was transferred into refrigerated (4°C) 0.02% EDTA-PBS solution. Whole blood for neutrophil isolation was collected from a privately owned healthy cat via jugular venipuncture. Serum aliquots were collected from all serum samples submitted to the Purdue University School of Veterinary Medicine Clinical Endocrinology Laboratory for measurement of serum total thyroxine concentration from September 2006 to July 2007 and stored at 80°C. Samples submitted included those from the Purdue School of Veterinary Medicine teaching hospital and unaffiliated privately owned veterinary hospitals. These samples originated from cats suspected of having hyperthyroidism, cats currently or historically treated for confirmed hyperthyroidism, and cats tested as part of a routine health screening protocol. All study protocols were approved by the Purdue University Animal Care and Use Committee.

RNA isolation and PCR assay—Total RNA was extracted from bone marrow within 1 hour after collection and stored at 80°C.a Specific primersb were used for gene-specific cDNA generation from RNA and amplification of target cDNA (Appendix). The cDNA was generated by use of a strand synthesis system,c and the fMPO cDNA was PCR amplified with recombinant DNA polymerased by use of a thermal cycler,e with the following cycling conditions: denaturation at 94°C for 2 minutes followed by 35 cycles of denaturation at 94°C for 30 seconds, annealing at the primer-specific temperature for 30 seconds, and extension at 72°C for 1 minute, with a 5-minute final extension cycle at 72°C. The PCR products were visualized on 1.5% agarose gels containing 0.1% ethidium bromide, purified with a purification kit,f and sequenced.g The PCR-product sequences were compared with known sequences of genes of other species and of the published 2X WGS assembly of the Felis catus genome25 by use of the Basic Local Alignment Search Tool algorithm of the National Center for Biotechnology Information and a commercial multiple sequence alignment program.h

Intracellular myeloperoxidase localization—Cytoplasmic distribution of myeloperoxidase was evaluated via IFA staining and electron microscopy. The protocol used for isolation and IFA staining of neutrophils was modified from a protocol reported elsewhere.14 Ten milliliters of whole blood was collected from a healthy human and a privately owned healthy cat, and neutrophils were isolated by use of a commercial isolation solution,i with the manufacturer's protocol modified to extend the initial centrifugation of the feline blood sample to 45 to 50 minutes because of the slower sedimentation rate of feline versus human RBCs. Isolated granulocytes were cytocentrifuged, and cells were blocked with 10% goat serum in PBS solution, primary antibodiesj−l were diluted 1:100 in 10% goat serum in PBS solution, and secondary antibodym was diluted 1:50 in PBS solution. Coverslipsn were applied to slides, and the slides were evaluated for positive immunofluorescence in arbitrary units (relative to immunofluorescence in cells labeled with diluent alone and no primary antibody) within 30 minutes.

For electron microscopy, isolated neutrophils were fixed with 4% paraformaldehyde and 0.1% glutaraldehyde in 0.01M phosphate buffer containing 2.7mM KCl and 137.0mM NaCl (pH, 4) for 15 minutes at 4°C. Cells were then washed, and cell pellets were enrobed with 1% agarose, dehydrated through a graded ethanol series, infiltrated through a graded ethanol and resino series, and embedded.o,p After polymerization for 24 hours at 55°C under nitrogen, blocks were cut into thin sections. The resulting grids were blocked in gelatin,q and primary antibodies in Tris-buffered saline (0.9% NaCl) solution plus 0.5% bovine serum albumin were applied and incubated overnight at 4°C. Sections were washed in Tris-buffered saline solution plus 0.3% Tween-20 and blocked, and secondary antibodyr,s was applied (1:50 in blocking solution) for 1 hour at room temperature (approx 27°C). After being washed and dried, sections were stained with 2% uranyl acetate in 70% ethanol for 10 minutes and viewed with an electron microscope.t

Neutrophil lysate preparation—Lysis of peripheral feline granulocytes was performed by use of protocols described elsewhere, with modifications.26,27 Sixty milliliters of whole cat blood in sodium citrate was diluted in 2 volumes of calcium- and magnesium-free 25mM EDTA and 0.01M PBS solution (pH, 7.4) and pelleted by centrifugation (800 X g for 20 minutes). The supernatant was discarded, and cells were washed in 25mM EDTA-PBS solution, resuspended in chilled (4°C) erythrocyte lysing buffer (NH4Cl [8.3 mg/mL], KHCO3 [1.0 mg/mL], and EDTA [90 mg/mL]), and held for 10 minutes on ice. Afterward, cells were pelleted (400 X g for 10 minutes), resuspended in fresh lysis buffer, and repelleted a total of 3 times. Cells were then resuspended in lysis solution (6.7mM NaH2PO4, 1mM μgCl2, and 3mM NaCl) with 15% (vol/vol) protease inhibitorst at a ratio of 1 mL of lysis solution to 2 × 107 cells.26,u The cell suspension was stirred vigorously for 3 hours at 4°C to ensure solubilization of myeloperoxidase and then centrifuged (20,000 X g for 20 minutes at 4°C) to pelletize the cell debris. Supernatant was resuspended in 100mM sodium acetate (pH, 6.3) and 100mM NaCl solution.

Peroxidase activity assay—The existence and relative concentration of myeloperoxidase during the isolation steps were monitored by means of the Bradley method.28 Ten microliters of the solution to be tested was added to 290 μL of myeloperoxidase assay solution (50mM NaPO4 [pH, 6.0], o-dianisidine dihydrochloride [0.0576 mg/mL],v and 3% H2O2 [0.167 μg/mL]), and optical density was immediately read with a spectrophotometerw at 460 nm, at 30-second intervals.

SDS-PAGE and western immunoblotting—Protein samples were separated via electrophoresis on a denaturing 15% polyacrylamide gel with SDS. After transfer, nitrocellulose membranes were blocked (10% milk protein) and then incubated with polyclonal rabbit anti-hMPOj (1:1,000 in 1% milk protein) at 4°C overnight. Membranes were rewashed, secondary antibodyx was applied (1:5,000 in 1% milk protein) for 1 hour at room temperature, and bound antibody was made visible.y Purified hMPO was used as a positive control sample.z

ELISA testing for anti-hMPO antibody—The ELISA platesaa were coated with hMPOz (5 μg/mL; 50 μL/well) and refrigerated overnight at 4°C. Afterward, the plates were washed and blocked with blocking bufferbb for 15 minutes (150 μL of buffer/well). Serum samples were rewashed, diluted 1:50 in blocking solution, applied to the ELISA plates (50 μL/well), and allowed to sit at room temperature for 2 hours. Alkaline phosphataseconjugated secondary antibodycc (1:5,000 in blocking buffer; 50 μL/well) was subsequently applied for 1 hour at room temperature, and then substrate was added for an additional hour.dd Afterward, optical density at 405 nm was measured with a microplate reader,ee and the resulting value was used to represent the degree of reactivity with the anti-hMPO antibody.

All serum samples were tested in duplicate, and results were averaged. Background reactivity in wells containing only coat buffer was subtracted from results for each sample. Each plate included an internal control sample of anti-hMPOj to serve as a positive control sample and to control for interplate variability by standardizing all results to a mean positive control optical density of 2.0. Immunoreactivity was defined as an optical density greater than the mean + 2 SD value for the nonhyperthyroid cat serum samples.

Statistical analysis—Analyses were conducted by use of a statistical software program.ff Median results for anti-hMPO ELISA reactivity (optical density) were compared between groups by means of the 2-sample Wilcoxon rank sum (Mann-Whitney U) test. Values of P < 0.05 were considered significant.

Results

fMPO cDNA and amino acid sequence—Generation of fMPO cDNA from total feline bone marrow RNA was first attempted with oligo(dT)20, followed by PCR assays of 6 overlapping myeloperoxidase cDNA fragments by use of reported primer pairs for hMPO29; however, only 1 fragment (No. 5) was successfully amplified with this technique. Alignment of the human primers by use of the Basic Local Alignment Search Tool suggested that the fMPO gene spanned 3 feline genome fragments (GenBank accession Nos. AANG01166014, AANG01671926, and AANG01671927), but on the basis of homology, it was determined that all 3 accessions were registered in GenBank in the reverse complement orientation. For subsequent cDNA generation and PCR amplification of the remaining 5 fragments, gene-specific fMPO primers were used (Appendix). The size of the resultant cDNA PCR products was similar to the reported size of hMPO cDNA fragments.29 Sequencing of the total overlapping fMPO cDNA composite resulted in a 2,493-bp partial fMPO cDNA sequence (Figure 1; GenBank accession No. EU491019). The longest open reading frame in this sequence was 2,160 bp in length, with noncoding mRNA extending 5a from the presumptive fMPO start codon by a minimum of 132 bp. It was not determined how many additional noncoding mRNA base pairs extended beyond the point at which the first fragment's reverse primer yielded sequence because our goal was solely to determine the fMPO CDS.

Figure 1—
Figure 1—

Complete nucleotide sequence and translated amino acid sequence of the presumptive coding portion of cDNA of fMPO (A) and schematic representations of the hMPO CDS and partial representation of fMPO CDS organization based on comparison of the fMPO cDNA sequence with published sequences of fragments of the feline genome (B). In the nucleotide sequence, noncoding mRNA may have extended farther in the 5′ direction to this sequence but was not characterized. In the schematic representations, breaks indicate unpublished or nonaccessioned portions of the feline genome and thus were introns of unknown length. Exon 8 (161 bp) likewise existed in an unsequenced portion of the feline genome. Rectangles represent exons, and lines connecting rectangles are introns; values in rectangles indicate the number of nucleotides in each exon, and values between rectangles indicate the number of nucleotides in each intron. The CDS exons 1 to 3 lay within the feline genome fragment submitted under GenBank accession No. AANG01671927, CDS exons 4 to 7 lay within GenBank accession No. AANG01166014, and CDS exons 9 to 12 lay within GenBank accession No. AANG01671926. * = Stop codon. M = Methionine. K = Lysine. L = Leucine. A = Alanine. G = Glycine. I = Isoleucine. V = Valine. P = Proline. Q = Glutamine. S = Serine. E = Glutamic acid. T = Threonine. C = Cysteine. R = Arginine. D = Aspartic acid. Y = Tyrosine. F = Phenylalanine. H = Histidine. N = Asparagine. W = Tryptophan.

Citation: American Journal of Veterinary Research 70, 7; 10.2460/ajvr.70.7.869

Comparison of the fMPO cDNA sequence with that of the feline genome confirmed that the fMPO gene spanned the 3 aforementioned accessioned fragments; however, the sequence reported here differed from the reported GenBank WGS project assembly by 6 nucleotides. The WGS assembly portion that encodes fMPO includes 4 incorrect insertions, 1 deletion, and 1 incorrect nucleotide (Table 1). Comparison of our fMPO cDNA sequence with that of the hMPO gene supported the validity of these corrections because without the respective insertions and deletions identified in the present study, there is no open reading frame of sufficient length to correspond to the mature fMPO protein.

Table 1—

Corrections to accessioned feline WGS assemblies based on results of fMPO cDNA amplification.

Feline GenBank accession No.Nucleotide (position in GenBank accession)CorrectionProposed new nucleotide (position in fMPO cDNA)
AANG01671927G (1,024)Insertion mistakeNucleotide removed (between 334 and 335)
AANG01671927C (1,074)Insertion mistakeNucleotide removed (between 384 and 385)
AANG01671927T (1,113)Incorrect nucleotideC (423)
AANG01166014(46)Deletion mistakeA (518)
AANG01671926G (3,218)Insertion mistakeNucleotide removed (between 2,433 and 2,434)
AANG01671926T (3,219)Insertion mistakeNucleotide removed (between 2,433 and 2,434)

By alignment of the final fMPO cDNA with that of the feline WGS assembly, the presumptive exon and intron splice sites for most of the complete gene were determined (Figure 1). The CDS was subdivided into a minimum of 12 exons, with a thirteenth noncoding 5a exon at least 130 bp in length. There remained an unsequenced fragment of genomic DNA in the WGS assembly that contained a central 161-bp portion of fMPO CDS and portions of the flanking introns.

The fMPO cDNA and predicted amino acid sequence were then compared with those of the hMPO gene and protein (GenBank accession No. DQ088846; Figure 2). The translated fMPO amino acid sequence was moderately similar to that of human, murine (GenBank accession No. X15378), and predicted canine (GenBank accession No. XM847352) myeloperoxidase sequences. The predicted feline amino acid sequence had 85%, 90%, and 82% homology with sequences of the other 3 species, respectively (excluding the additional 26 N-terminal amino acids resulting from translation of the initial 78 bp in hMPO).

Figure 2—
Figure 2—

Comparative amino acid sequences of feline, canine, human, and murine myeloperoxidase. Fully conserved amino acids are unshaded, residues with conservation of strong groups are shaded dark gray, residues with conservation of weak groups are in light gray, and residues with no consensus are shaded black. Numbers in the right column indicate the amino acid number of the last residue in the line from the methionine start codon. *Sequence of hMPO not displayed with 26 amino acids at the N-terminus because these amino acids lack homologous sequences in myeloperoxidase from other species.

Citation: American Journal of Veterinary Research 70, 7; 10.2460/ajvr.70.7.869

Primary granule localization of fMPO—Immunofluorescence antibody staining of intact feline granulocytes with anti-hMPO antibody resulted in a cytoplasmic staining pattern consistent with that expected from an intragranule protein and identical to results achieved with human granulocytes (Figure 3). Minimal to no staining was evident with rabbit serum, anti-polyhistidine rabbit polyclonal antibody, or PBS solution alone. Transmission electron microscopy confirmed that fMPO was confined to primary neutrophil granules, with a staining pattern that was identical to that detected in human neutrophils (Figure 4). However, the amount of antibody labeling detected in human neutrophils was significantly more than that in feline neutrophils, despite a concentration of primary antibody that was 5-fold as low (1:500 vs 1:100 in human and feline neutrophils, respectively).

Figure 3—
Figure 3—

Photomicrographs of IFA staining of human (polyclonal rabbit anti-hMPO antibody; A) and feline (polyclonal rabbit anti-hMPO antibody; B) granulocytes, indicating an identical diffuse cytoplasmic fluorescence pattern. Immunofluorescence antibody staining of feline granulocytes with polyclonal rabbit anti-polyhistidine antibody, rabbit serum, or PBS solution alone resulted in minimal to no background fluorescence. Bar = 10 μm.

Citation: American Journal of Veterinary Research 70, 7; 10.2460/ajvr.70.7.869

Figure 4—
Figure 4—

Electron micrographs of a feline neutrophil with immunogold (polyclonal rabbit anti-hMPO antibody [1:100]) labeling of primary granules (arrowheads; A) and a human neutrophil with an identical labeling (polyclonal rabbit anti-hMPO antibody of primary granules [1:500]) pattern (B). The labeling confirmed that myeloperoxidase was localized to primary granules. *Nucleus. Bar =1 μm.

Citation: American Journal of Veterinary Research 70, 7; 10.2460/ajvr.70.7.869

Heavy- and light-chain subunits of fMPO—Peroxidase activity was maintained at all expected steps during granulocyte lysate preparation, which was consistent with functional myeloperoxidase. Cell lysate contained immunoreactive myeloperoxidase that comigrated with purified hMPO (Figure 5). Major bands weighing approximately 60 and 14 kDa were recognized in the feline cell lysate by the anti-hMPO antibody, and these are the same molecular weights reported for the heavy- and light-chain subunits of hMPO.26 Minor bands at 40 and 20 kDa, respectively, in the hMPO and feline neutrophil lysate were also detected. Other western immunoblots in which the anti-hMPO antibody was used have revealed that these bands are autocatalytic protein products.30,31

Figure 5—
Figure 5—

Results of SDS-PAGE western immunoblotting of peroxidase-active feline granulocyte lysate (22.5 μL) indicating bands with similar molecular weights (left column; kDa) as those obtained with purified hMPO (22.5 ng). The myeloperoxidase heavy chain is visible at approximately 60 kDa, the light chain is visible at approximately 14 kDa, and autocatalytic fragments are visible at 40 and 20 kDa.

Citation: American Journal of Veterinary Research 70, 7; 10.2460/ajvr.70.7.869

Prevalence of anti-hMPO antibody in hyperthyroid cats—Serum samples from 112 hyperthyroid cats and 85 nonhyperthyroid cats were tested for anti-hMPO reactivity by means of ELISA (Table 2). Hyperthyroid cats included 47 cats of unknown treatment status, 14 untreated cats, 39 cats that had received or were still receiving methimazole, 4 cats that had been treated with radioactive iodine, 1 thyroidectomized cat, and 7 cats that had been treated with > 1 modality. The most common reason for cats undergoing multiple treatment modalities was administration of methimazole prior to treatment with radioactive iodine.

Table 2—

Anti-hMPO antibody reactivity (optical density) in serum harvested from blood samples of client-owned hyperthyroid and nonhyperthyroid cats, by treatment modality.

GroupNo. of samplesReactivity*
MeanMedianRange
Hyperthyroid112†0.1820.0350.03.58
   Untreated140.0620.0180.0–0.52
   Methimazole-treated390.2780.0450.0–3.58
   Radioactive iodine–treated40.0760.0510.03–0.17
   Surgical thyroidectomy10.2070.207
   Combination treatment70.1130.0680.0–0.29
Nonhyperthyroid850.1500.0380.0–1.65

Optical density measured at 405 nm, 60 minutes after addition of substrate. Interassay results were standardized by use of the optical density of a positive control sample.

Includes 47 cats of unknown treatment status at time of thyroxine measurement.

= Not applicable.

Anti-hMPO ELISA reactivity (optical density, 405 nm) in serum samples from hyperthyroid and nonhyperthyroid cats did not differ significantly (Figure 6). Mean + 2 SD anti-hMPO ELISA reactivity of the serum samples from nonhyperthyroid cats was 0.746; serum samples from 7 hyperthyroid and 5 nonhyperthyroid cats had an ELISA reactivity value that exceeded this value. When hyperthyroid cats were subclassified on the basis of treatment status, there was still no significant difference between groups with respect to anti-hMPO ELISA reactivity. Four of the 7 hyperthyroid-cat serum samples with anti-hMPO ELISA reactivity > 0.746 were from cats that had been treated only with methimazole.

Figure 6—
Figure 6—

Anti-hMPO ELISA reactivity (optical density, 405 nm) in hyperthyroid (n = 112) versus nonhyperthyroid (85) cats (A) and in hyperthyroid cats treated with surgery (1), methimazole (met; 39), radioactive iodine (4), a combination of treatments (combo; 7), or nothing (untreated; 14;B). No significant (P > 0.05) differences in reactivity were detected between or among groups. The mean + 2 SD optical density for the nonhyperthyroid cat group was 0.746.

Citation: American Journal of Veterinary Research 70, 7; 10.2460/ajvr.70.7.869

Discussion

In the study reported here, the cDNA sequence and presumptive exon and intron splice sites of fMPO were determined. In addition, a theoretic translation of the primary amino acid sequence was performed, and the approximate molecular weight and cellular location of mature myeloperoxidase were determined. Our reported partial fMPO cDNA sequence included the complete 2,160-bp CDS, and we detected 6 bp in the CDS that differed from the CDS of the accessioned partial feline genome. Comparison of the fMPO CDS with the hMPO sequence revealed 85% homology, including a likely identical number of exons.

The detection of a 6-nucleotide difference between the fMPO CDS reported here and that of the accessioned feline genome was not surprising. Certain types of systematic errors are inherent to PCR-assay and DNA-sequencing techniques; thus, complete genome reporting usually requires 3.6- to 7-fold coverage (ie, repeated sequencing of enough DNA to be equivalent to 3.6 to 7 times the entire length of the genome) before results are considered complete and accurate.25 The feline WGS assembly was published after only light coverage (1.9-fold) was performed because the National Human Genome Research Institute has endorsed preliminary reporting of genomes from several species as a method for stimulating research on larger-scale genome comparisons.25 Insertions, deletions, and incorrect nucleotides will continue to be reported as other feline genes are individually sequenced, as will lengths of unsequenced genome such as the one that includes the presumptive eighth fMPO exon.32

The hMPO CDS is comprised of 12 exons, with a thirteenth 5a noncoding exon. Exon lengths are highly homologous to those determined for fMPO in the study reported here, with only the first coding exon definitively differing in length between the 2 species. This difference was attributable to a 78-bp segment of noncoding mRNA 5a to the initiation codon in fMPO, which in humans encodes a 26-amino acid portion of myeloperoxidase with an initiating methionine start codon (GenBank accession No. DQ088846). This 26-amino acid N-terminal sequence is also lacking in murine myeloperoxidase, but it is undetermined in the canine sequence. The eighth coding exon in humans (161 bp in length) is homologous to a 161-bp portion of CDS in the fMPO cDNA that cannot be aligned with that of the feline WGS assembly. Given this homology, it is likely that this 161-bp span of CDS is a single exon in cats, but this cannot be definitely concluded on the basis of our findings.

Single nucleotide polymorphisms in the hMPO promoter region and the CDS have been reported.33–35 Although single nucleotide polymorphisms do not result in differences in amino acid sequence, they nevertheless have been associated with predisposition or protection from certain diseases such as systemic lupus erythematosus, some neoplasms, and cardiovascular disease.13,33–36 Single nucleotide polymorphisms have been identified throughout the feline WGS assembly as well, with as many as 1 in 600 nucleotides varying between the 2 alleles in the 1 cat used for the reported genome.25 One of the proposed corrections to the feline WGS assembly we reported here (1,113TmC; ie, change in position 1,113 in GenBank accession No. AANG01671927 from thymine to cytosine) may in fact be a single nucleotide polymorphism because this nucleotide change does not result in a change in the coded amino acid (alanine). Because the fMPO CDS reported here was generated from total bone marrow RNA from only 1 cat, we expect that additional single nucleotide polymorphisms will be identified in later studies.

Primary neutrophil granules are first synthesized at the promyelocyte stage of neutrophil development.23,37,38 Feline, human, and rabbit primary granules have peroxidase activity that is absent in the secondary granules synthesized later in the myelocyte stage.23 The close association between myeloperoxidase RNA transcription and translation and the neutrophil precursor stage is reliable enough that identification of myeloperoxidase RNA has been used in several species to characterize poorly differentiated hematologic neoplasms as granulocytic in origin.37,39,40 However, unlike in other species, peroxidase-active large granules in cats are also synthesized late in neutrophil development.23 Our confirmation and characterization of myeloperoxidase as the source of peroxidase activity in mature circulating feline neutrophils should thus allow better characterization of these late granules and use of myeloperoxidase as a marker for feline granulocytic neoplasms, as has been done in other species. Additionally, to our knowledge, our identification of myeloperoxidase as a constituent of primary granules in feline neutrophils represented the first time any protein has been localized to this cellular compartment. We had initially planned on performing confocal microscopy to confirm the intragranular location, but no other antibody was available that could be used as a positive control antibody for primary granule localization. Confirmation that fMPO is confined to neutrophil primary granules will therefore allow other investigators to use anti-myeloperoxidase antibodies and confocal microscopy (which is simpler and less expensive than transmission electron microscopy) to determine whether other proteins also localize to neutrophil granules.

Primary neutrophil granules are typically evenly distributed throughout the cytoplasm. During ethanol fixation, myeloperoxidase artifactually migrates to the perinuclear region; thus, humans with myeloperoxidase-specific ANCA most commonly have a so-called p-ANCA (perinuclear ANCA) staining pattern. The other antigen that is most commonly targeted by naturally developing ANCA in humans, proteinase-3, usually results in a so-called c-ANCA (cytoplasmic ANCA) pattern attributable to persistent pancytoplasmic distribution of the antigen after alcohol fixation.41,42 On the other hand, formalin-acetone fixation does not result in these differences in staining pattern.41 Interestingly, despite our use of ethanol fixation for feline and human granulocyte preparations, for unknown reasons our commercial anti-hMPO antibody resulted in a cytoplasmic ANCA immunofluorescence pattern in both species. Nevertheless, results of transmission electron microscopy confirmed that antibody binding was specific and that fMPO existed in primary neutrophil granules as hMPO does in humans.37

The mature hMPO protein is composed of 2 heavy- and light-chain dimers that are transcribed and translated from a single locus into promyeloperoxidase and then proteolytically cleaved into the 2 chains.26 Therefore, SDS-PAGE performed with reducing conditions is expected to result in 59-kDa (heavy-chain) and 13.5-kDa (light-chain) bands. We successfully identified the hMPO heavy and light chains and bands of similar molecular weight in feline granulocyte lysate, arguing that fMPO is similarly posttranslationally modified into a 2-chain protomer. Whether the mature feline protein incorporates a heme molecule or whether 2 heavy-light chain dimers associate to form the final protein quaternary structure similar to the process in humans is still unknown.26 However, based on similarities in myeloperoxidase from disparate species, the homology of fMPO with hMPO, and the peroxidase activity detected in feline neutrophil lysates, it is highly likely that the mature protein structure in cats is highly similar to that in other mammals.2,34

Despite the moderate (85%) homology between the murine and hMPO genes and the similar size and identical primary granule location of the mature proteins, studies14–16 in which rodents have been used to investigate anti-myeloperoxidase ANCA–associated diseases have involved the murine rather than the human protein. Humans with anti-myeloperoxidase ANCA develop antibody against various myeloperoxidase epitopes, and replacement of portions of the hMPO protein with murine myeloperoxidase abolishes some of this seroreactivity.15 Although we were able to detect some anti-hMPO seroreactivity in the hyperthyroid and nonhyperthyroid cat groups, results were not significantly different between these groups. Given results of similar studies14–16 involving mice, we therefore suspected that the 15% difference we found in the predicted protein sequence between cats and humans likewise prevented detection of some circulating anti-fMPO antibody, and future laboratory assays should be designed to rely on purified or recombinant fMPO rather than continuing to use hMPO-based detection systems.

Although a subset of methimazole-treated hyperthyroid cats with anti-hMPO antibody was detected, there was no significant difference in the prevalence of antibody between hyperthyroid cats and nonhyperthyroid cats or hyperthyroid cats treated with other modalities. We suspect that our inability to identify a high number of hyperthyroid cats with anti-hMPO antibody was because of a low prevalence of myeloperoxidase ANCA in this disease, as reported in humans.7–11 We initially attempted to retrospectively review patient medical records to determine the prevalence of methimazole-associated complications in these cats, but insufficient information had been recorded to allow this subclassification. Additional studies should be performed to specifically test hyperthyroid cats with adverse effects associated with methimazole treatment to increase the likelihood of anti-fMPO antibody detection and determine whether a specific drug reaction is more likely to be associated with anti-fMPO antibody development.

Results of the study reported here indicated that the sequence, structure, and intracellular localization of mature fMPO are likely moderately to highly homologous to those of hMPO. Our characterization of fMPO provides a reference for future studies of feline neutrophil granules and will assist in the development of fMPO-specific assays. On the basis of similarities between cats and humans with respect to several myeloperoxidase-associated diseases (particularly adverse effects attributable to treatment of hyperthyroidism with antithyroidal drugs), additional studies are warranted to determine whether anti-fMPO antibody represents a biomarker that can be adapted for clinical use. Although a significantly higher prevalence of anti-hMPO antibody in hyperthyroid versus nonhyperthyroid cats was not detected, we suspect that this was attributable to the only moderate homology with the human protein sequence. The 85% homology predicted between the hMPO and fMPO amino acid sequences is sufficient to warrant additional studies, but such studies may need to involve isolation of native fMPO protein or production of recombinant fMPO.

ABBREVIATIONS

ANCA

Anti-neutrophil cytoplasmic autoantibody

CDS

Coding DNA sequence

fMPO

Feline myeloperoxidase

hMPO

Human myeloperoxidase

IFA

Immunofluorescence antibody

WGS

Whole Genome Shotgun

a.

Purescript RNA isolation kit for blood and bone marrow, Gentra Systems, Minneapolis, Minn.

b.

Integrated DNA Technologies, Coralville, Iowa.

c.

SuperScript III first-strand synthesis system for reverse-transcriptase PCR, Invitrogen, Carlsbad, Calif.

d.

AmpliTaq DNA polymerase, Applied Biosystems, Foster City, Calif.

e.

Apollo ATC401 Thermal Cycler, GMI Inc, Ramsey, Minn.

f.

Qiagen PCR purification kit, Qiagen, Valencia, Calif.

g.

Genomics Core Facility, Purdue University, West Lafayette, Ind.

h.

CLUSTALW, Biology WorkBench, San Diego Supercomputer Center, University of California at San Diego, La Jolla, Calif. Available at: workbench.sdsc.edu. Accessed Oct 9, 2008.

i.

Polymorphprep, Axis-Shield, Oslo, Norway.

j.

Polyclonal rabbit anti-human myeloperoxidase, DakoCytomation, Glostrup, Denmark.

k.

Rabbit anti-polyhistidine, Santa Cruz Biotechnology, Santa Cruz, Calif.

l.

Normal rabbit serum, Bethyl Laboratories Inc, Montgomery, Tex.

m.

Fluorescein isothiocyanateconjugated polyclonal goat antirabbit IgG, Jackson ImmunoResearch, West Grove, Pa.

n.

ProLong Gold with DAPI, Invitrogen, Carlsbad, Calif; or Fluoromount G, Electron Microscopy Sciences, Hatfield, Pa.

o.

LR white resin, Electron Microscopy Services, Hatfield, Pa.

p.

BEEM capsules, BEEM Inc, West Chester, Pa.

q.

1% BSA-c with 0.1% cold water fish skin gelatin in BSA buffer, Aurion, Wageningen, The Netherlands.

r.

Goat anti-rabbit IgG (H+L), electron-microscopy grade 10-nm gold conjugate, Ted Pella Inc, Redding, Calif.

s.

Rabbit anti-goat IgG (whole molecule), electron-microscopy grade 10-nm gold conjugate, Sigma-Aldrich, St Louis, Mo.

t.

FEI/Philips CM-10 transmission electron microscope, 80 kV, FEI Co, Hillsboro, Ore.

u.

Protease inhibitor cocktail, Sigma-Aldrich, St Louis, Mo.

v.

Sigma-Aldrich, St Louis, Mo.

w.

ND-1000 spectrophotometer, NanoDrop, Wilmington, Del.

x.

Horse radish peroxidaseconjugated goat anti-rabbit IgG, Chemicon, Temecula, Calif.

y.

Amersham ECL western blotting system, GE Healthcare, Piscataway, NJ.

z.

Human myeloperoxidase, Calbiochem, San Diego, Calif.

aa.

Costar enzyme immunoassay/radioimmunoassay high-binding 96-well plates, Corning Inc, Lowell, Mass.

bb.

Superblock T20, Pierce, Rockford, Ill.

cc.

Alkaline phosphatase–conjugated goat-anti-cat antibody, KPL Laboratories, Gaithersburg, Md.

dd.

AP substrate kit, Bio-Rad Laboratories, Hercules, Calif.

ee.

Synergy HT ELISA microplate reader, BioTek Instruments Inc, Winooski, Vt.

ff.

Stata, version 10.1, StataCorp, College Station, Tex.

References

  • 1.

    Klebanoff SJ. Myeloperoxidase: friend and foe. J Leukoc Biol 2005;77:598625.

  • 2.

    Hansson M, Olsson I, Nauseef WM. Biosynthesis, processing, and sorting of human myeloperoxidase. Arch Biochem Biophys 2006;445:214224.

  • 3.

    Yang JJ, Pendergraft WF, Alcorta DA, et al. Circumvention of normal constraints on granule protein gene expression in peripheral blood neutrophils and monocytes of patients with antineutrophil cytoplasmic autoantibody-associated glomerulonephritis. J Am Soc Nephrol 2004;15:21032114.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4.

    Brennan ML, Penn MS, Van Lente F, et al. Prognostic value of myeloperoxidase in patients with chest pain. N Engl J Med 2003;349:15951604.

  • 5.

    Bonomini F, Tengattini S, Fabiano A, et al. Atherosclerosis and oxidative stress. Histol Histopathol 2008;23:381390.

  • 6.

    Shao B, Oda MN, Oram JF, et al. Myeloperoxidase: an inflammatory enzyme for generating dysfunctional high density lipoprotein. Curr Opin Cardiol 2006;21:322328.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 7.

    Bonaci-Nikolic B, Nikolic MM, Andrejevic S, et al. Antineutrophil cytoplasmic antibody (ANCA)-associated autoimmune diseases induced by antithyroid drugs: comparison with idiopathic ANCA vasculitides. Arthritis Res Ther 2005;7:R1072R1081.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8.

    Gumá M, Salinas I, Reverter JL, et al. Frequency of antineutrophil cytoplasmic antibody in Graves' disease patients treated with methimazole. J Clin Endocrinol Metab 2003;88:21412146.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9.

    Harper L, Chin L, Daykin J, et al. Propylthiouracil and carbimazole associated-antineutrophil cytoplasmic antibodies (ANCA) in patients with Graves' disease. Clin Endocrinol (Oxf) 2004;60:671675.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 10.

    Lionaki S, Hogan SL, Falk RJ, et al. Association between thyroid disease and its treatment with ANCA small-vessel vasculitis: a case-control study. Nephrol Dial Transplant 2007;22:35083515.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11.

    Sato H, Hattori M, Fujieda M, et al. High prevalence of antineutrophil cytoplasmic antibody positivity in childhood onset Graves' disease treated with propylthiouracil. J Clin Endocrinol Metab 2000;85:42704273.

    • Search Google Scholar
    • Export Citation
  • 12.

    Nauseef WM. Lessons from MPO deficiency about functionally important structural features. Jpn J Infect Dis 2004;57:S4S5.

  • 13.

    Gresner P, Gromadzinska J, Wasowicz W. Polymorphism of selected enzymes involved in detoxification and biotransformation in relation to lung cancer. Lung Cancer 2007;57:125.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14.

    Xiao H, Heeringa P, Hu P, et al. Antineutrophil cytoplasmic autoantibodies specific for myeloperoxidase cause glomerulonephritis and vasculitis in mice. J Clin Invest 2002;110:955963.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 15.

    Erdbrügger U, Hellmark T, Bunch DO, et al. Mapping of myeloperoxidase epitopes recognized by MPO-ANCA using humanmouse MPO chimers. Kidney Int 2006;69:17991805.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16.

    Heeringa P, Brouwer E, Tervaert JW, et al. Animal models of antineutrophil cytoplasmic antibody associated vasculitis. Kidney Int 1998;53:253263.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17.

    Patry YC, Nachman PH, Audrain MA, et al. Difference in antigenic determinant profiles between human and rat myeloperoxidase. Clin Exp Immunol 2003;132:505508.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18.

    Aucoin DP, Peterson ME, Hurvitz AI, et al. Propylthiouracil-induced immune-mediated disease in the cat. J Pharmacol Exp Ther 1985;234:1318.

    • Search Google Scholar
    • Export Citation
  • 19.

    Aucoin DP, Rubin RL, Peterson ME, et al. Dose-dependent induction of anti-native DNA antibodies in cats by propylthiouracil. Arthritis Rheum 1988;31:688692.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20.

    Peterson ME, Hurvitz AI, Leib MS, et al. Propylthiouracil-associated hemolytic anemia, thrombocytopenia, and antinuclear antibodies in cats with hyperthyroidism. J Am Vet Med Assoc 1984;184:806808.

    • Search Google Scholar
    • Export Citation
  • 21.

    Peterson ME, Kintzer PP, Hurvitz AI. Methimazole treatment of 262 cats with hyperthyroidism. J Vet Intern Med 1988;2:150157.

  • 22.

    Waldhauser L, Uetrecht J. Antibodies to myeloperoxidase in propylthiouracil-induced autoimmune disease in the cat. Toxicology 1996;114:155162.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 23.

    Fittschen C, Parmley RT, Bishop SP, et al. Morphometry of feline neutrophil granule genesis. Am J Anat 1988;181:195202.

  • 24.

    Bertram TA. Neutrophilic leukocyte structure and function in domestic animals. Adv Vet Sci Comp Med 1985;30:91129.

  • 25.

    Pontius JU, Mullikin JC, Smith DR, et al. Initial sequence and comparative analysis of the cat genome. Genome Res 2007;17:16751689.

  • 26.

    Apostolopoulos J, Ooi JD, Odobasic D, et al. The isolation and purification of biologically active recombinant and native autoantigens for the study of autoimmune disease. J Immunol Methods 2006;308:167178.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27.

    Cooray R, Petersson CG, Holmberg O. Isolation and purification of bovine myeloperoxidase from neutrophil granules. Vet Immunol Immunopathol 1993;38:261272.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 28.

    Bradley PP, Priebat DA, Christensen RD, et al. Measurement of cutaneous inflammation: estimation of neutrophil content with an enzyme marker. J Invest Dermatol 1982;78:206209.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 29.

    Ohashi YY, Kameoka Y, Persad AS, et al. Novel missense mutation found in a Japanese patient with myeloperoxidase deficiency. Gene 2004;327:195200.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30.

    Brown KE, Brunt EM, Heinecke JW. Immunohistochemical detection of myeloperoxidase and its oxidation products in Kupffer cells of human liver. Am J Pathol 2001;159:20812088.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 31.

    Taylor KL, Pohl J, Kinkade JM Jr. Unique autolytic cleavage of human myeloperoxidase. Implications for the involvement of active site MET409. J Biol Chem 1992;267:2528225288.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32.

    Green P. 2x genomes—does depth matter? Genome Res 2007;17:15471549.

  • 33.

    Bouali H, Nietert P, Nowling TM, et al. Association of the G-463A myeloperoxidase gene polymorphism with renal disease in African Americans with systemic lupus erythematosus. J Rheumatol 2007;34:20282034.

    • Search Google Scholar
    • Export Citation
  • 34.

    Chevrier I, Stücker I, Houllier AM, et al. Myeloperoxidase: new polymorphisms and relation with lung cancer risk. Pharmacogenetics 2003;13:729739.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 35.

    Zhang J, Zhu FY, Pu YP, et al. Analysis of multiple single nucleotide polymorphisms (SNPs) of myeloperoxidase (MPO) to screen for genetic markers associated with acute leukemia in Chinese Han population. J Toxicol Environ Health A 2007;70:901907.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 36.

    Pecoits-Filho R, Stenvinkel P, Marchlewska A, et al. A functional variant of the myeloperoxidase gene is associated with cardiovascular disease in end-stage renal disease patients. Kidney Int Suppl 2003;S172S176.

    • Search Google Scholar
    • Export Citation
  • 37.

    Borregaard N, Cowland JB. Granules of the human neutrophilic polymorphonuclear leukocyte. Blood 1997;89:35033521.

  • 38.

    Gullberg U, Andersson E, Garwicz D, et al. Biosynthesis, processing and sorting of neutrophil proteins: insight into neutrophil granule development. Eur J Haematol 1997;58:137153.

    • Search Google Scholar
    • Export Citation
  • 39.

    Villiers E, Baines S, Law AM, et al. Identification of acute myeloid leukemia in dogs using flow cytometry with myeloperoxidase, MAC387, and a canine neutrophil-specific antibody. Vet Clin Pathol 2006;35:5571.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 40.

    Liso V, Bennett J. Morphological and cytochemical characteristics of leukaemic promyelocytes. Best Pract Res Clin Haematol 2003;16:349355.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 41.

    Falk RJ, Jennette JC. Anti-neutrophil cytoplasmic autoantibodies with specificity for myeloperoxidase in patients with systemic vasculitis and idiopathic necrotizing and crescentic glomerulonephritis. N Engl J Med 1988;318:16511657.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 42.

    Savige J, Davies D, Falk RJ, et al. Antineutrophil cytoplasmic antibodies and associated diseases: a review of the clinical and laboratory features. Kidney Int 2000;57:846862.

    • Crossref
    • Search Google Scholar
    • Export Citation

Appendix

Primers for myeloperoxidase cDNA generation and amplification.

FragmentOrderTA (°C)Position*Sequence (5′ to 3′)
1F60−132TAG AGG ACA TAA AGG CCT GG
R327CGG TGA TGT TGA GGG TCC CTG G
2F64219CTG TCC TAC TTC AAG CAG CCG
R612GCA CGA TGG CGT TGG AGA CG
3F63618AAC GAG ATC GTG CGC TTC CCC
R886GTT GCG GAT GGT GAT GTT GCT CT
4F63831ATC AGG AAC CAG CGT GAC TG
R1,495ACC TTG CTG AGT GGA ACG CG
5F651,422TAC GGC CAC ACC CTC ATC CAA
R1,885GAG TGG GCC CAC GCG GCC TTT
6F601,843ATT GAC ATC TGG ATG GGC GGT
R2,340AAC ACA CAC ATA TGA AAT GCA AT

Nucleotide position based on number from adenine in the initiation codon.

F = Forward sequence. R = Reverse sequence. TA = Annealing temperature.

Contributor Notes

Supported by an equipment grant from Kindy French.

The authors thank Dr. Joe Anderson for assistance with PCR assays, Debra Sherman and Chia-Ping Huang for assistance with electron microscopy, and Dr. George Moore for assistance with statistical analysis.

Address correspondence to Dr. Pressler.