920 AJVR, Vol 61, No. 8, August 2000

Interleukin-1 (IL-1) plays a key role in initiating andmodulating the immune response associated with

inflammatory disorders, particularly sepsis and chron-ic inflammatory disorders.1 The IL-1 family consists of

3 distinct but structurally related proteins: IL-1α, IL-1β, and IL-1 receptor antagonist (IL-1ra). Interleukin-1α and IL-1β elicit physiologic and pathologicresponses,1 whereas IL-1ra acts as a receptor antagonistwith no known agonistic activity.2

The IL-1ra gene encodes a soluble protein synthe-sized by macrophages and monocytes in response tovarious inducers, including tumor necrosis factor-1,IgG, and lipopolysaccharides.3 Two isoforms of IL-1rahave been identified: secreted IL-1ra (sIL-1ra) andintracellular IL-1ra (icIL-1ra). Both isoforms havebeen cloned from human epithelial and blood cells4,5

and rabbit colonic mucosal cells6 and sequenced,whereas only the sIL-1ra isoform from mice,7 rats,8 andhorses9,10 has been characterized. Analyses of thegenomic structure of the human IL-1ra gene revealedthat both IL-1ra isoforms originate from the same geneby alternative splicing of the first exon.5

Previous studies3,11 have examined the potential ofusing recombinant IL-1ra (rIL-1ra) to mediate acuteand chronic inflammation.3,11 Administration of rIL-1raand gene transfer of IL-1ra have been found to reducethe severity of inflammatory responses,6,12 suggestingthat IL-1ra may be useful for treatment of inflammato-ry diseases. Until 1996, only the complementary DNA(cDNA) sequences for IL-1ra in humans, mice, rats,and rabbits were available. Therefore, the present studywas undertaken to determine the IL-1ra sequence forhorses and to compare messenger RNA (mRNA)expression of IL-1ra within and among various breedsof horses.

Materials and Methods Extraction of nucleic acids from horse blood cells—

Blood samples were collected from neonatal foals (1 to 19days old) and adult horses (3 to 7 years old) admitted to theTufts Large Animal Hospital because of various diseases andfrom healthy foals and adult horses at local breeding farms.Blood samples were centrifuged at 2,000 X g at 4 C for 10minutes, and total DNA and RNA were extracted from thebuffy coat by use of a guanidine isothiocyanate procedure.13

Determination of a portion of the genomic IL-1ra DNAsequence—A 220 base pair (bp) segment of the genomicequine IL-1ra DNA sequence was amplified by means of apolymerase chain reaction (PCR) procedure using DNAfrom an adult female Thoroughbred with colic. Degenerateprimers (sense, 5'-CAGGWZGTTAACATCACTGAYC-3';antisense, 5'-CTAYTXGTCYTCCTGGAAGTAG-3') based onthe conserved region of exons (amino acid positions 82through 152) in IL-1ra from humans, mice, and rats6 wereused. The amplified DNA was cloned,a and DNA from 3clones was sequencedb using the M13 forward and M13reverse primers derived from the vector sequence.

Received Sept 24, 1998.Accepted Sept 14, 1999.From the Departments of Environmental and Population Health

(Dhar, Thompson, Alcivar-Warren) and Biological Sciences(Paradis), Tufts University School of Veterinary Medicine,North Grafton, MA 01536. Dr. Dhar’s present address is SuperShrimp Group Inc, 1545 Tidelands Ave, Ste J, National City, CA91950.

Supported by an NIH short-term training grant (T35DK07635, toMST), the Dorothy Russell Havemeyer Foundation (MRP), theMassachusetts Horse Racing Fund (AA-W), and the TuftsUniversity School of Veterinary Medicine Program (AA-W).

The authors thank Dr. Lori Tumpowsky for technical assistance.Address correspondence to Dr. Alcivar-Warren.

Determination of the cDNA sequence and mRNA expression of interleukin-1 receptor antagonist in horses

Arun K. Dhar, PhD; Margret S. Thompson, DVM; Mary Rose Paradis, DVM; Acacia Alcivar-Warren, DMV, PhD

Objective—To determine the complementary DNA(cDNA) sequence of interleukin-1 receptor antagonist(IL-1ra) in horses and compare messenger RNA(mRNA) expression of IL-1ra among horses of variousbreeds.Sample Population—Blood samples from neonataland adult horses examined for a variety of diseases.Procedure—A polymerase chain reaction procedurewas used to amplify a 220 base pair (bp) portion ofthe genomic DNA. The upstream and downstreamregions of the cDNA sequence were determined bymeans of 5' and 3' rapid amplification of cDNA ends(RACE) procedures. Northern blot hybridization wasused to examine steady-state mRNA expression of IL-1ra.Results—The consensus sequence of the cDNAobtained with the 5'-RACE procedure and thesequence for the 220 bp portion of the genomic DNArepresented the putative sequence for secreted IL-1ra. The predicted secreted IL-1ra amino acidsequence contained 176 residues with an in-framestop codon; the N-terminal 25 amino acid residuesresembled the signal peptide reported for humansecreted IL-1ra. An approximately 1.3 kilobase pair (kb)band that represented a portion of the 3' end of thecoding region and the 3' untranslated region wasobtained by use of the 3' -RACE procedure. Northernblot hybridization detected a 1.6 kb transcript in bloodRNA from adult Arabian, Belgian, Thoroughbred, andStandardbred horses.Conclusions—Results suggest that the DNA forequine secreted IL-1ra has a short (29 bp) 5' untrans-lated region, a 534 bp coding region, and a long(approximately 1,080 bp) untranslated region. (Am JVet Res 2000;61:920–924)

920-924.qxd 10/21/2005 9:52 AM Page 920

AJVR, Vol 61, No. 8, August 2000 921

Determination of the IL-1ra cDNA sequence—Todetermine the cDNA sequence of the portion of the IL-1ragene upstream of the 220 bp genomic DNA sequence, RNAfrom an adult female Thoroughbred with colic and from afemale Thoroughbred foal with sepsis was used in a 5' rapidamplification of cDNA ends (5'-RACE) procedure. For thisprocedure, cDNA was synthesized from the horse’s and foal’sRNA using a primer designed from the 220 bp genomic DNAsequence (5'-CTGTTTGAGCGGATGAAGGT-3'). Homo-polymeric tailing of the cDNA was done using dCTP, and thecDNA was amplified by use of a PCR procedure using anantisense primer (5'-CAUCAUCAUCAUGTTCTCCTC-CTTGTTCTTGCT-3') synthesized from the 220 bp genomicDNA sequence that was 9 bp upstream of the primer used forcDNA synthesis and a universal anchor primerc as the senseprimer. Amplified cDNA in bands that corresponded toapproximately 300 bp and to approximately 400 bp were gel-purified and cloned.d Three clones were sequenced, using theSP6 and T7 primers derived from the vector.d Sequenceanalyses were done using the GCC program.e

To determine the cDNA sequence of the portion of theIL-1ra gene downstream of the 220 bp genomic DNAsequence, a 3'-RACE procedure was employed, using cDNAsynthesized with an oligo d(T)-adapter primer.f Followingreverse transcription, a PCR procedure was performed, usingthe same sense primer used to amplify the 220 bp portion ofthe genomic DNA and an abridged universal anchor primerf

as the antisense primer. The amplified cDNA was cloned.a

Evaluation of mRNA expression of IL-1ra—Northernblot hybridization13 was used to examine steady-state mRNAexpression of IL-1ra in blood samples from unrelated adulthorses. Aliquots of total RNA (0.5 to 3.0 µg) from adultThoroughbred (n = 8), Standardbred (8), Arabian (10), andBelgian (6) horses were electrophoresed in 1% agarose gels,blotted onto nitrocellulose membranes, and hybridizedovernight at 42 C in a formamide-base hybridization solu-tiong containing a 32P-labeled 220 bp IL-1ra DNA probe (2 X108 cpm/µg of probe DNA). Membranes were washed at 50 C,using the medium-stringency washing protocol,g andexposed to x-ray filmh at –70 C for 5 to 15 days.

ResultsThe sequence of a 220 bp portion of the genomic

DNA of IL-1ra in horses was determined (Fig 1;GenBank accession number AF072476). This portionof the genomic DNA represented the 220 bp upstreamof the stop codon for exon 4 of human IL-1ra. Thesequence of this portion of the genomic DNA was 78 to84% homologous with the corresponding sequencesfor human, mouse, rat, and rabbit IL-1ra.

The 5'-RACE procedures resulted in a faint smearyband of approximately 400 bp and an intense smearyband of approximately 300 bp (Fig 2). The cDNAsequence of the approximately 400 bp band overlappedwith the first 43 bases of the 220 bp portion of thegenomic DNA sequence (Fig 1), suggesting that thisband represented the cDNA upstream of the 220 bpportion of the genomic DNA sequence. The consensussequence of the cDNA obtained with the 5'-RACE pro-cedure (GenBank accession number AF088186) andthe sequence for the 220 bp portion of the genomicDNA represented the putative sequence for equine sIL-1ra. The predicted equine sIL-1ra amino acid sequencecontained 176 residues with an in-frame stop codon(TAA). All 3 clones that were sequenced (1 from a foalwith sepsis and 2 from an adult horse with colic) had

the same in-frame stop codon, suggesting it was not asequence artifact. The autoradiogram of the region sur-

Figure 1—The nucleotide and predicted amino acid (indicated bysingle letter codes) sequences of interleukin-1 receptor antago-nist (IL-1ra) in horses. The asterisk indicates an in-frame stopcodon; the putative N-glycosylation sites are marked in bold let-ters. � = Conserved cysteine residues. The solid line indicatesthe sequence of a 220 base pair (bp) portion of the genomicDNA. The dashed line represents the overlap between thesequence of this portion of the genomic DNA and the sequenceof the cDNA obtained by use of a 5’ rapid amplification of cDNAends (RACE) procedure. The arrow indicates the partialsequence of an approximately 300 bp band obtained with the 5’-RACE procedure.

Figure 2—Agarose gel electrophoretogram of the cDNA of IL-1raamplified by use of a 5’-RACE procedure. Lane 1 = Resultsobtained using total RNA from an adult horse with colic. Lane 2= Results obtained using total RNA from a foal with sepsis.Lane M = 100 bp Ladder.

920-924.qxd 10/21/2005 9:53 AM Page 921

922 AJVR, Vol 61, No. 8, August 2000

rounding the in-frame stop codon (Fig 3) did not indi-cate any sequence compression attributable to sec-ondary structure or GC content in this region of the IL-1ra sequence. This suggests that the in-frame stopcodon was not a result of sequencing artifact, but areflection of the cDNA sequence. Only a partialsequence (198 bp) was obtained from the approxi-mately 300 bp band obtained with the 5'-RACE proce-dure, and this sequence completely overlapped thesequence obtained for the approximately 400 bp band.

The cDNA and putative amino acid sequencesobtained for sIL-1ra were approximately 99% homolo-gous with previously published9,10 equine sIL-1racDNA sequences and 80 to 82% homologous withsequences reported for humans, mice, rats, and rabbits.The cDNA sequence obtained in the present study dif-fered from the sequence reported by Kato et al9 at posi-tions 104 (T vs C), 225 (T vs A), and 546 (T vs C). Thedifference at position 104 did not alter the putativeamino acid sequence, but the other 2 differences intro-duced a change from phenylalanine to leucine at posi-tion 546 and resulted in a change from lysine to an in-frame stop codon at position 225. The cDNA sequence

obtained in the present study differed from thesequence reported by Howard et al10 at positions 84 (Tvs C), 105 (T vs C), 225 (T vs A), 336 (A vs T), 337 (Avs T), and 546 (T vs C). The differences at positions105 and 337 did not result in alterations in the putativeamino acid sequence; however, the other 4 differencesresulted in a change from leucine to phenylalanine atposition 84, lysine to the in-frame stop codon at posi-tion 225, phenylalanine to asparagine at position 336,and leucine to phenylalanine at position 546.

An approximately 1.3 kilobase pair (kb) band wasobtained by use of the 3'-RACE procedure (Fig 4;GenBank accession number AF072535). To determinethe specificity of the 3'-RACE procedure, 2 internalcontrols were used for the PCR procedure. The firstinternal control consisted of amplification of a 220 bpfragment of cDNA from aliquots of the cDNA synthe-sized with an oligo d(T)-adapter primer and used forthe 3'-RACE procedure; amplification yielded a 220 bpband. The second internal control consisted of amplifi-cation of a 220 bp fragment of cDNA from the gel-puri-fied product of the 3'-RACE procedure. This confirmedthat the approximately 1.3 kb band obtained with the3'-RACE procedure represented a portion of the 3' endof the coding region and the 3'-untranslated region ofequine IL-1ra.

Northern blot hybridization revealed a single tran-script of approximately 1.6 kb in RNA from adultArabian, Belgian, Thoroughbred, and Standardbredhorses (Fig 5). Subjective assessment of hybridization

Figure 3—Autoradiogram of a portion of the cDNA sequence ofIL-1ra in horses, illustrating an in-frame stop codon (TAA).

Figure 4—Agarose gel electrophoretogram of the cDNA of IL-1raamplified by use of a 3’-RACE procedure. Lanes 1 and 2 =Internal control that consisted of amplification of a 220 bp frag-ment of cDNA from aliquots of the cDNA synthesized with anoligo d(T)-adapter primer and used for the 3’-RACE procedure.Lanes 3 and 4 = Internal control that consisted of amplificationof a 220 bp fragment of cDNA from the gel-purified product ofthe 3’-RACE procedure. Lanes 5 through 8 = Amplification of IL-1ra cDNA by use of the 3’-RACE procedure. Lane M = 100 bpLadder. Lanes 1, 3, 5, and 6 represented samples from an adulthorse with colic; lanes 2, 4, 7, and 8 represented samples froma foal with sepsis.

920-924.qxd 10/21/2005 9:53 AM Page 922

AJVR, Vol 61, No. 8, August 2000 923

signal intensity indicated that potential differences insteady-state mRNA expression of IL-1ra may existamong breeds and among individuals within a breed.However, to accurately determine differences in IL-1ramRNA expression levels, northern blots should behybridized with a constitutively expressed gene probe,such as 18s ribosomal RNA, as an internal standard sothat the corresponding signals of IL-1ra mRNA expres-sion can be compared.

DiscussionIn the present study, we determined the cDNA

sequence and putative amino acid sequence of sIL-1rain horses. Two reports9,10 on the cDNA sequence of IL-1ra in horses have been published since the presentstudy was initiated.i However, the present study identi-fied some unique features that were hitherto unknownfor the cDNA sequence of IL-1ra from any species,including horses. The N-terminal 25 amino acidresidues of the putative amino acid sequence obtainedin the present study resembled the signal peptidereported for human sIL-1ra,4 and the putative aminoacid sequence had the N-linked glycosylation site andthe 5 conserved cysteine residues (at amino acid posi-tions 25, 91, 94, 141, and 147) found in the previous-ly reported human, mouse, rabbit, and horsesequences. In humans and rabbits, the precursor pro-tein undergoes a proteolytic cleavage between cysteineand arginine (amino acids 25 and 26 of the human sIL-1ra sequence), resulting in a mature secreted IL-1raprotein of 152 amino acids. In the present study,although a cysteine residue was detected at amino acidposition 25, a histidine residue, instead of an arginineresidue, was detected at amino acid position 26, simi-lar to results reported by Kato et al9 and Howard et al.10

Purification and characterization of the equine sIL-1raprotein is needed to determine whether the precursorprotein undergoes proteolytic cleavage to produce amature protein.

Detection of an in-frame stop codon in the equinesIL-1ra open reading frame in the present study wasunexpected and unique. Mutations introducing prema-ture stop codons reduce the amount of functionalmRNA.14 Although the biological importance of this in-frame stop codon in the cDNA sequence of equine sIL-1ra is not known, we speculate that the IL-1ra mRNAin the diseased animals from which samples wereobtained for the present study may have undergoneposttranscriptional modification to introduce such an

in-frame stop codon. Determining the genomicsequence of the IL-1ra gene encompassing the stopcodon may help to elucidate whether the in-frame stopcodon arose as a result of posttranscriptional modifica-tion. Such modification may result in production of atruncated biologically inactive IL-1ra protein in dis-eased animals, which may result in disruption of thehomeostatic balance among IL-1α, IL-1β, and IL-1ra.

Use of 2 positive controls for the PCR procedure tocapture the 3'-end of IL-1ra provided a way to increasethe specificity of our 3'-RACE procedure. An alterna-tive method of increasing the specificity of the 3'-RACEprocedure involving changing concentrations of gener-al and transcript-specific primers used in the PCR pro-cedure has also been reported.15

Combining results for our 5'-RACE procedure,sequencing of a portion of the genomic DNA, and our3'-RACE procedure suggested that the DNA for equinesIL-1ra has a short (29 bp) 5' untranslated region, a534 bp coding region, and a long (approximately 1,080bp) untranslated region similar to other members ofthe IL-1 family of genes.1 Sequences of the long 3'-untranslated regions of the IL-1 family of genes weremore homologous than sequences of the codingregions,8 suggesting that the 3'-untranslated regionprobably exerts a similar regulatory mechanism for thisfamily of genes.

Future studies should be aimed at determiningwhether IL-1ra mRNA expression varies depending onphysiologic or health status of the animals, using aquantitative method of measuring mRNA expression.

apCR Script vector, Stratagene, La Jolla, Calif.bfmol DNA sequencing kit, Promega, Madison, Wis.cUniversal anchor primer, Gibco BRL, Rockville, Md.dpAMP1 vector, following the UDG cloning method, Gibco BRL,

Rockville, Md.eGCG program version 9.0, Genetics Computing Group Inc,

Madison, Wisc.fGibco BRL, Rockville, Md.g5Prime-3Prime Inc, Boulder, Colo.hBiomax, Kodak, EM Parker Co, Wilmington, Mass.iThompson M, Dhar AK, Paradis MR, et al. Molecular cloning and

mRNA expression of equine interleukin-1 receptor antagonist(abstr #B036). Anim Genet 1996;(Suppl 2)51.

References1. Dinarello EA. Interleukin-1 and interleukin-1 antagonism.

Blood 1991;77:1627–1652.2. Lennard AC. Interleukin-1 receptor antagonist. Crit Rev

Immunol 1995;15:77–105.

Figure 5—Composite figure of northern blot autoradiographs illustrating expression of IL-1ra mRNA in 8 Thoroughbred, 8 Standardbred,10 Arabian, and 6 Belgian horses. Aliquots of RNA were electrophoresed in 1% agarose gels, blotted onto nitrocellulose membranes,and hybridized with a 220 bp IL-1ra DNA probe. T = Control mouse testis RNA. The corresponding photographs of ethidium-bromidestained gels indicate amount of total RNA loaded per lane.

920-924.qxd 10/21/2005 9:53 AM Page 923

924 AJVR, Vol 61, No. 8, August 2000

3. Paris MM, Friedland IR, Ehrett S, et al. Effect of interleukin1 receptor antagonist and soluble tumor necrosis factor receptor inanimal models of infection. J Infect Dis 1995;171:161–169.

4. Eisenberg SP, Evans RJ, Arend WP, et al. Primary structureand functional expression from complementary DNA of a humaninterleukin-1 receptor antagonist. Nature 1990;343:341–346.

5. Haskill S, Martin G, Van Le L, et al. cDNA cloning of an intra-cellular form of the human interleukin 1 receptor antagonist associat-ed with epithelium. Proc Natl Acad Sci U S A 1991;88:3681–3685.

6. Cominelli F, Bortolami M, Pizarro TT, et al. Rabbit inter-leukin 1 receptor antagonist: cloning, expression, functional charac-terization and regulation during intestinal inflammation. J Biol Chem1994;269:6962–6971.

7. Zahedi K, Seldin MF, Rits M, et al. Mouse IL-1 receptor antag-onist protein. Molecular characterization, gene mapping, and expres-sion of mRNA in vitro and in vivo. J Immunol 1991;146:4228–4233.

8. Eisenberg SP, Brewer MT, Verderber E, et al. Interleukin 1receptor antagonist is a member of the interleukin 1 gene family:evolution of a cytokine control mechanism Proc Natl Acad Sci U S A1991;88:5232–5236.

9. Kato H, Ohashi T, Matsushiro H, et al. Molecular cloningand functional expression of equine interleukin-1 receptor antago-nist. Vet Immunol Immunopathol 1997;56:221–231.

10. Howard RD, McIlwraith CW, Trotter GW, et al. Cloning ofequine interleukin 1 receptor antagonist and determination of itsfull-length cDNA sequence. Am J Vet Res 1998;59:712–716.

11. Ohlsson K, Bjork P, Bergenfeldt M, et al. Interleukin-1receptor antagonist reduces mortality from endotoxin shock. Nature1990;348:550–552.

12. Makarov SS, Olsen JC, Johnston WN, et al. Suppression ofexperimental arthritis by gene transfer of interleukin 1 receptorantagonist cDNA. Proc Natl Acad Sci U S A 1996;93:402–406.

13. Alcivar AA, Hake LE, Millete CF, et al. Mitochondrial geneexpression of male germ cells of mouse. Dev Biol 1989;35:263–271.

14. Aoufouchi S, Yalamos J, Milstein, C. Nonsense mutationsinhibit RNA splicing in a cell-free system: recognition of mutantcodon is independent of protein synthesis. Cell 1996;85:415–422.

15. Bespalova IN, Adkins S, Burmeister M. 3' RACE: skewedratio of specific to general PCR primers improves yield and specifici-ty. Biotechniques 1998;24:575–577.

920-924.qxd 10/21/2005 9:53 AM Page 924