Gomparison of iatrogenic transmissionof Anaplasma marginale in Holstein steers via

needle and needle-free injection techniques

James B. Reinbold, DVM, PhD;Johann I :Jason S. Nickel l , DVM, PhD; Casey M. Riegel,

Coetzee, BVSc, PhD; LarryDVM; Julia A. Christopher,

C. Hol l is, DVM, MAg;MS; Roman R. Ganta, PhD

Objective-To compare iatrogenic transmissron ol Anaplasma marginale during sham vac-c inat ion between needle and needle- f ree in ject ion techniques.Animals-26 Holstein steers confirmed negative for anaplasmosis by use of a competit iveELISA (cELlSA) and an A marginale-specific reverse transcription (RT)-PCR assay.Procedures-An isolate ol A marginale was propagated to a circulating parasitemia of 2.0o/oin a splenectomrzed steer. Sham vaccinatron was performed in the left cervical muscles ofthe splenectomized parasitemic steer wLth a hypodermic needle fitted to a multiple-dosesyringe. The same needle and syringe were used to sham vaccinate a naiVe steer. This2-step procedure was repeated unti l 10 naive steers (group ND) were injected. Similarly,sham vaccinat ion of the r ight cerv ica l muscles of the splenectomized parasi temic steer andanothergroup of 10 naive steers (group NF) was performed by use of a needle-free injectionsystem. Five control steers were not inlected. Disease status was evaluated twice weekly for61 days by use of light microscopy, a cELlSA, and an A marginale-specific RT-PCR assayResults-latrogenic transmission was detected in 6 of .10 steers in group ND. Diseasestatus drd not change in tne NF or control steers. Sensitivity of l ight microscopy, cELlSA,and RTPCR assay was 100% on days 41 ,41 , and20 af ter sham vaccinatron, respect ive ly ;however, only cELISA and RT-PCR assay sustained a sensitivity of 100% thereafter.Conclusions and Clinical Relevance Needle{ree injection was superior to needle in-Jection for the control of iatrogenrc transmission of A marginale. (Am J Vet Res 2010;71.1178 -1188 )

/ f naplasmosrs, caused by Anaplasma marginale, isfa .onc o f thc n ros t p rcva lcn t t i ck - t ransmi t ted r i cke t t -s ia l d rscases o f ca t t le th roug,hout the wur ld . r - r

' l hc OIF

categorizes anaplasmosis as a reportable discasc as a rc-sult of socioccclnomic irnnact and international tradcres t r i c t ions .a I lowever , t l r c impor tance o l anap lasmos isis frequently underestirnated, compared with that ofother diseases, because ol 'seasonal outbreaks and disease s tab i l i t y in ende mic a reas , " C l in ica l s rgns in acu tc -ly in fec ted adu l t ca t t le inc lude, bu t a re no t l i rn i ted to ,anemia, lcver, icterus, and lethargy, and acutc infcct ion

Rcceived. . lu ly 30, 200i) .Acceptcd August .1I . 2009.[ : rom thc l )cpart r r re nrs oI l ) iagnost ic Me dic ine/ t ,athobiology

(Rernbold, Nickel l , Riegel , ( -hr isropher, Ganta) and (- l inr ta l Sci-enccs (Coetzcc). Col lege o[ Vcter inar) Mecl ic ine, and rhc Dc-partmcnt of Anirnal Scrences & Inclustry, , ( .o l lege of Agr ict r l turr :(Hol l is) , Kansers State Unive rs i ry, Manhattan, KS 66502.

Supponed by grants from Intervct lncorporated: USDA CooperativeStatc Rcscarch. Lducat ion, and Extension Service (AES projccr No.481851): and \at ional lnst i tute of i \ l lergy and lnfecr ious Diseases(No. A1070908).

Thc authors thank Drs. David A. Andcrson, Anarol l ,Luskorov, andBrian V Lubbers and Angela Baker, Susan Barnett. l-auren C-alland,Pi lar Ciunter , Kabel Robbins, Gina Scotr , Amanda Sherck, and KaraSmith Ior technica] . assistancc.

Address correspondence lo I)r. Coelzee (lcoctzcc@r,ct.k-state.cdu).

AsgRevrntlows165 subuni t o f rRNAComoet i t ive ELISAConf idence intervalCycle thresholdCoefficient of variationWor ld Organisat ion for Animal Heal thPercentage of parasitized erythrocytesReverse transcription

165 rRNAcELISACICrCVorEPPERT

may result in death." Moreover, abortion, high mortal-itv rates. reduced milk production. extensive treatmentcosts, and we ight loss are key economic considerationsof this disease. In 2003. it was eslimated that the costol'anaplasmosis to the US cattle industry was > $300mi l l ion/y. r

Historically, vaccination has been used to modulatedisease severity. In some countries, naive cattle are in-oculated IV with bovine blood infected with Anaplasmacentrale to reduce rnorbidity attributable to subsequentinfection with A marginale.s However, this strategy is norused or recommended in the United Sutes because of thepotcntial for the transfer of other blood-bome pathogensand production of erythrocytic isoantibodies.T A killed

1 1 l B AJVR, Vol 71, No. 10. October 2010

Anaplasma vaccine was previously available for use in cat-tle in the United Sutes; however, it currently is availablefor use only in special circumstances.d

Cattle that develop anaplasmosis alter natural in-fection and vaccination with l ive Anaplasmc spp remainlifelong carriers despite treatment with tetracl'cl ine. t.q- I2Carrier cattle are responsible for horlzontal, iatrogenic,and vertical transmission of anaplasmosls to naive cattleby providing a reservoir of infective blood for biologi-Cal , t ; t r mechanical , ro r t and in utero in fect ion. ln re

Appropriately timed application o[ insecticides isrecommended for reducing biological rransmission byhematophagous arthropods.r0-r2 Because of the lack ofsuccess with treatment strategies, vaccine availabil itvand problematic vector control, control strategies foranaplasmosis should primarily concentrate on esrab-l ished merhods o[ d isease Drevent ion.

Anaplasmosis can bc- t ransmirred dur ing rout in( 'animal husbandry pracrices.i52r Straregies tt i prevcnLtransmission of anaplasmosis associated with vaccina,tion have been considered impractical or uneconorni-cal or are potentially delererious to the success of thevaccination procedure.j.r+ In a survey, of biosecurit;,protocols in veterinary practices, it was indicatcd thatmost (38/55 169"/"1) vererinarians in bovine-onlv nrac-tices did not routinely change hypodermic needles'aftcrinjection of each cow (ie, used rhe samc needle for theinjection oI multiple cows).

Needle-free injection involves the use of a pneu-matlc-powered systent to deliver vaccines, and nicdlc-free injection techniques are efficacious for rhe deliveryo[ vaccines to cattle.25 27 However, it was suggesrcd inI study2s that there was rhe porential for the iransfer ofb lood products dur ing thc usc of necdlc- f ree in jccrorsfor the adnr in is t rat ion o[ in iect ions t ( ) c( )nsccut ivc ani -mals. The purposes of rhe study re portcd hcre were rcrcompare iatrogenic transmission of A margincrlc duringsimulated vaccinat ion bt ' twct 'n needle - f rec and t . t ,nvcn-tional needle injection techniques and to cvaluate thediagnostic efficicl ' of l ight miiroscopl,, a cELISA. andan A marginale-specific RT-PCR assali

Materials and Methods

Animals--lwenty-six healthy precondirioned Hol-slein steers were purchased from the Kansas Statc Uni-versity Dairy Teaching and Research Center in Manhar-tan, Kan. Steers were (mean r SD) 172.2 + 27.5 daysold at the time of the studlr All steers were screened forantibodies against A marginale hy usc o[ a commercial-ly available cELISA42e,ru'.b and for the detecrion of l65r\NA of A mar.ginale by use of an A marginale-specihcRT-PCR assaylrr Negative disease status was as.sessedthrough interpretation of results of the cEI_lSA and_RT-PCR assay in series.]r The study was approved bythe Kansas State University Institutional Animal Careand Use and Institutional Biosafetl ' Comrnittees.

Randomization, housing, and husbandry-steerswere assigned to I of 3 treatment groups: needle in-jection (group ND Itr = 10 steers]), needle-frce injcc-

l19.nlSroyp NF [0]), and a noninlected conrrol group(5). Briefly, steers were ranked in descendrne ordir onthe basis of body weight, assigned a random number

bv a random-number generator, ' and sort.ed by randomnumber in descending order. Sreers with the 5 largestrandom numbers were included in the noninjected con-trol group, and the remaining 20 steers were assignedto the ND and NF groups on an alternating basis. Steerswere individually housed in a biolevel 2 safety facil i ty.An insecticided was applied to each steer, in accordancervith the manufact.urcr.s reconrmendation, at the timeof entr,v into the biolevel 2 facil i t l l A total mixed rationwas fe d ar )..5oh of body weight (on an as-fed basis); theration for each day was divided into 2 equal portionsand fed twice claily Monensin (80 g of monensin/909 kgof thc tota l mixed rat ion) was the only ant imicrobia linc luded in the rat i t tn . \ \ 'arcr rvas avai lablc ad l ib i tum.When it was necessary to restrain a steer for any pro-cedure, a rope halter was used. Steers did not reciiveant imicrobia ls thar would potent ia l lv in ter fere wi thtransmission of anaplasmosis for rhc 30 days before rhestudy or cluring the sturlv period. Furthermore, proce-dures that werc not incluclcd in the exoerimental de-s ign werc not pcr formed c lur ing rhe study per iod.

Splenectomy protocol and procedure-()ne steerwas splenectomizecl and used to propagate an iso latcol A n'rarginale. A hand assisted, laparoscopic procedurewas usecl for the splenectclmyr l lr icfly, the steer was se-dated b1 ' lM in jcct ion of xy laz ine hydrochlor ide (0.1mg/kg), kctarnine lrydroel.rlerridc r0.l mgy'kg), and bu-torphanol tar t ratc (0.05 m/kg) . The sedated sreer wasrestrainccl in right latcral recumbencyr A 2olo solutiono[ ' l idocaine h1 'c l rochlor idc was in luscd as a local anes-thetic, ancl a 6-crn paracostal incision was rnade ovcrthe anesthctizecl area. Thc incision was centered overthc costochondral arch and extended approximately 3crn caudal to the I3th r ib .

- fhc abdominal cavi ty was

cntercd, and tl.rc surgeon inscrtccl a hand through thisinc is ion and b lunt l l ' d issccted the connect ive t issuc bc-tween thr sp lccn and rumen unt i l the splenic h i lus wasisolated. A 1.5-cm inc is ion wa\ lnade in rhe le f t f lank.and a laparoscopic s tapler , which lunct ioned as a l igating, c l iv id ing, ancl s tapl ing dcvicc, u 'as inserred inro rhcalldornirral cavity

-l 'hc surgeon used thc hancl rhat was

ins ide thc abclorn inal cavi ty to guide thc laparoscopicstapler around the vascular pedic le, and 2 staplcs wereappliccl around thc hilus.

' l-he splee n then was dissected

free from the rumen ancl rentoved from rhr abdomen.AI I inc is ions u 'ere c losed by use of a rout ine 3- layerclosurc. Prtstoperativc analgesia was provided via IVin ject ion of f lun ix in meglumine ( l mg/kg) . I ,en ic i l l in(i procainc (10,000 L;/kg, lM, q 24 h for 3 days.r wasadministered beginning on the day of slrrgerv. Skin su-tures were renrovt'd l4 days after surgery.

Inoculation and monitoring of splenectomizedsteer-A blood sample that contained a tick-transmis-sible \ ' irginia isolate ctf A marginale was collecred froman infccted cow by a rescarchir ar anorher institution,, 'p laced in hepar in anr icoagulant , and shipped on ice roour laboratory via overnight cor-rrier. Thts isolate wasfully characrcrized in 1978 and was originally obtainedby that researcher frorn the USDA Animal parasitologyInst i tu te in Bel tsv i l le , Md r r

l ' ive rnil l i l i ters of the heparinized blood sample wasused to inoculate the splenectomized steer. The steer

AJVF, Vo l 71 , No. 10 , October 2010 1179

was inoculated lV on dav 8 after splenectomy and wasthen monitored daily for clinical signs o[ anaplasmosis,including anorexia, lethargy. and fever. Blood sampleswere collected daily and immediarely used for deteimi-nation of the PPE and PCV

PPE determination-Blood fi lms werc made fromblood samples collected inro evacuated tubes conrain-ing K,-EDTA. An automated uniti was used to stainblood-fi lms with modified Wrichr srain.r{ r ' ' A Millerreticles (which has a large squirc u.trh an addition-al square inset thar is one-renth the srzc o[ the larqesquare) was used to derermine thc number c'rf rrarasir-ized erythrocytes. 16 Only parasitized eryrhroc),r;s \.\ 'crccounted in the large square: how'ever, all eryrhrocytes(parasitized and nonparasitized) u.cre countcd in thesmaller square The numbers o[ parasitized erythro-cytes and nonparasitized e r; ' throcytes were recclrded. Atotal o[ I ,000 erythrocytes were counted. The PPE rvasreported as a percentage and was calculared by use ofthe following equation, which was rnodified irom onequation16 used to measure the percentage o[ reticulo-cytes for s imi lar condir ions:

PPE = (number o[ parasitizcd erythrocl,tes in thc Iarge square/[number of erythrocyres in rhe srna]lrr square X gl) X 100

Measurement of PCV-Thc PCV w,as c{etermincclby partrally f i l l ing capil lary rubesr, w,ith blood sarnplesthat were col lected in to evacualcd tubes conrarnrngK,-EDTA.

' lubes werc centr i luged ar 12.600 X g for l0

minutes.]" The PCV was determined by rneasuiing' rheheight of the Ri lC porr ion and compar ing i t wi th thetotal height lor the samplc.

Experimental procedures-Transmissior.r cxpcri-mcnts were in i t ia ted whcn thc splenectomized stccrachieved a PPE o[ 2.0?,. The srerjr was ntcclicatccl by,IV in ject ion o l f lun ix in meglumine ( l rng/kg) . F i f teenminutes later, a halter was used to restrain the solcnectorn ized parasi t iz td s teer wirh i ts head extendcd andtied to I side to provide access to thc lcft side oi' i tsneck. Ten naive stecrs (group ND) were erach indiviclu-ally and serially restrained in a similar lnanncr adjace ntto the splenectomized parasitized steer. A rnultiple-dosesyringer was primed with a 50-ml. alicluot acquired fronta 1-L bag of sa l ine (0.97o NaCl) so lur ton.r Thc svr inscwas f i t ted wi th a s ingle usc.

. l .7 X 25-mm hypode.mic

needler and ad1usted to deliver a 2,rnL inje ction. A shamvaccination with 2 mL o[ stcri le saline solution was acl-min is tered IM into the cervrcal muscles on rhe lc l i s ideoI the neck of the splenectcimized parasitized steer. Thesame needle and syringe were rhen used within 60 sec-onds to administer a sham vaccination into the muscleson the left side o[ the neck ol a naive steer f rorn groupND.

-fhis 2-step procedure was repeated unril all i0 na-

ive steers in group ND were sham vaccinated via needleinjection.

The head o[ the splenectomrzed parasitized steerthen was moved to the other side to provide acccssto the r ight s ide of i ts ner .k . len naivJ steers (groupNF) were each individually and serially resrrainea in ;similar manner adjacenr ro rhe splenecionrized steer. Aneedle-free injection svstem* was prirned in accordance

with the manufacturer's recommendations with solu-tion frorn the same 1-L bag of steri le saline solution.kThe pneumatic pressure was adjusted to 55,160 Pa forIM delivery of 2 mL of saline solution ro calrle weighing< 227 .3 kg. The tip of the needle-free rnjecrion appara-tus was placed against the skin on the right side of rheneck of the splenectomized parasitized steer. The appa-ratus was agitated in a ctrcular motion to ensure therewas no hair bctween the apparatus tip and the skin.Manual, downward pressure was used to engage andread,v the apparatus [or inlection. and the injection trig-gcr was depressed to release the saline solution. Thisprocess was immediately repeated on the neck ol a steerin group NF lhis mulriple-stage procedure was repeat-cd unt i l a l l l0 naivc steers in group NF were sham vac-cinated by use of the needle-frec injection system.

None o[ the in jccr ion s i tes in er ther group wass*'abbed with disinfectant, and none c'rf rhe needles orthe nceclle-lrce apparatus rip was changed or disinfect-ed ciuring eithe r injection proccdurr. Di[[ercnr personsperformed all injections for thc steers in groups ND andNF Day o l sham vaccinat ion was designated as day 0.

Firre control steers served as sentinel anrmals: fur-thi:rrnore . lhesc steers \.f,e rc not sham vaccinated via ei,ther rnethod (nonin jecrec l conrro l group)

Stcers were rnonirored dail1, for signs oI i l lness for6l clavs. T)isease status was e valuated twice u'cekl1' (days0 , 2 . 6 , 9 . I 3 . 1 6 , 2 0 , 2 3 , 2 7 . 3 0 , 3 4 , 3 7 , 4 1 , 4 1 . 4 8 , 5 1 ,54,57. and 6 l ) . On thosc days, 2 b lood samples werccol lectcd l iom cach steer (a l - rnL samplc inro evacu-atcd tubes col ta in ing K.-EDTA and a 5-ml- sarrp le in toevacuatcd tr-rbes that contained no adclit ive f rc. scrumtubes]) . Samples in rhc serurn rnbes were centr i fugedat 5,000 X g for 5 minurcs at 4oC, and serum was har-vested ancl s torec l at -U0"C unt i l subscquent analvs is .Senrm sarnplcs we re evaluatecl by use of a cEl,lSA.

cELISA-A cttmmercial cllLlS,\r,was used to evalu-ate scruln samples; the cLl,lSA u,as used in accordanceui th the mcthod descr ibed b1 ' rhc Ol [1 and rccot l -mcnclcd b1, thc manufaclurcr.r ' l 'he optical density ol 'each u,c l lwas rneasLtrcd by use of an [ l . lSA p late readcrar a wavelengrh of 620 r r rn. 1 'he perce nragi inh ib i r ior toI each sarnplc was calcularec l by use of rhe Io l lowingequat ion: percenrage inhib i t ion = 100 - (sample opr i -ca l dcnsi ty X 100)/ rnean opr ica l densi ry o[ a negat ivecontrol sanrplc. Samples wirh a percentage inhibit ion< 30% werc recttrded as having ncgative resulls, where-as samplcs with a pe rcentage inhibit ion ) 30olo we re re-cordcd as hal ing posi t ive resul rs . ' 17 r '

RNA extraction and RIPCR assay-hxtractionof RNr\ I ront bkrod samples col lected inro evacuatedtubes containir.rg K, EDTA was pcrforrned b;- use ol'acommerciall l , available product.; extraction w,as per-frlrmed in accordance with the manufacturers recom-mendations. Brie{ly plasma was separatcd b1, centrifug-ing b lood sarnples at 2,750 X g lbr 5 minures at 4oa.Plasma was removed wirh a s ingle-use p ipetre. An a l i -quot (200 pL) of plasma-free blood was transferred ro amicrocentrifuge tubc. One rnil l i l irer of a monophase so-lution" of guanidine thiocyanate and phenol was addedto lvse the RBCs. The rnicrocenrrifuge tubes were vrgor-ously vortexed and then allowed to sit undisturbed for

11B0 A,JVR, Vo l 71 , No. 10 , October 2010

L0 minutes. Chlorolonn (200 trt]-) was added to eachtube, and tubes were vigorously vortexed tor l5 sec-onds. The solution was then allclwed to sit undisturbedfbr 10 minutes, after which it u,as centrifugecl at 12,000X g for I 5 minu tes ar 4oC. l 'he colorless upper aqueousphase of the solution was translerred to a new tr,rbe;500 pL of 2-propanol r,r 'as adcled to each tube, and thecontents were vortexed briefl1,. The solution then wasal lowed to s i t undisturbed lor l0 minutes. af ter whichi t was centr i fuged at 12.000 X g for l0 minutes at 4 'C.Supernatant was discarded, and each pellet was r,l 'asheclwith I mL of 75"/o ethanol; the samples were briefly vor-texecl, after which they w'ere cenlrifuged at 12,000 X glor 5 minutes at 4"C, Supcrnatant was discardcd. anclthe remain ing RNA pcl lc i rn each tube was a l lowccl toair dry. Nuclcase-free lvatcr (50 ;tL) was used to r(:sus-pend cach RNA pellet. Samples were storccl at -80oCuntil analysis by use of an RT-PCR assay.

A real - t imc quant i ta t ive Kl 'PCR assa1" ' was usedfor thc identif ication of 165 rRNA of A mcrrgtnuk hyuse o[ previously dcsigned't forward and reverscprimers a:nd a 1'aq polymerasc probc dcsigned as partof the prcscnt study. Thc lorward ancl rcversc prirrrerscquenccs" were 5 ' -CTC-AC;AA(-( ;AACG(-TC;( ; l 'and 5 ' -CATTTC- ' IAGT(; ( ;C.- IAIC(-C.- l ' , r t -spcct ive l ; , .The A marginalt probe scquence" was 5'-/16'l{l\/l/c G c A G C T T C ; C T G ( _ G ] ' G T A T C ; C ; r / l B r r Q _ I / - 1 ' , .A comrne rcially availahlc, 23-pl- RT-PCR assal' nrixturet'that irrcluded 10 pmol of eacl.r o[ thc l irrward and rc-verse primers, 5 pmol of circh deoxvnr.rcleotidc triphos-phate ,

. l 87.5 nmol of MgSO,, I I j pmol ol A marginult

probe, and 8 U of r ibonucleasc inhib i tor ' r was p laceclin a t t rbe, and 2 gL o[ t rn ip latc was aclc lcc l to achievea final volunre of 25 pl-. Tcnrpcraturc cy'clcs usccl forthe RT-PC-lt assay \^'ere an init ial complerncntarl, DNAgeneratir)n cycle at'1{J"C f or 30 rninutcs. t}re n 3 rninutcsat 94'C, f lolkru,ecl by 45 cl,cles rlf 94'C- for l5 sccur.rds,50oC for 30 seconds. and 60"C lirr 60 seconds.

An in vitro transcript of A murgiuLlt ' plasmicl D\Awas prepared lbr usc in clevclopmcnt ol 'thc RT-PC.R as-say. Plasmid DNA (3 Fg,) was digested with thc rcstric-tion enzyme ,Spt,l. I he .Spel-digersted plasmid DNA wasuse d tc'r generate re combinant transcril. l ts with a T7 RNApolymerase, as descr ibed in the inst ruct ions for the k i t .' ' l 'en- fo ld

ser ia i d i lu t ions ( ranging l rom l b i l l ion to Imolecule) of thc in v i t ro t ranscr ipt were analyzed intriplicate to optimize the assay in a commercially avail-able real - t ime quant i ta t ive l ' }CR systcm. ' Real- t ime [or-mation of the RI-PCR product was monitclred bv rnea-sur ing the emit tec l f luorcsccnce in the cxLension phaseof the PCR cycles wi th in the real - t i rnc PCR system. Areaction was assessed as positive fbr the templale whenit detected 7 fluoresccnt units for the cmission channelof rhe fluorescent probe, The PCR cycle at which fluo-rescence was evident, which is dependent on lhe con-centration ol the template in the reaction, was regardedas the Ct valLle.

Linear regression was used to quantifv the numberof l65 rRNA te mplate molecules in each 25-p L reactionon the basis of the corresponding Ct value by use o[ theequation Y = -3.4324x + 40.38, where y is the reportedCt value and x is the number of template molecules.The R2 for the regression equation was 0,9973. Efficien-

c1' of the RT-PCR assay was calculated to be 95.60lo.30reSamples from a cow known to be a carrier of A margi-nqle a d a naive cow were extracted and analyzed si-multaneously to monitor assay performance and qual-ity o[ the rRNA extraction technique.

Statistical analysis-Data were entered into asoftware package' for subsequent calculations and ma-nioulat ron. ( 'eometr ic mean and CV were calculatedfoi data acquired from recorded diagnostic test results.l) iagnostic test results were also converted to a bina-ry format (0 = negative and I = positive). Sensitivityand spcci f ic i t l 'w i th 95% CIs were calculated for l ightmicroscopy cELlSA. and RT-PCR assay for each sample( l t = 2 t ) a r c a c h t i m c p o i n t .

Agreemcnt between diagnostic results for eachtwicc-weekly sample was assessed by calculating a rstatistic.rt ' Results wcre conrpared by use of a softwareprogram' in a 2 X 2 contingency table to calculate rc viathc fo l lowrng c( luat lons:

l r P = ( l l a + b ) / n l . [ { a + c i l n ] ) * ( [ i c + d ] / n l . [ { b + d } /n l ) and r = ( l { a + d } /n l * EP) . ( l - EP)

where LI ' is the expccted proport ion of equal outcomesaccorcling lo chance ; a is the number of true-posirivercsults, b is thc numbcr of false-positive results, n isthe tota l nunrber of samnles. c is the number of fa lse-ncgat ive resul ts , d is thc numhcr oI t rue-negat ive resul ts ,( [a + d] /n) - LP is thc observed proport ion o[equal out-comes bc1'oncl chance, ancl I - EP is the maximal pro-por t ion o[ 'agreemcnt not at t r ibutable to chance. ThcK sl-atistic mcasLlres the agrecrnenl between tests on ascale f rorn 0 to l . wi th rbetween 0 and < 0.4 indicat-ing poor agrcement , r betwecn 0. ' l and < 0.75 indicat-ing good agrecment, ancl rc > 0.75 indicating excellentagrecrner l t . Whcn r could not be determined becauseol a lack o l .corrcordanr resul ts in > 2 of the 2 X 2 cel ls .an overall proportion of agreement was calculated byclivrding thc sum ol.concordant test results by the num-ber oI sarnnles lestecl.+()

Thc aisociation betr.r,een disease outcome and rc-sul ts lbr the 3 d iagnost ic tcst ing rcgimens was analyzedb)' .r.. of generalizcd l inear mixed models' ' and gen-cralized estimating equations.' Mcthods wcre used toaccolrnt lor thc lack o[ independence among repeatedobservat ions for the same animal over t rme. A semi-par:rrretric survival analysis" was performed to takeinto account the t ime point at which each respect ive d i -agnostic test l irst detected A marginale . A Kaplan-Meiersurvival anall 'sis* (nonparametric) was performed onthe rau' data to depict the amount of t ime elapsed altersham r. 'accination before a positive outcome for each ofthe 3 testing regimens. Values of P < 0.05 were used todetermine signi{icant differences.

Results

Prior to enrollment in the stud;i all steers wereconfirmed to have negative results for anapiasmosiswhen tested by use of the cELISA and an A marginale-specific RT-PCR assay. No adverse reaction attributableto surg,ery was detected in the splenectomized steer.The solenectomized steer was inoculated with 5 mL of

AJVR, Vo i 71 , No 10 , October 2010 1 i B '1

a hepar in ized b lood sample that had a PPE of 8.1% andPCV of 30%; however, an rRNA molecule count wasnot determined for the inoculunr becalrse of the hepa-rin in that blood sample .

Time elapsed from inoculation of the splenecto-mized steer unti l the development of a ttPE equal to 2olowas 34 days. The PPL, PCV and number of l65 rRNAmolecules were 2olo, 23%, and 8.9 X 10' molecules,respectivelx at thal t ime. No adversc reactions wcreobserved as a result of the lnjection rnethods usecl. Inaddition, the authors arc not ar,l 'are of anv indrcationsto suggest the use o[ f lun ix in rneglumine would predis-pose cattle to clotting disorders, and no such con.rpli-cations were detected a[ter adn-rinistration ol' f lunixinmeglun.rine to the splenectornized parasitized steer.The splencctonrized parasitized stcer was cuthanatized45 days after inoculation bccause of anorexia, lethargy,,and pyrexia, however, nonc of the other stee rs wcre re-moved from the studyl

At the end o[ thc stud] : 6 o[ l0 s teers in the NDgroup had positive results for the auaplasnrosis patho-gcn, A marginale , as determined on tl.rc basis o[ rcsultsoI thc cF-LlSA, l ight r-nicroscopy,, ancl R'l '-P(-R assay.

' l-his

represented thc first, second, Iourth. sixth. scvcnth,ancl tenth stccrs that wcre sharl vaccinated via necdleinjection. r\ l l steers in the NF and control groups haclnellative results Ior anaplasnrosis thrclughout rhe study.' l-hus,

the number of steers infected with A ntargirultdiffered signi{icantl1, between thc ND group and the NFar.rcl control groups.

The mean + SD predicted rnoclel-acljustcd proba-bi l i t ics of becorning in{e ctcd, compar ing thc ND groupwi th the NF and contro l groups, rvere 0.60 r 0.16:rnc l0.65 + 0. .18, respect ive l l , . 3cc.r " , te a l l cat t le in the NI 'and control groups had ncgativc results for A nrctrginolc,the rnodel-adjusted probabil itv of a positivr: tesr resultfor A nturginale was zero for each o[' thosc gror"rps. Asteer rn the ND group was significantl l ' rnttrc l ikcly ttrbecome infected wi t t r A marginale (odds rat io , 44.6,95" /o Cl . 19.5 to l0 l .U) than was a srcrr in rhe NF orcontro l groups.

In 25-prl- reactions with tcnrplate frclm the ]0-foldse rial dilutions, tht: standard curve ol the RT-PC-R assayranged l rom 100 to I b i l l ion tnolecules (F igure l ) . ThelLT-PCR assay identr f icd 120 molecules of l65 rRNA(Table l) in I stecr as early as 9 days after sharn vac-c inat ion (F igurc 2) . Sensi t iv i t ; 'o f rhe RT- l )CR assay wasconsistent in identi lying, steers infcctecl with A murgi-nale in successive sarnples.

' l 'hc pe ak sensirivity ( 100%,)

for the RI-PCR assay was at dav 20, and it was sustainedthrough the cnd of the stucl). The peak numbcr ol ' l65rRNA molecules recovered from 250 pL oI plasma,lreeblood was ].6 X l0'i molecules at da1' ,t l .

- l 'here l,,",erc no

positive results for the NF and conrrol groups by use oftl're RT-PCR assayr

The cELISA v ie lded posi t ive resul rs {or I sreer ar l3days after sham vaccination (l; igure 2). however. sensi-tivity of the cEI-ISA u'as incclnsistenr for the 4 succes-sive samples lrom that steer. No infected steer was iden-tif ied by use o[ the cEl-lSA at day I6. A sreer identif iedas infected with A marginale ar day 20 was nol the samesteer identif ied as infected at day 13. The cELISA didnot have consistent sensitivity unti l day 34. Peak sensi-

F gure 1-Sens t rv i ty and rnear i ty wi th RNA concentrat ions foran RT-PCB assay (A) and f luorescent emission irom serial dilutiontemplates (B) Serral lO-fo d dilutions of an In vitro transcript madefron Anaplasma ntargrnale plasmid DNA were used. The meanct values f rom 3 separate experiments were plotted against theog ! number of rRNA molecules (logrc 2 and log,o 9 correspond toTO0 and 1 b lon molecu es of 165 rRNA, respect ive ly) .The equa-t ion of the ine is y = -3 4324x+ 4038 (R? = 09973)

tivitv (100o/o) for: thc cELISA was at day 41, and it wassustained through the remainder of the study. This co-incided witl-r thc peak numbcr of 165 rRNA moleculesidentif ied by use of the RI-PCR assay and a decreasrngPCV (Table l ) .

The PPE determination (use of l ighr microscopy toexamine stained blood fi lms) detected I infected steeras early as i6 days after sham vaccination; however, thissteer did not consistently have positive results unti l day34. Interestingly. the day of peak sensitivity for l ightnricroscopy (day 4l) coincided with the day of peakscnsitivity for the cELISA, the peak in l65 rRNA mole-cules identif ied with the RT-l '}CR assav. and the decreasein P(.V (Table l). Furthermore. although thcy differedsignificantly, the odds of a positive test result for A mar-ginale by use of l ight microscopy were only 0.5 timesas l ikely as the odds lor a positive test result by use ofthe cEl-lSA.

Var iabi l i ty of the per formance of each d iagnos-t ic method and agreement among diagnost ic meth-ods dur ing the peracute, acute, and chronic s tages ofin fect ion were compared (Table 2) . Sensi t iv i ry wasinadequate among al l methods dur ing the peracutephase af ter sham vaccinat ion. Howeverr use of theRT-PCR assay, cELISA, and light microscopy identi-f ied an in fected steer on days 9. 13, and l6 af ter shamvaccination. respectively. Peak sensitivity for theRT-PCR assay, cELISA, and light microscopy was on

o o i A

.. I- - l

. o lo 2 5

E r n -

o 1 5 1I

r n l- l

s l

0 !1

\*-----^

\\ _*__*--*_____*

3 4 5 6 7

Log,o number o f mo lecu les

800

,600

400

.200

.000

800

600

400

200

0

1182 AJVR, Vo l 71 . No. 10 , October 20 '10

Table 1-Geometnc mean and geometric CV forfrom 6 steers in group ND that became inJectedlM intection with a needle.

each diagnostic test result and the PCV derived f rom the analysis of samples obtainedwith a Virginia iso ate of Anaplasma marginale after sham vaccination adm nistered by

Light microscopyl cELlSAt.9 Rr-PcR ll PCV

DaY ilean

l{o. olnonzero

CV valuest Mean cvNo. ol

nonzerovalues Mean cv

ilo. ofnonzerovaluest

No. ofnoIzero

CV valuestMean

R ql 2 c

000

r.2 x 10,1.8 x ltr1.9 X 103

4.2 X 10.4.8 X 1055 .7X l f f3.3 X l0/8.7 X l0?I 4 X ' 108

68.0 521 9.9 552.1 2

I

0

0 02 06 09 01 3 01 6 0 r 0

2 0 02 3 021 030 0,1034 0.4237 0.75

41 0.55u 0.2448 0.1 451 0.10v 0.1051 06 1 0

00000I

00t

r r9 . l 396.3 3

12.81 25.915.04 30.93.39 122,4

10.30 75.613.31 11.014.96 36.5

21.29 70.84 0 1 8 5 3 527.10 243.444.84 54.245 31 170-261.43 i3.1

69 34 43.28 5 5 1 1 8 988.55 r 0.7i9.57 8 479.57 I 1.382.59 9.784 65 12.1

666D

66

66o66D

6

66666

000I

666

6

6h

6

6o

r.6 x 1trl . 4 x 1 0 81.0 x 1088.4 X ' r0 /5.0 x t0i2.9 X 10r1 . 9 x 1 0 i

17 .019.0z t . d20.410.12.1

1 . 81 . 24.8

10.91 5 . 51 5 . 6\ t . 4

34.29 5.4 635.36 9.7 63 l .78 6 .1 63 l ,94 6 .8 631.30 5.3 630.1 1 6.8 6

32.14 4.6 629.98 4.2 630.63 5.3 630.41 8.6 630.44 6.8 62 8 4 1 8 6 6

25.92 14.8 625.60 8.0 627.06 9.9 62 7 3 7 r 0 9 628.53 I 0.9 629 03 0.7 630 26 7.3 6

All values reported represent percentages, except for the RT-PCR assay in which the geometric mean represents the number of 165 rRNAmolecules recovered lrom 250 pL of plasma-free blood.*Values reported representthe PPE. l lndicates the total number oi nonzero values included in the calculat ion ofthe geometric mean andgeometric CV tValues reported represent the percentage inhibit ion. SData were divided by 100 prior to calculat ing the geometric mean andgeomet r icCV;however , theqeomet r icmeanwasmul t ip l iedby l00betore i twasrepor ted . l lVa luesrepor tedrepresent thenumbero t l65rRNAmolecules recovered from 250 prL of plasma-free blood.

Day = g3t after sham vaccination (day of sham vaccination was designated as day 0). - = Not appl icable; the 95% Cl was not calculatedbecause of a lack of concordant results in > 2 cel ls in the 2 X 2 contingency table.

days 20 , 41 , and 4 l , r cspec t i ve l y . D iagnos t i c sens i t i v -i ty was susta ined throughout the rcmainder oI thestudy for the cEl , lSA and RT PCR assay; however,sensitivity ol l ight microscopy decreasecl to 0(/o byday 57. The IiT-PC.R assay was the only diagnos-t ic method that mainta ined a speci f ic i ty o l ' 100' /othroughout the study. Poor speci f ic i ty lor the c l l l - lSAwas caused by ft false-positive results for stecrs inthe contro l and NF groups on days 2 (n = I la lse-p o s i t i v e r e s u l t ) , 1 3 ( 2 ) , f 6 ( l ) , 2 3 ( 3 ) , a n d 3 0 ( l ) .Thc percentage inhib i t ion rccorded for each of rhescfalse-positive results was between 30o/o and 40'lo. I heodds that an A marginale-infected stecr would have aposi t ive resul t for the RT- l 'CR assay was 1.34 t imes asl ike ly as the odds for a posi t ive resul t lor thc cELISA;these odds d id not d i f fer s igni f icant ly (P = 0.07) .

Agreement between results ol the cFt-lSA andRT-PCR assay as well as among results o{ all I diagnos-tic methods was determined. However, a proportion ofconcordance (0.76) was calculated on days 0. 2, and 6alter sham vaccination bccause of a lack rlf concordanrresults in > 2 cells in the 2 X 2 contingency table. Sen-sitivity and specificity for each of these tesl days were0olo and l00o/o, respectively. The r was calculated for theremainder o[ results for days 9 through 6l. There wasperfect agreement on day 4l when comparisons weremade lor 2 and 3 tests; however, excel]ent agreementwas only susta ined through rhe remainder of i 'he srudywhen results were compared for the RIPCR assay andcELISA. At day 61, agreement berween resuhs for theCELISA and RT:PCR assay as well as among results for

Figure 2-Sensrtrvrty tor the RT-PCR assay (white squares),cELISA (whjte diamonds), and l ight mrcroscopy (white tr iangles){or detection of anaplasnrosis ln samples obtaaned f rom 6 steersthat became infected with a Virginra isolate of A rnarEnale aftersham vaccination administered by lM inlectron wtth a needle.Day of sham vaccinatron was designated as day 0.

a l l 3 tes ts was 1 .00 (95" / " C l ,0 .72 to t .28) and 0 .75(95o/o C l ,0 .5 l to 0 .97) , respec t ive ly .

Imprecision of diagnostic test resulls reported for the6 steers infected with A marginale in the ND group wascalculated as the CV of the geon.retric mean (Table l).The geometric CV for the RT:PCR assay at peak sensiriv-ity (20 days after sham vaccination) was ]7%. However.this estimate did not exceed 22olo throughour rhe stndyat any time point after inlected steers were detected. Thegeometric CV for light microscopy, the CELISA, and the

'100

90

BO

to

60

50

40

30

20

1 0

o l

s

.=ooat

AJVR. Vol 71, No 10. October 2010 1 183

Table 2-Sens ttvi ty (Se) and specif ici ty (Sp) for each of 3 dlagnostic methods used to detect anaplasmosis in samples obtained f romsteers during a 61-day period after sham vaccination.

Light microscopy

oay' Se (%) Sp (o/o) Se (o/o) sp (%lcE[1SA RI-PCR assay Agreement

sp (%) 2 tess (rlt All 3 tests (r)*Se (%)

I

l 3

1 6

20

L5

L I

30

34

37

41

48

5t

54

57

61

0 100{ - } ( - )0 100

{ - } t - }16.7 100

(0_46 s) (100_100)0 100

{ - ) ( - }

0 94,7(_l {84.7_100)0 94.7

(-) {84.7-1()()}16.7 94.1

(0-46.5) (84 7-roo)50 94.7

(ro.o-90.0) (84.7-ro())

50 94.7{r0.0-90.01 (84.7-r()())

100 r00(100-r00) (100-100)

83.3 100(53,F100) (100 r00)

33.3 100{0 7r .1) ( lm-r( )o)

s0 r00{r0,0-90) (100-100}

33.3 t00l0-7r.r) (100-100)

0 r00{ - ) { - )0 100

( l { - )

0 r00{ - } ( - }16.7 89 5

(H6.9) (75,7-100)0 94.7

(--) {84 7-100)16.7 100

(0.-46.5) (l(]()-t()o)

83.3 84.2(53.5-100) (67.8-r0o)

66.7 r 00{28.9-100) (100 100)

66.7 94.7{28.F1001 (84 7-r00)

83.3 100(53 5-100) (100*r00)

83.3 100(53 FI00) (100-r00)

100 100(100-r00) (100-100)

100 1001r00,100) (roo-'100)

100 100(r00 100) (100 r00)

'100 100{100,100) (100 r00)

100 r00(loo-1o()) (r00-1o())

r00 100(100 100) ( r00 100)

100 100(100-r00) {rm-r00)

0.12 0.08{H).zs) (H).17}

0.$ 0.31(o.lH.69I (0.i l{.50}

0.47 0.410.224.731 (0.2H.59)

0.68 0.49(0.42-{.94) {0.3H).69}

0.79 0.57(0.s2-1.r ) {0.3H) 80)

0.88 0.62(0.61-r.r6) (0 4(H).84)

0.83 0 64(0.55-1.il) (0.42{.86)

0.94 0.8rrc.67-1.771 (0.58-r.03)

0.94 0.81(0.6i-r .22) (0.58-1.03)

1.00 1.001072-1.281 10.77-1.23l.

r.00 0.9610.12 1.28l. (0.74-1.19)

1.00 0.84(0.72-1.281 (0.62-r.07)

1 00 0.88rc.72-1.28l {0.6Fi.r)'r.00 0.84(0.121.281 (0.62-r.07)

1.00 0 750.72-1.781 {0.s3--0.97)

r.00 0.75(0.72 r.28) (0.5H) 37)

16.7(H6.5)

66.7{28.9-r 00}

83.3(53.5-100)

100{rm-100}

100( 1 oo-1 00)

100( 1 00-r00)

100( 1 00-1 00)

100{r00 r00)

100( l 00-r 00)

t00(100-100)

100(I00-100)

100{I00 r00)

100(100,r00)

100('100-r00)

100(I00 100)

100(r 00-1 00)

100{ 1 00-r001

100( 1 00-1 oo)

100( 1 00-1 oo)

100(I00 r00)

100( 1 00-1 00)

100{ r 00-r00}

100{100 10o}

100( 1 00-1 00)

100( 1 00-100)

100{r00 100)

100{ r 00-1 00}

100{100 r00)

't00( 1 00-100)

100{ 1 00-r 00}

100{ r 00-l 00)

100( 1 00-r 001

Values in parentheses are 95% Cl.rValues for days 0,2, and 6 were omitted because of a lack of concordant results in : ' 2 cel ls in the 2 X 2 contingencytable. For each ofthese

days, the calculated proport ion of concordance was 0.76 (Se = 0% and Sp = 10070). tHepresents agreemenl between results forthe cELISA andBT-PCR assay. lRepresents agreement between results for i lght microscopy, cELISA, and RT-PCR assay.

SeeTable 1 for remainder of kev.

KLPCR assay at day 4I, which corresponded to the daywith the highest agreement among the 3 test,s, was 68o/n,43.2<'k, and l.8o/o, respectively. The lowest estimates wereat days 48,51, and 44 for l ight microscopy, the cELISA,and the KIPCR assay, respectively.

Results lor the semiparametric survival analvsiswere expressed as hazarcl iati<ls. Hazard ratios wcrc in-terpreted sinri larly as odds ratios, were assumed pro-portional ovcr t ime, and were representative of the el-fect of a unil change in the predictor on the lrequencyof the outcome.{{'The hazard ratio for cattle with a posi-tive test result for A marginale in the ND group by useof RT-PCR assay, compared with that for a positive testresult by use o[ the cELISA and light microscopy, was1.15 and 1.58, respectively. Similarly, the hazard raticrfor cattle with a positive test result for A marginale inthe ND group by use of the cELlSA, compared with thatfor a positive test result by use of the RT-PCR assay andlight microscopy, was 0.74 and 1.86, respectiveiy. Con-versely, light microscopy was the least likely to yieldpositive results during the study period; the hazard ra-tio for cattle with a positive test result for A marginaleby use of light microscopy, compared with that for apositive test result by use of the KI'-PCR assay and thecELISA, was 0.74 and 0.81, respectively.

Kaplan Meier survival analysis was performed onraw data to compare the 3 diagnostic regimens (Figure3). At day 0, the risk of infection for all steers exposedwas L However, this risk decreascd clver t ime as A mar-ginrrle-infected steers were identif ied.

Adclitionally, the semiparametric survival analysiswas used to test lor significant differences in the in-jection sequence arnong steers within the ND group.Once the needle was contaminated by injection of thesplcncctomized parasitized steer, the steers that subse-quently had a positive test result for A marginale werethc first, second. fourth, sixth, seventh. and tenth ani-mals in the ND group. l 'he hazard ratio for the sequenceof iniection (0.96) indicated that the risk of becominginfectccl with A marginale (as determined on the basisof a positive test result) was the same for all steers inthe ND group. Therefore, the sequence of injection wasnot significantly associated with a positive test result[or A marginale..

To further validate results lor the diagnostic meth-ods, the l5 naive steers from the NF and control groupswere each inoculated IV with 5 mL of blood obtainedlrom l ol the 6 ND steers iatrogenically infected withthe Virginia isolate of A marginale . All l5 steers becameinlected with A marginale as a result of IV inoculation,

1184 AJVR, Vol 71, No. 10, October 2010

10 15 20 25 30 35 40 45 50 55 60 65Time after injection (d)

quential time points after a single exposure o[ cattle to

A marsinale have been evaiuated.Tiansmission oI anaplasmosis from a known carrier

to susceDtible cattle via needles has been reported'r5 ln

that report. onlv I of 5 cattle became infected' Further-

nrore. the authors .rf that report described visual de-

tection of blood contamination ()n the needle between

iniections. ln the study reported here, 6 oI l0 'steers be-

r:a'rne infected. l l lood ioniamination on the needle was

visibie onlv prior to iniection of the last steer in the ND

group. Othe.ru'ise, the needle appeared to,be safe (ie,

no biood contaminatit ln) for injecting rnultiple cattle'

Bccause of the random pattern of transmrssion and lack

trf sienificancc associated with the in;ection sequence,

the r-cpetit ive usc of a needle among, cattle of unknown

clisease status should be regardcd as unacceptable' Fur-

thermorc, the fact that 6 o[ I0 steers exposed to an

A nrut t inale-{-'onlanlinated needle became infected alter

an lM'injection leads to the hypothesis that this route o[

infectrori may be extremely common in current produc-

tion svstenls.' l iansmission of blood components during needle-

{'ree injection techniques has been reported.:8 This ap-

pu..nri,t was negiigibie for the transmission of anaplas-

inosis for the conditions o[ this study reported here'

Sham-vaccination of the splenectomized parasitzed

steer prior to sham vaccination of each steer in the

Nf group robustly challenged the potential for iatro-

gcni.: rransnrissitrn of a Virginia isolatc of A marginale

i ' ia needle-free injection. However' transmission via

needle-free injection techr.riques is not known for situ-

ations in which anaplasmtlsis carrier cattl€ may havt-a

PPE > 2'lo. Similar ot -u.t. conditions are reasonably

unlikely to be encountered in lteld settings. However, it

rnay bc necessary to accounI for the temporal associa-

tion of previous vaccinations, disease prevalence, and

tlming o[ vaccinaticln in regard to the seasonal distribu-

ti.,n .ri cl inically affecteci cattle when considering the

tuse of needlc-free iniection tcclrniques. Furthermore,

it should be mentit lned that disease resistance among

hrceds has not been vcrif iecl.+' ar

The usc o[ nect l lc- f rec in jet t ion techniques in pro-

duction scttings wil l aid in the reductit-rn o[ biohazard

waste, alleviatJ operaltlr injury resulting from acciden-

tal neeil le l lunctures, and'eliminate the possibil i ty of

needle contamination attributable to vaccination in

consumable meat prodttcts while enabling producers to

nraintain rates for processing of cattle that are neces-

sary for minimizing handling and stress in the animals'

Fven though use oI needle-free injection techniquesis superior to use of needles for preventing iatrogenic

transmission ol A marginalc, it is recommended that

care be used to avoid unwarranted inoculation o[ cat-

tle that might result lrom improper removal of previ-

ouslv used vaccine products from the injection system

because oI poor cleining techniques.r+ Further studies

are necessary to fullv evaluate the use of needle-free in-jection techniques lor the control o[ other blood-borne

ciiseases of cattle.At the end of the study reported here, the preva-

lence of anaplasmosis was 24olo (6/25 steers). lt may be

arsued that ihe evaluation o[ sensitivity and specificity

u*ong the diagnostic methods included in this study

Ficure 3-Kaplan-Merer survival estimate derived trom the re-suits for the RT-PCR assay (black lrne), cELISA idashed line), andliqht rn,croscopy (gray l ine) for detection of anaplasmos s rn sampi.r oOta n"O iro6 6'steers that became infected with a Virgrnral"roatl oi A margtnale aft.er sham vaccination administered by lMini.it.n with a ieedle. Day of sham vaccination was designatedas day O

as cletermined by interpretatiolr o[ results of the cELlSA,

light microscopy, and RT-PCR assay in series. The 4 re-

*"ini"g steers in group N l) were not challenge e-xposed

because"they were enrolled in a subsequent studv'

Discussion

Anaplasmosis poses many problems to the cat t le

inc lust rybecause oI compl icat ions wi th d isease con-

t ro l . er id icat iorr , and t reatment . When cat t le of un-known disease status are being vaccinated, i t is h ighly

recommended i . t t t - tse hygienic techniqucs. Al though

repet i t ive use of needles to in iect cat l le is not recoln-

mencled, th is pract ice st i l l is in use. l t rnay be argued

that a cat t le populat ion wi th a stable prevale ne c oI

endemic disease rnay be advantagetlus as it rcsttlts

in min imiz ine c l in ica l d isease in adul t cat t le . How-

ever , th is appic ' ,ach docs not pe rmi t commingl ing o[

cat t le of u.kno*. d isea-se status. r \n t l ther d isadvan-

taqe would Lre the t ransnl iss ion oI other b lood-bornepulhog. t t t o f cat t le . I f the prevaler lce of anaplastnosis- . r . i l ln* .d to increase, cu l l ing pract ices may be

amnl i f ied and t rade restr ic t ions in tensi l ied between

enciemic and nonendemic countr ies because of the

current use of unre l iab le d iagnost ic methocls.' l-he

study reported here was designed to evaluate

the use of a needle-free iniection system for the controlof anaplasmosis transmission amtlng cattle during vac-

cination. Needle-free injection has been validatecl as a

tool for use in controll ing horizontal transmission ol

A marginule. Additionally, a novel K|-PCR assaY was

developed and evaluated for use in detecting A ntargi-

nale in bovine blood samples. This data set is clinicallyrelevant because of the pbtential spread of A marginale

to naive cat t le dur ing the per formance o l r ( )ut ine ani -

mal husbandry practices ai well as for identify' ing de-Iiciencies in the sensitivity and specificity o[ currentlyavailable diagnostic methods

'fo our knowledge, this is

the {irst repoit in which the use of needle-free injection

techniques for the control o[ iatrogenic transmisstono[ anaplasmosis as well as the performance of f irst-,

second-, and third-generation diagnostic methods at se-

AJVR, Vo l 71 , No. 10 , October 2010 1 1 8 5

should be interpreted with caution because of the smallstudy population and low number of infected steers;however, the accurate and precise diagnosis of ana-plasmosis has historically been problematic because ofdiagnostic methods that lack adequate sensitivlty andspecificity.l0 l i l{-16 Therefore, it is important to describethe inequality ol these diagnostic methods.

First-generation diagnostic methclds rely on thegrowth or visual identif icatron of the organism oI in-terest. These methods have limited sensitivity and lackadequate specificity to differentiate between morpho-logically similar pathogens, normal intracellular struc-tures, and stain artifacts. During the present study, l ightmicroscopy was proven to be unreliable because offalse-negative results. This was attributable to the lownumber of circulating rickettsial organisms encoun-lered. Even though al] stecrs infected with A margi-nale were accurately classil ied at some time cluring thestudy, I ight microscopy had a diagnostic sensrtivity rtfl00o/o only at day 41.

Second-generation methods, which rely on thcidentif ication o[ cel] cornponents, rnetabolic products,and detection of antigenic cornponents, arc currentlythe mclst commonly used techniques [or disease clas-sif ication in clinical medicir.re and research. One dis-advantage o[ these rncthods is the potential for cross-reactivity among coexisting diseases. Becausc olsimilarity among MSP5 surface proteins, cross-reacttvi-ty among A marginale, A c entr ale, and Anuplasma phago -cytophilum lbr the cELISA has been reportcd.]7'1"

-l-he

cELISA has had be t ter and more susta inable scnsi t rv i ty .compared with that of l ight microscr.rpyl l lowo,cr, thecLLISA did not have a sensi t iv i ty of l00o/o unt i l day 4 lfor the ND group. A disadvantage fclr this technique isthe cuto[[ value used to classify disease status. A cr"rto[[value o[ ) J0o/o inhibit ion was used in the nrcsent studvto clxs;;t a stccr as in{cr tcd wilh anapiarrrr,,sis. f his ualrrcled to multiple false-positive results for the NF and controlgroups. However, the scnsitivity of this assay at the car-l ier t ime poinLs would have been compromised if a cutoffvalue ol 407o inhibit ion had becn used.

Third-generatior.r methods use nucleic acid-basrcltcchniques for the classification oI discasc status. Thescmethods offer supcrior sensitrvity and specificrty ovcrfirst- and sccond-generation methods. This study rsnclt the Iirst in which investigators used a nuclcic acid-bascd technique Ior the diagnosis of A marginalc in bo-vine bloocl sarnples.r+2e47

't+ Howevcr, it is the first stucly

rn which a real - t imc ouant i ta t rve Kf-PCR method wasused to identify 165 ;RNA <>l A marginale in inle'ctcdcattle.

' fhe advanlages of this assay are thc enhanced

sensitivity for identifying rRNA targets that are pres-ent in higher quantit ies than a single copy ol DNA perorganisrn, the abil ity to quantify the genetic template.and elirnination o[ the need for PAGE. In acldit ion,this assay could serve as a substitute for the inocula-tion of splenectomized cattle with blood fronr cattle oIunknown disease status. The major disadvantage is thecost oI the reagents and equiprnent as well as the needfor equipment that may not be readily accessible .

The RT-PCR assay had a sensitivity of l00u/o byday 20. This was a noticeable improvement over theuse of i ight microscopy and cELISA for diagnosis in the

prepatent period; however, false-negative assay resultswere attributable to the inabil ity of the RT-PCR assayto detect disease in 250 pL of plasma-free blood ob-tained prior to day 20. Because of the perlormance ofthe diagnostic methods used during this study, the au-thors recommend repeated collection of blood samples{rom cattle of unknown disease status at 3- and 6-weekintervals when the RT-PCR assay and cILlSA, respec-tively, are used. However, l ight microscopy would notbe recommended for dete rminins disease in cattie withunknown disease sta l us.

Anaplasmosis is a cor-nplex and challenging diseasefor stakeholders in the cattle industry, loreign policy,and research arenas. Because of the lack o[ substantralsuccess with treatrncnt strategies and problems withvaccine availabil ity and vector control, anaplasmosiscontrol strategies should primarily conccntrate on cs-tablished methods {br disease Drevention. This data setis clinicall l ' relevant because o[ the potcntial spread ofA marginale to naivc cattle during performance o[. routrneanirnal husbandrv practices as well as the fact that it indi-cates deficiencies in thc sensitivlty and specificity oi cur-rcntly available diagnostic nlethods. Our results identif iednccdle-free injectton as a supcrior method lor controll ingthe iatrogenic transmission clIanaplasmosis. Furthermore,usc crf a novel A margindlespecific RI-PCR assay has thepote ntial to impact the luture of clisease classification pri-or to krcal. interstate. or internatronal movcment 0[ cattlebctwcen ende nric and nonendemic countries.

e. Andcrson DE, Si lv iera l : Survcv ol large animal Vctcr lnanansbrosctur i ty pr : rct iccs (abstr) . rn Pnxct ' r i i r tgs. 4 lst Annu ClonvAm Assor: Bovine I ' ract 20()8,24+.

b. Anaplasma ant iboclv lcst k i t . \ 'MRD Inc. Pul lman, Washc Microso[ t l :xcr l ] .1]07. Vrcrosol l ( -orp. Redmoncl , Wash.c l l , l t r a I l o ss , Sch r r i ng - I ' l ough , S r rmrn r t , N j .c Provrdtd b1' Dr. Kathcr i r rc M. Kocan. f )cpar lm( 'n l oI Vcte r inarv

I 'arhobiology' , ( cnte r Ior \ i ' tcr inarr ' I Ieal th 5cienccs. OklahornaState [ , n ivcrs i ty , , St i ] l rvatcr , ( )k la.

l . l k :ma - I r k , An rcs ( - r> , E l kha r t , l nd .g . M i l l e r r c t i c l c , K l a rn rann Ru l i ngs l nc . L i r chh t l d , \ l l .h. t l t 'n la lo C. lad pla in, Drummond Scicnt i f ic ( -o. Bnromal l , l 'a .i . Cr i tocaps mi(r( )hemir tocr i t capi l lary tubc rcadcr. \ {c(-ormrck

Sc i t n t i l i c . ! t Lou i s . Mo .lc lcal Ins l rumcnts l ' ro-Shot 10-rnl - p isto) gr ip syrrngt , DurvetInc Anrmal Hcal th I ' roducts, ISluc Spr ings, Mo.l l . [ ] raun Mecl ical Inc, l rvrnt : , ( -a l i [ .

Bccton, Dickrnson 6r Co. Frankl in Lakr:s, NJ.I ' u l se 250 nee t l l c - l r ec i n i ec t i on s ) s t c rn , Fc l t on l n t c rna l i ona l .I enexa . Kan .

n. I RI Rt 'agent, Sigma-Aldr ich. St l -ouis, lVlo.o . I n t eg ra ted DNA I cchno log ies l nc . ( - o ra l v i l l e , l ou ' a .p. \upcrScr ipt I I I revtrse t ranscr ip lase. Invi t rogen ("orp, ( -ar lsbad,

(-a i i f

11. I lecombinant RNAsin. Prornega (-orp, Madison, Wisr . .1 7 VE( iAscr ipt h igh-v ie ld t ranscr ipt ion k i t , Ambion lnc. Aus-

t i n . l e x .s. 5martCl , lc t ' r l l , ( -ephr:rd, Sunnyvale, Cal i f .t . Wintpiscope 2.0. Ct- lVE. Ecl inhurgh. Scot land.u. PROC (,hmmrx, SAS, vers ion L l , SAS lnst i tute Inc. Cary, ' r . - C.v I 'R()( - - Genmod. Sr\5, vers ion 9.1, SAS lnst i tute Inc. Car l i N(- .\ \ ' . Stata. vers ion 10. I , Stata Corp t - l l Col lcge Statron,

' lcx.

References

l ) i lenberg G. In(ernat ional col labclrat ive research: s igni f i canceof t ick-borne hemoparasrt ic d iseases to rvor ld animal heal th. VetP(rra.s i to l I 995:57: I9-41.

)

k .l

m

1186 AJVB, Vo 71 , No. 10 , October 2010

DumlerJS. Barbet AII Bekker CP et al. Reorganization o[ generain the families Rickettsiaceae and Anaplasmataceae in the orderRicket ts ia les: uni f rcat ion of sonre specie, of Ehr i ic l i rn wrrh anc-planna, Cowdriawlth Ehrli.hid and Ehrltchin* rrh Neorirhettsic.descriptions of six new species combinations and designatrcn ofEhr l ichia equi arrd ' l lGE agent 'as subject ive synonyms of f f t r -I ichia phagocytophi la. Int J 5vst Evol Microbio l 2001.51:2145-2 1 6 5Kocan KM, de Ia I 'uenteJ, Gughelmone AA, et a l . Anr igens andalternatives for control of Anaplasma marginale infection in car-t le. Cl in Microhio l Rcv 2003;16:698-712.OlE. Chapter 2.4.1: hovine anaplasmosis. ln Manual of stan-d.ards Jor diagnostic tests dnd vaccincs Jor trrestnal anrmcrls.Paris: OIE, 2008. Available al: ll 'w*:oie .int/eng/nornrcs/mmanu-aU Z?t)8/pdf /2.0 4. 0 I . bovine ;rnaplasnrosrs.pd f . Accessed I ul I 3,2009.Rogers RJ, Shiels IA. Epidcrnio logy and contnr l of anaplasmosrsrn Austra l ia. . / S AJr Vrt Assoc 1979;50:363-166.Radost i ts OM, Gay C(- , Blood D(- , et a l . Drscases causcd bv ar ,thropod parasites. ln: yrtsrindr.y rncdicinc: (i ttxLbool? ol the disease.s of cattk, sheep, pigr, goats and horst's. 9th rd. Sr I-ouis: WtsSaundrrs Co, 2000; l 40 I - l 405.Dcnnrs RA, Ol{ara Pf , Young Ml l c t a l . Neonatal immuno-hemolyt ic anenr ia and ic tcrus ol calves. J Anr Ver , ! led AssdcI 970: I 56: I 8b I -1869.

Luther I)Ci. Anaplasmosis vaccinc {rom l-hiverslr] Producr^s LLC..Available at: anaplasmosisvaccirre .com. Accesscd Mav l, 2009.Lincoln SD, Eckblad Wl l Vagonigle R4. Bovine : rnaplasnrosis:c l in ical , hcmatologic. and serologrc mani lestat ions ln cows glv-cn a long-act ing oxytetracycl inc formulat ion in thc prcpalcntper iod. r lm.f l t t Rcs l9f l2;41:1 160-1162.(-( )etzee JIr Aplev MD, Kocan KM. (-o lnpar ison of rhe el f icacy,of enrof loxacrn, rmidocarh, and oxytetracvcl inc lor c learancrof pers istent Anuplusmtt nnrgrnalc infcct ions ln cat t le. \ ' ( ' f I l iar2006:7:147 160Wilson Al , Parker R. Parkcr M. ct a l . ( -hemotherapl o l acurrbovinr analr lasmirs is. Aasl \ I t I 1979,51:7 | 7) .( .oetzec lF Aplry MD, Kocan KM. et a l . Clorupar ison ol lhrceoxytctrac)c l inc regrnres [0r thc t reatmcnt of pt ' rs is t i :nt Anaplas-nrn rnarginalc infect ions in heef caule. \ /et P.rrdsi lo i 2005. I 27:6 I7 t .Futsc J l r . L lct i M\\ / . Knowlcs DPJr, ct a l . Transrnission ol Arra-plasma moginol : b1, IJoophi lus miLroplus: rerenl ion () f vcct() rcornpetcncc i r r thc absence ol vector-pat l rogen iTr t r rar t i ( rnJ (-.Iirt NIir ntbiol 200 I ;4 I : 3U29-1814f r rks IS, Palmer Gl- i , McGurrr TC, ct a l . Dctect ion and quanrr tatron oI Anaplosm u nnrginuk in carr ier cat t le b) ' us ing a nucl t icacrd probe . J (lin Mrcrobrol l9ft9:27:279-2t14.Retves JD I l l , Swif t B[ , Iat rogenic r ransmission ol Anaplusmamargin le in bccI cat t lc . Vet Mtd Smal l Aninr ( - l i r r 1977.72:9l l -9 1 4ZauggJL, Kut t ler K[- . Bovrne anaplasrnosis: in utcro l ransn] I5-s ion and thc inrnrunologic s igni f icancc ol ingested colostra l an-t ibocl les. Am I \1 ' t Rcs I9U4l45:440-441.ZauggJI . Bovine anaplasntosis: l ransplacental t ransrniss ion asi t re lates to stage of gesrar ion. AnlJ \ t , t Rcs 1985:46:570,57 ) .Potgieter Fl ' , van Rensburg L The pcrs istencc of colosrra lAnuplasnn antrbodics and incidence of in u lero t ransmrssionol Anuplasma infections in calves under laboraror,v condirionsOnckrst tpoort J Vet Rcs t987:54:557-560.Norton JH, Parker Rf , Forbes-Faulkner JC. NeonataI anaplasmosis in a cal f . Atst ! 'e tJ 1983;60:348.Peter RJ, Van den Bossche P Penzhorn BL. et a l . T ick. I l1: andmosqui to contro l - lessons f rorn the past , solut ions for rhe fu-ture. Vr l Par usi to l 2005:I32:205-2I LDe Wal l D-I . Anaplasmosis contro l and diagnosis in Sourh Af-r ica. Ann N Y Acad Sci 2000:9 I 6:474-483.Rodr iguez-Vivas RI. Mata-Mendez Y, Perez-Gut ierrez E, e l a l .T he effect of management lactors on the seroprcvalence oI Ana-plasnta marginalc in Bos indicus cat t le in the Mexrcan t ropics.Trop Anim H e akh Prod 2004;16: I 15-1,13.Andrews AH. Lamport A. - { practrcal method of reduc-ing spread of d isease b1' hypodermic needles. \ t , r RecI 985 l l I 6 : 1 85 -1 86 .

Makoschey ll. Beer M. Assessment of the risk of transmission ofvaccine viruses by using insufficiently cleaned injection devices,Vet Rec 20Q4',1 55:563-564.I lo l l is LC. Smi lh JE Johnson BJ. ct a l . A compar ison of sero-logical responses when modified-live infectious bovine rhino-tracheitis r. irus vaccine and Mannheimia hatmolv tica bacterin-toxold arc administered n ' r th needle- f rce versus convent ionalneedle-based in jectron in year l ing feedlot steers. Boyrne Pract2005 r39 : IO6 -109 .lJuang Y. Babiuk t - { , van Drunen l - i t te l -van den l lurk S. lnr ,munization *'ith a bovine herpesvrrus I glycoprotern B D\Avaccrne induces cytotoxic T- iymphocytc responses in mice andcattle . J Gcn I'ilol 2005;ti6:887-t198.Manol S. ( i r iebei PJ, l labiuk I -A, et a l . Modulat ion of immune re-sponscs lo bovine herpesvrrus I in cat t le by immunizatron u ' i tha L)NA vac<:rnc cncoding glycoprotein D as a fusion protein wi thbovine Cll I 54. .lrntn unolog1 2004: I t 2 : 3 2tl-l-]8.Su'eat JM, Abd,v M. Weniger BG. et a l . Safery tesr ing of needlcfrec, jc t in jcctron devices to detect contarninat ion ui th b loodand other t rssur l lurds. ,4nn N Y Accd 5tr 20001916:6Ul-682.Tononi r lc Echaic lc S. Knowles Dl ' , Mc( iu i re T(- , et a l . Detcr(ronof cat t lc nat t r ra l ly in{cctet l wi th Anapl . lsmd marginale in a rcgionof endernic i ty lx , ncsted P(.R and a conlpet i t ivc rnzl ' rne- l inkeclimmunosorbent assa1, using recombinant lnajor surfacr protc in1. J CIvt Mrcrobiol 1998 )6:777 7IL.( .oetzce JE Schrnidt I ' l . Aplcy MD, et a l . ( -ompar ison of thecomplernent f rxal ion tesl ancl c<;nrpet i t ive Lt ISA t i r r serodrag-nosis o l Anaplasnu morginrr l r : r r r lcct ion in expcr imrntal lv ln-lectccl s lerrs. Arn J I 'c t Rcs 2007;6t l :872-t t7f l .RcinboldJB. L- .oetzee Jt ;Sir ig i reddy KR, et a l . Dercct ion oIAla-

ltlosnta margtnalt ancl A. phagotttopltilum in buvrnc pcripheralb lood samples b1, t luplcx real- l i rnt ' r rverse t ranscr iptasc [ 'CR as-sar: J ( - l i r r Microhio l 20I0148:2424-24.12.Dohoo I , Mart in \ [ Strvhn H. Scrt 'cning and dragnosr i i : lcsts. In:I t lo inar t t 'p i l rnr io iogi< rcsct t t r h. Char lot t r - tou n, PI : , ( -anada:AVC Inc , ) t lO7 : I 0 1 -102 .dc la Fucnt t - J . Blouin I : Iq Kocan Klv l . ln fectron exclusion of thcrrckct ts ia l palhogen Arr( .p lasma rnarginalc in thc t rck vcctor I )crnl : ( (n lu r r r r iabi l is . Cl in Diagn Lah Immunol 2003;10: I82-I t l4.Tcrrasaksi lp S, Wiwani tk i t \1 Lckngam l l C.omparat ive study, ofb lood ccl l s ta in ing wi th Wright-g ie l l tsa sta ln, f le ld star in. and anew rrrodi l icd sta in. Lub I l tmatol 2005:1 l :76 7i t .Kutarsh \ . r \utonratrd sta in ing ol bone marrow and ptr iphcral blood by a rnodilicd W'righr',s technique . Am J CIin Pothollq82:77:- l | 9-12( lRi ic i RS, Ben-[ jzra lM, Goel R, ct a l . Rct icul t )cvtes and rct iculo-cv l e cnun le ra t i on . J ( . l i n Lah Anu l 2001 '15 :267 -294 .Str ik Nl . . { l ieman AR. Uarbct Af t et a l . ( .haracrcr izat ion of Ana-plcsnra phago. l tophi lum ma;or surface protc in 5 and the rxtentol rts cross rcactivir), with A nrarginalt'. Chn Vrccint Intmunol2007 :14 :2o2 -26U .Sir ig i rcc ldv KR, ( ianta RR. Mulrrp lcx derecr ion of lhr l ichia anoAnuplonna specics pathogens in penpheral b lood by real- l imrrc\ :ersc t ranscr iptase-polymcrasr chain reacl ion. J N{ol Diagn2005 .7 : J0 f i - l 16 .( i inz ingcr D( i . Genc quant i f icat ion using rcal- t ime quant i tat lvrP(- l l : an ernerging tcchnology hr ts thc mainstream. I : r ; r Hlnratol2002;3t) :503-5 I 2.Le C f. Prolrabilitv ancl probabiliry models. ln: Intrttducton bio-stdt is t rcs. Hoboken, NJ: Wi ley- lnterscrencc, 2001:1 I8- ' l 19.Wi lson AJ, Parker R. l rueman KI: Suscept ib i l i t t o l Bos indicuscrossbred and 8o.s tdur.Lls cattle to Anaplasnn rlargrnale infec-tion. Trdp Anim Health Prod 1980;12:90-94.Bock RE, Kingston lG. De Vos AJ. Ef fect ot breed o[ cat t le oninnate resistance to in lect ion wl th dndpldsrnd mdrgindle t rans-nrr t ted bv Er.r t rphi ius microplus. Au.st VdrJ 1999:17:748-751Jonsson NN. Bock RF. Jorgensen WK. Product iv i ty and heahheffects o l anaplasmosrs and babesiosrs on Bos indicus cat t le andtheir crosscs, and the effects oi differing intcnsity oI rick conrrolin Australra. Vt't Paresitol 200ill I 55: l-9.Gonzalez Eli Long Rll Todorovic RA. Contparisons of the com-plement-fixation, indirect fluorescent. anribody, and card agglu-t inat ion tests for the c l iagnosis oI bovine anaplasmosis. Am J VetRes 1978 .19 : | 518 - i i 41 .

24

z5

26.

) 7

28

29

30

t 0

l l

t )

J )

l l .

) l

) )

l6

3 7

Iri

i l

t2

t 1

t 6

I 5

t 7 .l9

+0

4 l

A L

t 8

1 9

20

2l

22

z )

AJVR, Vol 71, No. '10, October 2010

aa

1 1 8 7

5 l

52,

45

46

47

4u

Bradway DS. Torioni de Echaide S, Knowles Dl er al. Sensrrrv-ity and specificity of rhe complement fixarion test for derectionof cattle persistently infected with Anaplasma marginale. I VttDiagn Invest 200I;I3:79-€LDreher UM, de la Fuente J, Hofmann-Lehmann R. er al. Serolog-ic cross-reactivity between Ana plasma marginale and Anaplasmaphagocytophilum. CIin Diagn Lab Immunol 2005;I2: | 177-t t8l.Carel l i G, Decaro N, Lorusso A, et a l . Detect ion and quanr i f ica-tion o[ Anaplasma marginale DNA in blood samples o[ catrle byreai-time PCR. Vet Microbiol 2007 : 124:t 07 -l 1 4.Decaro N, Carel l i G. Lorusso E, et a l . Duplex real- r ime poly-merase chain reaction for srmultaneous detection and quann,fication o[ Anaplasma marginale and Anr.iplarma trntrali. 1 VttDiagn Invest 200tt120:606-6 I I.F igueroaJ{ Alvarez JA, RamosJA, et a l . Bovine babesiosis andanaplasmosis follow-up on cattlc rclocated in an endemic arrafor hemoparasi t ic d iseases. Ann N YAcadSri l99t l ; t i , t9: l -10.Ge I ' lL , Kocan KM. Ewing SA, et a l . Use o[ a nonradioacrrvc

DNA probe for detection of Anaplasma marginale infection infield cattle: comparison with complemenl fixation serology andmicroscopic examination. J Vet Diagn Invest 1997;9:3943.Ge NL, Kocan KM, Murphy GL, er al. Detecrion of Anaplasmamarginale DNA in bovine erythrocyres by slot-blot and in siruhybridrzation with a PCR-mediated digoxigenin-labeled DNAprobe. J Vet Diagn Inrest 1995,7:465472.Goff WL, Stiller D, Roeder RA. et al. Comnarison of a DNAprobe, complemenr-fixation and rndirect immun<lfluorescencetests for diagnosing Anaplasma marginale in suspected carriercattle. Vct Microbiol I 990:2.1:38 I-390.Hoar BR, Nieto NC, Rhodes DM. er a l . Evaluar ion of seouent ia lcoinlection with Anaplasma phagocytophilum and Anaplasntamarginale in cattle. Am I Yt:t Res 2008;69:l I 7 t-I178.Molad l - , Mazuz ML, Fle iderovi tz t , . et a l . Molecularand serological detect ion of A centra le- and A marginale-infected cat t le grazing wirh in an rndcmic area. Vet Microbki2006 : I I 3 : 55 -62 .

q a

l +49

50.

1 1 8 8 AJVR, Vo l 71 , No. 10 , October 2010