Comparative study of antipneumocystis agents in rats by using a Pneumocystis carinii specific DNA probe to quantitate infection

0095-1137/92/112968-07$02.00/0

CopyrightC1992,American SocietyforMicrobiology

Comparative Study

of

Antipneumocystis

Agents

in

Rats

by

Using

a

Pneumocystis

carinii-Specific DNA

Probe

To

Quantitate

Infection

P. A.

LIBERATOR,`*

J. W.ANDERSON,1 M. POWLES,2L.A. PITTARELLI,2M. WORLEY,3 M. BECKER-HAPAK,3 D. C.GRAVES,3ANDD. M. SCHMATZ2

Departments of Genetics andMolecularBiology' and BiochemicalParasitology,2MerckResearch

Laboratories, Rahway, New Jersey 07065, and Department ofMicrobiologyandImmunology,3

Universityof Oklahoma HealthSciencesCenter, OklahomaCity, Oklahoma 73190

Received 20 December1991/Accepted27August 1992

Arepetitive genomicDNA clone (B12-2) thatspecificallyhybridizestoPneumocystis cariniiDNAhasbeen

identified. No cross-hybridizationtogenomicDNApreparedfrombacteria,otherfungi, protozoa,ormammals wasobserved. Clone B12-2 ismultiplyrepresented inthe P. cariniigenome. BydirecthybridizationtoDNA

prepared from thelungsofimmunosuppressedrats, theprobecandetect theequivalentoffewer than1,000P.

carinii organisms.Ahybridizationassayemployingclone B12-2has beendevelopedtoquantitate organismload in the ratmodel forP. carinii. Application of theassay totrackthe accumulation oforganisms duringthe

immunosuppression regimen as well as tomonitor the efficacy of twodrug therapies used clinicallyforthe treatmentofP.carinii pneumoniaisdescribed here. The cloneB12-2hybridizationassayfor thedetermination ofP. carinii organism load possesses several advantageous features and thus should serve to complement

conventionalstaining andimmunohistochemical methods.

Pneumocystis carinii pneumonia

(PCP)

is the most

com-mon opportunisticinfection in AIDSpatients intheUnited States (17). Current treatments forthis disease havemajor

limitations because ofahighincidence of adverse reactions. Asaresult,there isaneed for new, safer

therapeutic

agents. Evaluation ofpotential

antipneumocystis

agentsis hindered

bythe lack ofdependablemethods of

culturing

P. carinii in

vitro, although somelimited progress has been made in this

area(4, 7).Asaresult,mostlaboratories evaluate potential drugsin animmunosuppressedratmodel for PCP similarto

that originally described by Frenkel et al.

(12).

Rats from

colonies known tohave latent P. carinii infections or rats

inoculated with P. carinii-infected lung tissue(3)aretreated with immunosuppressive drugs for 6to 8 weeks to induce

acute PCP, after which they are treated with candidate compounds with continued immunosuppression. At the

completion oftherapy, the extentof disease is determined eitherbyexamination ofhistological sections of thelungor

by microscopic examination of stained slides of processed lung tissue orimpression smears.

There are two knowndevelopmental stages of P. carinii. The cyst form is 5 to 8 p,m in diameter and contains up to

eight intracystic bodies, sometimes referred to as sporozo-ites. The trophozoite form of the life cycle varies markedly in size, ranging from 2 to 5 R,m. Usually the degree of P.

carinii infection isdetermined by examining slides of histo-logical sections or stained cysts, and a qualitative score from 0to5+ isassigned (2, 16, 31). Only rarely is the actual cyst load determined. The number of trophozoite forms is deter-minedevenlessfrequently because of both their small size andthe fact that they commonly aggregate in large clumps. Evaluation of the trophozoite load in rat lung tissue is

important, since this stage accounts for a substantial com-ponent of the P.

carinii

infection. Organism load

assign-* Correspondingauthor.

mentsbased merely onmicroscopic cyst counts are poten-tially tremendously misleading. For example, novel antipneumocystis therapies which selectivelytarget tropho-zoites might go unnoticed if only cyst load is monitored. Quantitation of both cyst and trophozoite forms may also lead to a clearerunderstanding of how these stages relate to each otherduring the course ofaP. carinii infection.

There are numerous reports in the literature of pathogen-specific DNAprobesbeing usedtoquantitateorganism load in various tissues (1, 29). In the present paper,a repetitive genomic DNA clone(33) is demonstrated to bespecific in its ability to hybridize to P. carinii DNA. An assay with the probe has been developed to quantitate the progression of the P. carinii infection in theimmunosuppressed rat model. To furthertest theutilityofthisprobe, immunosuppressed rats treated with trimethoprim-sulfamethoxazole (TMP-SMZ) or pentamidine to clear the P. cannii infection were assayedvisuallyandby DNAhybridization. Theresults of these studies indicate that the antipneumocystis activityof drug treatments can be effectively monitored inthis model with a DNA probe.

MATERIALS AND METHODS

Animal model andevaluation ofcompounds. A rat model similartothatoriginallydescribed by Frenkel et al. (12) and

Hughesetal. (15)wasused. Briefly, male Sprague-Dawley rats weighing approximately 250 g each, obtained from

Sasco, Inc. (Omaha, Nebr.), were maintained on a

low-protein diet (8%) and immunosuppressed with

dexametha-sone (Phoenix Pharmaceuticals, St. Joseph, Mo.) in the

drinkingwater (2 mg/liter) to induce P. carinii pneumonia.

Tetracycline (1 g/liter; Vedco, Inc., St. Joseph, Mo.) was alsoaddedto the drinking water to prevent bacterial infec-tions. Inthetime course study to monitor the onset of the P.

carinii infection, six rats were sacrificed each week for 9 weeks.

2968

on April 12, 2020 by guest

http://jcm.asm.org/

Generally, after6 weeks ofsteroid therapy, therats had a

significant organism load. In the drug evaluation studies, several animals were sacrificed at this point to confirm the presence of P. cannii. Theremaininganimals were divided into six to eight animals per treatmentgroup. Steroid pres-sure was continued during the treatment period, which

continuedfor as long as 21days. A long-term therapy study was conducted tocompare the efficacies of the two major treatmentsfor PCP in humans. Four treatmentperiods (4,7, 14, and 21 days) were compared to determine the relative ratesofcystclearanceand theoverallreductionof P. carinji organisms. TMP-SMZ (Interchem Corporation, Paramus, N.J.) wasadministered continuouslyinthe drinking water at a dose equivalent to at least 2.5 times the recommended

human dose (50 mg of TMP and 250 mg of SMZ per kg of bodyweight).Pentamidine(SigmaChemical Company) was

administered intravenously daily at a dose of 10 mg/kg (6 mg/ml in water). This was equivalent to 2.5 times the

recommended dose for humans. Parallel groups of control

animalsweresimilarly immunosuppressed, treated with the

appropriatevehicle, andsacrificedateach timepoint. At the

completionof the respective treatments, all rats were

sacri-ficedbyexposure to carbon dioxide gas, and the lungs were removedand homogenized with a Brinkmann homogenizer (Brinkmann Instruments,Westbury, N.Y.) in 10 ml of saline and processed for quantitation as described below and in Schmatz et al. (22). In some experiments, tissue homoge-nates fromheart, intestine, and kidney were similarly

pre-pared.

Visualevaluationofinfected lungtissues.The lung homoge-nateswere washedonce with 10 ml ofsaline by centrifuga-tionat1,700xgfor 10 min. Theerythrocyteswere lysed by

resuspending thepelletin 5 ml of0.85% ammonium chloride and incubated for 5 min at 37°C. The samples were then

washed two additional times in saline, and the final pellets weresuspended in 2 ml of saline. A 5-,ul aliquot was taken fromeach sample and dried ontoTeflon-coated microscope slides with a fixed surface area (11-mm circles; Carlson

Scientific, Peotone, Ill.). The extent of disease for each

animal was determinedby microscopic analysis of stained

slides. The total numbers of cysts per animal lung were

determinedby counting the numbers ofcysts in 50 micro-scope fields (magnification, x1,000) of homogenized lung tissue on slides fixed with ether-sulfuric acid and stained with toluidine blue(6). Numbers of total nuclei were deter-mined fromduplicate slides stained with Diff-Quik

(Ameri-can Scientific Products, McGaw Park, Ill.). Briefly, the

slideswere sequentially dipped five times inDiff-Quik

Fix-ative, eight times in Diff-Quik Solution I, and five times in Diff-Quik SolutionII. The slides were then rinsed in distilled waterand allowedtoairdry.The total numbersoforganisms

per ratlungweredeterminedasafunctionof thesurfacearea

ontheslide, the volumeofthe

applied

sample, and thetotal volume of theprocessed homogenate.

Quantitation of nuclei with the P.

carinii-specific

DNA

probe.

Cloning

and initialcharacterization of

genomic

clone

B12-2 from P.

carnii

have been described elsewhere (33).

Clone B12-2was one ofa number of clones which

specifi-cally

hybridized

to

multiple

preparations

of P. cannii ge-nomic DNA andnot togenomicDNA

prepared

fromeither rathepatictissueortheratGH3 tissue culture cell line. All

experiments which utilized B12-2 as a

hybridization probe

have been done aftertwice

gel

purifying

the insertaway from vector sequences. Other

hybridization probes, including

a highly repetitive rat satellite DNA clone

(20)

and a clone

corresponding

tothe

intervening

sequence of P.

cainni

small

subunit rRNA(ssrRNA)genes (24), were similarly twice gel

purified.

Whole rat lungs were homogenized with a Brinkmann homogenizer as described earlier. The portion of the ho-mogenate that was not used for microscopic evaluation of cysts was mixed with an equalvolume of 2x proteinase K

digestionbuffer(1x proteinaseKdigestion buffer is 200 mM

Tris-HCl

[pH 7.6], 50 mM EDTA, and 0.5% sodium dodecyl sulfate

[SDSJ).

Two to three aliquots ofproteinase K (at a final concentration of at least 50 ,ug ofhomogenateper ml) were added over the course of a 24- to36-h digestion period at55°C. A

600-pl

aliquot of the digested homogenate was

extractedtwice withphenol-chloroform-isoamyl alcohol (25: 24:1) and then once withchloroform-isoamyl alcohol (24:1). The DNAcontentofthe final aqueous phase wasdetermined

by afluorescence bisbenzimideassay(18). Treatmentof the lunghomogenatewithNovozyme234 (28) orlysing enzymes

(SigmaL-2265) prior toproteinaseKdid notincrease either the yield of genomic DNA or the intensity of the P.

carinii-specific hybridizationsignal.

Anequivalent amount of DNA (usually 50

p,g)

from each

experimental animal was diluted to 500 ,ulwith TE (10 mM

Tris-HCl

[pH7.6], 1 mMEDTA).Three serial dilutions were

similarly preparedin TE. Each sample was then adjusted to 0.3 N NaOH. The RNA was hydrolyzed and the genomic DNA wasdenatured byincubationfor 30 to 60minat 70°C. Thesampleswere cooled to roomtemperatureand neutral-ized by the addition of an equal volume of 2 M ammonium

acetate (pH 7.0). The samples were then immobilized on

duplicate nylon

filters

(Nytran; Schleicher and Schuell, Keene, N.H.) in a Minifold II slot blotapparatus.The filters were airdried, baked for 2 h at 80°C undervacuum, wet in 6xSSPE

(lx

SSPE is 180 mM NaCl, 10 mM NaH2PO4 [pH

7.4], and 1 mM EDTA), and prehybridized at 44°C for at least 4 h in 6x SSPE-1%

SDS-10x

Denhardt's solution (lx Denhardt's solution is Ficoll, polyvinylpyrrolidone, and

bovine serum albumin each at

0.02%)-100

,g ofdenatured

salmon sperm DNA per ml. The prehybridization solution wasremovedandreplaced with a buffer whosecomposition

was 6x SSPE-1%

SDS-50%

deionizedformamide-10% dex-transulfate.Hybridizationprobes wereradioactivelylabeled with

[a-32P]dCTP

(3,000

Ci/mmol;

Amersham Corporation)

by random

primer

labeling(10, 11). The specific activity of the probes was routinely greater than 109 cpm/,ug. The radioactive probe was denatured along with salmon sperm DNA(equivalentto 100,ug/ml ofhybridization solution)and mixed with the hybridization buffer. Hybridization was

conductedat 44°C for 16 to 20 h. The filters were washed to afinal stringency of

0.1x

SSPE-0.1%

SDS at65°C, rinsed

several times in 2x SSPE at room temperature, air dried,

andexposedtoKodakXR-4X-rayfilm.

After an adequate exposure, the filters were cut into

blocks corresponding to the slots in the manifold and the amount of radioactive probe hybridizing to each slot was

quantitated by liquid scintillation counting. Alternatively (and always priortoliquid scintillation

counting

for compar-ison), signals werequantitated directlyfrom the filterswith the Betascope 603 BlotAnalyzer(Betagen, Waltham,

Mass.)

or thePhosphorImager400E

(Molecular

Dynamics,

Sunny-vale, Calif.).In ourhands,the datacollectedboth

by

liquid

scintillation counting and with the Betascope603

typically

exhibited very narrow linear ranges for standardization purposes. Because of this, at least three serial dilutions (from25

pug

to25ng of

genomic

DNA)

of each

experimental

samplewere slotted and

hybridized.

On the other

hand,

the PhosphorImager 400E

displayed

amuchbroader linear range

on April 12, 2020 by guest

http://jcm.asm.org/

1 2 3 4 5 6 7 8 9 10 I

I

11 12131415 16 171819 2021

-- 21

.23-4.97 3.53

2.03 1.58 -

1.38--

0.95--0.59\

FIG. 1. Specifichybridization of clone B12-2toP. cariniigenomicDNA. Either5(lanes1to10)or2(lanes11to21)p,gofgenomicDNA fromavarietyoforganisms wasdigested with EcoRI, fractionatedbyagarosegelelectrophoresis, transferredtoanylon membrane,and hybridizedwith32P-labeled B12-2.GenomicDNAwasfrom chicken(lane1),ratGH3cells(lane 2),P. carinii(lane3),T. brucei(lane 4), T. gondii(lane5), Eimeria tenella (lane6),S. cerevisiae(lane7),C. albicans(lane8),C.neoformans (lane9),A.nidulans(lane

10),

heart(lane

11), intestine(lane12),kidney(lane13),andlungtissues(lane14)fromanimmunosuppressed rat, E.coli(lane15),C. kutscheri(lane16),

P.multocida(lane17),P.aeruginosa(lane18),S.pneumoniae(lane19),A.fumigatus(lane20),andcytomegalovirus-infectedhuman foreskin fibroblasts(lane21). Relativemigrations of molecularweightmarkers (inkilobases) areindicated between thetwopanels.

and therefore required fewer dilutions. Radioactive emis-sionswere converted to organism number with standards

included ineach hybridizationreaction mixture.

Organismsdestined foruseasstandardswerepreparedby modification(33)ofthe stomacherprocess(7).Thisincluded

collagenase, hyaluronidase, and DNase I digestions to

re-move asmuch debrisaspossibleso asto allow foraccurate

microscopic quantitation. Followingvisualquantitation,

ge-nomic DNAwas extracted fromknown numbers of

organ-isms. Weassumed aquantitative recovery of nucleicacids from the organism preparation when we assembled the standard curve. Since this is not likely, the experimental

numberspresented hereareprobablyoverestimates. Genomic Southern zoo-blots. The following organisms and/ortissueculture cell lineswerekindly providedtous:rat

GH3 cells, Trypanosoma brucei, Toxoplasma gondii,

Cryp-tococcus neoformans, Corynebactenium kutscheri, Pas-teurella multocida, Pseudomonas aeruginosa, Streptococ-cus pneumoniae, Aspergillus fumigatus, Aspergillus nidulans, and cytomegalovirus-infected human foreskin fi-broblasts. Unless statedotherwise, genomicDNAwas iso-latedby proteinase Kdigestionandorganic solvent

extrac-tions as described earlier for P. caninii-infected tissue. C. neoformans spheroplasts were first prepared with lysing

enzymes(Sigma L-2265)asdescribed by Edman and Kwon-Chung (9) and thenhandled inasimilarfashion. Aspergillus genomic DNAwas isolated by the method of Raeder and Broda(21). GenomicDNAwasprepared fromC. kutscheri, P. multocida, P. aeruginosa, S. pneumoniae, and

Esche-richiacoliby the followingprotocol. Cellsweresuspended

in50 mMglucose-25mMTris (pH8.0)-10mMEDTA-50mg

oflysozyme ml-' and incubated for 60 min at 37°C with

shaking. After the suspension was adjusted to 0.5% SDS,

digestion with proteinase K was carried out as described

earlier. The suspension was then mixed with an equal

volume of 5 M guanidinium thiocyanate-0.1 mM

EDTA-0.5% SDS and incubatedonicefor 5min. An equalvolume ofcold7.5 Mammoniumacetatewasadded, and the mixture

was incubated on ice for 10 min. Samples were extracted

once with 4 volumes of cold phenol-chloroform-isoamyl

alcohol(25:24:1) and then extracted asdescribedearlier for

P. carinii-infected tissue. Preparations of genomic DNA fromSaccharomyces cerevisiae and Candida albicans were kindly provided to us. Genomic DNA was restricted with EcoRI, fractionated by agarose gel electrophoresis,

trans-ferred by diffusion (25) to Nytran, and prehybridized, hy-bridized, and washed as described earlier.

RESULTS

Specific hybridization of clone B12-2 to P. carinni genomic ,DNA. Steroid-induced immunosuppression renders the rat

host susceptible to a plethora of opportunistic organisms includingviruses, bacteria, fungi, and protozoa. In order to validate a DNAprobe assayto quantitateorganism load in this model system, it is imperative to demonstrate specific hybridization toP. carinii genomic DNA. The isolation of clone B12-2 from a P. carinii-enriched genomiclibrary has beendescribed elsewhere (33). The clone isapproximately 5 kb inlength and doesnothybridize torattissueculture cell genomic DNA(33) (Fig. 1, lane 2). To extend the test for specificity, genomic DNAs fromawidevarietyof organisms

were probed with clone B12-2. Hybridization was entirely restricted to P. carinii genomic DNA(Fig. 1). In particular, there was no cross-hybridization to A.fumigatus, C. albi-cans, C. neoformans,C. kutscheri,P. multocida,P. aerug-inosa, or S. pneumoniae. Each oftheseorganisms is asso-ciatedwithpulmonary pathologyinhumans or rats.

In light of the fact that the B12-2 hybridization probe is preparedas anEcoRIfragment, the hybridization profiles in Fig. 1 indicate that the clone is multiply represented in the P.

carinii genome. The majority of the EcoRI-generated ge-nomic fragments which hybridize with clone B12-2 fall within the 1.4- to 12-kb size range. The repetitive nature of thispieceof DNAaffords us increased sensitivity when it is used as ahybridization probe.

Limits of detection for clone B12-2. The results of an

experimentdesigned to define the limits of detection of this hybridization probe are shown in Fig. 2. Enriched

prepara-tionsof P. cannii(7, 33) from the lungs of heavily immuno-suppressedrats werequantitated by microscopic analysis of stained slides. Total nucleic acids were extracted from 5 x

on April 12, 2020 by guest

http://jcm.asm.org/

-l al i's A B C

1rO, ....-

-I0 -- -9

A B C

_

*SxW

T-earn.ent

3

10D.

A

S*erCo DoseimgL]

0 2 4

B

Steroid Dose (rg

_-O 2 4

< (ij

;.,

FIG. 2. Organism content in an enrichedpreparation of P.

carinji

isolated from the lungs of heavily immunosuppressed rats was quantitated histochemically. Genomic DNA was prepared,serially

diluted from107to101genomeequivalents withquantitative recov-eryassumed, and then immobilized on a nylon filter. ColumnsAand Bareidentical, except that each dilution in columns B was spiked with 15

jig

ofgenomic DNA prepared from the GH3 cell line. The material in columns C was derived from the same preparation of organisms, the exception being that organisms were serially diluted prior to the extraction of nucleic acids. Control samples include5

pLg

ofratGH3tissue culture cellgenomic DNA (slots A, ii), 15 ,ug of rat livergenomic DNA (slots B, ii), 10 ng of pUC119 (slots C, ii), 1 pg of B12-2 as arecombinant in plasmid pUC120 (slots C, i). The filters werehybridized withP.cariniigenomic clone B12-2 (left panel) and a ratsatellite DNAprobe (rightpanel).

109

organisms. Thefinalaqueous phase was seriallydiluted andappliedtoduplicatenylon sheets in slots containing

107

to 10 organism equivalents (Fig. 2, columns A). Alterna-tively, incolumnsC, organismswereserially diluted prior to

nucleicacid extraction, andthen, following extraction, nu-cleic acids were immobilized on themembrane. Hybridiza-tion of this titraHybridiza-tion withB12-2(Fig. 2, left panel) indicates that as few as 1,000 organisms may be detected. This assumesthe quantitative recoveryofgenomic DNA during the extraction process. Samples in Fig. 2, column B, are

identicaltothose incolumns A, except for the fact that they arespiked with 15

p,g

ofgenomicDNApreparedfrom the rat

GH3 tissue culture cellline. The inclusion of rat genomic

DNA to more closely approximate the composition of

nu-cleic acids extracted from a lunghomogenate did not

sub-stantially decrease the signal-to-noise ratio and therefore still

allowsfor the detection ofapproximately 1,000organisms. Extraction of nucleic acids from organisms after serial dilution does appear to result in less-efficient recovery

(compare Fig. 2, columnsAand Bwith columnC).

A duplicate membrane was hybridized in Fig. 2 (right panel)with ahighlyrepetitive ratsatellite DNAprobe (20).

This piece of DNA has been estimated tobe present in at least10,000 copiesin the ratgenome. Inspiteof the fact that this preparation was enriched for P. carinii, it is apparent thataconsiderableamountofratgeneticmaterial

copurified

duringthe

purification

procedure.

Accumulation of P. carinii in the ratmodel.

Microscopic

detection of P.

carinii

inthe rat model is not

reproducibly

seen until after 4 weeks of steroid treatment

(20a).

The

accumulationof P. carinii ingroups ofratstreated with

0,

2,

or 4 mg of dexamethasone per liter of

drinking

water was monitoredby

hybridization

withclone B12-2. Animalswere sacrificed after3, 6, and 9 weeks of treatment, andnucleic

acidswere

prepared

from

lung

homogenates.

No P. carinii couldbedetectedinratswhich didnothavesteroid in their

drinkingwater

(Fig.

3),

evenafter

being

housed for 9 weeks

6 _ _ (i)

' (ivO)

-

-9

- -

~~~(II)4

FIG. 3. Ratsadministereddexamethasone in the drinking water atdoses of 0, 2,and4mg/liter were monitored at 3-week intervals for the presence of P. carinii organisms. DNA extracted from whole-lunghomogenates was quantitated, and 25 (i), 2.5 (ii), 0.25 (iii), and 0.025(iv)

pLg

weredenatured and immobilized on duplicate filters. Filters were hybridized with clone B12-2 (A) or a portion of theP. carinii16S-like rRNA gene (B).

inthesame rooms with the rats from the treatment groups.

Steroid treatment at either 2 or 4 mg/liter did result in the

accumulation of P. carinii over time, as detected by the

hybridization probe. By this assay approach, organisms

wereobvious at 3weeks but were notreproduciblydetected any time sooner. Polymerase chain reaction analysis with

amplification primers locatedwithinclone B12-2 (5) was able tospecifically detect P.

carinii

earlier during the

immuno-suppression regimen (datanot shown).The intensityof the hybridization signal indicated thatorganisms progressively accumulatedateach ofthethreetimepoints assayedinthis experiment. After 6weeks oftreatment, animalsdosedwith 4 mg of steroid per liter had a greater organism load than those treated with 2 mg/liter. However, at 9 weeks both

levels of steroid resultedin a

similar

P.

carindi

burden. The sensitivity ofgenomic clone B12-2 (Fig. 3A) as a

hybridization probewascomparedwith that of a second P.

carinii-specific

DNA clone. The ssrRNA gene from P.

carinii

contains an intron

(24)

which, like clone

B12-2,

hybridizes specificallytoP.

carinii

genomic

DNA

(18a).

By analogy to ssrRNAgenes from other

organisms

which are

multiply

represented

in thegenome

(19)

and from the

obser-vation that most if notallssrRNAgenes in P. cariniicontain

this intron(18a),we

predicted

thatanintron

probe

wouldbe

both

specific

and sensitive. The results shown in

Fig.

3B indicate that clone B12-2 is

approximately

10-fold more sensitiveas a

hybridization probe.

Since thetwo

probes

used

inthis experiment had

equivalent

specific

activities,

it fol-lows that clone B12-2 is

represented

more often in the P.

cariniigenome than the ssrRNA gene.

Accordingly,

B12-2 is amoredesirable

hybridization probe

for this assay.

on April 12, 2020 by guest

http://jcm.asm.org/

Cu

a

Cu

la

0 co

ci

IC

16-10.

Cu

1... ...

6 4 5 6 7 8 9 10

Weeks of

Immunosuppression

9~~~~~~~~

FIG. 4. Groups ofratstreatedwith dexamethasoneat adoseof 2mg/literinthedrinkingwaterforupto9 weeksweresacrificedat

weekly intervals andassayed forP.cariniiorganism loadbystaining

of cysts with toluidine blue(-),byDiff-Quikstainingof nuclei(0),

andby hybridizationwithclone B12-2(0).

Rats immunosuppressed for 9 weeks were sacrificed at

weekly intervalsstartingat3weeks.Aportion of

lung

tissue

from each animal ineachgroup(at least four andgenerally eight animalsper

group)

was

processed

for visual P. carinii

cystand nucleuscounts.

Organism

load in the balance of the tissuewasevaluatedbyhybridization withclone B12-2. The

accumulation of P. carinii over thecourse of the

immuno-suppressivetreatmentperiodasmeasuredbythetwo

meth-ods isdepictedinFig.4.Therewas a

gradual

increasein the number of cysts andnuclei (intracysticbodies and tropho-zoites) as determinedby

microscopic quantitation

with the

respective stains. Cyst and nuclei appeared to increase

proportionally, suggesting a link between these two

devel-opmental stages. The

utility

ofthe DNA

probe

as anassay tool is also demonstrated by this

figure. Organism

load determinedby the DNA hybridization assay

closely

paral-leled the data obtainedbymicroscopicevaluation. This fact isvividlydemonstrated byplotting the correlationbetween P. cariniiorganism load determinedby thetwo assay pro-cedures(Fig. 5).

Useof clone B12-2 to monitorclearance of P. cariniiwith

knownantipneumocystis agents. The efficacies oftwo

thera-peuticregimensusedto treatP. carinii pneumonia clinically (14, 30) were tested in the rat model and monitored both

visuallyandby DNA hybridization with clone B12-2. Ani-malsimmunosuppressedfor 6 weeks to establish a P. carinii infection were treated intravenously with pentamidine or with -a combination of TMP and SMZ administered orally. Lung tissue was assayed in both experimental and control groups ofanimals after 4, 7, 14, and 21 days. Cyst counts

determinedbystainingwith toluidine blue are plotted in Fig. 6Aasafunction of the treatment period. Assay of organism loadby DNAhybridization is shown in Fig. 6B. Both assays indicated that the two forms of chemotherapy effectively cleared the P. cariniiburden from rats. A 99% reduction in organism load was realized by 21 days of treatment. The

kineticsof P. cariniiclearance was virtually the same for the

two forms oftherapy, and this was demonstrated by both assay approaches. By deduction, the results generated by the DNA-based assay, which measures both documented stages of the P.

carinii

life cycle, suggest that trophozoite

9

8

7

6

5 6 7 8 9 10

Log Number using DNA Probe Quantitation

FIG. 5. Scatter plot to illustrate the correlation between P. canniicountsdeterminedmicroscopicallyand those obtainedby the DNA hybridization assay.Organismcountsgeneratedin Fig. 4by Diff-Quik stainingwereplotted againstthe hybridizationdata

col-lected from the same experiment. The diagonal line represents a strict 1:1 correlation.

elimination parallels cystclearance forTMP-SMZ and

pen-tamidinetreatments.

DISCUSSION

The data presented here highlight the utility of using a

DNAhybridization probetomonitorP. cariniiburden inthe experimental rat model system. Thisapproach to

quantita-tion is desirable because each stage in an organism's life cycleisintheory quantifiable. Immunohistochemicalassays

forassessing parasiteburdenarecomplicated bythe

poten-tialforstage-specificantigen expression.Microscopic quan-titation of P. carinii cysts and trophozoites with selective

strains istedious andrequiresmultiple staining procedures.

Accurate determination of the trophozoite contribution is compromised by their small size. Consequently, they are easily confused with debris from the lung homogenate. Trophozoite forms also tend to aggregate together in large clumps.Dataobtained withimpressionsmears aresubjectto thecriticismthatonlyaportion of the lung is sampled.This mightnotaccuratelyreflecttheP.cariniiburden in the entire

tissue. In situ hybridizationofP. carinii rRNA, which has

been proposed as a method for diagnosis (13), would be subject to the same criticism if it were to be used to

quantitateorganism load.Moreimportantly, in situ

hybrid-izationisverydemanding technically and notapplicableto

large numbers of experimental samples.

We have shown that genomic clone B12-2 hybridizes

specificallytoP. carini DNA. Acaveat of this conclusion becomesobvious whenoneconsiders themannerin which

the P. carinii infection is induced. Immunosuppression of rats is known to permit the expansion of populations ofa

widevarietyofopportunisticorganisms (23). Insitu hybrid-ization oflung sections prepared from immunosuppressed

6bo

0

0 0

0

on April 12, 2020 by guest

http://jcm.asm.org/

u

a

0

u)

e

0

c

*

0 2 4 6 8 10 12 14 16 18 20 22

Time (daysof treatment)

FIG. 6. Rats immunosuppressed for 6 weeks to induce a P.

cariniiinfectionweretreatedintravenouslywith pentamidine (0),

orally with TMP-SMZ, (U), orwith the respective vehicles (0).

Groups of animals during the treatment period were assayed to

determine P. cariniiload visuallyafterstaining withtoluidine blue

(A) and byhybridizationwithclone B12-2(B).

rats with clone B12-2 (or aportion thereof) will

unequivo-callyprovethatthehybridizationwhichwe aremeasuring is specific. Lacking this result, we remain confident that hy-bridization with clone B12-2 is specific for the following

reasons. No hybridization ofclone B12-2to genomic DNA

preparedfromorganisms commonly foundin hosts

display-ing pulmonary pathology was detected (Fig. 1). These in-clude A. fumigatus, C. albicans, C neofonnans, C.

kut-scheri, P. multocida, P. aeruginosa, and S. pneumoniae.

Secondly, therewas astrict correlation betweenthe results generated by DNA hybridization and visual counts (Fig. 4 and5). Furthermore, thequalitative hybridization intensity

realizedwithcloneB12-2asaprobe parallels thatseenwith

aP. caninii-specific (8, 24) ssrRNAgeneprobe (Fig. 3, for example). Finally, the efficacyof documented

antipneumo-cystis therapeutic regimens was accurately predicted by hybridizationwithclone B12-2(Fig. 6).

CloneB12-2representsarepetitiveDNA element in theP. carinii genome. The P. carinii-specific ssrRNA intron was less sensitive when used as a hybridization probe in a side-by-side experiment (Fig. 3). On the basis of this, we

contendthatcloneB12-2 hasacopynumber at leastasgreat

as that of the ssrRNA gene family of P. carinii. Partial nucleotidesequence analysis of clone B12-2 (18a)indicates

that itdoesnotcorrespondtoanyP. cannii rRNAsequence

describedin the literature (8, 27, 32). Inaddition, restriction

enzyme mapping of B12-2 differentiates it from another

DNArepetitive element cloned fromP. carinii(26). Making the assumption that 100% recoveryofDNAwas

achieved fromaP. caindi preparation thatwas quantitated

microscopically, the results in Fig. 2 indicated that clone B12-2 can detect approximately 1,000 organisms in the

current assay format. However, losses of nucleic acids

during purification undoubtedly occur, which renders this

assumption inaccurate. Consequently, it is safe to predict

that the assay is sufficiently sensitive todetect fewer than 1,000 P. canniiorganismsevenin thepresenceofa

tremen-dous excess ofratgenomic DNA. Although the assay can

specifically detect fewer than 1,000P. cainni organisms, the limits ofdetection fororganism load in anentire animalare

in the orderof5 x 1 Becauseofalimitedsurfacearea,the

maximum amountof DNA thatcanbeimmobilizedthrough

the slot blot manifold onto the nylon membrane is 25 p,g.

This is onlyafractionofapercentageof the DNA in alung

homogenate. Furthermore, thevastmajority of thisDNAis

derivedfromthehost,especiallyinalightinfection.Theraw

dataoutputfrom the assayisfirstcomparedwithastandard

curve that is generated by using DNA extracted from puri-fiedP. carinii organismswhosenumberhas beendetermined

bytoluidine blue and Diff-Quik staining. This allows for the assignment of organism number to the portion of the lung homogenate whichwasassayed. From this,the total number of P. carinii organisms in the lungs of each experimentalrat

is projected.

Therelationship between trophozoite andcystforms of P.

carinii is not well understood, although it is generally as-sumed that the two forms are interdependent. Indeed, the life cycle for this organism remains obscure. If P. carinii

does infactgothrough otherstagesin its lifecycle,theseare

not yet recognized microscopically. Barring dramatic

changes in DNA content or composition, it is safe to

conclude thataDNAhybridization-basedassaywillcapably detect each developmentalstagein anorganism's lifecycle. This, of course, also assumes that physical access to the

genomic DNAcanbe achieved ateach stage during the life

cycle. Because ahybridization assay doesnot discriminate

among developmental stages, it possesses a tremendous

advantage over traditional approaches to quantitation. This

isespeciallytruefororganismslikeP. cariniiwhosebiology is not thoroughly characterized. By the employment of a DNAhybridization assay in screens fornovel antipneumo-cystis drugs, stage-specific therapeutic interventionwill be

detected. Bywayofexample, compounds which selectively interfere with the trophozoite form of P. cannii might go

undetected if cyst load as determined by toluidine blue

staining was the sole assay to measure their efficacy. In summary, the clone B12-2hybridization assayfor the

deter-mination of P. carinii organism load offers several advan-tages and therefore should serve to complement

conven-tional staining andimmunohistochemical methods.

ACKNOWLEDGMENTS

WethankKarenKirk, Myra Kurtz,JeanMarrinan,John

Thomp-son, KenBartizal, Ed Hayes, Anne Gurnett, Joseph Waner, and

Sandy Gouldforproviding organisms and/orgenomicDNA prepa-rations; DaveLinemeyerfor assistancewith theMolecular Dynam-ics PhosphorImager; and Kim Likowski for preparation of the

manuscript.

This work was supported in part by grants from the National

Institutes of Health (AI20940) and the Oklahoma Center for

Ad-vancementof ScienceandTechnology (HS16-C413-081) to D.C.G.

A

7

6

5

4 ,

0 2 4 6 8 10 12 14 16 18 20 222

on April 12, 2020 by guest

http://jcm.asm.org/

REFERENCES

1. Barker, R.H.,L.Suebsaeng,W.Rooney,G. C.Alecrim,H.V.

Dourado, andD. F. Wirth. 1986. SpecificDNA probeforthe diagnosis ofPlasmodiumfalciparummalaria. Science 231:1434-1436.

2. Bartlett, M. S., J. A. Fishman, M. M.Durkin, S. F.Queener, andJ. W. Smith.1990.Pneumocystiscarinii:improved models to study efficacy of drugs for treatment or prophylaxis of

Pneumocystis pneumoniain the rat(Rattus spp.) Exp.Parasitol. 70:100-106.

3. Barlett, M. S., J. A. Fishman, S. F. Queener, M. M.Durkin, M. A. Jay,andJ.W. Smith. 1988. Newratmodel of

Pneumo-cystis carinii infection. J. Clin. Microbiol. 26:1100-1102.

4. Bartlett, M.S., J. J. Marr,S. F.Queener,R.S.Klein,andJ.W.

Smith. 1986. Activityof inosine analogs againstPneumocystis cariniiin culture. Antimicrob.Agents Chemother. 30:181-183.

5. Becker-Hapak, M.,P.Liberator,and D. Graves. 1991. Detection

of humanPneumocystiscariniibythepolymerase chain

reac-tion. J. Protozool.38:191S-194S.

6. Chalvardjian, A., and L.Grawe. 1963. Anewprocedureforthe

identification ofPneumocystis carinii cysts in tissue sections

andsmears.J. Clin. Pathol. 16:383-384.

7. Cushion,M. T., J. J. Ruffolo, M.J. Linke,and P. D. Walzer.

1985.Pneumocystis carinii: growthvariables and estimates in

A549andWI-38 VA humancell lines.Exp.Parasitol. 60:43-45.

8. Edman, J. C., J. A. Kovacs, H. Masur, D. V. Santi, H. J. Elwood, and M. L. Sogin. 1988. Ribosomal RNA sequence

showsPneumocystis cariniitobeamemberof thefungi. Nature (London) 334:519-522.

9. Edman, J. C., and K. J. Kwon-Chung. 1990. Isolation of the

URAS gene from Cryptococcus neoformans var. neoformans

anditsuseas aselectivemarker fortransformation. Mol. Cell.

Biol. 10:4538-4544.

10. Feinberg, A. P., and B. Vogelstein. 1983. A technique for radiolabeling DNA restriction endonuclease fragmentsto high specificactivity. Anal. Biochem. 132:6-13.

11. Feinberg, A. P., and B. Vogelstein. 1984. Addendum: a

tech-nique forradiolabeling DNA restriction endonuclease fragments tohigh specific activity. Anal. Biochem. 137:266-267. 12. Frenkel, J., J. Good,and J.Schultz. 1966. Latent Pneumocystis

infection of rats, relapse, and chemotherapy. Lab. Invest.

15:1559-1577.

13. Hayashi, Y., J. Watanabe, K. Nakata, M. Fukayama, and H.

Ikeda.1990. Anoveldiagnostic method of Pneumocystis carinii:

in situhybridization of ribosomalribonucleic acid with biotiny-latedoligonucleotide probes. Lab. Invest. 63:576-580. 14. Hughes, W. T., S. Kuhn, S. Chaudhary, S. Feldman, M.

Verzosa, R. J. A. Aur, C. Pratt, and S. L. George. 1977. Successfulchemoprophylaxis for Pneumocystis carinii

pneumo-nitis. N.Engl. J. Med. 297:1419-1423.

15. Hughes,W. T., P. C. McNabb, T. D. Makres,andS. Feldman. 1974. Efficacy of trimethoprim and sulfamethoxazole in the

preventionandtreatmentof Pneumocystiscariniipneumonitis. Antimicrob. Agents Chemother.5:289-293.

16. Kim, C. K., J. M. Foy, M. T. Cushion, D. Stanforth, M. J.

Linke, H. L. Hendrix,and P. D.Walzer. 1987. Comparisonof histologic and quantitative techniquesinevaluation oftherapy

forexperimental Pneumocystiscarinii pneumonia.Antimicrob. AgentsChemother.31:197-201.

17. Kovacs,J.,and H.Masur. 1988.Pneumocystiscannii pneumo-nia:therapyandprophylaxis.J. Infect. Dis. 158:254-259. 18. Lebarca,C.,and K.Paigen.1980. Asimple, rapidand sensitive

DNAassayprocedure. Anal.Biochem. 102:344-352. 18a.Liberator, P.,andJ.Anderson.Unpublisheddata.

19. Long, E.O., and I. B.Dawid. 1980. Repeatedgenesin eukary-otes.Annu. Rev. Biochem. 49:727-764.

20. Pech, M.,T.Igo-Kemenes,and H. G.Zachau. 1979. Nucleotide sequenceof ahighly repetitivecomponent of rat DNA. Nucleic Acids Res. 7:417-432.

20a.Powles,M.Unpublished data.

21. Raeder,U.,and P.Broda.1985.Rapidpreparation of DNAfrom filamentousfungi. Lett.Appl.Microbiol. 1:17-20.

22. Schmatz, D. M., M. A. Romancheck, L. A. Pittarelli, R. E. Schwartz,R. A.Fromtling,K. H. Nollstadt,F. L. Vanmiddles-worth, K. E. Wilson, and M. J. Turner. 1990. Treatment of Pneumocystiscariniipneumoniawith1,3-beta-glucan synthesis inhibitors. Proc. Natl. Acad. Sci. USA87:5950-5954.

23. Sethi, K. K., K. Salfelder, and J. Schwarz. 1964. Pulmonary fungus flora in experimental pneumocystosis of cortisone-treated rats.Mycopathol. Mycol. Appl. 24:121-129.

24. Sogin, M.L., andJ. C. Edman. 1989.Aself-splicing intron in the small subunit rRNAgeneofPneumocystis carinii. Nucleic Acids Res. 17:5349-5359.

25. Southern, E. M. 1975. Detection ofspecific sequencesamong DNAfragments separated by gel electrophoresis.J. Mol. Biol. 98:503-517.

26. Stringer, S.L., S.Hong, D.Giuntoli,and J.R.Stringer. 1991. Repeated DNA in Pneumocystis cannii. J. Clin. Microbiol. 29:1194-1201.

27. Stringer, S. L., J. R. Stringer, M. A. Blase, P. D. Walzer, and M. T. Cushion. 1989. Pneumocystis carinii: sequence from ribosomal RNA implies a close relationship with fungi. Exp. Parasitol.68:450-461.

28. Taft, C., T. Stark, and C. P. Selitrennikoff. 1988. Cilofungin (LY 121019) inhibits Candidaalbicans (1-3)-beta-D-glucan synthase. Antimicrob. Agents Chemother.32:1901-1903.

29. Tenover, F. C.1989.DNAprobes forinfectious diseases.CRC Press, Inc., Boca Raton, Fla.

30. Walzer, P., D. Perl, D. Krogstad, P. Rawson, and M.Schultz. 1974. Pneumocystis carinii pneumonia in the United States. Epidemiologic, diagnostic and clinical features. Ann. Intern. Med.80:83-93.

31. Walzer, P. D., C. K. Kim, J. Foy, and J. Zhang. 1991. Furazoli-doneandnitrofurantoinin the treatment of experimental Pneu-mocystis cannii pneumonia. Antimicrob. Agents Chemother. 35:158-163.

32. Watanabe, J., H. Hori, K. Tanabe, and Y. Nakamura. 1989. Phylogenetic association of Pneumocystis carinii with the "Rhizopoda/Myxomycota/Zygomycota group" indicated by comparison of5S ribosomal RNA sequences. Mol. Biochem. Parasitol.32:163-168.

33. Worley,M.A., M. H. Ivey, and D. C. Graves. 1989. Character-ization and cloning of Pneumocystis carinii nucleic acid. J. Protozool. 36:9S-11S.

on April 12, 2020 by guest

http://jcm.asm.org/