1)11 hIL.L COUYAD_ _

CONTRACT NO: DAMD17-87-C-7242 AD- A225 114

TITLE: MOLECULAR CHARACTERIZATION OF MEFLOQUINE RESISTANCEIN PLASMODIUM FALCIPARUM

PRINCIPAL INVESTIGATOR: Stephen C. Merritt, Ph.D.

CONTRACTING ORGANIZATION: University of North Carolinaat Chapel Hill

School of Public Health360 Rosenau Hall, 201H, $7400Chapel Hill, NC 27599

REPORT DATE: March 22, 1990 FI. C -JUL 31199

TYPE OF REPORT: Final Report

PREPARED FOR: U.S. ARMY MEDICAL RESEARCH AND DEVELOPMENT COMMANDFort Detrick, Frederick, Maryland 21701-5012

DISTRIBUTION STATEMENT: Approved for public release;distribution unlimited

The findings in this report are not to be construed as anofficial Department of the Army position unless so designated byother authorized documents.

40 -0 30 119

uhclassifiedSECURITY CLASSIFICATION OF THIS PAGE

Form ApprovedREPORT DOCUMENTATION PAGE OMB No. 0704-0188

la REPORT SEC.UfRIT CLASSIFICATION Ib RESTRICTIVE MARKINGSUnc assifieA

2a. SECURITY CLASSIFICATION AUTHORITY 3 DISTRIBUTION/AVAILABILITY OF REPORTApproved for public release;

2b. DECLASSIFICATION /DOWNGRADING SCHEDULE distribution unlimited

4. PERFORMING ORGANIZATION REPORT NUMBER(S) 5. MONITORING ORGANIZATION REPORT NUMBER(S)

6a. NAME OF PERFORMING ORGANIZATION 16b OFFICE SYMBOL 7a NAME OF MONITORING ORGANIZATION

university of North CarolinaI (If applicable)at Chapel Hill I

6c. ADDRESS (City, State, and ZIP Code) 7b. ADDRESS (City, State, and ZIP Code)910 Raleigh Rd.Chapel Hill, NC 27599

8a. NAME OF FUNDING/SPONSORING 8b OFFICE SYMBOL 9. PROCUREMENT INSTRUMENT IDENTIFICATION NUMBER6§G TIRdical Research and (If applicable) DAMD17.87.-C-7242

DOvel nnment Command I8c. ADDRESS (City, State, and ZIP Code) 10. SOURCE OF FUNDING NUMBERS

Fort Detrick PROGRAM PROJECT |TASK |WORK UNITNO NO. ' NO 'ACCESSION NO.Frederick, MD 21701-5012 OT 62770A870 AF 024

11 TITLE (Include Security Classification)

Molecular characterization of mefloquine resistance in Plasmnod',um aciprum

12. PERSONAL AUTHOR(S)

Merritt. Stephen .13a. TYPE OF REPORT 13b. TIME COVERED J14. DATE OF REPORT (Year, Month, Day) 15. PAGE COUNT

Final I FROM87 09,1 TOU-_.L 1990 March 22 2216. SUPPLEMENTARY NOTATION

17. COSATI CODES 1,PSUBJECT TERMS '(C t nI-n-e rWinees-ar-y-nu ide,,ry fb, b umbe.r)FIELD GROUP -SUB 61OUP / . .FE GO S R / A LiBiotechnology;DNA structure;Drugs;Infectious diseases;

nr, i i I Malaria ;Mefloquine;Plasmodium Parasitology ,1%, ISTRACT (Continue on reve if necessary and identify by block number)

he molecular basris for mefloquine resistance in the human malaria parasite, Plasmodiumalciparum was investigated using recombinant DNA techniques, The parasites were Yoojhd'to'e capable of manifesting extremely high levels of resistance to the drug in vitro, Resistantines were found to be rapidly growing, impervious to various treatmentS, and significantlytimulated to grow by low levels of mefloquine, At the molecular levels, it was found thathe genes for tubulin, a calmodulin-like protein, and a P-glycoprotein-like protein were allmplified in a quantitative relationship to the degree of resistance, Clones for each ofhese genes were isolated and characterized. Attempts were made to induce the amplificationf tubulin genes in a sensitive parasite to see whether this event would confer resistance.igh levels of vinblastine resistance were achieved but without evidence of gene amplificationence, the exae't relationship between the amplification event and drug resistance remainsnresolved. 'A hypothetical relationship was developed to explain the amplification of theseenes together possibly leading to the resistance trait.

20. DISTRIBUTION/AVAILABILITY OF ABSTRACT 21 ABSTRACT SE.C ITY CLASSIFICATION0 UNCLASSIFIED/UNLIMITED 0 SAME AS RPT I DTIC USERS unc/assi ed

22a. NAME OF RESPONSIBLE INDIVIDUAL 22b TELEPHONE (Include Area Code) 22c. OFFICE SYMBOLMrs. Virginia M, Miller 301-663-7325 SGRD-RMI-S

DD Form 1473, JUN 86 Previous editions are obsolete. SECURITY CLASSIFICATION OF THIS PAGE

unclassified

FORENM

opinions, interpretations, cdclusions and recommendations are those of theauthor and are not necessarily endorsed by the U.S. Army.

Mhere copyrighted material is quoted, permission has been obtained touse such material.

Mere material from docunents designated for limited distribution is,quoted, permission has been obtained to use the material.

Citations of commercial organizations and trade names in this report donot constitute an official Department of the Army endorsement or approval ofthe products or services of these organizations.

In conducting research using animals, the investigator(s) adhered to the"Guide for the Care -and Use of Laboratory Animals," prepared by the Omitteeon Care and Use of Laboratory Animals of the Institute of Laboratory Animalresources, National research buncil (NIB Publication No,. 86-23, evised 1985).

For the protection of human subjects, the investigator(s) have adheredto--plicies of applicable Federal law 45CFR46.

X In conducting research utilizing recombinant Mh technology, the

Te-veigator(s) adhered.to current guidelines promulgated by the NationalInstitutes of Health.

Accesion For PI Signature Date;TIS CRA&I I i Stephen C, Merritt 90-03-22

DTIC TAB El

ii. . . .. .. - .- _- -:. - -

0 1IC

Ifly

A'j ,,i . tI

Introduction

The studies conducted here were designed to identifythe genetic events accompanying the- emergence of the resistancetrait in vitro and to define the nature of mefloquine resistanceat the molecular level in parasites in which the trait has beeninduced and in human isolates demonstrating natural mefloquineresistance.

It is difficult to exaggerate the impact that the humanmalarias continue to have on the hundreds of millions of peopleliving in endemic areas and the even greater risk which infectionposes to individuals who must enter these regions without eventhe partial immunity conferred by previous exposure. Despiteconcerted efforts at vector control, chemoprophylaxis, and rapiddiagnosis and treatment of suspected cases, the infection remainsa risk to the health and well-being of roughly half of theworld's population (1).

Of the four principal species of malarial parasitesinfecting humans, Plasmodium falciparum presents the most seriousthreat. Because of its potential lethality, this disease must betreated when detected, but the emergence and spread of drugresistant strains have increasingly prevented the use of whatonce were the most effective and reliable antimalarials. Inaddition, regions with overtly drug resistant strains oftenoverlap with areas in which the AnDheline vectors of malariahave developed resistance to the most commonly usedinsecticides(2). While this has been occuring, significantprogress has been made toward the identification of relevantprotective antigens and the ability to produce these antigens bygenetic engineering for the potential development of effectiveimmunoprophylaxis. Undoubtedly, this approach holds a great dealof promise for the future, but until such vaccines can bedeveloped and the logistical problems of testing, production, anddeployment have been met, the other alternative, the derivationof new antimalarial drugs, will continue to play an essentialrole in practical malaria control (3).

The most promising member of the new generation ofantimalarials is the synthetic quinolinemethanol derivative,mefloquine (WR 142,490), prepared during the intensive drugdevelopment program of the U.S. Army Medical Research andDevelopment Command. This blood schizontocide, a structuralanalogue of quinine, is effective against many strains ofPamodium falcipam resistant to other drugs and shows very lowtoxicity in humans (4). Mefloquine is also the only drugcurrently available for the treatment and prophylaxis ofmultidrug-resistant malaria and has recently received FDAapproval for use in the U.S.

In nature, it is clear from previous studies that resistantparasites can exist in a clinical isolate in the absence ofprevious drug pressure and that these parasites can be selectedby drug pressure to become the predominant population.

As with several other antimalarials, however, we know verylittle about the mode of action of mefloquine, less about thekinetics of the induction of resistance Th vitro and in XivQ, andvirtually nothing about the molecular mechanisms which form the

2

basis of the resistance trait.It is obvious that the induction of a resistant clone from a'

clonally-derived sensitive population requires a fundamentalgenetic difference between the two cell types. Conceivably, thisdifference could give rise to a transitory altered-cell state ora relatively permanent change conferring resistance. In eithercase, however, changes in genetic expression or the genome itselfleading to altered expression must be invoked to explain thedifference. Even events such as post-translational modificationsand peptide half-life differences must ultimately have a geneticbasis.

The studies reported here were predicated on a simplehypothesis: the production of stable, heritable, mefloquineresistance in clones derived from a previously sensitive cloneinvolves fundamental genetic changes which lead to altered geneexpression conferring the resistance trait. These geneticchanges would not have to lead to a continual expression of theresistance trait, but at the very least, they would have toprovide significant changes in the parasites' ability to respondto drug pressure. The corollary of this hypothesis is thataltered gene expression should be reflected in quantitative orqualitative changes in the products of the genes involved.Accordingly, these changes should be detectable as differencesbetween sensitive and resistant cells at the molecular levelproviding that the means of detection is sufficiently sensitive.The detection of such differences should be possible by means ofthe specificity and sensitivity inherent in recombinant DNAtechniques. The identification of gene products which differwill provide specific probes for the gene or genes which providethe mechanistic basis for resistance.

METHODOLOGY:

In the beginning of our analysis of W2 and W2MEF at themolecular level, we generated a genomic library for each cloneusing the bacteriophage vector Lambda gtll (5). Our initialcharacterization of falciparum DNA with restriction endonucleasedigestion indicated that conventional strategies for genomipcloning would likely not be adequate for our purposes.Specifically, the high A/T content of the DNA (approximately 81%)makes the use of enzymes often used .for the fragmentation ofgenomic DNA for cloning less suitable. Accordingly, we used theenzyme Dra I which recognizes the restriction site, TTTAAA, andgenerates a blunt-ended fragment upon cutting. Although thisenzyme recognizes a hexamer site, with the Plamoiua DNA, itcuts frequently (approximately evexy 220 nupleotides) and thus,acts like the 4-base restriction enzymes commonly used tofragment genomic DNA. The DNA from each clone was cut overvarying time points with varying amounts of enzyme and thedifferent fractions were pooled and subjected to sizefractionation by stringent isopropanol precipitation andchromatography on Sephacryl S-1000. Fragments ranging between2000 bp and 7000 bp were collected and methylated using Ego RImethylase. Synthetic Eco RI linkers were ligated to thesefragments followed by EcQ RI digestion and ligation into gtll.

3

The recombinant phage DNA was packaged into viable phages invitro and the resulting libraries of recombinants were amplifiedon ,. coli Y1090(r-) bacteria. Approximately 160,000recombinants were present in the unamplified packaging mixtures.If we assume that the estimate s of the plasmodial genome size are-correct (approximately 2 X 10 base pairs) a random gene libraryconsisting of 15,000 recombinant clones carrying 6 kb insertsshould contain all clonable sequences with a probability of 99%.Our libraries contain ten times this number of clones and should,therefore, be.representative of the genomes of the respectiveparasite clones. Although our choice of gtll as a vector limitsthe size of the possible recombinant DNA insert (approximately7.2 kb maximum), screening the larger number of clones did notrepresent a problem. In preliminary screening, 15,000 clonescould be screened on a single 9 cm plate. For analysis of thegenetic context of clones of interest we also generated librairesfrom both the sensitive and resistant parasite clones containinglonger stretches of contiguous sequences in the phage vector EMBL4 which accepts inserts up to 22 kb.

We also compared W2 and W2-MEF at the DNA level by directrestriction endonuclease digestion and agarose gelelectrophoresis. No obvious differences were found between thetwo, but they were markedly different from other clones examined.This finding did not rule out the possibility of geneamplification being involved in the development of resistance,but it did indicate that if amplification occurs concommitantlywith resistance it does not involve the generation of more thanabout 50 copies of the amplified' sequence unlike the case inother drug resistant protozoa where amplified sequences can bedirectly visualized on gels (7).

We then did some prel.iminary screening of the librarieswith radiolabeled genomic DNA (for repetitive sequences) andcDNA (for gene-specific sequences). Both libraries havecomparable numbers of cDNA positive clones which have beenqollected into sub-libraries, but the W2MEF library appears tohave roughly 10 times the number of repetitive sequence-specificclones when screened with W2MEF DNA. We know that some of thesesequences differ between the two clones but we have not had timeto do an in depth analysis of eaci of the roughly 200 clonesisolated in this way. Instead, we concentrated on the fewrecombinant clones which we could demonstrate distinctquantitative differences in the resistant and sensitiveparasites which appeared to correlate with the degree ofresistance found in each (yide infra).

Subsequently,we also reapplied drug pressure to theresistant parasites in vitro to attempt to increase the level ofresistance. This was done for two reasons:,. First, we wanted tobegin to examine the limits of resistance that can be expressedto this drug. Secondly, we wanted to maximize our chances ofdetecting changes associated with resistance by increasing theoverall resistance level. We were able to demonstrate that theparasite can adapt to as much as 2.4 ug/ml whereas the normalIC-50 for the sensitive- parasite is about 5 ng/ml. We have keptthe parasites under this level of continuous drug pressure for aslong as 16 days with continued schizogony. We then began to

4

grow the parasites in quantities sufficient to allowthe isolation of-mRNA for cDNA cloning, We prepared the mRNA byoligo-d(T)-cellulose chromatography. This then was used for thepreparation of two sets of cDNA libraries for cross-screeningbetween the mefloquine sensitive clone W2 and the mefloquineresistant clone W2-MEF. One set was constructed using thebacteriophage expression vector gtlt while the other librarieswere made using the vector gtlO (5). These two separateconstructions were made to allow us to screen the libraries witheither DNA probes or antibody to parasite proteins (gtll) or toensure positive selection for recombinants (gtl0). Both sets oflibraries had active phage titers of greater than 1 millionplaque-forming units per milliliter. This is at least a hundredtimes more than would be needed to represent even the leastfrequently expressed mRNA and hence all the libraries appear tobe useful for our purposes. We have isolated DNA clones from thelibraries representing, alpha and beta tubulin and a calmodulin-like sequence.. These were isolated from the library usingheterologous probes for the genes prepared fromTryDanosoma gambiense from other studies in my laboratory.

These clones are of particular interest because we arebeginning to suspect that both microtubules and calmodulin mightbe involved in the mechanisms behind at least certain aspects ofdrug resistance. During the Southern blotting experiments wedetermined that there are quantitative and qualitative changes inboth the tubulin gene and the calmodulin gene dosage that appeared tocorrelate with the degree of resistance. In particular, we havecompared three separate clones-: D6 (naturally resistant tomefloquine but senstive to other antimalarials)', W2 (sensitive tomefloquine but resistant to many other antimalarials), and MEF-2.4 (highly resistant to mefloquine and to most antimalarialdrugs) . The blotting experiments indicated that as the degreeof multidrug resistance increased the number of ,rubulin-and calmodulin genes increased coordinately (Figure 2 and 3).Despite the use of heterologous probes, the signal strength ofthe clones isolated was very good even under stringent conditions(Figures 4 and 5). In addition, we have found, that both tubulingenes and calmodulin are linked in DNA isolated from theparasite. We determined this by Southern blotting of genomic DNAas well as isolation of genomic recombinant clones using-tubulinor calmodulin probes (Figures 6 and 7) In each case, clonespositive for the tubulin probe were also recognized by thecalmodulin probes under high stringency (Figure 8.).

We have also done studies looking at the reversal of invitro resistance with antimicrotubule agents (vinblastine andvincristine - ref. 9), calcium channel blockers (verapamil - ref8), and calmodulin antagonists (calmodazolium andchlorpromazine ref. 9). Each of these agents was able to restore theImefloquine sensitivity to the resistant clone MEF-2.4 but less soto the more sensitive clone D6 or W2 (Figures 9, 10, and 11). Wehave found that this this clone grows faster than all of theother clones and strains which we maintain in the laboratory andthat it is able to survive various treatments which aredeleterious to other clones (heat shock, endotoxin, microtubuleinhibitors, and calmodulin antagonists). Interestingly, although

5

highly resistant to quinine and moderately resistant tohalofantrine, this line shows enhanced sensitivity to chloroquine(iC-50 of 26 ng/ml). Although this is still a measure ofresistance, it is well below the level manifested by the parentalline W2-MEF.

We have also found that calcium channel blockers, calmodulinantagonists, and microtubule inhibitors all reverse theresistance to. mefloquine in this line in a manner similar to thatreported by the WRAIR group for chloroquine. Since this patternof reversal is very reminiscent of multi-drug resistance in othereukaryotic cells types, we began to examine the parasites for thepresence of a 'P-glycoprotein-like' entity which might mediatethe efflux response. Initial screening using a consensussequence oligonucleotide indicated that there weresequences which bore structural similarity to the P-glycoprotein(9) and we isolated representative clones for further analysis.We wanted to use a molecular probe homologous to the P-glycoproteinto ask whether or not this gene (if present) is amplifiedcoordinately with the manifestation of resistance. As wecontinued- our work, it became clear that they were as were genesfor the tubulins, alpha and beta isotypes, and a calmodulin-likesequence (Figure 12).

We have also placed the non-multidrug resistant clone D6under yinblastine pressure jn vitro and have obtained parasiteswhich can survive high levels of this microtubule inhibitor. Aswe isolate lines which manifest increasing levels of resistanceto this compound, we want to ask whether or not the lines willalso be more resistant to known antimalarials in a manner similarto the multidrug resistant clones which have amplified thetubulin genes. This would help to confirm the participation ofthese genes in the development of resistance as well as therelative coordinate degree of amplification.

We have also continued the pressure on MEF-2.4 such that itcan now survive at least 6 days of continuous pressure with2.4 ug/ml. This is 480 times the IC-50 for the sensitive cloneW2.

We then continued our study of mefloquine resistance inPlsmodium falciparu in addition to our examination of themultidrug resistance phenomenon. As noted above, we havedetected an increase in the copy number for the alpha and betatubulin genes and a calmodulin-like Sequence. Previously, we havebeen referring to this latter gene as calmodulin since it showsstrong hybridization characteristics with a trypanosomecalmodulin-specific cDNA probe. We have now sequence about 500bases of this gene and have yet to detect any sequence similaritywith calmodulin. This is. predictable however if, as in other§ystems, this cDNA clone with which we are working is flanked onboth ends by long 5'- and 3'"non-coding regions (10). Since theprobe recognizes only the coding.,domain of calmodulin, this wouldmean the probe is binding internally to the regions sequencewhich have-been at both ends of the clone. Alternatively, thisgene might not be calmoduli i at all. It is interesting to notein this regard, that a sequence which is found amplified in humandrug resis'ance cell lines, sorcin, is aPparently a c'Icium-binding protease with substantial: homology to calmodulin (11).,

.6

Regardless of the identity of this sequence, it is still clearthat it is amplified in the more resistant malaria clones andthat, as we reported earlier, it is truly contiguous with the twotubulin isotype genes. Thus, it remains of great interest eventhough we cannot yet identify it.

In an effort to facilitate this, we have sub-cloned uniquefragments of genomic clones that are specific for alpha tubulinand beta tubulin, respectively. This should allow more rapidcharacterization of each gene's context in the genomic DNA.

The evidence that all three genes were contiguous andcoordinately amplified suggested to us that one might be able toderive a multidrug resistant clone from a previously sensitiveclone by selectively pressuring the parasites with agents whichact on either microtubules or calmodulin as opposed +o standardantimalarials. If the pressured gene were amplified in responseto the presence of the agent, then the entire locus of genesshould be co-amplified in a sub-set of the clones. If thisconfers multidrug resistance to, a previously sensitive cellpopulation, one can inferentially implicate these genes in themanifestation of resistance.

Accordingly., we maintained clone D6 under increasingvinblastine pressure, since this agent is highly toxic to theparasites and since it binds preferentially to the microtubulescausing their disassembly. We now have a cell line which canwithstand 5X the normal IC-50 for this compound. Much like ourobservations of the highly mefloquine resistant clone, this linegrows better in vitro, tends toward large vacuole formation, andappears to have lost or suppressed the capacity for gametocyteproduction.

Our vinblastine resistant clone has been examined in moredetail and we have characterized the nature of the resistance.Rather than simply surviving the pressure of the drug in shortbursts with long recovery times, we now have a population whichrecovers in about a week from levels of this agent that shouldkill the entire population. The parasites appear to be veryhealthy and they grow remarkably well compared to the progenitorclone D6 which tends to be somewhat fastidious. This isimportant in the context of resistance to current or futureantimalarials eve' though this compound is not used clinicallyfor the infection. This agent usually acts on Plasmodium in avery toxic and irreversible manner. With sensitive cells, onceyou have applied the drug in vitro, you cannot remove thetoxicity by washingor pulsing With vinblastine analogues. Thus,it is very clear that the parasites have the genetic andbiochemical means to overcome the metabolic lesions induced byeven highly toxic compounds.

These studies have been encouraging in terms of deriving aresistant line, but the results of our drug testing in vitro haveUi but disproved our original hypothesis behind the derivationof this vinblastine resistant population. After testing theresistant parasites in the standard radioactive hypoxanthineuptake assay, we have determined that the cells remain sensitiveto all the antimalarials tested. In addition, the measuredIC-50's are within the range expected for the parental clone D6.This could be explained in a couple of ways although we do not

7

know what the reason is. First, the resistant parasites couldform a minority subset of the entire population and in theabsence of pressure, their numbers may diminish substantially.This would reduce the population IC-50 even though resistantparasites are present. It is also possible that the resistant inthis new line is not based on tubulin gene amplification. All itwould take would be one or a series of point mutations whichsubstantially reduced the binding of the drug to themicrotubules. This has been reported in yeast beta tubulinsequences. It is also possible that there are two pumps for thedifferent drugs or two binding sites on a P-glycoprotein-likemolecule. Changes leading to accelerated efflux of vinblastinemay not affect the way in which the cell handles other agents.

We have also analyzed the amplified genes in the multidrugresistant parasites comparatively with the sensitve clones. Itis evident that one or more genes with sequence homology to the3'-most end of the P-glycoproteins of mouse, human, and hamsterare amplified in these resistant clones. The degree ofamplification parallels that seen with the other genes which areamplified and the quantitative pattern is the same. We believe,but have not proven, that at least one of these P-glycoprotein-like sequences is contiguous with the tubulins and calmodulin-like sequence. If this is correct, we have probably detected an'amplicon' which is selectively increased in copy number inresponse to drug pressure. This makes the idea of intracellularcooperation of these elements in the manifestation of resistanceeven more inviting. It might also prove to be an appropriateregion for which to probe by the polymerase chain reaction orconventional hybridization for a quantitative estimate of drugresistance.

Summary and ConclusionsThe problem of drug-resistant malaria is an acute one from

the military perspective. The absence of effective prophylaxisand treatment represents a major consideration affectingdecisions about troop and personnel deployment into endemic areasevea for short periods of time. Most of the potentially volatilearea3 of the world where the necessity for a U.S. militarypreseace might arise are malarious zones as are many otherregions where advisory and support personnel are presentlysituated or may be stationed in the future. The spreadingprevalence of chloroquine resistance concommitantly withresistance to other antimalarials may necessitate the increasedclinical use of mefloquine in these ,situations, but the parasiteswe will use in these studies illustrate something of thepotential for widespread resistance to this drug as well.Certainly, information about the mechanisms, potential, andextent of resistance would be significant for the future use ofthis drug and as fundamental information about the molecularstrategies employed by the parasites to circumvent the toxicactions of antimalarials in general.

Our results strongly indicate that the parasites have thecapacity to develop very high levels of resistance to variousagents and that multidrug resistance as a phenomenon similar tothat seen with mammalian cells appears to be present in some

parasite lines. As noted in previous reports on this work, theresistant parasite lines which we have derived are exceedinglyhardy. They will outgrow any other strain or line which we havein culture in our laboratory and they will withstandenvironmental assaults (accidental heat shock, endotoxintreatment, etc) more readily than other parasite lines. inaddition, these lines appear to be stimulated by low levels ofmefloquine. This stimulation results in an overall increase inthe number of parasites but we do not know whether this is achange in division rate, a shortened cell cycle, enhancedefficiency of red cell penetration, or the like. What is clearhowever, if one can extrapolate from these findings to what mightoccur in the field setting, parasite strains might arise whichare resistant to prophylactic and therapeutic levels of this

drug, which may be harder to treat once established because oftheir overall hardiness, and which may be stimulated to fasterdevelopment by low levels of the drug as might be achieved bychemoprophylactic use. This is especially likely given thatnaturally resistant strains have been detected in various partsof the world. One potentially positive aspect of this work isthe finding that the Parasite appear in -some cases to developcollateral sensitivity to certain agents upon the development ofhigh levels of mefloquine resistance (particularly, chloroquineand gentamycin).

The discovery of gene amplifications which appear tocorrelate with the degree of resistance suggests that some degreeof coordinate activity among the various genes products may leadto the manifestation of the resistance trait. One couldenvision, for example, the interplay between vesicular transportwithin the cell along microtubule-defined paths, calcium-mediatedenzyme activities,, and a p-glycoprotein-like activity leading tocompartmentalization of the drugs within the cell limiting theirtoxicity and ultimately leading to the enhanced efflux of drugsfrom the cell surface.

In addition, although the specific application to malariaremains speculative, from these studies it may be possible todevelop either DNA probes or antibodies specific for theresistance trait allowing for the rapid identification of thedrug-resistance trait in cells without the necessity for culture.If any or all of these genes turns out to be significantly andconsistently amplified in resistant cells, it may be possible touse the gene we have identified to develop a rapid diagnosticprocedure for the presence of the resistance trait. If thecomponents of the mult-drug resistance network are expressed onthe surface of infected cells as they appear to be in tumorcells, it may also be possible to develop humoral or cellularimmunity to these components by vaccination. Thus thoseparasites not responding to chemotherapy my be cleared moreefficently by immune mechanisms than would normally be the case.This might provide the 'edge' required by the host to curtail thedevelopment of the infection and in so doing, modify its5lethality if not eliminating the parasites altog&ther.

9

LITERATURE REFERENCES:

1. Wernsdorfer, W.H. (1980) The importance of malaria in theworld. In: Malaria, Vol. 1, (Kreier, J., ed.), AcademicPress:New York, pp. 1-93.

2. Peters, W. (1985) The problem of drug resistance in malaria.Parasitology, 90:705-715.

3. Hopkins, D.R. (1985) Vaccine delivery in developingcountries. Am. J. Trop. Med. Hyg., 34:1013-1016.

4. Wolfe, A.D. (1983) Mefloquine. In: Antibiotics, VolumeVI,(Hahn, F.E., ed.), Springer-Verlag:New York, pp. 108-120.

5. Young, R.A. and Davis, R.W. (1983) Efficient isolation ofgenes by using antibody probes. Proc. Natl. Acad. Sci. (USA),80:1194- 1198.

6. Karr, J., Brenner, S. and Barnett, L. (1983) Newbacteriophage lambda vectors with positive selection for clonedinserts. Meth. Enzymol., 101:3-19.

7. Beverley, S.M., Coderre, J.A., Santi, D.V. and Schimke, R.T.( 114) Unstable DNA amplifications in methotrexate-resistantLeishmania consist of extrachromosomal circles which relocalizeduring stabilization. Cell, 38:431-439.'

8. Martin, S.K., Oduola, A.M.J., and Milhous, W.K. (1987)Reversal of Cloroquine Resistance in Plasmodium falciparumby Verapamil. Science, 235:899-901.

9. Scheibel, L.W., Colombani, P.M., Hess, A.D., Aikawa, 1.,Atkinson, C.T. and Milhous, W.K. (1987) Calcium andcalmodulin antagonists inhibit human malaria parasites(Plasmodium falciarum: Implications for drug design. Proc.Natl. Acad. Sci. (USA) 84: 7310-7314.

10. Christian Tschudi, Yale University, Personal Communication.

11. de Bruijn, M.H.L., Van Der Bliek, A.M., Biedler, J.L. andBorst, P. (1987). Differential amplification anddisproportionate expression of five genes in threemultidrug-resistant Chinese hamster ovary cell lines. Mol.Cell. Biol. 12:4717-4722.

PERSONNEL PARTICIPATING IN THIS WORK

Stephen C. Merritt, P.I.Sheila A. Peel, MSPH conferred, Ph.D. candidateVictoria Mann, MSPH, Ph.D. conferredElizabeth Reinoehl, MSPH conferredTamara K. Bennett, Research technician II

10

FIGURE 1

Figure I. Ethidium bromide stained gel comparing W2 and W2-MEF.Lane 1 and Lane 14 - MW marker at 1 kilobase intervalsLane 2 and Lane 3 - W2 and W2-MEF respectively cvt with Alu ILane 4 and Lane 5- W2 and W2-MEF respectively cut with Bam HILane 6 and Lane 7 - W2 and W2-MEF respectively cut with Dra ILane 8 and Lane 9 - W2 and W2-MEF respectively cut with Eco RILane 10 and Lane 11 - W2 and W2-MEF respectively cut with Hind IIILane 12 and Lane 13 - W2 and W2-MEF respectively cut with Pst i

F 1 ",

C444)

Southern blot of falciparum genomic DNA of parasite clone W2,W2-Mef, and D6. Blot was probed with ac;-tubulin-specif ic 1140bp Kpn-Eco RI fragment of trypanosome tubulin repeat. Lanes1, 2. arnd 3 were digested with Dra I while Lanes 4, 5. and 6were cuT. with Mbo I.

FIGURE 3

C 4 0 "44

-4.

!*

4'm

Southern blot of faiciparum genomic DNA of parasite clonesW2. W2-Mef. and D6. Blot was probed with 0-tubulin-specific970 bp trypanosome subclone. Lane 1. 2 and 3 were Dra idigests while Lanes 4. 5. and 6 were cut with Eco RI and Sspi.

p ITUB 1.4

a b c d a 1 4:1 h i k I m n

Southcrn of recombinant phage pf TUB 1. 4 probed with nick-translated Kpn I/Eco RI fragnent of pTB81 specific for a-

ubu 1ifi . The restriction digestions are as follows:Lari( a. Bgl 1i Lane h. Eco RI/Hind IIILayin b, Bg! Il/Dra i Lane i. Eco RI/Ssp ILan c Bgl ll/Ssp I Lane j. Hind IIIuane d. Dra I Lane k. Hind III/Bgl IILani' e. Eco RI Lane 1. Hind ITI/Dra ILan t F Eco RI/Bgl Ii Lane m, Hind II/Ssp ILanei g Eco RI/Dra 1 Lane n. Sin I

pf TUB 1.4

a b c d f g h i j k I m n

9

40

-;uThern of recombinant phage pf TUB 1. 4 probed with nick-translated pP2. a subclone containing the 970 bp Bgl Iifragment of pTB81. The restriction digestions are asfollows:

Lane a. Bgl iI Lane h. Eco RI/Hind IIILane b, Bgl II/Dra i Lane i, Eco RI/Ssp ILane c, Bgl iI/Ssp i Lane j, Hind IIILane d. Dra i Lane k. Hind III/Bgl IiLane e. Eco RI Lane 1. Hind III/Dra ILane f. Eco RI/Bgi Ii Lane m. Hind II/Ssp i.ane g. Eco RI/Dra I Lane n. Sin I

Mef EMBL-10

a b c d,e'

Recc'minant genornic clone Mef EMBL-10 blotted tonitrocellulose and Probed with nick-translated pTB8IcoriLraining tryparnosome A- and a-tubulin genes.

Lane a. Bgl I! Lane d, Kpn iLane b. Eco R- Lane e, Sma 1iLane c. Hind I!

Mef EMBL-1O

a b C d * f g f

i !"I

, I

Lan bBlI aeh p

I

It

Southernl blot analysis of clone Mef EMBL1-IO with 900 bp .

liciparum cDNA clone selected with trypanosome calmodulinpr'obe. Restriction digestions were as follows:

Lane a. Bamn HI Lane g, Hind IIiLane b. Bgl II Lane h. Kpn I

Lane c. Eco Ri Lane i, Kpn I/Hind IIILane d. Eco RI/Bam H! Lane j, Sac IILane e. Eco RI/Hind !I! Lane k. Sac If/Hind IIILane f. Eco RI/Ssp I

FIGURE 8

Duplicate nitrocellulose f ilters of recombinant phagepf TUB 1.4. Replica filter pf TUB 1.4a was screened with nick-translated pTV23 specif ic for A- and w-tubulin and filterpf TUB 1.4b was screened with nick-translated pCAL1, a cDNAfrom P. falcip~aru with strong sequence similarity tocalmodulin.

FIGURE 9

ACTIVITY OF P.DULATION AGENrS ON RESPONSE OF CLONE )6 TO NTIrLAARIALS -

IN VITRO tIICRODILUTION ASSAY ASSESSPEWJT OF REVERSAL

ALONE + VERAPA11L (1 ut) + CAUDDIMOLILF.1 (0.5 utE) + VINBLASTINE (501

(FIFTY PERCETI INHIBITORY CONCENTRATIONS BASED ON HYPOXMNTHINE UPTAKE)DRUG

NGA'L_

CHLOROQUINE 2,1 2.4 1.9 2.3

rEFLOUIRE 26.7 25.9 25.0 26.8

.-IODIAGUINE 0.89 0.92 ND ND

UININE 10.1 9.5 8.5 9.7

QUINACRINE 12.3 B.4 ND ND

NOT DETERMINED

FIGURE 10

ACTIVIYOF IM TION Ai M ON RE NSE OF am 1w210 IMAIAIAL$ -

IN VITRO IaMDILWTION ASSAY ASSES9W OF IFSA

ALNE + WRAPAMIL (1 1D + AI ILIM 10.5 UM) +VIILIASTINE (50,M).

(FIFY PERCET INIIBITORY, W D TIONSSB)W ON lftIW(MIIE IFTME)DRUG

01LOIROINE 90.3 16.8 28.4 25,3

IEFLOWUINE 4,5 5.1 4,3 4,1

AIODIAUIfE 2.8 1.1 NO ND

QUININE 63.5 17.4 21.6 44,2

QUINACRINE 107.6 27.9 N ND

NOT DETERMINED

ACTIVITY OF VMDULATION AGENTS ON RESPOISE OF CLONE 12rEF-2.4 TO /VI ARIALS -IN VITRO- MICRODILUTION ASSAY ASSMItT OF REVERSAL

ALONE + VERAPAIIL (1 ui) + CALfIDIZOLIMl (0.5 UM) + VIrBLASTI:

(FIFTY PERCENT INHIBITORY CO IE"ATIONS BASEDON HYPOXANTHIINE UPTAKE)NG/11t

DRUG

CHLOROQUINE 46,8 11.0 17.1 23.4

[EFLOQJUINE 29.0 6.8 6,2 6.9

AffDDIAQUINE 3.2 1.2 ND ND

QUININE 78.4 14.9 23.5 25.7

QUIlACRINE 111.7 31.7 ND ND

OT IEIB11IB

FIGURE 12

DENSI TE SCAIING DETEIRINATION OF ESIIMED GENE COPY

-1,l0 FOR PLASrWIAL MB IN DRUG EI1TIVE e MILTIDRUG

RESISTNT CL S - IN TERLA VARIATION IN IWIATIVE N1.140

RIATIMEM1 J1OIORN 1 D INIBITIES (SINGLE NYLlON SOURIE]M)PARASITE CLIM LH 'U ... ~A.WN AtUI -LOiJE

D6 1.05 1.15 1.1 1,0

3.4 3.1 16 1.8

12nF-2,4 6.1 6.3 3,4 3.7

SCSNNING WA CARRIED OUT ON AIMORADIOGRA"M USING AN E-C APPIM11U CORP, DENSIIETER WITH INIE TOR, BECAUOF DIFFEFCES IN PROBE LENGTH AND SPECIFIC ACTIVITY, ONLY IMN URK COPARISONS FOR THE SAME GENE ARE VALID.

liESE VALUES DO NOT REPRE T 11E A ME CflPY NUBER PER GENE BIJIVALENr ALTHOUGH TEE ARE, PROBABLY 2ALPHA AND ThO BETA TIBILIN GENES IN EACH HNAP.ID NUCLEUS (MIaISIED OBSEIATIONS).