genetic transformation of coccidioides immitis facilitated by

2106

CONCISE COMMUNICATION

Genetic Transformation of Coccidioides immitis Facilitated by Agrobacteriumtumefaciens

Raed O. Abuodeh,1,2,4,a Marc J. Orbach,1,3

M. Alejandra Mandel,1,3 Anath Das,5

and John N. Galgiani1,2,4

1Valley Fever Center for Excellence; 2Medical and ResearchServices, Southern Arizona Veterans Affairs Health Care System;

and 3Department of Plant Pathology, College of Agriculture,and 4Department of Internal Medicine, College of Medicine, University

of Arizona, Tucson; 5Department of Biochemistry, Molecular Biology,and Biophysics, University of Minnesota, St. Paul

Agrobacterium tumefaciens was used to facilitate genetic transformation of Coccidioidesimmitis. A gene cassette containing the gene encoding hygromycin phosphotransferase (hph)was cloned into a T-DNA vector plasmid and introduced into A. tumefaciens, and the resultantstrain was used for cocultivation with germinated arthroconidia. This procedure producednumerous colonies 60- to 1500-fold more resistant to hygromycin than untransformed mycelia.Both polymerase chain reaction and Southern blot analysis of the transformants indicatedthat all contained hph, usually as a single genomic copy. A transformation frequency of 1 per105 arthroconidia was obtained by varying the germination time prior to cocultivation andaltering the bacterium : fungus ratio. This approach requires no special equipment that mightcomplicate biocontainment. Furthermore, transformation does not require digestion of fungalcell walls, further simplifying this procedure. A. tumefaciens–facilitated transformation shouldmake possible the development of tagged mutagenesis and targeted gene disruption technologyfor C. immitis and perhaps other fungi of medical importance.

Genetic analysis of Coccidioides immitis would undoubtedlycontribute to our understanding of the disease it causes [1].Progress in this regard, however, has been impeded by the lackof methods to transform this fungus efficiently. For example,although Yu and Cole succeeded in introducing DNA into C.immitis by use of a biolistic method [2], their studies producedonly 3 transformants.

We have approached this problem by borrowing a commonlyused technique from plant geneticists. Agrobacterium tumefa-ciens is a Gram-negative bacterium that causes crown galls bytransferring a part of its tumor-inducing (Ti) plasmid DNAinto the plant nuclear genome [3]. The transferred DNA (T-DNA) is flanked by a 24-bp imperfect direct repeat sequenceknown as the border sequence, and the intervening DNA can

Received 13 December 1999; revised 14 February 2000; electronicallypublished 5 June 2000.

Presented in part: 100th annual meeting of the American Society forMicrobiology, Los Angeles, 24 May 2000.

Financial support: US Department of Veterans Affairs; California HealthCare Foundation; Purdue University (grant U.S.D.A. Prime 96-34340-2711).

a Present affiliation: College of Health Sciences, University of Sharjah,Sharjah, United Arab Emirates.

Reprints or correspondence: Dr. John N. Galgiani, Valley Fever Centerfor Excellence (1-111), 3601 S. Sixth Ave., Tucson, AZ 85723 ([email protected]).

The Journal of Infectious Diseases 2000;181:2106–10q 2000 by the Infectious Diseases Society of America. All rights reserved.0022-1899/2000/18106-0041$02.00

be as much as 150 kb in size. A second region of the Ti-plasmid,the vir region, is also essential for DNA transfer. Recently, A.tumefaciens–facilitated gene transfer was extended to filamen-tous fungi [4, 5]. Stimulated by these reports, we tested thisapproach for transforming C. immitis. Our success, combinedwith the relative simplicity of the procedure, makes A. tume-faciens–facilitated transformation appealing for this and per-haps other medically important fungi.

Materials and Methods

Fungal and bacterial strains. Arthroconidia ( or75 3 10) from C. immitis (strain Silveira) [6] were inoculated into85 3 10

100 mL of glucose yeast extract (GYE) liquid medium in 500-mLErlenmeyer flasks and allowed to germinate at 377C with agitationfor 24 h or other time periods, as indicated in the text. Germlingswere collected by centrifugation, resuspended, and used in numbersas indicated in the Results. For selected experiments, spheroplastswere made from germlings by enzymatic digestion of fungal cellwalls. All A. tumefaciens strains used were derivatives of A. tu-mefaciens EHA105 [7].

Plasmid construction. The gene encoding hygromycin phos-photransferase (hph) was cloned into the transforming plasmid in2 steps. The hph cassette was first cloned between T-DNA borders.A second fragment was then added that contained a wide-host-range replicon and a mutant virG gene to allow plasmid mainte-nance in A. tumefaciens and inducer-independent expression of thevir genes, respectively [8]. These steps were carried out as follows.

The plasmid pAD1310 is an intermediate Ti-plasmid vector that

Dow

nloaded from https://academ

ic.oup.com/jid/article/181/6/2106/2191691 by guest on 17 D

ecember 2021

JID 2000;181 (June) Transformation of C. immitis with A. tumefaciens 2107

contains the octopine Ti-plasmid pTiA6 left and right borderregions of the T-DNA. Plasmid pAD1308 was constructed by de-leting the EcoRI-BamHI polylinker region of plasmid pUC118.pAD1310 was constructed by cloning the 3.5-kb HindIII fragmentof pEndI [9] into plasmid pAD1308.

For the construct used in most studies, the hph cassette wasderived from plasmid pMP6 [10]. This cassette, as a 2.9-kb HindIII fragment, contained a modified cpc-1 promoter of Neurosporacrassa, the hph gene from Escherichia coli, and the Aspergillus ni-dulans trpC terminator [10]. After its ends were filled in, using theKlenow enzyme, the hph cassette was cloned into the filled-inBamHI site of plasmid pAD1310 to produce pRATi2.9. For com-parison, another hph cassette was derived from plasmid pCB1004and contained the E. coli hph gene under the control of the A.nidulans trpC promoter [11]. This was isolated as a 1.4-kb HpaIfragment and cloned into the filled-in BamHI site of plasmidpAD1310 to produce pRATi1.4.

After pRATi2.9 and pRATi1.4 were constructed, a 7.5-kb frag-ment from plasmid pAD1378 [12] was inserted into each. Thisfragment contained the wide-host-range replicon 75D2 and the con-stitutive vir gene mutant virGN54D [8]. The 7.5-kb EcoRI fragmentof pAD1378 was cloned into the SalI site of plasmid pRATi2.9and the XbaI site of plasmid pRATi1.4 to construct plasmidspAD1625 and pAD1624, respectively. Plasmids pAD1625 andpAD1624 were introduced into A. tumefaciens EHA105 by elec-troporation [13] to construct A. tumefaciens A974 and A973,respectively.

Procedure for transforming C. immitis. A. tumefaciens A973or A974 was grown overnight in 2 mL of agrobacterium broth(AB) liquid medium with carbenicillin and tetracycline at 287C withshaking, to A600 1.0–1.5. An aliquot (0.25 mL) of this culture wasdiluted 15–20-fold in induction medium and grown for another 24h, until an A600 of 0.5–1.0 was obtained. The bacteria were collectedby centrifugation at 4000 rpm for 10 min, washed once, and re-suspended in 0.5 mL of induction medium (AB medium plus 25mM 2-[N-morpholino]ethanesulfonic acid, pH 5.8).

Arthroconidial germlings were mixed in sterile microfuge tubeswith A. tumefaciens in varying ratios. The cell mixtures were cen-trifuged briefly, the supernatant was removed, and the pellet wasdispersed with a sterile spatula and placed on sterile 0.45-mm ni-trocellulose filters on solid induction medium and incubated for 2days at room temperature.

Selection of transformants. After incubation, filters were trans-ferred to tubes with 1.0 mL of normal saline. Cells were dislodgedand were plated on GYE agar containing 200 mM cefotaxime or50 mg/mL kanamycin for counterselection against A. tumefaciensand containing hygromycin at concentrations as indicated in Re-sults. Plates were incubated at 377C until discrete colonies emerged.They were enumerated, and representative colonies were subcul-tured for further study.

Analysis of transformants. Genomic DNA [6] was analyzed bypolymerase chain reaction (PCR), using primers specific for hphor for a specific gene of C. immitis [6]. The 2 reactions were carriedout simultaneously. Southern blot analyses were performed withgenomic DNA from putative transformants and untransformed C.immitis. The 4 probes used were the hph gene cassette, the p-Bluescript vector, and the T-DNA left and right border fragments.

Results

Transformation. Cocultivation of A. tumefaciens A973 orA974 with C. immitis germlings (grown for 24 h), with71 3 10or without prior spheroplasting, resulted in the growth of dis-crete fungal colonies on selection plates containing hygromycin(10 mg/mL). Transformation with A. tumefaciens A973 pro-duced 22 colonies (10 from spheroplasts and 12 from non-spheroplasted germlings), and that with A974 produced 38 col-onies (26 from spheroplasts and 12 from nonspheroplastedgermlings). When subcultured to solid medium containinghigher concentrations of hygromycin, all putative transfor-mants were resistant to >150 mg/mL, and 46 were resistant to500 mg/mL (the highest concentration tested). All 14 putativetransformants with lower resistance (150 or 300 m/mL hygro-mycin) were derived from cocultivation with A. tumefaciensA973, in which hph is under the control of the trpC promoter.In control experiments, spontaneous resistance to hygromycin(2.5 mg/mL) was not observed.

Mitotic stability of all 60 putative transformants was testedby growth for 3 days in liquid GYE with hygromycin (10 mg/mL, 377C with shaking), subculture onto GYE agar withouthygromycin for 2 weeks, and screening for resistance by growthon GYE agar containing 150, 300, or 500 mg/mL of hygro-mycin. The level of resistance to hygromycin after this proce-dure was found to be identical to that obtained previously.

Molecular characterization of transformant strains. Tostudy whether the hph gene was present in the transformants,genomic DNA from 23 putative transformants, obtained usingA. tumefaciens A974, was analyzed by PCR. A band specificfor hph was observed with DNA from all transformants butnot with DNA from untransformed C. immitis (figure 1A).

Southern blot hybridization analysis was performed to studythe nature of the transforming DNA insertions. If transfor-mation was due to mobilization of the T-DNA from A. tu-mefaciens, followed by insertion into the C. immitis genome,only those sequences between the T-DNA borders will be trans-ferred. The simplest fate of the transferred DNA would be asingle insertion of the T-DNA, as diagrammed in figure 1C.Although all transformant DNAs hybridized to the hph geneprobe, none exhibited homology to plasmid vector sequences(data not shown), which indicates that transformation occurredvia T-DNA–mediated transfer of DNA. With the hph geneprobe (figure 1B), 17 of the 21 samples in which the EcoRIdigestion was successful had 2 bands, as would be expectedwith a single T-DNA insertion (figure 1C). One of the bandswas a 2.3-kb internal T-DNA fragment that included most ofthe hph gene and the terminator (figure 1C). The second bandin each transformant represents a junction fragment that in-cludes the cpc-1 promoter, the T-DNA right border, and ad-jacent C. immitis genomic DNA. The variability in the size ofthis fragment is due to differences in the size of flanking ge-nomic DNA. That the junction fragments vary in size amongthe different transformants indicates that the transforming

Dow



ecember 2021

Figure 1. Molecular analysis of hygromycin-resistant transformants of Coccidioides immitis. A, Polymerase chain reaction (PCR) analysis oftransformant DNAs. DNA of the untransformed recipient strain Silveira and 23 hygromycin-resistant strains was subjected to PCR analysis usingoligos (forward, 5′-AGTGTCACGTTGCAAGACCTGCCA-3′; reverse, 5′-GTGACGGTGTCGTCCATCACAGTA-3′) that should generate a 221-bp hygromycin phosphotransferase (hph) gene band (open arrowhead) and oligos (forward, 5′-CGTTCTCGTCCGTTAGAC-3′; reverse, 5′-CA-GCACCTGGGAACTCAGCT-3′) that should generate a 731-bp band from the gene encoding a proline-rich antigen specific for C. immitis(closed arrowhead). Strains were generated through cocultivation of Agrobacterium tumefaciens with 24-h C. immitis germlings, either without(lanes 1–12) or with (lanes 13–23) enzymatic digestion of the fungal cell wall. The 50-mL PCR reaction solution contained 0.2 mM of each dNTP,1.5 mM MgCl2, 1 nM of each primer, 0.5 U Taq-2000 polymerase (Statagene, La Jolla, CA), and 100 ng of template DNA. The reaction conditionswere 947C for 3 min (947C for 1 min, 557C for 2 min, and 727C for 2 min), 35 cycles, and 727C for 7 min. B, Probing of hygromycin-resistantC. immitis strain DNAs with the hph gene. DNAs of the same 23 hygromycin-resistant strains in panel A were digested with EcoRI, electrophoresed,and transferred to a nylon membrane. The DNAs in lanes 4 and 6 failed to be digested. The filter was probed with an hph gene probe. Probeswere labeled with 32P dCTP. The DNAs are in the same order as in panel A. C, Diagrammatic representation of a transferred DNA (T-DNA)insertion in a C. immitis chromosome. The thin lines represent C. immitis genomic DNA, and the thick black line represents T-DNA sequences.The left border (LB) and right border (RB) are indicated by boxes. The hygromycin-resistance cassette is likewise indicated by boxes: the Neurosporacrassa cpc-1 promoter (Pcpc-1) and the Aspergillus nidulans trpC terminator (TtrpC) are indicated by open boxes, and the hygromycin-resistancegene is indicated by a cross-hatched box. Restriction enzyme sites indicated include EcoRI (E), BglII (Bg), BamHI (B), and HindIII (H). Thevariability in the positions of the C. immitis–flanking EcoRI sites is represented by the zigzag between them and the T-DNA ends. The linesbelow the map indicate the size of the predicted bands from the hph gene (as an HpaI fragment of pCB1004) and the LB and RB probes, whichwere isolated from pAD1310. The LB probe is a 1.6-kb EcoRI-HindIII fragment that corresponds to nucleotides 602–2212 of T-DNA (GenBankaccession no. X00493), and the RB probe is a 0.3-kb BamHI-SacI fragment that corresponds to nucleotides 13780–14090 of the T-DNA.

Dow



ecember 2021

JID 2000;181 (June) Transformation of C. immitis with A. tumefaciens 2109

Table 1. Effect of mycelial germination times and the ratio of bacteria to germlings on thefrequency of hygromycin-resistant colonies.

Mycelialgrowth beforecocultivation (h)

Arthroconidia ofCoccidioides immitis

(3 107)

Agrobacteriumtumefaciens

(3 107)

No. of hygromycin-resistant colonies

of C. immitis

Transformationfrequency

(3 106 arthroconidia)

24 0.5 None 0 05 0.5 10 0 0

24 0.5 10 2 0.448 0.5 10 51 10.224 1.0 None 0 024 1.0 1 4 0.424 1.0 10 4 0.424 1.0 100 11 1.124 1.0 500 40 4.0

NOTE. Germlings were cocultivated with or without A. tumefaciens and subcultured onto agar containing20 mg/mL of hygromycin.

DNA integrated into diverse sites within the C. immitis genome.The hybridization patterns of 4 transformants gave results sug-gesting a more complex integration event. In 2 of them, 2 bandswere present, but the 2.3-kb internal band was absent, whichsuggests a rearrangement of the T-DNA (figure 1B, lanes 1 and9). Two other transformants had 12 hybridizing bands, whichsuggests that 11 copy of the T-DNA had been inserted. Thehph probe did not hybridize to DNA from untransformed C.immitis (figure 1B).

Hybridization with the right border fragment identified, ineach transformant, >1 bands that comigrated with the variableband(s) seen with the hph gene probe (data not shown). Hy-bridization with the left border probe (figure 1C) also indicatedthat the majority of transformants had a single T-DNA in-sertion, with others having a more complex pattern (data notshown).

Effects of other factors on transformation frequency. Bothgermination time and the ratio of A. tumefaciens to germlingshad a major effect on the transformation frequency (table 1).Arthroconidia germinated for 5 h prior to cocultivation pro-duced no transformants, whereas those germinated for 24 hand 48 h prior to cocultivation produced 2 and 51 colonies,respectively, from cells resulting from germinating 65 3 10arthroconidia. Also evident was that changing the germ-ling : bacteria ratio from 1 : 1 to 1 : 500 led to a 10-fold increasein transformation efficiency.

Discussion

In this report, we demonstrate a simple and efficient methodfor transforming C. immitis by use of A. tumefaciens–mediatedDNA transfer. Transformation with A. tumefaciens requires aminimum of manipulations and no special equipment. Thesefeatures make the technique very attractive for studies with C.immitis and potentially with other pathogenic fungi that mustbe handled with special biocontainment procedures. The plas-mid constructs that we used had 2 mutations, one that permitsconstitutive expression of all vir genes and another that allowshigh plasmid copy numbers [8]. These features improved trans-

formation efficiency in plants [14] and may be responsible forthe greater frequency of transformants and higher degree ofresistance produced by A974. A. tumefaciens A974 producedgreater hygromycin resistance than did A973. From other stud-ies, we expect this finding is due to differences in promotersused to construct the hyg cassettes [10]. In most transformants,Southern blot analysis demonstrated that the transformingDNA inserted into the C. immitis as a single T-DNA fragment,and, in all cases, transformation resulted in stable genomicintegration of hyg and phenotypic resistance to hygromycin.

A significant increase in transformation efficiency was ob-served by either extending the germination time or increasingthe ratio of bacteria to mycelia. Cultures that were germinatedfor 48 h resulted in mycelial mats that were difficult to disperseinto homogeneous suspensions. Therefore, germinating arthro-conidia for 24 h may be preferable. On the other hand, we didnot determine a maximum ratio of A. tumefaciens to germlings.It is possible that the use of more bacteria may further improvetransformation efficiency.

In addition to the introduction of foreign genes, use of A.tumefaciens–facilitated transformation can facilitate other mo-lecular studies. For example, this approach could be the basisof insertional mutagenesis gene tagging in C. immitis, as hasbeen done with other fungi [15]. Also, as with other fungi, theaddition of homologous sequences to the transforming DNAmay allow targeted gene disruption [3, 5]. Mycelia of C. immitisappear to be haploid [16]. Thus, single disruptions could pos-sibly produce altered phenotypes. Because both of these ap-plications could be useful to the study of other pathogenicfungi, our findings may not be restricted solely to C. immitis.

Acknowledgments

We thank Maria Lourdes Lewis (Research Service, Southern ArizonaVeterans Affairs Health Care System, Tucson) for maintenance of fun-gal cultures and production of genomic DNA. We thank also Dr. S.Gelvin (Department of Biological Sciences, Purdue University, WestLafayette, IN) for providing A. tumefaciens strain EHA105 for use inour studies.

Dow



ecember 2021

2110 Abuodeh et al. JID 2000;181 (June)

References

1. Galgiani JN. Coccidioidomycosis: a regional disease of national importance;

rethinking approaches for control. Ann Intern Med 1999;130:293–300.

2. Yu J-J, Cole GT. Biolistic transformation of the human pathogenic fungus

Coccidioides immitis. J Microbiol Methods 1998;33:129–41.

3. Das A. DNA transfer from Agrobacterium to plant cells in crown gall tumor

disease. In: Biswas BB, Das A, eds. Plant microbe interactions. London:

Plenum, 1998:343–63.

4. de-Groot MJ, Bundock P, Hooykaas PJ, Beijersbergen AG. Agrobacterium

tumefaciens–mediated transformation of filamentous fungi. Nat Biotech-

nol 1998;16:839–42.

5. Gouka RJ, Gerk C, Hooykaas PJJ, Bundock P, Musters W, Verrips CT,

et al. Transformation of Aspergillus awamori by Agrobacterium tume-

faciens–mediated homologous recombination. Nat Biotechnol 1999;17:

598–601.

6. Peng T, Orsborn KI, Orbach MJ, Galgiani JN. Proline-rich vaccine candidate

antigen of Coccidioides immitis: conservation among isolates and differ-

ential expression with spherule maturation. J Infect Dis 1999;179:518–21.

7. Ritchie S, Lui C-N, Sellmer J, Kononowicz H, Hodges T, Gelvin S. Agro-

bacterium tumefaciens-mediated expression of gusA in maize tissues.

Transgenic Res 1993;2:252–65.

8. Gubba S, Xie Y-H, Das A. Regulation of Agrobacterium tumefaciensvirulence

gene expression: isolation of a mutation that restores virGD52E function.

Mol Plant Microb Interact 1995;8:788–91.

9. Klee H, Yanofsky M, Nester EW. Vectors for transformation of higher plants.

Bio/technology 1985;3:637–42.

10. Orbach MJ. A cosmid with a HyR marker for fungal library construction

and screening. Gene 1994;150:159–62.

11. Carroll AM, Sweigard JA, Valent B. Improved vectors for selecting resistance

to hygromycin. Fungal Genet Newslett 1994;41:22–22.

12. Das A, Anderson L, Xie Y-H. Delineation of the interaction domains of

Agrobacterium tumefaciens VirB7 and VirB9 by use of the yeast two-hybrid

assay. J Bacteriol 1997;179:3404–9.

13. Mersereau M, Pazour G, Das A. Efficient transformation of Agrobacterium

tumefaciens by electroporation. Gene 1990;90:149–51.

14. Hansen G, Das A, Chilton M-D. Constitutive expression of the virulence

genes improves the efficiency of plant transformation by Agrobacterium.

Proc Natl Acad Sci USA 1994;91:7603–7.

15. Brown JS, Holden DW. Insertional mutagenesis of pathogenic fungi. Curr

Opin Microbiol 1998;1:390–94.

16. Reichard U, Yu J-J, Seshan KR, Cole GT. Transformation of Coccidioides

immitis and molecular evidence for haploidy of this fungal pathogen [ab-

stract F22]. In: Proceedings of the general meeting of the American Society

for Microbiology, Chicago, 30 May–3 June 1999. Herndon, VA: ASM

Press, 1999:299.

Dow



ecember 2021

genetic transformation of coccidioides immitis facilitated by

Documents