Double-Stranded RNA Virus in the Human Pathogenic ... - NCBI - NIH

4 downloads 15 Views 1MB Size Report
Aug 1, 1994 - infection has been tenuous at best. Reports ... The viral particles (VPs) were isolated as described ... The VPs from strain 6071, therefore, were.

Vol. 68, No. 11

JOURNAL OF VIROLOGY, Nov. 1994, p. 7554-7558

0022-538X/94/$04.00+0 Copyright © 1994, American Society for Microbiology

Double-Stranded RNA Virus in the Human Pathogenic Fungus Blastomyces dermatitidis SHIGERU

KOHNO,1t TSUTOMU FUJIMURA,2 SHEN RULONG,3 AND K. J. KWON-CHUNGl*

Clinical Mycology Section, Laboratory of Clinical Investigation, National Institute ofAllergy and Infectious Diseases,' and Section of Genetics on Simple Eukaryotes, Laboratory of Biochemical Pharmacology, National Institute of Diabetes and Digestive and Kidney Diseases,2 Bethesda, and Membrane Biology Section, Laboratory of Mathematical Biology, National Cancer Institute-Frederick Cancer Research Center, Frederickl3 Maryland Received 20 April 1994/Accepted 1 August 1994 Double-stranded RNA viruses were detected in a strain of Blastomyces demaitidis isolated from a patient in Uganda. The viral particles are spherical (mostly 44 to 50 nm in diameter) and consist of about 25% double-stranded RNA (5 kb) and 75% protein (90 kDa). The virus contains transcriptional RNA polymerase activity; it synthesized single-stranded RNA in vitro in a conservative manner. The newly synthesized single-stranded RNA was a full-length strand, and the rate of chain elongation was approximately 170 nucleotides per min. The virus-containing strain shows no morphological difference from virus-free strains in the mycelial phase. Although the association with the presence of the virus is unclear, the virus-infected strain converts to the yeast form at 37°C, but the yeast cells fail to multiply at that temperature.

isolates, 7 from Africa, 5 from North America, and 1 from India, were used in the study. The isolates either were obtained from the American Type Culture Collection or were from the culture collection of our laboratory. The African isolates were 6066 (South Africa), 6071 (Uganda), ATCC 48089 (Zaire), ATCC 56214 (Mozambique), ATCC 56217 (South Africa), ATCC 56218 (Algeria), and ATCC 56220 (Angola). The five American isolates were ATCC 18187 (Wisconsin), ATCC 14112 (?), 6059 (?), B-3227 (Tennessee), and B3228 (Arkansas). One isolate from India, ATCC 48938, was from the lung of a bat. Nucleic acids were extracted from mycelial cultures (28) grown in YEPD (1% yeast extract, 2% peptone, 2% glucose) broth at 30°C on a shaker for 10 days and electrophoresed in 1% agarose gel. Three African isolates, 6071, ATCC 56217, and ATCC 56220, showed extrachromosomal bands of 5 kb or less on an ethidium bromide-stained gel (not shown). No such bands were observed for other isolates. The largest and most prominent band (5 kb) was in strain 6071, which was isolated from a patient in Uganda (not shown). To determine the nature of the band, an aliquot of the nucleic acid sample from 6071 was electrophoresed with or without RNase and DNase treatment. The band disappeared with RNase treatment while it remained in the sample treated with DNase, suggesting that the band is RNA. To test whether the RNA is single stranded (ss) or ds, the nucleic acid sample from strain 6071 was treated with pancreatic RNase A under high and low salt concentrations. The RNA band persisted in the high-salt condition even if the RNase concentration was 1 mg/ml. Under the low-salt condition, however, the band disappeared within 30 min (not shown) with addition of 0.1 mg of RNase per ml (not shown). Resistance of pancreatic RNase in a high-salt condition is characteristic of dsRNA. Purification and characterization of the viral particles. We suspected that strain 6071 might be infected with a dsRNA virus, as there is a wide range of reports on virus-infected fungi (21). The viral particles (VPs) were isolated as described previously (11) with modifications. Mycelial mats of 6071 harvested by filtering 10-day-old broth cultures through a Whatman filter (no. 1) were washed with buffer A (1 M sorbitol, 0.1 M Tris-Cl [pH 7.6], 1 mM EDTA), resuspended in the same buffer with 48 mM 2-mercaptoethanol, and incubated

The occurrence of virus has been reported in over 60 species from more than 50 genera of true fungi. The majority of these fungi are plant pathogens and saprophytes, and only a few species are known to cause disease in humans (21, 34). In human pathogens, however, proof of the occurrence of viral infection has been tenuous at best. Reports have been based on the observation of virus-like particles on electron microscopy (1) or on the production of killer toxin (15, 16, 24). Virus-like particles were isolated from a strain of Aspergillus flavus, but biophysical characterization failed to detect any nucleic acid (35). Blastomyces dermatitidis is a dimorphic fungus that causes blastomycosis, which is most prevalent in the southeastern United States and the Ohio-Mississippi river valleys (20). The disease is also known to be endemic to wide geographic areas of Africa. The African isolates of B. dermatitidis are antigenically different from American isolates (17), and in tissue, the yeast cells of North American strains are mostly spherical while those of African isolates tend to be more ovoid (20). The conversion to the yeast form in vitro is much more readily achieved with North American than with African isolates (20). African isolates and American isolates fail to produce teleomorphs (the Ajellomyces state) upon sexual cross (19). Although this trend may be associated with the host rather than the fungus, Segretain noted that patients with African blastomycosis tend to develop more subcutaneous abscesses and fewer ulcerated cutaneous lesions than those with the North American disease (26). During our comparative study of rDNA restriction fragment length polymorphism between seven African isolates and five American isolates of B. dermatitidis, we detected doublestranded (ds) RNA virus in one of the African isolates. To our knowledge, this is the first report of the purification and characterization of a virus from fungi pathogenic to humans and animals. Detection of a dsRNA in B. dermatitidis. A total of 13 * Corresponding author. Mailing address: LCI/NIAID, Bldg. 10, 11C 304, NIH, Bethesda, MD 20892. t Present address: Second Department of Internal Medicine, Nagasaki University Medical School, Nagasaki, Japan.

7554

NOTES

VOL. 68, 1994

for 15 min at 37°C. The mycelia were then treated with a mixture of Zymolyase-20T (1.7 mg/ml; ICN Biochemicals Inc., Costa Mesa, Calif.) and mureinase (3.3 mg/ml; United States Biochemical Corp., Cleveland, Ohio) for 2 h at 30°C (5). The enzyme-treated sample was centrifuged for 5 min at 4,500 x g (4°C), and the pellet was suspended in an equal amount of buffer B (50 mM Tris-Cl [pH 7.6], 150 mM NaCl, 10 mM MgCl2, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride) and then vortexed with an equal volume (usually 5 ml) of sterile glass beads (0.45 mm) for 2 min at 4°C. The solution was then centrifuged for 1 min at 4,500 x g, 4°C, and the supernatant collected was centrifuged for15 min at 12,000 x g, 4°C. After the cell debris was removed, the VPs were pelleted by centrifugation for 2 h at 110,000 x g. The pellet was suspended in buffer B and centrifuged at 13,000 x g for 15 min. Cesium chloride was added to the supernatant (1.35 g/ml) and centrifuged for 20 h at 160,000 x g. The samples were fractionated (0.8 ml) in 14 tubes, and all fractions were dialyzed for 3 h against buffer C (50 mM Tris-Cl [pH 7.6], 250 mM NaCl, 5 mM EDTA [pH 8.0]) containing 20% (vol/vol) glycerol. The dsRNA content of each fraction was examined by electrophoresis in an agarose gel stained with ethidium bromide. To detect viral protein, each fraction was dissolved in electrophoresis loading buffer (1% sodium dodecyl sulfate [SDS], 10% glycerol, 10 mM Tris [pH 7.6], 0.005% bromophenol blue, 1% 2-mercaptoethanol) and boiled for 3 min. Electrophoresis was carried out in a 7.5% acrylamide gel with 0.1% SDS, and protein was stained with Coomassie brilliant blue. In this test for the presence of dsRNA as well as proteins, fraction 5, with a density of 1,425 g/cm3, was the only fraction that showed the 5-kb RNA band on an ethidium bromide-stained gel (Fig. 1A) and two major protein bands on an SDSpolyacrylamide gel electrophoresis (PAGE) gel (Fig. 1C). The protein bands were 90 and 81 kDa with similar densities. In a repeat experiment, however, the 90-kDa band was clearly denser than the 81-kDa band. This suggests that the 81-kDa protein is a degradation product of the 90-kDa protein. Fraction 5 was also the only fraction that produced a single peak of RNA polymerase activity, as shown in Fig. 1B (see below). A pellet obtained from fraction 5 by ultracentrifugation was subjected to electron microscopic study. The pellet contained numerous spherical particles, mostly 44 to 50 nm in diameter (Fig. 2). The particles consisted of a shell surrounding an electron-dense core, which is typical of mycoviruses found in various filamentous fungi and yeasts (4). Composition of VPs. The content of dsRNA was estimated according to the following equation (22): F = Bn/Bv [(Bv Bp)/(Bn - Bp)], where Bn = the density of RNA in CsCl (1.9 g/ml) (27), Bp = the density in CsCl of VP protein (empty capsid, 1.31 g/ml) (25), and Bv = the density of VPs in CsCl (1.425 g/ml). The VPs from strain 6071, therefore, were estimated to have 25% dsRNA and 75% protein by weight. If we assume that the virions contain one 5-kb dsRNA per particle, the molecular masses of the dsRNA and protein per particle can be calculated as 3.5 x 106 and 10.5 x 106 Da, respectively. From the latter value, it can be estimated that the virions contain about 118 molecules of the 90-kDa coat protein per particle (assuming that the 81-kDa protein is a degradation product of the 90-kDa protein). These features are quite similar to that of Saccharomyces cerevisiae L-A viruses. The L-A virus contains one 4.6-kb dsRNA and about 120 copies of the major coat protein (76 kDa) per particle (7). The L-A virion structure is icosahedral, with T = 1, and a dimer of the major coat protein forms the asymmetric unit as deduced from cryoelectron micrography and image reconstruction (27a). Sucrose gradient centrifugation revealed that the sedimenta-

7555

A

Top

Bottom

2 3 4 5

6 7 8 9 10 11 12 13 14 Fraction No

2 3 4 5

6

B 14

12

I 10 10

CD

1

7

8 9

10 11 12 13 14

Fraction No

97 66 A

* .

1

2 3 4 5

6

;*,*

W:AS

7 8 9 Fraction No.

toobeF J.

10 11 12 13 14

FIG. 1. CsCl gradient fractions showing VPs from isolate 6071. (A) RNA was extracted from each fraction of the CsCl gradient (total, 14

fractions), separated in an agarose gel, and stained with ethidium bromide. Fraction 1 corresponds to the bottom of the gradient. Only fraction 5 contained an RNA band of 5 kb. (B) RNA polymerase activity in the same fractions. (C) Coomassie brilliant blue-stained SDS-PAGE gel of the protein showing two major bands of 90 and 81 kDa, most markedly in fraction 5.

tion coefficient of the VPs isolated from fraction 5 was 180S as opposed to 160S for the L-A virus. From their sedimentation coefficient (180S) and the diameter of the virus (44 to 50 nm), one can calculate (22) that the virions have a mass of about 15 X 106 Da, in good agreement with the value obtained above. Therefore, the virus from B. dernatitidis consists of one 5-kb dsRNA and about 120 copies of the coat protein per particle. RNA polymerase activity of the VPs. After CsCl equilibrium density gradient centrifugation, aliquots of each fraction were incubated with an RNA polymerase reaction mixture and the trichloroacetic acid-precipitated reaction products were measured. Three microliters of VP samples was used for an RNA

7556

J. VIROL.

NOTES 1

2

1

2

1 2

4wl

:

FIG. 2. Electron micrograph of the VPs isolated from CsCl gradishowing spherical particles with a shell surrounding an electron-dense core (bar = 100 nm). After being washed with phosphate-buffered saline (pH 7.0) and fixed with 1.5% gluteraldehyde, the pellet on Formvar and carbon-coated copper grids was negatively stained with either 2% uranyl acetate or 2% phosphotungstate (pH 7.0) (14). Micrographs were taken at a magnification of x70,000 on a Philips 410 electron microscope.

.

ent fraction 5

polymerase assay. The RNA polymerase reaction mixture (33) contained 50 mM Tris-Cl (pH 7.6); 5 MM MgCl2; 0.1 mMM EDTA; 20 mM NaCl; 5 mM KCI; 10 mM 2-mercaptoethanol; 40 mg of bentonite per ml (9); 0.5 mM (each) ATP, CTP, and GTP; and 20~z.M [ct_32p]UTP (New England Nuclear, Boston, Mass.). The reaction mixture (25 LIi) was incubated for 1.5 h at 30'C. A peak of RNA polymerase activity was found in fraction 5, which contained VPs (Fig. iB). The phenol-extracted RNA polymerase product of the VPs was ss as judged by its sensitivity to pancreatic RNase in a high-salt condition (Fig. 3). Furthermore, time course and pulse-chase experiments showed that, with increasing incubation time, the ssRNA products grew in length, reaching full length by 30 min, but that the dsRNA template was not labeled during the experiments (not shown). These results indicate that the ssRNA is transcribed in a conservative manner. We measured the length of ssRNA transcripts in glyoxal denaturing gels (23). The in vitro transcripts comigrated with genomic RNA labeled with [32p]pCp and T4 RNA ligase, indicating full length (not shown). Thus, the average rate of chain elongation in ssRNA synthesis in vitro calculated from these data is about 170 nucleotides per min. To test whether the ssRNA synthesized in vitro is homologous to the dsRNA of VPs, two Northern (RNA) experiments were performed according to the previous 1

2 3

4 5

FIG. 3. Effect of RNase treatment on the RNA synthesized in vitro by the VPs from 6071 . 32p-labeled RNA products were made in vitro by the VPs and treated with RNase A under high- or low-salt conditions (11). After treatment, the products were phenol extracted and analyzed in an agarose gel. An autoradiogram of the gel is shown. Lanes 1, no RNase; 2, 20 p.g of RNase per ml under a high-salt condition; 3, 4 pRg of RNase per ml under a high-salt condition; 4, 20 p.g of RNase per ml under a low-salt condition; 5, 4 p.g of RNase per ml under a low-salt condition.

A B C FIG. 4. dsRNA of the VPs from isolate 6071 (lanes 1) and a mixture of the L-A dsRNA and the positive-strand ssRNA (lanes 2) were separated in an agarose gel. Ethidium bromide staining of the gel is shown in panel A. After electrophoresis, the RNAs were denatured in the gel, blotted onto a nylon membrane, and hybridized with either 32P-labeled polymerase products of the VPs from 6071 (B) or 32p_ labeled L-A positive-strand-specific probe (C). Hybridization was detected by autoradiography. Panel A contains HindIll-digested lambda DNA marker in the first lane. method (10) with minor modifications; the dsRNA from purified VPs was separated in an agarose gel (1.8%) and denatured by incubating the gel in 5% formamide-9% formaldehyde-10 mM MOPS (morpholinepropanesulfonic acid) (pH 7.0)- 1 mM EDTA for 30 min at 55°C. The gel was then washed for 20 min in 18x SSC (lx SSC is 0.15 M NaCl plus 0.015 M sodium citrate) (pH 7.0), and the RNAs were transferred onto a Nytran membrane (Schleicher & Schuell, Keene, N.H.). The 32P-labeled ssRNA of the RNA polymerase product as described above was used as a probe. We also included the L-A dsRNA and the plus-strand ssRNA as a control. The L-A plus-strand-specific probe was made by T3 RNA polymerase with PvuII-cut pLM1 (12). The blotted RNAs were hybridized with either 32P-labeled in vitro transcripts from the VPs or the L-A plus-strand-specific probe. Following hybridization, the membrane was washed twice with 6x SSPE (lx SSPE is 0.18 M NaCl, 10 mM NaH2PO4, and 1 mM EDTA [pH 7.7])-0.2% SDS for 15 min and once with 1 x SSPE-0.2% SDS at room temperature. The final wash was with the same buffer for 1 h at 55°C. As shown in Fig. 4, the 5-kb dsRNA from purified VPs hybridized only with the in vitro transcripts and not with the L-A probe. On the other hand, the L-A specific probe detected denatured dsRNA of the L-A as well as the L-A plus-strand ssRNA but not the 5-kb dsRNA. These results confirmed that the ssRNA was synthesized by the VPs. When the same radiolabeled in vitro transcripts were used as a probe for hybridization with the RNAs extracted from the original 13 isolates, the probe hybridized only with the 5-kb dsRNA from strain 6071 and not with the smaller bands (