Restriction analysis of the prototype strain of enteric adenovirus type 41 using exonuclease III

Journal of Virological Methods, 38 (1992) 25-38 0 1992 Elsevier Science Publishers B.V. / All rights reserved / 0166-0934/92/$05.00

25

VIRMET 01328

Restriction analysis of the prototype strain of enteric adenovirus type 41 using exonuclease III Tim H. Scott-TayloriT

Gurmukh

Ahluwalia2

and Gregory

W. Hammond’,2

‘Department of Medical Microbiology, University of Manitoba, and 2Cadham Provincial Laboratory, Winnipeg, Manitoba, Canada (Accepted

1 I December

1991)

Summary Enteric adenoviruses 40 and 41 (Ad40 and Ad41) are a prominent cause of gastroenteritis in young children. Diagnosis of these enteric types by conventional means is complicated by their fastidious growth characteristics. Enteric adenovirus growth was enhanced by cocultivation. Typing of enteric isolates currently entails analysis of the extracted viral DNA with restriction enzymes. Restriction endonuclease fragments of the Ad41 strain Tak genome were ordered by (i) double digestion, (ii) release of restriction fragments from plasmids containing 84% of the Ad41 genome in EcoRI fragments A, B and C, (iii) hybridization of Southern blotted Ad41 fragments with EcoRI fragment containing plasmids and various segments of the Ad2 genome, (iv) sequential reduction of the genome beginning with terminal restriction fragments with exonuclease III and Sl nuclease. The termini of adenovirus genomes are difficult to clone and the use of exonuclease III is a useful alternative to conventional restriction mapping. DNA restriction patterns, fragment sizes and restriction maps of the Ad4 1 strain Tak with enzymes BumHI, BgLII, CM, EcoRI, HindIII, PstI, &n, SmaI and XhoI are presented. Prototype strain restriction maps should enable better understanding of adenovirus type 41 and its epidemiology. Restriction analysis; Enteric adenovirus type 41; Exonuclease III

Correspondence to: G.W. Hammond, Department of Medical Microbiology, University of Manitoba, Winnipeg, Manitoba, Canada R3C 3Y1.

26

Introduction Two different types of adenovirus have been identified as the fastidious agents seen with the electron microscope in the majority of cases of adenoviral gastroenteritis in children (De Jong et al., 1983; Uhnoo et al., 1984). Since the advent of semi-permissive cell lines (Kidd and Madeley, 1981; Takiff, Straus and Garon, 198 1) Ad41, along with Ad40, has been implicated as a principle cause of gastroenteritis, second only to rotavirus, in children up to 5 years of age (Uhnoo et al., 1984; Brandt et al., 1985). Ad40 and Ad41 are now classified as a separate subgenus F of human adenoviruses and are called ‘enteric’ to denote their exclusive association with gastroenteritis (Petric et al., 1982). The fastidious growth characteristics of enteric adenoviruses make diagnosis involving culture difficult. The banding patterns produced by DNA restriction analysis have become the preeminent method of typing fastidious isolates. A variety of genomic variants of Ad41, with different restriction patterns in gel electrophoresis, have been demonstrated (Kidd, 1984; Buitenwerf et al., 1985; Hammond et al., 1985; Shinozaki et al., 1988; Van der Avoort et al., 1989) and changes determining the prevalence of strains are becoming apparent (Van der Avoort et al., 1989; Scott-Taylor et al., 1990). Great potential to link the distribution of these variants with the epidemiology of adenoviral gastroenteritis is emerging. With continued description of new isolates and genomic analyses from different locations it is possible that the geographical extent and movement of circulating strains of Ad41 can be established. Restriction maps for two variant strains of limited distribution (Takiff et al., 1984; Allard et al., 1985) and a number of the less complicated maps of the prototype strain (Van Loon et al., 1985) have already been defined. Restriction maps of the Ad41 variants strains vary quite considerably and discrepancies have been noted (Scott-Taylor et al., 1990) in restriction maps representing the prototype strain (Broker et al., 1984; Akusjarvi and Wadell, 1987). It is the intention of this study to eliminate the present confusion of conflicting Ad41 maps with restriction analysis of a well characterized prototype strain and to present a new method that may help prevent further discrepancies. Restriction analysis of the prototype strain of Ad41 should fill the requirement for base line information for comparison of circulating strains in epidemiological investigation of infantile gastroenteritis and further molecular studies.

Materials and Methods Ad41 strain Tak and Ad2 strain Adenoid 6 were obtained from ATCC (American Type Tissue Collection, 2301 Parklawn Drive, Rockville, MD 20852) and propagated in monolayers of 293, or A549 and HEp2 cells (ATCC), respectively. Viral and cell passage number were kept to a minimum, 293 and HEp2 were regrown from frozen stock before passage 50 or 300, respectively, while viral stocks were maintained within 4 passages of receipt. Batches of ten

27

150-mm* flasks with semi-confluent monolayers were infected with the virus suspended in L15 medium (Flow Labs. Inc., No. lo-51 1, Mississauga, Ont.) at a multiplicity of 1 to approximately 10 TCIDS~ per cell. The viral overlay was replaced after 1-2 h with medium containing 2% fetal bovine serum and changed every third day. Viral titre was assessed as by Reed and Muench (1983) from application of log dilutions of stock culture supernatants to cells in 4 duplicate wells of 24-well plates. Manipulation of Ad2 and Ad41 nucleotide sequences for sequence comparison and restriction endonuclease maps and sites was performed with Pustell Sequence Analysis software (National Bioscience, 725 Tower Drive, Hamel, MN 55340) on sequences stored by GenBank (NIH, Bethesda, MD). Cocultivation

of fastidious

viruses

Trypsinized infected cells were mixed with equal quantities of fresh 293 or A549 cells in a small buffered volume, 20 ml per 150-cm* flask, of 20% fetal bovine serum for 30-60 min by magnetic stirrer. Mixed cells were split 1:3 and incubated in L15 medium supplemented with 10% fetal bovine serum for 5 days or until the development of 3-4+ cpe. Virus concentration was assessed by evaluation of TCID 50 in A549 cells exposed for 1 h to diluted virus supernatant. Endpoint dilution of fastidious adenoviruses was confirmed by electron microscope and enzyme immunoassay (Cambridge Bioscience, Worchester, MA) evaluation of freeze-thawed, clarified cell supernatant. Preparation

of viral DNA

Cells were exposed to several cycles of freeze/fracture. Cell lysate was clarified by centrifugation and virus was concentrated at 30 000 rpm for 90 min in a type 35 (Beckman No. 330926, 3108 Beta Avenue, Burnaby, B.C.) rotor over a cushion of 1.8 g/ml cesium chloride (CsCl). The interphase was homogenized with an equal volume of freon (trichloro-trifluoro ethane; Fischer Scientific) in a (Sorval Ltd, Norwalk, CT) omnimixer. Cell lysate was centrifuged for 10 min at 800 x g to separate the freon layer and extracted lipid. Virions were then purified by density gradient centrifugation, first in a preformed 1.2-1.5 g/ml continuous CsCl gradient and last by isopycnic centrifugation in 1.344 g/ml CsCl. Virus bands were dialysed against 100 mM Tris-100 mM MgCI. Viral DNA was purified by treatment with 50 pg/ml proteinase K and 1% SDS for 1 h at 37°C and extracted with equal vols of phenol and chloroform before precipitation in 2 vols of ethyl alcohol. Cloning and electrophoresis

EcoRT restriction endonuclease fragments A, B and C of the Ad41 genome and Hind111 fragments A, B, C, D, E, G, H, I, J, of the Ad2 genome were cloned as in vector pGem 32 (Promega Biotech, 2368 Durwin Drive,

28

Mississauga, Ont.) to provide abundant DNA of 85 and 90% of the respective genomes for analysis. Ligated plasmids were used to transform competent E. coli HBlOl and transformants were distinguished as white colonies on LB plates supplemented with ampicillin, isopropyl thio-D-galactoside and X-Gal (BRL, Gaitersburg, MD). Large plasmid preparations were purified on CsCl gradients after alkaline lysis of bacteria. 0.5-2 fig of viral or plasmid DNA was digested with at least 10 units of restriction endonuclease in buffers recommended by the manufacturer (Boehringer Mannheim). Electrophoresis was performed overnight at 25 V/5 mA in 150-ml slabs of 0.8% agar in a DNA subcell apparatus (Bio-Rad, 3140 Universal Dr., Mississuaga, Ont.) with 0.89 M Tris-borate-10 mM EDTA/O.S pg per ml ethidium bromide buffer. Gels were photographed with Type 57 Polaroid film using a MP-4 land camera under UV light with a red filter. Terminal fragments of the Ad2 genome were electroeluted from slices of 0.5% Seakem (FMC Bioproducts, 5 Maple St., Rockland, ME 0481) agarose at 100 V and collected in small volume from coated dialysis tubing. Terminal fragments were further purified through 3% agarose before use as probes to substantiate end fragments. Blotting and hybridization

A modification (Reed and Mann, 1985) of the Southern blot was used to transfer’the DNA to nylon membrane (0.45 pm mesh, Micron Separations Inc.) which were stored at 4°C after baking at 68°C for 6-12 h. Membranes were prehybridized for 3-6 h in 0.5% SDS/6 x SSC (1 x = 0.15 M NaCl + 0.015 M Na citrate)/5 x Denhardt’s solution (0.1% Ficoll 400, bovine serum albumin and polyvinyl pyrrolidone) buffer containing 100 mg/ml sheared, denatured herring sperm DNA at 65°C. Hybridization, under stringent conditions, was done overnight in fresh buffer with the addition of at least 1 x lo7 cpm/ml of nick-translated probe (Amersham kit No. N500, 505 Iroquois Shore Rd., Oakville, Ont.) before washing and exposure to film with an intensifying screen at - 70°C. Exonuclease

III mapping

The Ad41 genome was degraded by exonuclease III in the 3’-5’ direction along both strands. With a series of increasing concentrations of enzyme the progressive reduction of the genome was shown in a series of fragments lost from the termini. The range of concentrations of exonuclease III and Sl nuclease (Boehringer Mannheim), used to reduce the single-stranded tails created by Exonuclease III, were first established in checkerboard titrations with Ad2 DNA. The rate of digestion of 1 fig or 0.42 pM aliquots of genomic Ad2 DNA in buffers recommended by the manufacturer varied from 25 to 17 base pairs/min per unit of exonuclease III over 30, 60 and 90 min. Then, 1 pg quantities of Ad41 DNA in 20-~1 volumes were digested in l-25 units of exonuclease III for 30 min at 37°C expanded to 30 ~1in Sl nuclease buffer with

29

3 units of Sl nuclease and incubated for an additional 30 min at 22°C. Samples were extracted with equal volumes of phenol and chloroform before the DNA was precipitated with the addition of two volumes of ethanol at - 70°C for ten min. After being washed twice in 70% ethanol, the DNA was dried, resuspended and digested with the relevant restriction enzyme.

Results Substantial increase in yield of the fastidious enteric adenovirus virions was achieved by mixing trypsinized infected cells with equal quantities of fresh 293 or A549 cells. An increase in viral titer of one to two logs was typically seen with Ad2 and Ad41 while a 3-log increase could be observed with Ad40 in cocultivated cells compared to conventional passage. In no instance was a decrease in titer observed after cocultivation with cells infected with Ad40 although Ad40 passaged conventionally in A549 or 293 cells frequently dropped in titer or failed to grow. Genomic Ad41 DNA, digested with restriction endonucleases BumHI, CluI, EcoRI, HindIII, ,%a1 and X/z01 as in Fig. 1, produces characteristic banding patterns in gel electrophoresis. For the banding patterns of enzymes BgIII and PstI see Fig. 5. The fragment sizes were determined by comparison to lambda phage DNA standards flanking each gel. The size of large EcoRI, MI and X/z01 fragments were estimated by digestion with two enzymes. The sizes of discernible fragments in base pairs are presented in Table 1. Restriction fragments of most digests combined to form an estimated genomic length of between 34.4 and 35 kb, averaging 34678 base pairs. The total size of Hind111 TABLE 1 Size of each restriction

fragment of Ad41

Fragment

sizes in base pairs of fragments

Estimated BarnHI

A : D E : H I J

9,100 8,850 7,560 4,200 1,320 1,305 750 620 590 486 160

CIUI

EcoRI

Hind111

6,450 5,010 4,680 3,890 3,890 3,125 2,940 2,290 1,950 440

15,330 10,970 4,575 3,940

18,830 6,440 4,370 2,630 2,550

9,400 5,050 4,200 3,150 2,800 1,975 1,780 1,680 1,340 1,200 750 570

7,245 6,680 4,730 4,365 3,850 2,540 2,425 1,500 1,110 525

15,140

34,665

34,8 15

34,820

33,895

34,970

34,645

BgfII

: M Total

34,941

from PstI

SalI

11,400 6,650 1,455

SfflUI

XhoI

7,330 4,500 4,300 3,300 2,680 2,680 2,190 1,620 1,480 1,330 1,225 950 815

15,590 10,940 8,425

34,400

34,955

30

fragments was considerably less and may indicate the existence of several additional restriction sites within several hundred base pairs of each other.The map for SmaI, in particular, has undergone radical revision with every version and for each new strain described. The positioning of SmaI fragments was investigated by hybridization of blotted DNA fragments with plasmids containing part of the Ad41 genome in cloned EcoRI fragments A, B and C, shown in Fig. 2. The reaction of Ad41 DNA in gel A, blotted and hybridized with the EcoRI plasmids shown in panel B, is compared to hybridization with a genomic probe in the lower panels. EcoRI A fragment in plasmid p41EA evidently overlaps SmaI C, D, both E/F, G, H, I, J and K fragments and these

X size

in kbp 21*2-

Fig.

1. Restriction

patterns

of Ad41 strain

Tak DNA with frequently electrophoresis.

used restriction

enzymes

in gel

31

A.

A site in Kb 23*69,t 6.6~ 4.3”

Ad 41 probe 2 34

~41 EA probe 234234

p41 EB probe

p41 EC probe 2 3 4

Fig. 2. In~es~gation of the Ad41 SmaI rest~c~on map by ~yb~di~~o~ with pl~mids enclosing 85% of the Ad41 genome. (A) Ad41 EcoRI, S&I and Smal restriction fragments in lanes 2, 3 and 4. (3) Ad41 EcoRI fragment A and 6, A, B, and C plasmids. In the lower panel gel A was byb~dized with {C) genomic Ad41 DNA, ctoned k&RI fra~ent con~i~j~g p~a~~ds, (D-F) cloned EcoRI fra~e~t co~~inj~g plasmids.

32

Fig. 3. Dual digestion of Ad41 DNA to i~v~tigat~ the intestine enzymes.

of the Smal restriction map by

other

- . _ 88. - ‘3-6 Fig. 4. Exonuckase fli reduction of Ad41 genomic DNA with El; Osu, E2; 2u, E?; SU, E4; IOU,E5; 25~; E6, 5011for 30 min prior to digestion with HindHI.

33

A 0

w” Ad41+Bql + -u Exo III

A

A II

;~IbJ:--

;

Ad41 + Pst

4 ,Exolll .E UJ 1

I ; 4 3 .5

A size in

kb

:3:“9 1.5 ,1.3

0.9 ,0.8 .0.6

Fig. 5. Use of exonuclease III digestion to analyse BglII and PHI restriction maps.

SmaI fragments were ordered by dual digests in Fig. 3. Only SmaI fragment L did not appear in Fig. 2; its placement in the ElA region is determined, contrary to previous maps, by reduction with BarnHI in which cleaves 4 small fragments from the left end of the Ad41 genome. Fragment J hybridizes lightly indicating cleavage by EcoRI, consistent with co-hybridization with both p41EC and p41EA and intersection with EcoRI in Fig. 3. Ordering fragments at the right end of the Ad41 Hind111 restriction map proved difficult by dual digestion, with few enzymes having intersection sites in that area. Hind111 fragments were ordered by sequential loss in exonuclease III and Sl digestion. Fig. 4 shows progressive reduction of fragments B then A at one Hind111 terminus and L followed by, in order, G, H, K, J and D from the other terminus in increasing concentrations of exonuclease. Exonuclease III reduction was subsequently used to order fragments on the BgZII and MI restriction maps in Fig. 5. The first panel shows the concurrent reduction of fragments A and both D/E fragments at either end of the BgflI restriction map. With the enzyme MI, fragment J disappears before the reduction of E is initiated whilst C is degraded at the opposite terminus of the Ad41 genome. The SmaI map, together with BumHI, Clal, HindIII, MI, SaZI and X/z01

Genome

mw

units

KW

L

e

I

10

0

C

A

F

A

I

24.9

B

I

1

32,7

EdF

30

I

C

G

I

4o

I

1

1

C

II

65.8

1

15

I 50

I

20

C

I

J M

63.4

C

68-7

65

I

59.5

E/F

t

I

63~4

F/G

I

60-6

25

1 I 70

D

t

734

A

I

75.9

I 85

83.0

I

83.2

8

I

I

30

87,1

H

1 90

B

4

100

34 6

J

I

1

98.5

GI

98.3

J

E

I

93.1

E

87~5

JKH

Ill

FIG

A

0

82.3

i

0 as4

v2T5

O/E

t

A

I

I

D/E

88.8

77.6

6

74.1

I

t

B

F

74.7

6i.5

G

I

K H

I

51.7

0

f&3

I

56.5

FI

t

56.7

63

I

I D

71.1

59.1

44-b

II

43.4

47.0

I

A

1

52.1

t

I

E

50-V

I

i

35.8

39.1

I

32.9

1

I

32.4

IO 20

A

42.6

I B

56.0

I

)A,6

I

HJI

Ii

49.0 17.7

A

24.5

I

B

c

I

I

5

I

J

15.3

I3.f

.. t

1A.V

I

c

B

I

76

I

11,3

A

41.1

32.1

I

33.0

1

GE

ii

29.1 27. i

l&6

c

2.B

D

D

5,4

KWJI

Xho I C&TTCGAG

Sma I CCC,GGG

3.7

l.?

tll’

a.5

35

cleavage maps are shown in Fig. 6, complete with restriction sites demarcated in map units. Discussion

A large part of the difficulty of analysis of enteric adenoviruses is due to their fastidious growth characteristics in vitro. The growth restriction was partly alleviated with the cocultivation of infected and trypsinized cells, the mixing of which improves the growth of other fastidious viruses (Benyesh-Melnick et al., 1964). Mixing of cells facilitated the spread of enteric adenoviruses and cytopathic changes progressed much more rapidly in cocultivated cells. Infection with enteric adenoviruses tends to spread to adjacent cells in infected foci (Retter et al., 1979; Kidd and Madeley, 1981) and factors that promote cell contact could enhance the spread of cell-associated virus (Benyesh-Melnick et al., 1964). The EcoRI restriction map with few large internal fragments and relatively small terminal fragments provided a simple means to clone and produce 85% of the Ad41 genome in quantity for restriction analysis. Characterization of the EcoRI clones disclosed discrepancies with the restriction maps of supposedly homogeneous strains (Broker et al., 1985; Akusjarvi and Wadell, 1987) and revealed differences with most other strains of Ad41 identified of one or two restriction sites on average per restriction enzyme (Takiff et al., 1984; Allard et al., 1985). This corresponds with the level of diversity observed in the restriction patterns of different clinical isolates (Kidd, 1984; Van der Avoort et al., 1989; Scott-Taylor et al., 1990). The restriction patterns shown here for enzymes BarnHI, BgnI, Hind111 and SmaI correspond to those drawn by Adrian et al. (1986) for the prototype strain except that a characteristically smeared but distinguishable SmaI B band is present. The additive size of restriction fragments averaged 34 678 base pairs which is between previous estimates of the Ad41 genome length (Takiff et al., 1984; Van Loon et al., 1985). The maps bear some resemblance but do not always conform exactly to the maps defined for other strains of Ad41 examined. Ad41 strain D389 (Allard et al., 1985) lacks a pair of BumHI restriction sites and possesses an extra EcoRI site in comparison to the prototype strain, but apparently has an identical SulI map. Strain G1105 has an EcoRI map that complies with strain D389 but the BumHI map has further variations (Takiff et al., 1984). The restriction maps representing type Ad41 in a collection of genomic animal maps (Broker et al., 1985; Akusjarvi and Wadell, 1987) have undergone some revision and may need further amendment. The less intricate restriction maps match those found for the prototype strain. The more complicated HindIII, PstI and SmuI maps, however, resemble those of strain G1105 and may represent a similar strain. Ordering of fragments towards the termini of adenoviruses can be problematical (Kitchingman, 1982). The terminal protein attached to the 5’

36

terminal nucleotide of both strands (Rekosh et al., 1977) resists complete degradation with proteases and interferes with end labelling and cloning. Gel retardation, a method utilising the intact terminal protein to prevent the entry of terminal fragments into an agarose gel (Kitchingman, 1982), was found prone to partial digestion and difficulty with DNA solubilization. Exonuclease III circumvents the terminal protein by degrading the 3’ strand and exposing a single-stranded sequence to Sl nuclease. Enzymatic reduction of sequential restriction fragments was found an effective, rapid and reproducible means of examining the order of terminal fragments. Furthermore, exonuclease III could reduce very large sequences of DNA at a steady rate and could be used to order restriction fragments in the entire genome. The variability and the large number of variant genomic restriction patterns observed (Kidd, 1984, Shinozaki et al., 1988; Van der Avoort et al., 1989) emphasizes the need to establish a convention for classification of new isolates and variant strains. Adrian et al. (1985) have suggested a system based on a panel of 6 enzymes. Restriction patterns of the prototype strain of Ad41 should fill the requirement of a standard for definition and cataloguing of variants in epidemiological investigation of infantile gastroenteritis. Prototype strain restriction maps should enable clearer classification and better understanding of the genetic drift affecting genomic variants. Prototype strain Tak of species Ad41 is currently isolated relatively infrequently and only reported from the Netherlands (Buitenwerf et al., 1985; Van der Avoort et al., 1989). Ad41 strain Tak may have been the prevalent Dutch isolate when first described in 1973 (De Jong et al., 1983). Other genomic variants, particularly the strain designated D12, now appear to be on the ascendance in that country (Van der Avoort et al., 1989). However, a Taklike strain has recently been recovered from stool from central Canada (ScottTaylor et al., 1990) and its distribution may be as yet incompletely mapped.

References Adrian, T., Best, B. and Wigand, R. (1985) A proposal for naming genome types, exemplified by adenovirus type 6. J. Gen. Viral. 66, 2685-2691. Adrian, T., Wadell, G., Hierholzer, J.C. and Wigand, R. (1986) DNA restriction analysis of adenovirus prototypes 1 to 41. Arch. Virol. 91, 277-290. Akusjarvi, G. and Wadell, G. (1987) Genetic maps of human and animal adenoviruses. In: S.J. O’Brien (Ed), Genetic Maps 1987, Vol. 4. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, pp. 75-78. Allard, A.K., Wadell, G., Evander, K.M. and Lindman, G.K.K. (1985) Specific properties of two enteric adenovirus 41 clones mapped within early region la. J. Virol. 54, 145-150. Benyesh-Melnick, M., Dessy, S.I. and Fernbach, D.J. (1964) Cytomegaloviruria in children with acute leukemia and in other children. Proc. Sot. Exp. Biol. Med. 117, 624630. Brandt, CD., Kim, H.W., Rodriguez, W.J., Arrobio, J.O., Jefferies, B.C., Stallings, E.P., Lewis, C., Miles, A.J., Garder, M.K. and Parrot, R.H. (1985) Adenovirus and pediatric gastroenteritis. J. Infect. Dis. 151, 437443. Broker, T.R., Keller, C.C. and Roberts, R.J. (1984) The adenovirus 2 cytoplasmic RNA transcripts.

37 In: S.J. O’Brien (Ed), Genetic Maps 1984, Vol. 2. Cold Spring Harbour Laboratory, Cold Spring Harbour, NY, pp. 196203. Buitenwerf, J., Louwerens, J.J. and De Jong, J.C. (1985) A rapid and simple method for typing adenovirus 40 and 41 without cultivation. J. Virol. Methods 10, 394. De Jong, J.C., Wigand, R., Kidd, A.H., Wadell, G., Kapsenburg, J.G., Muzeric, J.C., Wermenbol, A.G., Madeley, C.R. and Firtzlaff, R.G. (1983) Candidate adenoviruses 40 and 41: fastidious adenoviruses from infant stool. J. Med. Virol. 11, 215-231. Hammond, G.W., Mauthe, G., Joshua, J. and Hannan, C. (1985) Examination of clinical isolates of human adenoviruses by restriction analysis. J. Clin. Microbial. 21, 61 l-616. Kidd, A.H. (1984) Genome variants of adenovirus 41 (subgroup G) from children with diarrhoea in South Africa. J. Med. Virol. 14, 49-59. Kidd, A.H. and Madeley, C.R. (1981) In vitro growth of some fastidious adenoviruses from stool specimens. J. Clin. Pathol. 34, 213-216,Kitchingman, G.R. (1982) Restriction mapping and molecular cloning of adenovirus type 4 (subgroup E) DNA. Gene 20, 205-210. Petric, M., Kajden, S., Doubnia, N. and Middleton, P.J. (1982) Enteric adenoviruses. Lancet i, 10741075. Reed, K.C. and Mann, D.A. (1985) Rapid transfer of DNA from agarose to nylon membranes. Nucl. Acids Res. 13, 7207-7221. Reed, L.J. and Muench, A. (1983) A simple method of estimating 50% endpoints. Am. J. Hyg. 27, 493. Rekosh, D.M.K., Russel, W.C., Bellet, A.J.D. and Robinson, A.J. (1977) Identification of a protein linked to the ends of adenovirus DNA. Cell 11, 283-295. Retter, M., Middleton, P.J., Tam, J.S. and Petric, M. (1979) Enteric adenovirus: detection, replication and significance. J. Clin. Microbial. 10, 574578. Scott-Taylor, T.H., Alhuwalia, G., Klisko, B., Hammond, G.W. (1990) Prevalent enteric adenovirus variant not detected by commercial monoclonal antibody enzyme immunoassay. J. Clin. Microbial. 28, 2797-2801. Shinozaki, T., Araki, K., Kobayashi, M., Fujita, Y. and Abe, T. (1988) Genome variants of human adenoviruses 40 and 41 (subgroup F) in Japan. J. Clin. Microbial. 26, 2567-2571. Takiff, H.E., Straus, S.E. and Garon, C.F. (1981) Propagation and in vitro studies of previously non-cultivable enteral adenoviruses in 293 cells. Lancet ii, 832-834. Takiff, H.E., Reinhold, W., Garon, CF. and Straus, SE. (1984) Cloning and physical mapping of enteric adenoviruses (candidate types 40 and 41). J. Virol. 51, 131-136. Uhnoo, I., Wadell, G., Svensson, L. and Johannson, M. (1984) Importance of enteric adenovirus 40 and 41 in acute gastroenteritis in infants and young children. J. Clin. Microbial. 20, 365-372. van der Avoort, H.G.A.M., Wermenbol, A.G., Zomerdijk, T.P.L., Kleijne, J.A.F.W., van Asten J.A.A.M., Jensma, P., Osterhaus, A.D.M.E., Kidd, A.H. and de Jong, J.C. (1989) Characterization of fastidious adenovirus types 40 and 41 by DNA restriction analysis and by neutralizing monoclonal ntibodies. Virus Res. 12, 139-158. van Loon, A.E., Rozijn, T.H., de Jong, J.C. and Sussenbach, J.S. (1985) Physiochemical properties of the DNAs of the fastidious adenovirus species 40 and 41. Virology 140, 197-200.