Molecular epidemiology of adenoviruses isolated in Iceland during 1988–1990

Journo~ of Virological methods, 42 (1993) 193.--206 cn 1993 Elsevier Science Publishers B.V. / All rights reserved / 0166-0934/93/$04.00

193

VIRMET 01472

Molecular epidemiology of adenoviruses isolated in Iceland during 1988-1990 Mats E. JohanssorP,

Maria Ahrel-Andersson”

and Sigrun Gudnadottirb

“Department of Clinical Microbiology, Sertion of Virolog_v, Karolinska Hospital. Stockholm (Sweden) and ‘Department of Microbiology, Vu-us Laboratories, Universit_v of Iceland, Reykjavik (Iceland) (Accepted

I I November

1992)

Summary All human adenoviruses isolated in Iceland during 1988-1990 have been subjected to restriction endonuclease analysis. Of 55 isolates altogether, subgenus C (Adl, Ad2, and Ad5) predominated with 42 isolates followed by subgenus B (Ad3 and Ad7) with 12. Analysis of the 9 Ad1 isolates revealed 6 DNA-variants. Among these the established DNA-variants D4, D7, and D10 were recognized. The remaining 3 DNA-variants were primarily found in Iceland. Among the 22 Ad2 isolates, 7 DNA-variants could be distinguished. D2 predominated with 15 isolates whereas the prototype was isolated only once. The novel 5 DNA-variants of Ad2 were all closely related to D2. Analysis of the 11 Ad5 isolates revealed 6 DNA-variants, 2 of which (D2 and D.5) were already established. Ice2 and D3 were the most common occurring DNA-variants of Ad5. Ad5 showed the highest degree of genomic variability within subgenus C, both in terms of the low degree of pair-wise comigration of restriction fragments and the number of principal variants of RE-patterns. Analysis of the 9 Ad3 isolates revealed 3 DNA-variants: D3, DlO, and Ice1 (a novel DNA-variant that resembles 010). The DNA-variants D3 and DlO were each represented by 4 isolates. The three Ad7 isolates belonged all to the DNA-variant D5. Adenovirus; (Iceland); DNA variant; Genetic variability; Restriction endonudease

Correspondence to Karolmska

Mats E. Johansson,

Hospital, S-10401 Stockholm,

Department Sweden.

of Chmcal Microbiology,

Section of Virology.

194

Introduction Since Rowe et al. 1953 discovered adenovirus, an array of 47 serotypes has been distinguished to date on the basis of virus neutralization and hemagglutination-inhibition (Hierholzer et al., 1991). The large number of DNA-variants among human adenoviruses (Wadell, 1984) have been detected by restriction endonuclease (RE)-analysis. Though analyzing only a minor part (1%) of the virus genome, the technique allows in a feasible way proper recognition of adenovirus DNA-variants. RE-analysis also allows reliable typing of most adeno types and is widely used now as a complement to virus neutralization. Among subgenus D isolates, however, only Ad8, Ad19, and Ad37 can be efficiently typed by RE analysis. To date there are comprehensive maps for Ad12, Ad18 and Ad31 of subgenus A (Tharner et al., 1992; Johansson et al., 1991); Ad3 and Ad7 (Li and Wade& 1986 and 1988a; Adrian et al., 1989b and 1990a), Adll, Ad34, and Ad35 (Li et al., 1991), and Ad21 (Avoort et al., 1986) of subgenus B; Adl, Ad2, Ad5 and Ad6 of subgenus C (Adrian et al., 1985, 1989); Ad8, Ad19 and Ad37 of subgenus D (Kemp and Hierholzer, 1986; Kemp et a1.,1983; Adrian et a1.,1988); Ad4 of subgenus E (Li and Wadell, 1988’0); Ad40 and Ad41 of subgenus F (Avoort et al., 1989). Only limited information is available on the molecular epidemiology of the most common adenoviruses (i.e., Adl, Ad2, Ad3, Ad5, and Ad7) recovered from restricted areas and periods of times. The report on the adenovirus strains from the Hannover area during 1981 is the most thorough study within this field (Adrian et al., 1989~). In addition, Ad3 strains from Glasgow and London from 1981 have been analyzed (O’Donell et al., 1986), as has the genetic heterogeneity of Ad3, Ad4, and Ad7 among isolates from Manchester in the UK during 1984 (Bailey and Richmond, 1986). Another study comprises adenoviruses from patients with acute lower respirator tract infection isolated in Buenos Aires, Argentina between 1984 and 1988 (Kajon and Wadell, 1992). Iceland is an appropriate area for molecular epidemiological studies in view of its position as a fairly isolated island in the Atlantic Ocean, the limited number of inhabitants (about 250 000), and virological service provided mainly by one laboratory. The present study provides information about adenovirus strains isolated there during 1988-1990, and extends the knowledge of the worldwide molecular epidemiology of human adenoviruses.

Materials

and Methods

Cells Most adenoviruses were originally isolated in HEp-2 cells. For preparation of DNA in this study, they were propagated in HeLa, A549, or human embryonic lung diploid fibroblast cells. Eagle’s minimum essential medium (MEM) containing 10% fetal bovine serum was used for outgrowth of the cells.

195

The maintenance medium contained ml) and streptomycin (100 pg/ml).

2% fetal bovine serum, penicillin (100 IE/

Virus isolates The study comprises all human adenoviruses isolated in Iceland during 1988-1990 (Table 1). Of the 810 specimens (from 701 patients) sent to the laboratory for routine virus isolation, 55 Ads were recovered. Of these, 51 were derived from stools, five of which also accompanied by a positive throat specimen. However, each patient is represented here by only one isolate. Of those infected with DNA-variant D2 of Ad2, 3 were positive also for rotavirus. The patient excreting Ice1 of Ad5 in faeces was positive for RS-virus as well in a concomitant nasopharyngeal specimen. Preparation

of adenovirus DNA, cleavage with RE and separation of fragments

Adenovirus DNA was extracted from Hirt supernatants (Hirt, 1967) and the concentration of DNA was estimated semiquantitatively (Grillner and Blomberg, 1984). RE cleavage of DNA and electrophoretic separation of the fragments were performed as previously described (Johansson, et al., 1989) except SmaI digestion of Ad5 that in addition were separated in 3% NuSieve agarose gel (100 V, 45 min). Molecular weight markers (HindIII-cleaved lambda DNA) were included in each run. The enzymes and lambda DNA were purchased from IBI, New Haven, CT, USA, except for BstEII that was purchased from USB, Cleveland, Ohio, USA. The fragments were stained with ethidium bromide (1 ,ug/ml) in TBE buffer (0.089 M Tris, 0.89 M boric acid, 0.002 M EDTA pH 8.3) for 30 min, visualized by UV light, and photographed with Polaroid positive/negative film type 665. Denomination

of Ad-strains

We have employed the naming system proposed by Adrian et al. (1985). This system provides a unique numeric code for each DNA-variant within a particular serotype. The code is composed by the principle pattern for each of seven endonucleases listed in alphabetic order: BarnHI, BglII, BstEII, EcoRI, HindIII, KpnI, and SmaI (except for Ad3 and Ad7 for which EcoRI has been omitted according to Adrian et al. (1989a,b)). Hence, the prototype (Dl) of each serotype has the code: 1111111. The novel RE-patterns are preliminary designated x or y for respective enzyme. The novel DNA-variants of each serotype have preliminary been named Icel-5. Pair-H!ise comigrating

restriction fragment

The genetic relatedness between estimated by pair-wise comigration

(PCRF)

analysis

Ad-strains of a particular serotype was of restriction fragments (PCRF) as was

196 TABLE 1 Serotype, spectmen, and chmcal feature of 55 Icelandtc adenovirus ______

.__-.I

_

Enteritis

._.~

_~_.

Respiratory

-___

isolates from 1988 to 1990

symptoms

Other”

____.-.__

.__-

Sub,tprrs c Ad1 (n=9) Faeces Throat Ad2 (n = 22) Faeces Colon specimen Ad5 (n= 11) Faeces Throat

5

2 2

13

3

5

I (fatal) 3 1

I

5

Szrbgenus B Ad3 (~7~9) Faeces Ad7 (n=3) Faeces Subgenus F Ad41 (n= 1) Faeces

3

3 3

1

_---

“Including cases with rash, conjunctivitis unknown clinical condition.

.“. or lethargy

as major symptoms

_~._.

-

and two cases with

expressed in percentages. For each pair of DNA-variants of Adl, Ad2, and Ad5 116-120, 119-124, and 120-128 fragments, respectively, were compared.

Results We have examined the RE-patterns of 55 Icelandic adenovirus isolates. Serotype and DNA-variant assignment were made by comparing the resulting RE-profiles with those already established (Adrian et al., 1986, 1989a-b, 1990). In addition, 30 of 55 isolates were typed by virus neutralization (Hierholzer, 1989) and full concordance between the two methods were obtained. The isolates are presented in Table 1, showing distribution into serotypes, type of specimen, and clinical condition. A striking feature was the predominance of faecal samples (92.7%). The median ages were 1.5, 1.0, and 1.5 year for Adl, Ad2, and Ad5, respectively, and 3 and 11 yr for Ad3 respective Ad7. There was no infant under one month, while 13 (24%) were under one year, all of which infected with subgenus C viruses. Molecular epidemiotog~~ of Adl, Ad2, und Ad5 isolates (subgenus Ci RE analysis of 9 Ad1 isolates revealed 6 DNA-variants, 3 of which (D4* D7 and DlO) previously were reported by Adrian et al. (1990). The novel DNA-

197

TABLE 2 DNA variants, representative DNA-variant

strains, and enzyme code of adenovirus types I, 2, and 5 of subgenus C

Representative strain

Origin of isolate

Enzyme code”

No. of isolates

place

year

Ad71 123 1039 309 S2821 51042 SlflO

Wash., Illinois D.C. Seattle Argentina Iceland Iceland Iceland

64 54 66 67 88 89 89

2221111 1111111 2121111 2131111 212x111 21211x1 xl2lxyl

03 2 1 1 1 1

~~en~v~r~s type 2: Dl Ad6 D2 V678 Ice1 S2108 Ice2 S2907 Ice3 S1582 Ice4 Sl634 Ice5 52313

Wash., D.C. Arizona Iceland Iceland Iceland Iceland Iceland

54 59 88

1 15 1

:: 89 90

1111111 1121111 3121111 6121111 1121x11 llxllll 1124111

Ade~ovirus type 5: Dl Ad75 :: 37 1

Wash.. D.C. Arizona Georgia

53 60 62

1111111 2222222 2242222

0 31

Ice1 Ice2 Ice3 Ice4

Iceland Iceland Iceland Iceland

88 88 90 90

x2x1212 62422x2 2722221 32422x2

Adenovrrus type 1: :: D7 DlO Ice 1 Ice2 Ice3

S821 s1015 s55 S1429

; 1 1

:

1 1

Principal variants of patterns are shown in Fig. 1 {a and b). “In alphabetic order: BarnHI, 8gfiI, BstEII, EcoRI, IlindIII, &XII, and SmaI.

variants of Ad1 (Icel-3) are based on unique codes (Table 2) derived from principle patterns schematically displayed in Fig. 1. The DNA-variants Icel-2 are closely related to D7, digressing by one enzyme only. Ice3 showed the lowest percentage (83%-92%) of comigrating fragments when compared with other DNA-variants of Ad1 as well as with the prototype. Among the 22 Ad2 isolates, 7 DNA-variants were distinguished. The prototype was isolated once. D2 was the prevailing DNA-variant accounting for 68% of all Ad2 isolates. The novel 5 DNA-variants, designated Icel-5, were represented by one isolate each, except Ice3 that had two. Analysis of the 11 Ad5 isolates revealed 6 DNA-variants. D2 and D5 were represented by one and three isolates, respectively. The novel DNA-variant Ice2 (62422x2) was frequently isolated up to July of 1989 but not found thereafter. The closely related DNA-variant Ice4 (32422x2) arose as a singleisolate during the ensuing year. The four patients infected with Ice2 differed significantly in age from those infected with other DNA-variants of Ad5 (mean ages: 15.5 yr and 1.3 yr, respectively; Student’s I-test and Mann-Whitney test: P

198

BamHl

(A) Ad1

Ad2

12x 2Q.v 10:.

---

-__

1236x

-_-

a----

Ad1

__-_

--_ --

Ad5

136

-

(, --cL5,0-* f! 4m

Bglll

10

-

--

-

390, -

l.O-*

1,O.

0,5-

0.5-

‘-

-----

--

=

SPI ___

--

-

---

EcoRl Ad1 1 x

Ad5 124x

-

Ad2 14

--

===

---_-

---

--

-__ ---=

10:. I

p 5,0-. 3 4,0*, 3,0 2.0,

,,

1,0*, 0,5_

Ad5 12

20..

20..

10..

--

-

BstEll Ad2 12x

7

--I --== = ---

2,O

Ad1 123

Ad5 12

--0

50 8480

--

3,O 2,0%,

1

20 1

--_

-

Ad2

12

-

---

-

--

-

--

3,o. 2,O.

___ s31

-_ ---

--II ----

-__ ___

w-w --_

--__ --__

____

--_ ---

-___ _---

< 0.01). Ice1 was recovered Stockholm 13 months later.

Molecular epidemiology

--

p 50.. 2 4,0*

as a single-isolate

-

----

--

l,O-.

0,5-

in 1988 but reappeared

in

of Ad3 and Ad7 isolates (subgenus B)

RE-analysis of 9 Ad3 isolates revealed 3 DNA-variants: D3, DIO, and Ice1 . The DNA-variants D3 and DlO were each represented by 4 isolates, whereas Ice1 had only one (Fig. 4). Ice1 is distinguished from DlO by an extensively altered BamHI pattern. The three Ad7 isolates (all from Kopavogur, a village

199 Hindlll Ad5

Ad2

Ad1

Kpnl Ad1

20

== --l t

50 $ 40 Q

z=

3,0

zo ,,o

03

--

--

--

--

_-

-=

EE -

a,_

--

--

--

__

_-

--

5,o ,% 480 33

-

2.0

1.0

-

====---

-

Ad5 12x

-=--= =-_ we-

---

--*

Ad2 ’

10

a

l -- --

1

I’XY

-

-a---

---

0.5,

Smal Ad1 1

=

Ad2

Ads

1

1 2

---

-

--

-

== ----

=

51 --

-

--

=-

+

t

Fig. 1. (a and b). Restriction endonuclease patterns of Adl, Ad2, and Ad5 strains. For each enzyme, the pattern of the prototype strain is designated 1, other patterns 1-7. The novel patterns x and y for respectwe enzyme, are orlginally found among the Icelandic Ad-strains.

not far from Reykjavik) were all recovered within a 6-wk period during the fall of 1990. All had identical RE-patterns (5323 13), consistent with DNA-variant D5 of Ad7 (Adrian et al., 1989a).

200

Adenovirus type 1 Ice1 Ice2 Ice3 A AA

Other DlO

A A

D7

A

A

D4

‘J-M

A-J

J-S

0.O’J-M

A AJ

1988

A

J-S

O-D ’ J-M

A-J

1989

J-S

O-D ’

1990

Adenovirus type 2 Ice1 Ice2 AA

Other Ice 3

Ice4 A

Ice5 A

A

A

: D2

i

A

AA r:

:t

Dl

A ‘J-M.AJ.J-S.0.D’J-M.AJ.J-S.O-D’J-M.A-J’J-S.0.D’

1988

1989

1990

Adenovirus type 5 Other

Ice1 A .

Ice2

.

*

.

A

D5

Ice4 A

Ice3 A A

A A

D2

‘J-M

A-J

J-S

0.D’J-M

1988 Fig. 2. Temporal

distribution

AJ

J-S

O-D ‘J-M

AJ

1989 of DNA-variants

J-S

O-D’

1990 of Adl.

Ad2. and Ad5

Discussion Adenoviruses belonging to subgenera A, D, and E were altogether absent among the Icelandic isolates. These subgenera were expected to account for 7% of the isolates according to worldwide statistics reported to the World Health Organization (WHO) (Schmitz et al., 1983). On the other hand, the high proportion (76%) of subgenus C (Adl, Ad2, and Ad5) isolates in Iceland as compared to Stockholm (40%) (Johansson et al., unpublished data), Chile (30%) (Kajon and Suarez, 1990); Argentina (51%) (Kajon and Wadell, 1992), and worldwide (59%) (Schmitz et al., 1983) was another notable feature. The cause of the high proportion of Adl, Ad2, and Ad5 isolates (subgenus C) is unclear. This may be a consequence of favourable endemic spread of these

201

Adenovirus type 3

BamHl

Bglll

1 2x

125

BstEll

Hindlll

Kpnl

1

1

12

Smal 13

20-. -10:.

--

---

-

c

5,0.. 4,o.

-== ==-

3,o. 2,0,

-., -

l,O..

-

11’ -

Fig. 3. Restriction

=--

--

= -

--

= -

--

--

-----

0,5-

--

--

=

=

--==--=--== ==-------

-

-

--

endonuclease

patterns

-

of Ad3.

serotypes in this particular population as well as absence of extensive epidemics caused by other serotypes. The genetic variability within subgenus C was previously suggested (Adrian et al., 1990) to be more extensive than within serotypes of other subgenera. In terms of numerous DNA-variants, our results also support an extensive genomic variability within subgenus C. On the other hand, most DNA-variants within both Ad1 and Ad2, are closely related when assessed by pairwisecomigration of restriction fragments. Among the DNA-variants of Adl, only Ice3 had figures low enough (83%-92%) to be indicative of a separate origin. For Ad2 the similarities between the DNA-variants are even more apparent. Adenovirus type 3

.

Ice1 A

DlO

A

AA

AA

D3

1..

J-M AJ

-

I

*.

J-S O-D J-M AJ

I

*.

distribution

‘I

J-S 0-D J-M AJ J-S O-O

1989

1988 Rg. 4. Temporal

’

A

A

of DNA-variants

1990 of Ad3.

202 TABLE 3 Percentages of comigrating (subgenus C) Adenovirus type I: D4 Dl D4 D7 DlO Ice1 Ice2

Ice1

98

of adenovirus type 1,2, and 5

DlO

Ice1

Ice2

Ice3

94 96 97

90 ;z

92 F/

92

97 92

84 84 90 92 83 86

Ice2

Ice3

Ice4

95 98 97

96 z!:

95 98 95

96

Adenovirus type 5: D2 Dl D2 D5 Ice1 Ice2 Ice3

D7

93

Adenovirus type 2: D2 Dl D2 Ice1 Ice2 Ice3 Ice4

fragments between separate DNA-variants

82

D5

Ice1

80 98

84 89 92

Ice2

Ice5

95 95 96 95

Ice3

Ice4

82 92 93 85 89

94 9”: 86

Actually, all Ad2 isolates from Iceland belonged either to D2 or were D2-like, i.e., RE-pattern deviating in only one enzyme of seven. Ad2 displayed a degree of pair-wise comigration of restriction fragments that ranged from 95% to 98% (Table 3). Thus in Iceland, D2 seems to be a major DNA-variant of Ad2 from which closely related variants frequently emanate. Most of these DNA-variants probably disappear quickly. Occasionally, strains may survive as shown by the two Ice3 isolates: the first strain was isolated from a 1-yr-old child living in a village in the country (Stykkisholmur); the second was recovered 15 months later from an occasion of sudden infant death, a 6-mth-old baby residing in a neighbouring farm (Myrasysla). In Hannover (Adrian et al., 1989~) as well as worldwide (Adrian et a1.,1990), also D2 and D2-like strains prevail accounting for around 50% of the DNA-variants of Ad2, whereas D2 accounts for 25% or less (if any) among Ad2 strains from patients with acute lower respiratory tract infections in Buenos Aires, Argentina (Kajon and Wadell, 1992). Whether this discrepancy is due to another epidemiological situation in that area or mirrors differences in tropism or virulence is unclear. In this study Ad5 possessed the highest genomic variability within subgenus C as indicated by the low degree of PCRF between different DNA-variants;

203 TABLE 4 DNA variants, representative Iceland during 1988-1990 DNA-variant

strains, and enzyme code of adenovirus

Representative strain

Origin of isolate

types 3 and 7 isolated in

Enzyme codea

No. of isolates

place

year

Adenovirus type 3: Dl GB

Maryland

53

111111

0

D3 DlO Ice1

Georgia Illinois Iceland

64 77 90

251123 221123 x51123

f

California New York

54 70

111111 532313

0 3

1244 119 s2225

Adenovirw type 7: Dl Gomen D5 375

1

aIn alphabetic order: BumHI, Bg/II, BstEII. HindIII, KpnI, and SmaI (EcoRI has been omitted according to the panel of endonucleases for Ad3 and Ad7 used by Adrian (1989b, 1989a)).

PCRF is considered by us being a more valid measure for genomic diversity than the number of DNA-variants. Also in view of the number of principle variants of RE patterns, Ad5 shows higher degree of genomic variability than Ad1 and Ad2: 17 RE patterns compared with 13 each of Ad1 and Ad2. Indirectly, our data confirms the results by Adrian (1990) that suggest Ad5 to have the highest number of varying restriction sites within subgenus C. The DNA-variant AdSa, attributed to an extensive outbreak in the CSSR (Bruckova et al., 1980) were not seen in Iceland; neither were DNA-variants Ad5c and AdSd (Webb et al., 1987). The Ad5 isolates from Argentina (Kajon and Wadell, 1992) had all been placed into 3 novel DNA-variants. However, scrutini~ng the published photographs does not allow proper disc~mination between pattern no:1 and no:2 of S~na1. As distinguishing of these very similar SmaI patterns are essential, there are uncertainty whether the DNA-variants 5* (2222221) and 5# (3242221) are novel or belong to the established DNAvariants, D2 (2222222) and D17 (3242222), respectively. Subgenus B, represented in this study by Ad3 and Ad7, constituted 22% of the Icelandic isolates, as compared to 33% worldwide. At the end of the past decade, an extensive Ad3 epidemic was encountered in Scandinavia and other European countries. This epidemic began in Iceland in July, 1989 and was on the whole moderate. The DNA-variants D3 and DIO co-circulated whereas the prototype (Dl) of Ad3 was absent. In a concomitant Ad3 epidemic in Stockholm, Dl was found to be as common as D3 and DlO, as reported by others from Scotland (Done11 et al., 1986). It is noteworthy that D3 currently is isolated both in Iceland and in Stockholm (Johansson et al., unpublished data): D3 prevailed almost exclusively in the United States in a comprehensive Ad3 study with strains from the US, East Asia, Germany, and other places recovered from 1954 through 1984 (Adrian et al., 1989).

204

Acknowledgements

We wish to thank Marianne Brundin for skillful technical assistance. This work was supported by grants from the Karolinska Institute, Stockholm, Sweden.

References Adrian, T., Best, B. and Wigand, R. (1985) A proposal for naming adenovirus 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 prototype 1 to 41. Arch. Virol. 91, 277-290. Adrian, T., de Jong, J.C., van der Avoort, H.G.A.M. and Wigand, R. (1988) Genome type analysis of adenovirus 37 isolates. J. Med. Virol. 25, 77-83. Adrian, T., Becker, M., Hierholzer, J.C. and Wigand, R. (1989a) Molecular epidemiology and restriction site mapping of adenovirus 7 genome types. Arch. Viral. 106, 73-84. Adrian, T., Best, B., Hierholzer, J.C. and Wigand, R. (1989b) Molecular epidemiology and restriction site mapping of adenovirus type 3 genome types. J. Clm. Microbial. 27, 1329-1334. Adrian, T., Wigand, R., Knocke, K.W., Schafer, G. and Grundmann, M. (1989c) Genome type analysis of adenoviruses: isolates from one year from the Hannover area. Arch. Virol. 105. 89101. Adrian, T., Sassinek, .I. and Wigand, R. (1990) Genome type analysis of 480 isolates of adenovirus types 1, 2, and 5. Arch. Virol. 112, 235-248. Bailey, AS. and Richmond, S.J. (1986) Genetic heterogeneity of recent isolates of adenovirus types 3, 4, and 7. J. Clin. Microbial. 24, 30-35. Bruckova, M., Wadell, G., Syrycek, L. and Kunzova. L. (1980) An outbreak of respiratory disease due to type 5 adenovirus identified as genotype 5a. Acta Virol. 24, 161-165. Grillner, L. and Blomberg, I. (1984) Restriction enzyme analysis of human cytomegalovirus using DNA extracted from infected cells. J. Med. Virol. 14, 313-322. Hierholzer, J.C. (1989) Diagnostic procedures for viral, rickettsial and chlamydial infections. In: N.J. Schmidt, and R.W. Emmons, (Eds). Adenoviruses. American Public Health Assoc.. Washington DC, pp. 219264. Hierholzer, J.C., Stone, Y.O. and Broderson, J.R. (1991) Antigemc relationships among the 47 human adenoviruses determined in reference horse antisera. Arch. Virol. 121, 1799197. Hirt, B. (1967) Selective extraction of polyoma DNA from infected mouse cell cultures. .I. Mol. Biol. 26, 365-369. Johansson, ME., Brundin, M., Adamson, L., Grillner, L., Landqvist, M., Thiirner, A. and Zweygberg Wirgart, B. (1989) Characterization of two genome types of adenovirus type 31 isolated in Stockholm during 1987. J. Med. Virol. 28, 63-68. Johansson, M.E., Brown, M., Hierholzer, J.C., Thiirner, A., Ushijima, H. and Wade& G. (1991) Genome analysis of adenovirus type 31 strains from immuno~ompromised and immunocompetent patients. J. Infect. Dis. 163, 293-299, Kajon, A.E. and Suarez, M.V. (1990) Molecular epidemiology of adenoviruses isolated from hospitalized children with severe lower acute respiratory infection in Santiago, Chile. J. Med. Virol. 30, 294297. Kajon, A.E. and Wadell, G. (1992) Molecular epidemiology of adenoviruses associated with acute lower respiratory disease of children in Buenos Aires, Argentina (19841988). J. Med. Virol. 36, 292-297. Kemp, M.C. and Hierholzer, J.C. (1986) Three adenovirus type 8 genome types defined by restriction enzyme analysis: Prototype stability in geographically separated populations. J. Ciin. Mi~robiol. 23, 4699474.

205 Kemp, M.C., Hierholzer, J.C., Cabradilla, C.P. and Obijeskt, J.F. (1983) The changing etiology of epidemic keratoconjunctivitis: Antigenic and restriction enzyme analyses of adenovirus types 19 and 37 isolated over a IO-yr period. J. Infect. Dis. 148, 2432. Li, Q.-G. and Wadell, G. (1986) Analysis of 1.5different genome types of adenovirus 7, isolated on live continents. J. Virol. 60, 331-335. Li. Q.-G. and Wadell, G. (1988a) Comparison of 17 genome types of adenovirus type 3 identified among strains recovered from six continents. J. Clin. Microbial. 26, 1009-1015. Li, Q.-G. and Wadell, G. (1988b) The degree of genetic variability among adenovirus type 4 strains isolated from man and chimpanzee. Arch. Virol. 101, 65-67. Li, Q.-G., Hambreus, J. and Wadell, G. (1991) Genetic relationship between thirteen genome types of adenovirus 11, 34. and 35 with different tropism. Intervirology 32, 338-350. O’Donnell, B., Bell, E., Payne, S.B.. Mautner, V. and Desselberg, V. (1986) Genome analysis of species 3 adenoviruses isolated during summer outbreaks of conjunctivitis and pharyngoconjunctival fever in the Glasgow and London areas in 1981. J. Med. Virol. 18, 213-227. Rowe, W.P., Huebner, R.J., Gillmore, L.K., Parrot, R.H. and Ward, T.G. (1953) Isolation of a cytopathogenic agent from human adenoids undergoing spontaneous degeneration in tissue culture. Proc. Sot. Exp. Biol. Med. 84, 570. Schmitz, H., Wigand, R. and Heinrich, W. (1983) World-wide epidemiology of human adenovirus infections. Am. J. Epidemiol. 117, 455466. Van der Avoort, H.G.A.M., Adrian, T., Wigand, R., Wermenbol, A.G., Zomerdijk, T.P.L. and De Jong, J.C. (1986) Molecular epidemiology of adenovirus type 21 in the Netherlands and the Federal Republic of Germany from 1960 to 1985. J. Clin. Microbial. 24, 10841088. 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 enzyme analysis and by neutralizing monoclonal antibodies. Virus Res. 12, 1399158. Wadell, G. (1984) Molecular epidemiology of human adenoviruses. Curr. Top. Microbial. Immunol. 110, 191-220. Webb, D.H., Shields, A.F. and Fife, K.H. (1987) Genomic variation of adenovirus type 5 isolates recovered from bone marrow transplant recipients. J. Clin. Microbial. 25. 305-308.