Genetic relationship between thirteen genome types of adenovirus 11, 34, and 35 with different tropisms.

® 1991 S.Kargcr AG, Basel 0300 - 5526 / 91/0000 0338 S2.7 S/0

Intervirology 1991;32:338-350

Genetic Relationship between Thirteen Genome Types of Adenovirus 11, 34, and 35 with Different Tropisms Quan-gen L/'ab, Johan Hamhraeusb, Goran Wadellb J Department of Virology, 302nd Hospital, Beijing, People’s Republic of China; hDepartment of Virology, UrneS University, UmeS, Sweden

Keywords. Adenovirus 11,34,35 • Genome type • Genomic cluster • Pairwise comigrating restriction fragments

Introduction Subgenus B of human adenoviruses (Ad) has been divided into two DNA homology Address inquiries to: Goran Wadell, MD, Depart ment of Virology, Ume4 University, S— 90185 UmeS (Sweden) Received: February 18, 1990 Accepted: October 24, 1990

groups: B1 and B2. Ad3, Ad7, Adl6, and Ad21 are the members of group Bl; whereas Adll, Adl4, Ad34, and Ad35 are the members of group B2 [1-4]. DNA restriction analysis of Ad3 and Ad7, the two members of group Bl most frequently associated with outbreaks of respiratory disease, revealed 17 and 19 ge nome types, respectively. They could be grouped into three genomic clusters per sero type [5,6],

Downloaded by: Lund University Libraries 130.235.66.10 - 11/23/2018 8:43:26 AM

Summary. Eleven genome types of adenovirus serotype 11 (Adi 1) were identified among 20 strains isolated from healthy pregnant women and patients with urinary tract infections, respiratory tract infections, or pharyngoconjuctival fever by use of 13 restriction endonuc leases: Bam W , Bcl\, Bgl\, BglII, BuEII, £coRI, ////idIII, Hpal, ft/1, Sail, Smal, Xhal, and Xhol. All genome types could be grouped into three genomic clusters according to their genetic homology expressed as pairwise comigrating restriction fragments. The genome types within a genomic cluster were very closely related. They shared on an average pairwise comigrating restriction fragments of 91.6-97.7%. The Adll strains of genomic clusters 1and 3 were isolated from urine, whereas all the Adll strains isolated from the respiratory tract were identified as members of the genomic cluster 2. One genome type of Ad34 and one genome type of Ad35 were identified from a hemorrhagic cystitis patient and an organ transplant recipient, respectively. Both were closely related to Adll. The genome type of Ad35 could be located in the Adll genomic cluster 1.

Strains of Adll, Ad34, and Ad35 differ from the members of subgenus Bl, since they cause persistent infections of the urinary tract, including the kidney [7, 8], The Adi 1/34/35 strains are also strongly overre presented among isolates from bone marrow transplant recipients [9]. Adi 1in a few reports has also been associated with outbreaks of respiratory disease [10]. An extensive analysis of the genetic relationship between Adll, Ad34, and Ad35 strains was, therefore, per formed. Altogether eleven different genome types of Adll were newly identified by use of 13 restriction endonucleases (REs) after analyz ing the DNA of 20 strains isolated from Africa, Asia, Europe, and North America. The strains were recovered from urine or throat swabs. The urine strains were grouped into genomic clusters 1 and 3, whereas all respiratory isolates of Adi 1were identified as members of the Adll genomic cluster 2.

Materials and Methods Virus Strains Thirteen Adll strains from children with welldocumented respiratory disease were isolated in Bei jing from 1965 to 1985. They were compared to the eleven characteristic urinary isolates from transplant recipients with hemorrhagic cystitis and healthy preg nant women (table 1). The Adll prototype strain Slobitiski was isolated from a stool specimen of a child with poliomyelitis. It cannot be excluded that this virus strain originated from urine. A certain degree of serological cross-reactivity between Ad7 and Adll can be demonstrated. This may lead to spurious typing results; six of the alleged Adi 1strains sent to us were demonstrated to represent the Ad7b genome type by DNA restriction analysis. Neutralization Test All strains were typed by neutralization test in their laboratories of origin. We confirmed the sero-

339

typing of representative strains of Adll, Ad34, and Ad35 by neutralization test. Rabbit antisera against Adll, Ad34, and Ad35 were kindly provided by Dr. Lennart Svensson, National Bacteriological La boratory, Stockholm, Sweden. Serial twofold dilu tions of antisera in 0.15-ml volumes were incubated with 0.15 ml 100 TCDS0 (50% tissue culture infectious dose per 0.05 ml of virus in the A549 cell line) at 22° for I h. Incubation was continued overnight (18 h) at 4° and for I h a 22°. Two tubes with A-549 cells were then each inoculated with 0.1 ml of the mixture of virus and rabbit antisera. Final reading was per formed when the TCDS0 of the virus control reached 100. Preparation o f Viral DNA All strains were propagated in the A-549 cell line. The intracellular viral DNA was extracted by the method described by Shinagawa et al. [II], with some modifications. Briefly, virus-infected cells of 125 cm2 were har vested 2-6 days after infection. The cells were pel leted at 3,000 rpm for 10 min at 4° and suspended in 1.5 ml TE (10 mM Tris-HCI, 1 mM EDTA, pH 7.9). Then, 0.1 ml of 10% SDS was carefully added to lyse all cells; finally, 0.4 ml 5 MNaCl was added. Most of the cellular DNA was precipitated by incubation at 4° overnight, followed by centrifugation at 18,000 rpm (Beckman centrifuge, model J2-2I, rotor JA-20) for 30 min. The supernatant was extracted with an equal volume of phenol by mixing in a Vortex blender and centrifugation at 5,000 rpm (Heraeus Christ centri fuge, model Biofuge A) for 1 min. The phenol phase including the interphase was extracted twice with an equal volume of TE to remove the remaining RNA and fragmented DNA. 1.5 vol of ethanol (95%) was added and kept at -20° overnight. The precipitate was obtained after centrifuga tion at 3,000 rpm (Beckman centrifuge, model TJ-6, rotor TH-4 5-89) for 10 min at 4°, rinsed twice with 80% ethanol and ether successively, then suspended in 0.5 ml TE-containing 100 mM NaCI and 0.5% SDS, and incubated with 50 pg/m l Proteinase K (Boehringer Mannheim, FRG) at 37° until the insol uble pellet disappeared. Then, it was extracted twice with an equal volume of phenol and ether succes sively at 5,000 rpm for I min in an Eppendorf centri fuge (Biofuge A) to remove the protein bound to the 5'-terminal end o f adenovirus DNA and the remain ing phenol.


Genetic Relationship of Adi I, Ad34, and Ad35

Li/Hambraeus/W adell

340

Table 1. Origin of 24 Adi 1, Ad34 and Ad35 strains identified as 13 different genome types Genome type“

Representative strain

Place of isolation

Year of isolation

n

Specimen

Clinical diagnosis

lip llp l

Slobitski Numata

USA Japan

1954 1966

1 1

stool urine

1Ip2

5647/80

England

1980

1

urine

poliomyelitis hemorrhagic cystitis pregnant woman

lla lia i

BC34 BC3403

China China

1965-85 1981

9h

1 Ia2

BC4824

China

1985

1

1la3

BC3738

China

1981-82

2

lib

529/78

England

1978

1

throat swab throat swab throat swab urine

lie

deLange

1978

1

urine

llc l

24/78

South Africa England

1978

1

urine

1Ic2

Jennifer

1979

1

urine

Ad34a

259

South Africa Japan

1975

1

urine

Ad35p

12221/80

England

1980

1

urine

Ad35p

S-763

Sweden

1986

2

urine

1

Bronchopneumonia URTI URTI pregnant woman renal transplant pregnant woman pregnant woman hemorrhagic cystitis renal transplant bone marrow transplant

URTI = Upper respiratory tract infection. * The genome types were designated according to Li and Wadell [5]. b Two strains were isolated from eye swabs from patients with pharyngoconjunctival fever, five strains were isolated from throat swabs from 4 patients with URTI and I patient with pharyngoconjunctival fever, and two strains were isolated from feces from I patient with URTI and 1 patient with pharyngoconjunctival fever.

DNA Restriction Cleavage All the 13 REs, BamHl, BelI, Bgll, BglU, ÄsfEII, EcoRI, Hpal, ///n d lll, Pstl, Sail, Smal, Xbal, and

Xhol, were purchased from Boehringer Mannheim. All enzyme reactions were carried out according to the method of Maniatis et al. [12], Phage lambda DNA digested by /findII I was used as a size standard for determination of the molecular weight of the frag ments. Agarose Electrophoresis o f DNA Restriction Fragments The DNA fragments were separated by electro phoresis in 0.8—1.2% agarose (HGT, Sea Kern) hori


Viral DN A was precipitated from the water phase by addition of an equal volume of isopropanol fol lowed by incubation at -20° overnight. The purified adenovirus DNA precipitate was obtained after cen trifugation at 12,000 rpm (Eppendorf centrifuge 5,414) for 15 min at 4° rinsed twice with 80% ethanol, and, finally, dissolved in 100-150 pi TE.

341

Genetic Relationship of Adll, Ad34,and Ad35

Determination o f Pairwise Comigrating Restriction Fragments (PCRF) The migration distances were measured after elec trophoresis of DNA restriction fragments. Conse quently, the approximate molecular weight of the fragments could be obtained when the distances of migration were compared with the migration dis tances of restriction DNA fragments of phage lambda and OX 174. The molecular sizes of these fragments are derived from their nucleotide sequences. A pair wise comparison of genome types involves summa tion of the total number of RE fragments obtained after cleavage with all REs used, followed by determi nation of the number of fragments of the same size in the restriction fragment profiles of both members of the pair. This number of comigrating fragments is divided by the total number of fragments. The result is expressed as a percentage of PCRF. Computer Program fo r Comigration Analysis The data obtained were stored and compared us ing the ‘Comig 4’ computer program for comigration analysis of related genomes. The program was devel oped by Johan Hambraeus in the operating environ ment of MS-Dos 3.10 using the compiler Turbo-Basic (version 1.0; Borland International). Comig 4 is both a database system, designed to store the results from previously performed analyses on genome types, and a program to compare different strains with established genome types after cleavage with a maximum of 38 different REs. Terminology The results of the analyses of Adll, Ad34, and Ad35 will be discussed using the concepts of genome type, genomic cluster, and subgenus of adenoviruses that are defined in the following way: Genome Type: A distinct viral entity within a sero type identified by DNA restriction site patterns [2]. Genomic Cluster. A group of genome types that are closely related. A statistically significant difference

should be demonstrated in a comparison between the mean of PCRF within a group and the mean of PCRF between two groups [5, 6). Subgenus. Each subgenus is defined by a DNA homology of more than 50% between members within a subgenus and less than 20% between members of different subgenera [see refs. 34 and 35],

Results Identification o f Eleven Genome Types o f Adll. One Genome Type o f Ad34, and One Genome Type o f Ad35 Analysis of DNA extracted from 20 Adll strains, one Ad34 strain, and three Ad35 strains with the 13 REs (see above) revealed eleven genome types of Adll, one genome type of Ad34, and one genome type of Ad35. The Adll genome types were designated as Adllp, Adllpl, Adllp2, Adlla, Adllal, Adlla2, Adlla3, Adlib, Adllc, Adllcl, and Adllc2 according to the denominating sys tem previously described [5]. The BamH \ and Smal restriction patterns of strain 259 were different from the prototype (strain Comp ton) of Ad34 [1, 2], It was tentatively desig nated as genome type Ad34a. The restriction patterns of the three Ad35 strains after cleavage with BamH\, Bgl\, ifa/EII, EcoRI, //t'ndlll, Hpal, Pst\, and Smal were very similar to the prototype (strain Holden) of Ad35 [1, 2, 13, 14], It was tentatively designated genome type Ad35p. The schematic presentation of the DNA restriction patterns of these 13 genome types is shown in figures 1 and 2. Neither £coRI nor Sal\ and Xhol can distinguish the genome types, since they yield only one pattern. For diagnostic purposes six REs, BamHl, Bell, Bgl\, Hpa\, Pstl, and Xbal, are required to distinguish all 13 different ge nome types.


zontal slab gels which were prepared and run in 89 m M Tris-borate buffer (pH 8.3) with 2.54 mM EDTA at 50 V. The 2% agarose (Sea Plaque) gels were used for resolution of fragments in the size range of 200 bp. Gels were stained in ethidium bromide (0.5 pg/ml). Photography was performed with TM 36 UV light (wavelength 3,020 A, intensity 8 mW /cm!), using Polaroid 4 x 5 Land film type 55 or 57.

Sam HI p khn

KDP 30-

o

Bd I

b e

P 1 o1 P2 a 2 «3

A d 34a c 1 Ad35p c 2

P

pi p2

°

al o3

Bgl II

Bgl l

Pi °

a 2 Ad34a b c 1 c2

p

a 1 ,o 2 a3,b,c c 1,c2

c

Ad35P 20-

Hindill

EcoRI

p o Ad34ac pi ol cl p2 a2 c2 b a3 Ad35p

a ll

p P1 P?

Ad35p

a Ad34 c ol 3 c2 a2 a3 b c1

30 20

Ad35p

—

—

10

—

10

54 ■ 3 -

2

-

05 -

- 0.5

Fig. 1. Schematic presentation of restriction patterns obtained after cleavage of DN A from 13 genome types of Adll, Ad34, and Ad35 with BamH\y Bcl\y Bgl\y Bgl\I, EcoRI, and //rndll I. The asterisks indicate double fragments. E s lE II

kb p 30

20

P e pi el p2 c2 a o1 a2 o3 b Ad 34a Ad35o

H pa 1 P Pi p2

P s t\

a b c a 1 cl a2 Ad34a o3 Ad35P

c2

P pi p2 b

a a1 a2

S a lt

a 3 Ad34Ad35p e 3 c1 c2

o il

p2

— —

a a1 a2 03

b c cl c2 Ad34a A d35p

10 —

Xbal

Sm al P

~

--------------. —

—

-----

—

—

=

1 11

-- --_

--___

=

1=1

------- —

“

—

z — —

—

—

—

-

5 4

—

—

—

__ __

3 2

—

— ----- -----

11 11

11

----- =

10

—

—

1 1 II

— =

30 20

—

•

a ll

_ _ _ __

_ ____

----—

—

—

Xho 1

a a1 c a2 c2 p2 a 3 Ad35p c 1 b Ad 34a P

—

1

Fig. 2. Schematic presentation of restriction patterns obtained after cleavage of DNA from 13 genome types of Adll, Ad34, and Ad35 with Bs/Ell, Hpal, Pstl, Sail, Sma\, Xbal, and Xhol. Asterisks indicate double fragments.


0.5 L

Analysis o f 13 Genome Types The approximate size of each DNA res triction fragment was determined (table 2). Subsequently, the degree of the genetic rela tionship expressed as percentage of PCRF was calculated. Between 190 and 206 frag ments were compared in each pair of genome types (table 3). The relationship between 13 genome types is presented in a phylogenetic tree (fig. 3). Obviously all eleven genome types of Adll could be grouped into three clusters. They were designated as genomic cluster 1 containing Adllp, A dllpl, Adllp2, and Adlib: genomic cluster 2 containing Adlla, A dllal, Adlla2, and Adlla3, and gen omic cluster 3 containing Adllc, Adllcl, and Adllc2. A genomic cluster-specific restriction pat tern could be noted after cleavage with BglU. Genome type Ad34a constitutes one separate genomic cluster. Genome type Ad35p is so closely related to Adlib (93% PCRF) that it is considered a member of genomic cluster 1. To ascertain the relevance of suggested clusters, the mean PCRF values of each ge nome type within one cluster and between two clusters were analyzed by the chi-square method. A highly significant difference (p < 0.001) was observed. The genome types within a cluster were relatively conserved. The mean PCRF values were 91.6, 97.7, and 94.7% within genomic clusters 1, 2, and 3, respectively. In the genomic cluster 1, only one or two restriction sites of Bgl\ differed in each pair of genome types Adllp, A dllpl, and Adllp2. A dllpl was isolated in Japan 12 years after the isolation of Adllp in America, and Ad 11p2 was isolated in England 25 years after the first isolation of Adllp. Thirteen strains of the genomic cluster 2 were isolated in the same limited area, Bei

343

jing. Nine strains, recovered during a 20-year period (1965-1985), could not be distin guished from each other and were identified as Adlla. All four genome types in genomic cluster 2 were closely related. Only one to three restriction sites differed in the pairwise comparison of these four genome types. In the genomic cluster 3, Adllc, Adllcl, and Adllc2 shared 94-96% PCRF. The four to five different restriction sites of BelI, Hind lll, Hpal, and Xba\ were found in each pair of those three genome types. A dllc and Adllc2 were isolated in South Africa, but Adllcl was isolated in Europe during the same period (1978-1979). It is interesting that all genome types of the genomic clusters 1 and 3 of Ad34 were isolated from urine. All genome types of gen omic cluster 2 of Adll were isolated from throat swabs, eye swabs, or feces from pa tients with respiratory symptoms. The three strains isolated from renal or bone marrow transplant recipients were indistinguishable and identified as Ad35p. Genome types Ad34a and Adlib share 87% PCRF. They differed only in six restric tion sites; two Bell sites, two Pstl sites, and two Xbal sites. Confirmation o f Serological Typing o f Serotypes Adll. Ad34. and Ad35 When the close genetic relatedness be tween Adlib, Ad35p, and Ad34a had been demonstrated, we performed an additional neutralization assay to confirm the original serological typing of these strains. The re sults confirmed that the strain 529/78 of Adlib was serotype Adll, the strain 259 of Ad34a was serotype Ad34, and the strains 12221/80 and S-763 of Ad35p were serotype Ad35. Strain 12221/80 showed a slight cross reaction (1:20) with Adll.


Genetic Relationship of Adll, Ad34, and Ad35


344

Table 2. Molecular size of DNA of 13 genome types of Adi 1, Ad34 and Ad35 after digestion with 13 REs Bam HI

Pattern

1

2

3

Genome type

P.pl. P2

a,al, a2,a3

b, c,cl, Ad34a, c2 Ad35p

Size of fragment, bp

9,200 5,300 4,300 4,150 2,650 2,000 1,620 1,620 1,500 1,170 960 750 720

Bel I

9,200 5,400 4,300 3,400 2,650 2,550 2,000 1,620 1,620 1,500 960 720

4

9,200 5,300 4,300 4,150 3,400 2,000 1,620 1,620 1,500 1,170 960 720

9,200 5,400 5,300 4,300 3,400 2,000 1,620 1,620 1,500 960 720

Bgl I

1

2

3

4

P.pl, p2,c, Ad35p

a,al, a3

a2

Ad34a b,cl, c2

7,200 5,930 5,930 4,980 4,400 3,280 3,200 1,070

10,400 12,000 6,150 6,150 5,930 5.930 4,400 4,400 3,280 3,280 3,200 3,200 1,550 1,070 1,070

5

7,200 6,150 5,930 4,400 3,280 3,280 3,200 1,550 1,070

1

2

P

p l.a.al, a2,a3,b. c,cl. c2,Ad35p

7,200 13,300 13,300 5,930 12,000 11,000 5,930 4,500 4,500 4,400 4,170 4,170 3,280 2,080 2,080 940 3,280 3,200 1,550 1,070 220

Hpa I

Eco RI Hind III

Restriction enzyme

Bst Eli

Pattern

1

2

1

I

2

Genome type

P.pl. p2,a. al,a2, a3,b, Ad34a, Ad35p

c,cl c2

all

P.pl. P2

Ad34a c,c2 a.al. a2,a3. b,Ad35p, cl

Size of fragment, bp

10,500 7,600 6,900 4,350 3,250 2,600 760

10,500 23,000 11,300 11,300 7,600 7,100 5,800 5,800 6,900 5,900 5,200 5,200 3,500 3,500 4,350 3,250 3,350 3,450 2,600 3,350 1,450 2,280 2,280 1,150 1,400 1,400 760 800

3

4

4

1

2

3

P.pl. P2

a.al, a2,a3

b.cl, c Ad34a, Ad35p

11,300 11,300 10,800 10,800 10,800 10,800 5,800 5,800 7,400 8,850 7,400 7,400 5,200 5,200 6,800 7,400 7,400 4,900 3,500 4,700 4,900 4,900 4,900 3,700 3,450 3,500 2,580 2,580 2,580 3,500 2,300 3,350 1,500 1,500 1,500 2,580 1,450 1,500 2,280 2,280 1,450 1,450 1,400 650 1,070


Restriction enzyme

345

Genetic Relationship of Adll, Ad34, and Ad35

Pst I

3

4

1

2

3

4

1

2

3

4

5

P2

Ad34a

p,pl,p2, b,Ad35p

a,al, a2,a3

Ad34a

c,c, c2

P,pl, p2,b

a,al. a2

a3

Ad34a

Ad35p c,cl, c2

13,300 11,000 6,580 4,170 940

11.000 10,500 4,500 4,170 2,800 2,080 940

13,700 6,000 4,300 2,900 2,300 1,950 1,850 1,150 1,020 730

12,000 6,000 4,900 4,500 2,300 1,850 1,620 1,020 980 730

Sal I

Sma I

5

1

1

2

3

1

2

c2

all

p,pl P2

a.al. a2,a3

b,c, cl,c2. Ad34a, Ad35p

P.pl. p2,b

a,a2, a1 a3,cl. Ad34a

c,c2, all Ad35p

10,800 7,400 4,900 3,700 3,600 2,580 1,620 1,450

12,000 9,600 9,000 5,400

24,600 3,600 3,300 2,900 840 750

24,600 24,600 3,600 4,350 3,300 3,300 1,880 1,880 1.020 1,020 840 840 750

24,600 5,180 3,600 1,020 840 750

9,800 9,000 6,700 3,300 2,500 1,950 1,800 1,070

8,800 6,000 4,500 3,200 2,900 2,300 1,950 1,850 1,620 1,020 980 730

8,800 6,000 5,500 3,200 2,900 2,300 1,950 1,850 1,620 1,020 730

7,700 6,800 4,250 3,450 2,780 2,600 2,080 1,340 1,340 1,070 1,050 880 610

12,400 7,700 4,250 2,780 2,600 2,320 1,340 1,070 880 680

8,950 7,700 4,250 3,450 2,780 2,600 2,320 1,340 1,070 880 680

Xba 1

9,800 9,000 6,700 3,300 2,500 1,950 1,070 920 880

9,000 9,000 6,700 3,300 2,500 1,950 1,070 920 880 800

7,700 6,800 4,250 3,450 3,250 2,780 2,600 2,080 1,070 1,050 970

6

7,700 6,800 5,600 2,780 2,780 2,600 1,550 1,340 1,340 1,070 1,050 880 610

8,800 5,350 4,250 3,450 3,450 2,780 2,600 1,820 1,430 1,070 980

Xho I

3

4

1

11,200 8,300 6,700 5,800 2,700 1,400


Bgl II

346


1Ipl 99 lip lip ! 1Ip2 lib Ad35p 1la lla l 1Ia2 1Ia3 lie llc l lc2

1Ip2

lib

Ad35p

1la

lla l

1la2

1 1a3

lie

llc l

1 Ic2

Ad34a

97 99

89 91 89

86 88 86 93

76 77 76 80 79

74 76 74 78 77 98

76 78 76 79 79 98 97

76 78 76 80 78 98 97 97

74 76 74 80 81 76 74 76 77

74 76 74 85 81 80 78 79 80 94

72 73 72 81 79 75 73 74 76 96 95

77 79 77 87 83 81 79 80 82 80 85 80

Distribution o f Genome Types o f A dll Africa. Adi lc and Adllc2 of genomic clus ter 3 were isolated in South Africa in 1978 and 1979, respectively. They are closely related to each other and share 96% PCRF.

Asia. All the four genome types of gen omic cluster 2 were identified in China. They are very closely related and display 97-98% PCRF. Adlla predominated in the Beijing area fo r20 years (1965-1985). The nine Adlla


Tabic 3. Percentage of PCRF of 13 genome types of Adi I. Ad34 and Ad35 obtained after cleaving with 13 REs

347

Genetic Relationship of Adi I, Ad34, and Ad35

Table 4. PCRF analysis performed between the 57 genome types of Ad3 and Ad7 (subgenus Bl group), Adi 1, Ad34 and Ad35 (subgenus B2 group), and Ad4 (subgenus E)a Ad3 Ad 3 Ad 7 A dll Ad34 Ad35 Ad4

110.8/164.8 22.0/169.0 22.9/174.3 22.9/168.3 2.8/179.5

Ad 7

Adi 1

Ad34

Ad35

Ad4

67

13 17

13 16 81

14 17 82 83

2 1 7 7 7

28.8/171.6 28.8/176.9 28.8/170.9 1.7/182.0

164.2/203.1 163.7/199.1 13.6/209.0

170.0/204.0 15.8/213.9

14.8/209.9

strains were associated with upper respira tory tract infections or pharyngoconjunctival fever. A dllal, isolated from a throat swab, was the causative agent of bronchopneumo nia. Adlla2 and Adlla3 were also associated with upper respiratory tract infections. A strain identified as Adllpl of genomic clus ter 1 was isolated in Japan 1966. Europe. Adllp2 and Adlib of genomic cluster 1 were isolated in 1980 and 1978, respectively. Adllp2, Adlib, and Adllc of genomic cluster 3 were isolated from pregnant women without any clinical symp toms. Analysis o f the Genetic Relationship between Serotypes Ad3 and Ad7 (subgenus Bl), Adll, Ad34 and Ad35 (subgenus B2), and Ad4 (subgenus E) To estimate the genetic relationship be tween Ad3, Ad4, Ad7, Adi 1, Ad34, and Ad35, we have subjected each identified genome type to pairwise analysis using at least 12 REs. Comparisons of the mean values of the

PCRF of all identified genome types within and between each serotype have been per formed (table 4). A limited pairwise compar ison of PCRF of the prototypes of all adeno virus serotypes belonging to subgenus B has been previously described [1,2,4], This exten sive analysis of 57 different genome types of Ad3, Ad4, Ad7, Adll, Ad34, and Ad35 with 12- 16 REs, involving 1,682 pairs, corrobo rated the conclusion that human adenovirus subgenus B should be divided into two groups. The mean of PCRF was 67% between genome types of Ad3 and Ad7 and 81-83% among genome types of Adll, Ad34, and Ad35. Ad3 and Ad7 should be placed in the same group subgenus Bl, whereas Adll, Ad34, and Ad35 should be in another group subgenus B2. The mean values of PCRF be tween subgenus Bl and subgenus B2 were 13- 17% which indicated that the genetic rela tionship was distant. It is interesting that the relationship between subgenera E and B2 was closer than the relationship between E and Bl.


“17 Ad3,8 A d4,19 A d7,11 Adi 1,1 Ad34,and 1 Ad35 genome types were compared. Upper right: percentage of PC RF. Lower left : the numerator is the mean of comigrating fragments ; the denominator is the mean of total compared fragments in one pair of genome types.

Discussion Adi I was first isolated from a fecal speci men from a child diagnosed as having pol iomyelitis in 1954; poliovirus type 1 was also recovered [15,16]. Adll has been reported to be associated with upper or lower respiratory illness [10,17-19], hemorrhagic cystitis in chil dren and young adults, especially in renal transplant recipients [7, 8, 20-24], and ep idemic hemorrhagic conjunctivitis [25,26]. Adll and the closely related Ad34 and Ad35 accounted for less than 1% of the 24,184 adenovirus strains reported to the World Health Organization from 1967 to 1976 [27], The prevalence of neutralizing antibodies against Adll was around 2% among453 Italian children [28], However, Adll strains were more frequently isolated in China. A total of 313 adenovirus strains were isolated at the 302nd Hospital, Beijing, during a period of 21 years (1965-1985). Thirty-five (11.2%) of them were identified as Adll by neutralization test in primary human embryo kidney cells. The frequency of Adll isolates was only super seded by Ad3 (39.6%), Ad7 (18.2), and Ad2 (11.8%). An extensive survey of the prevalence of antibodies against Adll among 5,405 sera, the majority from children below the age of 7 years, from 12 provinces and areas in the east ern part of China revealed that 4.4% displayed HI antibodies specific for Adll [29]. Adll, Ad34, and Ad35 can cause persis tent infections of the kidney and possibly also of the urinary tract. They have been isolated from renal transplant recipients, bone marrow transplant recipients, and AIDS patients [7, 13, 30-33]. Adll probably can cause latent infection in the urinary tract, because shedding of this virus in the urine has been seen in healthy pregnant women and children [21].

In this study, the genome types of genomic clusters 1 and 3 of Adll and genome types Ad34a and Ad35 seem to be associated with tropism for the urinary tract, whereas all ge nome types of genomic cluster 2 of Adll seem to be associated with tropism for the respira tory tract. Out of 35 strains from the 302nd Hospital, Beijing, 25 were isolated from throat swabs and 2 from patients suffering from pneumonia. From the winter 1964 to the spring 1980, 34 cases of Adll-associated pneumonia were studied at the Institute of Pediatrics, Chinese Academy of Medical Sciences, Beijing [18]. Ad34a and Ad35 were both so closely re lated to all the three Adll genomic clusters that they would be regarded as members of the same group of genomic clusters, be it not for the fact that they represent two serotypes distinct from Adll. We interpret this observation so that the genomic clusters are senior, forming con served entities that together constitute an adenovirus subgenus. The continuing evolu tion of new genetic variants is mirrored by the numerous genome types that can be identi fied in most genomic clusters. The serotype is apparently defined by strongly conserved epitopes in the hexon, since serotype-specific epitopes are expressed in genome types of serotypes 3, 4, 7, and 11, belonging to several different genomic clusters [5,6; see ref. 36]. This means that strains of the same sero type represent numerous different genome types that could be grouped into different genomic clusters. The information presented in this paper infers that the serotype entity may result from minor alterations resulting in the definition of new serotypes, although the genetic dis tance between different serotypes may be small. In fact, Ad35 and the Adll genomic



348

cluster I are more closely related than any pair of Adll genomic clusters. The early analysis of the DNA homology between Ad3, Ad7, and Adll, performed by liquid hybridization, demonstrated no differ ence between the homology obtained for Ad3 compared to Ad7 (80.9-92.6%) and the homology measured between Ad7 and Adll (84.1-88.6%) [35], The PCRF analysis based on 48 genome types was more efficacious, allowing identification of two different groups of genomic clusters within subgenus B which gave 67% PCRF within the subgenus B1 group, 81-83% PCRF within the subgenus B2 group, and 13-17% PCRF between subge nus B1 and subgenus B2. In the analysis of DNA homology be tween members of different subgenera, DNA of Ad4 (subgenus E) hybridized to Ad7 and to Adll to an extent of 16.3—19.8 and 6.8-8.9%, respectively (35]. The PCRF analysis using 57 genome types displayed only

Terminal dilution purification of adenovirus prototypes, and the antigenic relationship between types 4, 16 and 14.

Conserved sequences of the adenovirus genome for detection of all human adenovirus types by hybridization.

The relationship between different types of dissociation and psychosis-like experiences in a non-clinical sample.

In vitro characterization of human adenovirus type 55 in comparison with its parental adenoviruses, types 11 and 14.

Microleakage between different types of base materials.

Genome stability of adenovirus types 3 and 7 during a simultaneous outbreak in Greater Manchester, UK.

Sensitive analysis of genetic heterogeneity of adenovirus types 3 and 7 in the Soviet Union.

Analyzing Relationship Between Monoclonal Gammopathy of Undetermined Significance (MGUS) with Different Types of Neuropathy: An Observational Study.

Relationship between vacA Types and Development of Gastroduodenal Diseases.

Different types of secondary information in the genetic code.

Genetic Relationship between Schizophrenia and Nicotine Dependence.

Genetic difference between Europeans and Indians: tissue and blood types.

Relationship between physical attributes and heat stress in dairy cattle from different genetic groups.

Relationship Between Different Types and Forms of Anti-Asthmatic Medications and Dental Caries in Three to 12 Year Olds.

Relationship between the formant structure of different types of singing voices and the dimensions of supraglottic cavities.

Are there differential relationships between different types of childhood maltreatment and different types of adult personality pathology?

The sequence of the genome of adenovirus type 5 and its comparison with the genome of adenovirus type 2.

Talonavicular joint coverage and bone morphology between different foot types.

Whole-genome sequence-based genomic prediction in laying chickens with different genomic relationship matrices to account for genetic architecture.

Relationship between inflammation and sex hormone profile in female patients receiving different types of renal replacement therapy.

Alternation between different types of evidence attenuates judgments of severity.

Comparison between efficacy of four different types of orthodontic separators.

Importance of distinguishing between different types of health app.

Genetic relationship between acylpeptide hydrolase and acylase, two hydrolytic enzymes with similar binding but different catalytic specificities.