J. gen. Virol. (1987), 68, 1971-1982.

Printedin Great Britain 1971

Key words: BVD virus~bloodmononuclear cells/persistent infection

Demonstration of Bovine Viral Diarrhoea Virus in Peripheral Blood

Mononuclear Cells of Persistently Infected, Clinically Normal Cattle
By H. B I E L E F E L D T O H M A N N , I * ~ " L. R O N S H O L T 2 A N D I . B L O C H 1
1Department of Veterinary Virology and Immunology, The Royal Veterinary and Agricultural
University, 13 Bulowsvej, DK-1870 Frederiksberg C and 2The State Veterinary Institute for
Virus Research, Lindholm, Kalvehave, Denmark

(Accepted 23 March 1987)

SUMMARY
Peripheral blood mononuclear cells (PBL) from cattle known to be persistently
viraemic with bovine viral diarrhoea virus (BVDV) following a foetal infection, were
examined for the presence of viral antigens and cell-associated infectious virus. Using
immunocytochemical techniques, physical separations of PBL subsets and virus
isolation techniques (directly and by cocultivation) it was found that infection occurred
Downloaded from www.microbiologyresearch.org by

in B and T lymphocytes, monocytes, and a group of cells designated null cells for lack of
more specific classification. The latter three groups also supported viral replication, as
infectious virus could be isolated from enriched cell populations. BVDV-like particles
On: Tue, 06 Sep 2016 21:46:32

in cytoplasmic vesicles of PBL subsets were detected by electron microscopy.

IP: 120.188.92.144

INTRODUCTION
Bovine viral diarrhoea virus (BVDV) is the aetiological agent of mucosal disease (MD) in
cattle, and may cause a whole range of other clinical conditions, now included in the BVD
syndrome (St6ber, 1984; Bielefeldt Ohmann & Babiuk, 1986b). BVDV is currently classified as
a member of the Pestivirus genus of the non-arbo togaviruses (Westaway et al., 1985). The virus
has a marked tropism for lymphoid and epithelial (keratinized and mucosal) tissues (Bielefeldt
Ohmann, 1983) and in vitro it can replicate in cell cultures derived from such tissues.
For decades the pathogenesis of MD remained a complete enigma. Animals succumbing to
MD, which is an inevitably fatal course of the infection, are viraemic and virus can be isolated
from almost every tissue. However, whether the course of the disease is acute or more
protracted, perhaps even chronic, extending over several months, the animals usually fail to
develop BVDV-specific neutralizing antibodies. This may suggest some kind of virus-specific
tolerance. Very recently some understanding of the mechanisms involved in the disease have
emerged. Thus, it has been shown that MD only develops in cattle persistently infected with
non-cytopathic BVDV, established by foetal infection during the first 3 months of gestation
(McClurkin et al., 1984; Roeder & Drew, 1984; Brownlie et al., 1984). This early infection
induces an apparent tolerance to the virus and the virus persists and continues to replicate. The
tolerance appears to be virus strain-specific, but whether it is absolute and invariably permanent
seems controversial (Gardiner et al., 1983; Terpstra, 1981 ; Westbury et al., 1979; L. R~nshoit,
unpublished data). The question that is central to the understanding of the pathogenesis of MD,
i.e. exactly what precipitates the disease in apparently healthy, persistently infected animals,
has also created much controversy. Various hypotheses for these precipitating factor(s) have
been put forward such as superinfection with heterologous BVDV strains (Liess et al., 1983;
Brownlie et al., 1984; Bolin et al., 1985), the emergence of a cytopathic mutant of the persisting
virus strain (Brownlie et al., 1986), hormonal changes related to puberty (Roeder & Drew, 1984)
or superinfection with an agent unrelated to BVDV (Littlejohns & Walker, 1985).

t Present address: Veterinary Infectious Disease Organization, 124 Veterinary Road, Saskatoon,
Saskatchewan, Canada S7N 0W0.

0000-7559 © 1987 SGM

 1972 H. BIELEFELDT OHMANN, L. RONSHOLT AND B. BLOCH
Questions p e r t i n e n t to the u n d e r s t a n d i n g o f the p a t h o g e n e s i s c o n c e r n the m e c h a n i s m ( s )
g o v e r n i n g the e s t a b l i s h m e n t and m a i n t e n a n c e of the persistent virus infection, a n d the identity
of the cell(s) h a r b o u r i n g and replicating the virus. Virus can be isolated f r o m buffy c o a t cells o f
the p e r i p h e r a l blood and, being circulating cells, these m a y serve as vehicles for the
d i s s e m i n a t i o n o f virus to o t h e r target organs. T h e a i m of the p r e s e n t i n v e s t i g a t i o n s has b e e n to
identify the cell type(s) in the p e r i p h e r a l blood o f persistently infected, clinically healthy cattle
that harbours and p e r h a p s replicates the virus.

METHODS
Animals. Eleven calves (four non-infected and seven persistently infected following an experimental infection
with a mixture of cytopathic and non-cytopathic BVDV strains in the first 3 months of foetal life) were bled
regularly over a 9-month period (age 10 to 19 months). Two of the persistently infected calves succumbed to
clinical MD during the investigation period. The remaining nine (four non-infected and five infected) were
healthy at the age of 18 to 20 months. A separate, persistently viraemic blood donor calf was the offspring of a cow
inseminated with a BVDV-containing semen sample (A. Meyling, unpublished data). Blood samples were also
obtained from two clinically healthy yearlings, naturally infected (viraemic) with BVDV and identified as such
during a control examination. Additional blood samples were drawn from clinically healthy, seropositive cattle of
all ages and different breeds.
Peripheral blood mononuclear cellpreparations. Blood was drawn by venipuncture either into syringes containing
citrate-dextrose or into syringes and defibrinated mechanically with glass beads; the peripheral blood
mononuclear cells (PBL) were isolated by density centrifugation on Ficoll-Hypaque (Pharmacia) as previously
Downloaded from www.microbiologyresearch.org by

described (Bielefeldt Ohmann et al., 1983). Adherent (AD) cells were depleted by incubating 5 x 107 to 10 x 107
cells, suspended in 5 ml RPMI 1640 (Gibco) with 10~ horse serum (HS; Gibco) in 25 cm 2 tissue culture flasks
(Nunc) which, prior to cell seeding, had been coated by incubation for 1 h at 38 °C with heat-inactivated foetal
On: Tue, 06 Sep 2016 21:46:32

bovine serum (found to be devoid of BVDV-specific antibody and BVDV).

After incubation for 2 h at 38 °C, the non-adherent (NA) cells, resuspended in the medium, were collected and
IP: 120.188.92.144

pooled with the first aliquot of wash fluid (RPMI). Further depletion of monocytes (as well as some B cells, see
Table 5) from the plastic-NA cells was accomplished by Sephadex G-10 column passage as previously described
(Bielefeldt Ohmann et al., 1983) except that nylon wool was used in place of glass beads.
The plastic-AD cells were washed several times in cold Ca 2+- and Mg2+-free phosphate-buffered saline (PBS)
and incubated for 5 rain with ice-cold EDTA. Final detachment of the ceils was accomplished by vigorous
pipetting. They were then washed twice in cold PBS, counted and employed in the various assays. The viability of
the AD cells was always more than 98~, as indicated by trypan blue exclusion.
Depletion of subpopulations by treatment with monoclonal antibody (MAb) and complement. Complement (C')-
mediated lysis of PBL subpopulations, defined by reactivity with MAb (see below and Table 1) was accomplished
as described previously (Bielefeldt Ohmann et al., 1985). The complement source was rabbit serum (Glapco Aps,
Aarhus, Denmark or Cedarlane Laboratories, Hornby, Ont., Canada). After determining the proportion of dead
cells by staining with trypan blue, the dead cells were removed by differential centrifugation on Ficoll-Hypaque.
Treatment with L-leucine methyl ester (L-LME). PBL were treated with L-LME (Sigma) in order to deplete the
monocytes, following a protocol adapted for the bovine system (Bielefeldt Ohmann et al., 1985). The L-LME was
used at a final concentration of 10 mM. In some experiments an attempt was made to remove damaged and dead
cells by centrifugation on Ficoll-Hypaque.
The purity of the cell populations obtained by the above methods are given in Table 5. Cell yields varied with the
method employed (see cited literature).
MAbs. Murine MAbs with specificity for bovine leukocyte populations were obtained from Dr W. C. Davis
(Washington State University, Pullman, Wash., U.S.A.) and Dr A. J. Teale (ILRAD, Nairobi, Kenya). Table 1
summarizes the reported properties of these MAbs. CHI28A, CH61A and B26A(4) detect the same cell
population. However, B26A(4) shows poor reactivity on fixed cells, but binds complement well. For similar
reasons PIg45A, reactive with bovine B cells, was only employed for the antibody-complement lysing procedure
(see above). Three MAbs, TH14B, TH8IA and H42A, react with bovine major histocompatibility complex
(MHC) class II antigens (Ia-like antigens) which are the equivalents of human DR, DQ and DP, respectively.
These antigens are present on B lymphocytes and on most monocytes.
Histochemistry and immunocytochemistry (ICC). Cytospins (cytospin centrifuge; Shandon Southern Products
Ltd., Runcorn, U.K.) were made on cleaned poly-L-lysine (mol. wt. 500K; Sigma)-coated glass slides, from all cell
preparations. One set was fixed in formaldehyde and stained for non-specific ~-naphthyl butyrate esterase (Koski
et al., 1976), in order to detect cells of the mononuclear phagocyte lineage (Bielefeldt Ohmann et al., 1983).
Another set of cytospin preparations for ICC were fixed in acetone~chloroform (1 : 1) for 8 min with thorough air
drying both before and after fixation. The slides were stored at - 20 °C in sealed boxes until staining could be
performed.
 BVDV in blood o f p e r s i s t e n t l y i n f e c t e d cattle 1973

T a b l e 1. R e a c t i v i t y o J M A b s
Designation
of MAb Isotype Reactivity Source/reference
BIg73A IgG~ Bovine IgM, B cells W.C. Davis (personal
communication)
PIg45A IgG2B IgM Davis et al. (1987)
CH128A IgGl All T cells (E rosette receptor) Davis et al. (1987)
CH61A IgGl All T cells Davis et al. (1987)
B26A IgM All T cells Davis et al. (1987)
DH59B IgG~ Monocytes, granulocytes Davis et al. (I987)
TH14B IgG2A MHC-II (equivalent to human DR) Davis et al. (1987)
TH81A IgG2A MHC-II (equivalent to human DQ) Davis et al. (1987)
H42A IgG2A MHC-II (equivalent to human DP) Davis et al. (1987)
IL-A12 IgG2A BoT4, 30~* Baldwin et al. (1986)
Morrison et al. (1986)
IL-A17 IgG, BoT8, 20~* Morrison et al. (1986)
IL-A23 IgG ~ Macrophage differentiation antigen; 1-11 ~* Ellis et al. (1986)
*Percentages mentioned are the proportion of peripheral blood leukocytes in normal, adult cattle positive
for the antigen, as determined by fluorescence-activated cell sorting analysis.
Downloaded from www.microbiologyresearch.org by

The protocols for single and double immunolabelling for detection of either virus or leukocyte antigen or both,
respectively, were as described in detail elsewhere (Bielefeldt Ohmann, 1987). Briefly, BVDV antigens were
detected by indirect immunoperoxidase staining of cytospin preparations following blocking of endogenous
On: Tue, 06 Sep 2016 21:46:32

peroxidase activity and non-specific antibody binding. The primary antibody was purified swine anti-BVDV
immunoglobulin (Ig)G and the conjugate was rabbit anti-swine Ig-horseradish peroxidase (RAS-HRP,
IP: 120.188.92.144

Dakopatts, Glostrup, Denmark). The specific binding was visualized with a 0.05~ diaminobenzidine (DAB)
substrate. On some occasions a fluorescein isothiocyanate-conjugated rabbit anti-swine Ig (Dakopatts) was used
for visualization of infected cells.
A triple-step alkaline phosphatase-anti-alkaline phosphatase (APAAP) system was used for leukocyte antigen
detection. Initially, slides were incubated with the appropriate dilution of MAb (Table 1), followed by incubation
with rabbit anti-mouse immunoglobulin (Dakopatts) and finally with the APAAP complex (Dakopatts, code no.
D 651). Washing sequences of 4 x 5 rain in PBS were performed after each incubation. The reaction was
visualized by incubating the slides with a chromogen substrate (2 mg naphthol AS-MX phosphate in 0.1 M-Tris-
buffer pH 8.2, 1 mu-levamisole and 10 mg fast red TR salt). After counterstaining with methyl green the cell
preparations were mounted in aqueous gelatine (Glycergel, Dakopatts).
Double immunolabelling for both virus and leukocyte antigen was performed by applying the sequences
mentioned for the APAAP system and including swine anti-BVDV lgG in the first and second step and RAS-
HRP in the third step (Bielefeldt Ohmann, 1987). For visualization of the two enzyme conjugates the slides were
first incubated with the DAB substrate followed by washing in PBS and then incubation with the naphthol-fast
red substrate.
Specificity controls for BVDV detection consisted of adsorption of the antiserum to BVDV-infected monolayer
cultures, fixed in 100% acetone for 10 min. As controls for both the virus and the leukocyte antigens were also used,
omission of either of the antibodies, replacement of the primary Ab with a nonsense Ab (swine anti-rotavirus
serum in the case of BVDV, MAbs to human T4 or T8 antigens in the case of the leukocyte antigens) or with
normal serum from the species of origin. Negative controls comprised cell preparations from BVDV-free animals.
Two or more slides were stained for each antigen. The preparations were read blind, and a minimum of 300 cells
per slide were recorded.
Virus isolation and titration. Madin-Darby bovine kidney (MDBK) cells, free from BVDV and mycoplasma
(kindly provided by Dr A. Meyling) were grown in Eagle's minimum essential medium (MEM) supplemented
with L-glutamine (292 mg/1, Gibco), sodium bicarbonate, gentamicin (50 ~tg/ml, Gibco), and HS (7% for
outgrowth and 2 ~ for maintenance). Samples of the various cell preparations were disintegrated by three cycles of
freezing and thawing and the TCIDs0 of lysed material, blood plasma (procured after the initial centrifugation of
blood for PBL isolation) or wash fluid from the PBL isolations was estimated by twofold titration in a microtitre
system (Nunc) (Bielefeldt Ohmann, 1981). Using three or four wells per sample dilution, a suspension of MDBK
cells (1 × l0 s cells per well) was incubated with the inoculum in a total volume of 100 ~tl per well for 1 to 2 h at
38 °C. Following addition of 100 ~tl MEM supplemented with 8 % HS to each well, incubation was continued for 72
h. At this time the cultures were checked for c.p.e, and then terminated. Following a brief wash in PBS, the
cultures were dried (1 h, 38 °C), fixed in 20%0 acetone in 0.85% NaC1 with 0.02% bovine serum albumin, again
 1974 H. BIELEFELDT OHMANN, L. RONSHOLT AND B. BLOCH
Table 2. Frequency of virus antigen-positive cells among PBL and yieM of infectious virus
Virus-positive
Animal no. cells* (~o) TCIDso/106 cellst n:~
6 18.3 + 3.5 32-512 8
KI0 16-3 ± 1.5 160-512 4
TI0 13-6 _+ 2.3 128-254 5
13 9.3 + 1-4 64-188 4
14 30.3 ± 3.3 128-256 7
144 24.4 _+ 3.4 128-1024 9
* Determined by ICC (see Methods), percentage of non-fractionated PBL (mean ± S.D.).
t Direct isolation from lysates of PBL.
Number of independent samplings during an 8 to 9 month period.

dried for 2 h at 38 °C, and then either stained immediately for BVDV antigen or stored in sealed bags at - 20 °C to
be stained later.
For immunocytochemicaldetection of BVDV antigen, cultures were incubated with swine anti-BVDV IgG for
20 rain at 38 °C, followed by a 15 rain incubation with RAS-HRP. The plates were washed five times in PBS with
l ~oTween 80 after each antibody incubation. The immunolabelling was visualized by incubating with substrate
containing 2 mg 3-amino-9-ethylcarbazoleper 5 ml 0.1 l-acetic acid/acetate buffer pH 5, and 0-015% H,O2.
Downloaded from www.microbiologyresearch.org by

Evaluation of the labelling was either performed immediately at the end of a 20 to 25 rain incubation period, or
substrate was replaced by a 4 ~ formaldehyde solution and the plates were stored at 4 °C for reading later. Most
assays were performed twice.
Virusdetection by cocultivation.Since subpassage of cocultures of leukocytes and MDBK cells did not increase
On: Tue, 06 Sep 2016 21:46:32

the detectability of virus-replicating cells (data not shown), we turned to the use of a microsystem. The leukocyte
samples were seeded in quadruplicate sets of two-fold dilutions in 96 well tissue culture plates, from 1 × 10~ cells
IP: 120.188.92.144

per well to several dilutions beyond one cell per well. MDBK cells were then added and the cocultures, in a final
volume of 200 ~1 of RPMI : MEM/5 ~ HS, were incubated in a 5~ COz atmosphere at 39 °C for 96 h. At this time
the cultures were checked visually for c.p.e., fixed and stained for virus antigen as described above.
Electronmicroscopy. Pellets of the various cell preparations were prepared for transmission electron microscopy
as previously described (Bielefeldt Ohmann & Bloch, 1982). In some cases the leukocytes were labelled in the
living state for surface antigens by an immunogold technique employing the MAbs listed in Table I and
subsequently prepared for electron microscopical examination (Bielefeldt Ohmann & Bloch, 1982; H. Bielefeldt
Ohmann, B. Bloch, W. C. Davis & J. Askaa, unpublished results).
Statistical analysis. Where appropriate, comparison of data from non-infected and infected animals was done
using Student's t-analysis.

RESULTS
Non-cytopathic BVDV was readily isolated from serum, plasma or cell lysates (after freezing
and thawing) of calves with known foetal exposure to the virus and lacking virus-specific
neutralizing antibodies at the age of 6 months and onwards. The cell-related virus was not "carry-
over' from the plasma as indicated by testing the third washing solution following Ficoll-
Hypaque isolation of the PBL. This solution was always negative for virus even after four
passages (data now shown). The results of the direct virus isolation from PBL were supported by
immunocytochemical detection of intracellularly located viral antigen in 5 to 3 6 ~ of the PBL.
Table 2 summarizes the results obtained from six calves (omitting the data from a large n u m b e r
of repeatedly virus-negative animals). Whereas the frequency of antigen-positive cells appeared
to be fairly constant for a particular animal during the investigation period (8 to 9 months),
differences between the individual calves were evident (Table 2). The amount of infectious virus
rescued from cell lysates of PBL varied both for the individual animal between samplings
(ranges shown in Table 2) and between animals, and could not be directly related to the
frequency of infected cells (Table 2). There was no direct correlation between the frequency of
virus antigen-positive cells as detected by cocultivation of PBL with M D B K cells (data not
shown). At no time was cytopathic virus isolated from any of the calves and no reversion to
cytopathogenicity was seen with any of the isolates during multiple cell culture passages (more
than 12).
 BVD V in blood of persistently infected cattle 1975
Table 3. Distribution of virus-infected cells among PBL subpopulations in persistently infected
calves
Positive cells (~)t
Phenotypic marker (PM) r ~
(monoclonal antibody)* PM-/V- PM+/V- PM-/V + PM+/V+ n:~
B cell (IgM) 63-0 20.1 16.0 0.91 16
(BIg73A) (49.0-78-5) (12.0-29-0) (4-0-29.0) (0-0-4-5)
All T cells 49.4 30.1 12-2 8-3 7
(CH128/CH61A) (42-0-56.7) (21-0-41.7) (8.7-17-0) (2,7-14.0)
BoT8 63-4 16.8 14.6 5.1 10
(IL-A17) (57-0-80.5) (8.5-26.5) (6-5-25.5) (2.0-9.0)
Monocytes 68.4 12.7 14.8 4.1 13
(DH59A) (57.5-81.0) (6.5-18-5) (3.5-26.5) (0-5-14.0)
Monocyte subpopulation 80.9 3.6 12-5 3.0 7
(ChI6A/CH137) (65.5-94.5) (0.5-7.0) (4.0-22-0) (0.0-6.5)
MHC-II (DR equivalent) 58.2 24.7 11.6 5.5 11
(TH14B) (43.0-69.8) (15.1 3 1 . 0 ) (3.5-22.0) (1.0-11.0)
MHC-II (DQ equivalent) 53.4 22.5 10-9 3.2 10
(TH81A) (57.0-70.5) (15.5-28.5) (9.0-22-0) (0.5-5-5)
MHC-II (DP equivalent) 59.0 22.8 13.3 4.9 9
(H42A) (50.0-73.0) (16.0-29.6) (7.8-19.5) (2-7 8-5)
Downloaded from www.microbiologyresearch.org by

* The codes in parentheses refer to the MAbs used for the detection of phenotypic markers, as listed in Table 1.
Percentage of non-fractionated PBL. The total frequency of virus-positive cells (i.e. PM-/V + plus PM+/V+) is
not exactly the same in all sets because these were not generated from the same calves in all cases. The averages
given should, therefore, only be taken as indicative.
On: Tue, 06 Sep 2016 21:46:32

:~Number of animals examined (includes repeated examinations of some animals).

IP: 120.188.92.144

Identification of virus-infected cells by immunocvtochemical procedures

Double immunolabelling procedures, employing routine MAbs against bovine leukocyte
antigens (Table 1) and a polyclonal swine IgG against BVDV using two enzyme-conjugated
secondary antibodies, revealed that infected cells could be found within all four major subgroups
of cells in PBL, i.e. monocytes, T and B lymphocytes and null cells (Table 3 and Fig. 1).
However, each subgroup did not contribute equally to the infected pool. Based on data included
in Tables 2 and 3 it can be estimated that T cells (CH128A+/CH61A + cells) constituted 40 to
50~, B cells (BIg73A*/MHC-II +) less than 4~o, monocytes (NSE +, DH59B÷/MHC-II +) 17 to
2 4 ~ and null cells (non-T, non-B, non-macrophage cells) 24 to 4 0 ~ of the virus antigen-positive
cells.
The persistent infection did not cause any significant changes in the total leukocyte numbers
in the blood (not shown), but shifts in the frequencies of the mononuclear cell subpopulations
were apparent (Table 4). Thus, the frequencies of B lymphocytes (BIg73A + cells) and monocytes
(DH59B +, NSE + cells) were elevated, the latter significantly (P < 0.05), while the frequency o f T
lymphocytes was significantly lowered (P < 0.05) in the persistently infected calves compared to
non-infected calves. From this it can also be concluded that the frequency of null cells (non-T,
non-B, non-macrophage cells) was elevated in the infected calves. From the data in Tables 3 and
4 it can be inferred that in the animals included in this study, approximately 2 0 ~ of all T-cells,
4 ~ of the B cells, 2 4 ~ of the monocytes and 5 3 ~ of the null cells contained virus antigen.
Within the infected T cell population BoT8 + cells, the suppressor/cytotoxic cell phenotype,
accounted for approximately 60 ~ and the BoT4 ÷ cells, the helper cell population, for 40 ~ of the
infected cells.

Subpopulations of PBL replicating B VD V

A variety of experiments were performed to estimate the virus production by the various
subpopulations of mononuclear cells in peripheral blood. As infected cells occurred amongst all
the major subgroups and each of them contributed different numbers of positive cells and yields,
it appeared exceedingly difficult to obtain a clear picture of this aspect. However, as depicted in
Table 5, where the different physical, chemical and immunological cell separation procedures
 1976 H. BIELEFELDT OHMANN, L. RONSHOLT AND B. BLOCH
Downloaded from www.microbiologyresearch.org by

On: Tue, 06 Sep 2016 21:46:32

-¢~ A

Fig. 1. Double immunolabelling of PBL from a persistently viraemic calf to demonstrate virus infection
IP: 120.188.92.144

in phenotypically defined cell subsets. The cytospin preparation was stained for the BoT8 and BVDV
antigens using a combination of the APAAP and HRP/DAB systems. The BoT8 ÷ cells and BVDV
antigen-positive cells cannot be clearly distinguished in this black and white photomicrograph.
However, cells positive for both antigens stand out distinctly as very dark stained cells (arrows). Bar
marker represents 10 ~tm.

significantly reduced the various target subgroups of mononuclear cells, it appeared that
monocytes (expt. I no. K 10 and 14, and expt. II no. 3) and T cells (expt. V, no. 6 and 13 and expt.
III, no. 12) were able to produce infective virus. In contrast, B cells (expt. V, no. 6 and 13)
contributed very little or not at all to the virus propagation. The latter finding supported the
productive role of monocytes inferred from experiments III (calf no. 12) and IV (calves no. 144
and 6).

B VD V infection of glass-adherent cells from persistently infected cah,es

The number of blood monocytes had significantly increased in persistently infected calves
(Table 4) and the cells appeared 'activated', i.e. were larger, more vacuolated with more and
longer pseudopods than in monocytes from non-infected animals (not shown). The monocytes
also constituted a large proportion of the virus-infected cells, and because of the inherent nature
of monocytes to migrate to all kinds of tissues and differentiate into mature macrophages, a few
studies were conducted to characterize this population further. PBL suspensions from infected
and non-infected calves were seeded in a fivefold dilution row into duplicate wells of 24 well
tissue culture plates with cleaned glass coverslips and incubated for 14 h. Non-adherent cells
were then rinsed off thoroughly and the coverslips transferred to new trays or fixed for
immunocytochemical studies. In the former case, MDBK cells were seeded onto the
macrophage cultures and the cocultures incubated for 4 days at which time virological assays
were conducted. This series of experiments revealed that whereas in non-infected animals only
one cell out of 2000 NSE + cells in a PBL preparation adhered and developed in vitro into a
mature macrophage (i.e. 0.05 ,%), one cell per 35 to 140 NSE + cells would do so in the persistently
 BVDV in blood of persistently infected cattle 1977
Table 4. Frequencies of mononuclear cell subpopulations in peripheral blood of non-infected and
persistently B VD V-infected calves
~o positive among PBL
Calves persistently
Subpopulation of PBL Non-infected infected with BVDV
B lymphocytes 19.8 (15.0 30.3)]" 23.0 (14.7-40.0)
(BIg73+/PIg45+)* (n = 17)~ (n = 45)
T lymphocytes 59.8 (37.5-77.4) 46.8 (31.7-63.7)
(CHI28+/CH61a ÷) (n = 14) (n = 14)
BoT4 cells 31.3 (29.0-32.3) 31.7 (27.0-39.7)
(IL-A12) (n = 5) (n = 6)
BoT8 cells 23.2 (17.7-27.5) 19.5 (12.0-29.5)
(IL-A17) (n = 9) (n = 17)
Monocytes 11.6 (9-0-17.5) 1@5 (8.2-28-5)
(DH59A) (n = 6) (n = 27)
Monocyte subpopulation 6.9 (0.0-14.0) 8.5 (3.0-12.5)
(CH137A) (n = 5) (n = 7)
Monocyte subpopulation 10.8 (8.5-14.0) 8.8 (1.5--13.5)
(CH16A) (n = 5) (n = 7)
Monocyte subpopulation 7.5 (5.0-10.0) 17.5 (10.5 20.5)
(IL-A23) (n = 3) (n = 6)
Downloaded from www.microbiologyresearch.org by

Monocytes (NSE +) 11.2 (5.0-18.5) 17.8 (11.5 25.1)

(n = 18) (n = 28)
MHC-II (DR equivalent) 32.9 (26-0-41.0) 32.3 (20.9-47.0)
(THI4B) (n = 13) (n = 33)
On: Tue, 06 Sep 2016 21:46:32

MHC-II (DQ equivalent) 28.8 (17.0-37.1) 25.7 (16.5 34.0)

(TH81A) (n = 13) (n = 23)
MHC-II (DP equivalent) 31.5 (20.040,5) 27.1 (20.2--35.1)
IP: 120.188.92.144

(H42A) (n = 11) (n = 17)

* MAbs used for the detection of the phenotypic cell marker (see Table 1).
t Range of frequencies.
n, Number of separate determinations (different animals and different days).

infected calves (i.e. 0.7 to 2.8 ~). However, no significant differences were found between the
adhering cells in "the two animal groups with respect to expression of M H C - I I antigens or the
macrophage differentiation antigens defined by DH59B and IL-A23 (data not shown). Another
notable finding in this series of experiments was that whereas the proportion of virus antigen-
positive cells among the adherent NSE ÷ population in most cases corresponded quite well to the
estimates previously made and correlated well with the total frequency of virus antigen-positive
cells in the PBL-isolates (Table 6), only a small proportion of the cells, i.e. from 5.0 to 12.5 ~ of
the adherent cells were productively infected (Table 6) as detected by cocultivation with M D B K
cells. If the monocytes and macrophages were left to mature in vitro for 5 to 6 days before
cocultivation, up to I00 To of the cells could appear to be productively infected (data not shown).

Electron microscopy
Examination of ultrathin sections of cell preparations by electron microscopy, either non-
fractionated PBL or enriched/depleted populations, including monocyte cultures, revealed
BVDV-like particles with a size of 45 to 55 nm. The particles were found in small cytoplasmic
vesicles in both typical lymphocytes and in monocytes (Fig. 2). In no instance were virus
particles found attached to the cell surface, thus supporting the control experiments for non-
specific virus 'carry-over' described above. By labelling subpopulations of cells by the
immunogold technique and employing the MAbs listed in Table 1, cells containing viral
particles could be phenotypically identified as T and B cells as well as monocytes. This
corroborated both the results of the immunocytochemical studies and the findings presented in
Table 5.
 1978 H. BIELEFELDT OHMANN, L, RONSHOLT AND B. BLOCH
Table 5. Virus antigen and infectious virus in enriched populations of mononuclear cells from
peripheral blood*
Expt. & Frequency (~)
animal r x ~ ~ BVDV-antigen- TCIDs0/106
number Treatment T cellst B ceUs:~ M§ positive cells cellsll
I 2 None 66.0 17-3 11.5 0 0
L-LME 72.0 19.5 0-5 0 0
K10 None 55-5 16-0 17.5 15-0 160
L-LME 70.3 10.5 2.5 17.0 5
14 None 55-0 26-3 22.0 25.0 128
L-LME 57-7 24.5 7-5 29.0 20
II 1 None 46-5 27-0 27-0 19-9 1024
Defibrinated blood 62-5 24.7 <0-7 27.3 ND¶
3 None 48-3 32.4 14-3 27.0 512
Defibrinated blood 72-7 15.0 2-5 27-5 256
4 None 49.3 27-0 16.8 22.6 256
Defibrinated blood 69.0 17.7 1.7 25.7 128
III 7 None 45.7 27.0 20-5 30.4 512
Sephadex NA 71-3 11.7 <0.01 32.9 256
AD ceils 6.7 1.3 80-0 18.1 256
12 None 48-3 22.0 13.3 22-0 128
Sephadex NA 57.7 16.3 <0.01 30.3 512
Downloaded from www.microbiologyresearch.org by

AD cells 7.0 2.3 88.3 24.3 256

14 None 44-7 18.5 26.3 28.8 256
Sephadex NA 72.4 13.0 1-0 29.8 256
AD cells 2.7 1.3 94.3 35-0 ND
On: Tue, 06 Sep 2016 21:46:32

IV 144 None 31-9 30.5 11.7 24.7 171

C' control 40.7 29.3 9.3 24.3 160
IP: 120.188.92.144

THI4B+C" ( < l X TH14 +) 60-7 <0.3 <0-3 29.0 40

6 None 43.3 20,3 21.7 15.0 64
C' control 48-0 22,5 21.3 19.7 64
TH14B+C' 93.0 <0.3 2-3 20.1 32
8 None 56.0 20-7 13.0 0 0
C' control 58.0 20.7 13.0 0 0
TH14B+C' 83.0 <0-2 <0.7 0 0
V 5 None 52-0 15.1 5.0 0 0
C' control 60.0 13-5 4-0 0 0
Pig45 + C' 86-6 < 0.3 4.3 0 0
B26A4 + C' 1.3 28-0 ND 0 0
6 None 43-3 28-3 17-3 23.3 128
C' control 51.7 18.3 14-2 22.3 40
Pig45 + C' 67-0 3"0 13-7 19-0 80
B26A4+C' 1.7 38-3 ND 10.0 40
13 None 58.3 26.3 16.3 9.7 128
C' control 65-5 15.0 12-0 8.3 80
PIg45+C" 73.0 0.3 11.3 11.7 160
B26A4 + C' 5-0 34-0 12.0 5'0 20
* Representative examples from a series of more than 35 separate experiments.
t Cells positive for the CH128-defined antigen.
Cells positive for the BIg73-defined antigen.
§ NSE + cells (M, monocytes/macrophages).
IrInfectious virus isolated directly from cell lysates.
¶ NO, Not determined.

DISCUSSION
In this study it was found that in vivo, BVDV infects and replicates in subpopulations of all the
four major cell groups in the peripheral blood, i.e. T and B lymphocytes, monocytes and null
cells. Of these, the B cell population seems to make only a minor contribution. This finding is
interesting in the context of the apparent tolerance to the persisting virus, as defined by the
complete absence of neutralizing antibody production, which characterizes these animals.
Several explanations for this apparent paradox might be considered. Thus, the virus-specific
 B V D V in blood of persistently infected cattle 1979
Downloaded from www.microbiologyresearch.org by

On: Tue, 06 Sep 2016 21:46:32

Fig. 2. Electron micrograph of BVDV-like panicles in a small cytoplasmic vesicle in a lymphocyte from
the peripheral blood of a persistently infected calf. The cells were fixed in 2-5~ glutaraldehyde only. Bar
IP: 120.188.92.144

marker represents 100 nm.

Table 6. B VD V infection in glass-adherent cells from persistently infected calves*

Glass-adherent cells BVDV-antigen- Virus-producing BVDV ÷ cells
Calf as ~ of NSE + cells positive cells (~)t cellst (~) in PBL (~)
6 0.7 17.9 6.2 15.7
K10 1.4 14.4 8.5 16.5
T10 2.8 21.6 6.2 13.8
13 0.8 8.8 5-0 9.5
14 1.4 24.1 12-5 27.3
* The results are averages from four or five independent experiments in each calf.
t The proportion of those NSE ÷ cells which had adhered after a 14 h incubation of PBL.

lack of response could be due to an indirect effect on B cell functions caused by a defect in virus
antigen presentation by macrophages or in the production of some essential factor(s) by
macrophages and/or T lymphocytes. Alternatively, the infection of T cells and monocytes could
induce virus-specific suppressor cells amongst either or both cell types. Indeed, Larsson (1986)
did report the presence of a Fc receptor-bearing suppressor cell in PBL from persistently
infected calves. However, the possibility o f a direct effect o f the B V D V on the B cells cannot be
ruled out either. Thus, the very few virus-infected B ceils encountered m a y represent vestiges o f
a much more extensive B cell infection, perhaps at the bone marrow level, causing clonal
deletion of the virus-specific (precursor) cells. Future studies should be a i m e d at clarifying these
aspects.
The variability of the levels of infection between individual calves was notable (Table 2). The
infection level could not be related to any aspect of the health status of the animals, as all except
two remained clinically normal throughout the investigation period stretching over more than 8
months. In those two calves, which succumbed to MD, no increase in infection level o f PBL was
 1980 H. BIELEFELDT OHMANN, L. RONSHOLT AND B. BLOCH
observed in or around the time of disease development and death (unpublished data). It is
therefore by no means obvious what kind of factor(s) might govern the persistence and
replication of BVDV in the PBL. Stimulation of the PBL with mitogens or interleukin 2 failed to
enhance virus rescue by cocultivation (H. Bielefeldt Ohmann, unpublished data), perhaps
suggesting that it is 'fixed' differentiation subsets of cells that replicate the virus. Thus, cell
division may provide no or only a minor contribution to virus persistence, but rather the
infection is maintained by continuous infection of newly recruited cells within each group of cell
types. It will therefore clearly be of interest to explore what is happening in the primary
lymphoid tissues, i.e. thymus, bone marrow and perhaps the gut-associated lymphoid tissues, in
the persistently infected, clinically normal calves. Furthermore, among the candidate factors
with potential for influencing establishment and maintenance of virus persistence are the
interferons (Jacobson & McFarland, 1982; Matthews & Vorndam, 1982; von Rheinbaben et al.,
1985). Their role in persistent BVDV infection requires some attention. The demonstrated
activation of the monocyte population of infected animals, i.e. increased numbers, enhanced
membrane ruffling, vacuolation and increased propensity to adherence, also among cells not
containing detectable viral antigen, could potentially be due to interferon exposure (Bielefeldt
Ohmann et al., 1984, 1986; Bielefeldt Ohmann & Babiuk, 1986a). However, the possibility of
the recruitment of a completely new subset of the macrophage lineage during the course of
infection should not be ruled out at present (Narayan et al., 1984; Gendelman et al., 1985).
Downloaded from www.microbiologyresearch.org by

Although based on a relatively small number of samples it was striking to find, by electron
microscopy, virus-like particles, similar to those previously described in the tissues of calves
suffering from MD (Bietefeldt Ohmann & Btoch, 1982) in a much larger proportion of cells than
that expected from the direct isolations from the corresponding cell lysates. This could of course
On: Tue, 06 Sep 2016 21:46:32

be due to low recovery efficiencies, either because of suboptimal sensitivity of the MDBK cells
used in our isolation assay or due to the presence of intracellular inhibitors liberated
IP: 120.188.92.144

simultaneously with the intracellular virus particles. Alternatively, it might indicate that only an
exceedingly small proportion of the virions produced are actually infectious, a well known
phenomenon in virology (Richman et al., 1984). This could also explain the low virus-titres
usually achieved with BVDV both in vivo and in vitro despite widespread virus infection (up to
100K in vitro) as detected by immunocytochemistry (Horzinek, 1981 ; Bielefeldt Ohmann, 1981).
Finally, the infection of the null cell population should be briefly commented on. This cell
type, which may contribute up to 20~ of PBL in normal cattle (W. C. Davis, personal
communication; Morrison, 1986; H. Bielefeldt Ohmann, unpublished observations and the
present work, see Table 4) appeared to contain up to 4 0 ~ of all the virus antigen-positive cells in
a persistently infected animal, with approximately 50~ of the null cell population being
infected. At least two, not mutually exclusive, possibilities exist. The null cell population may
include precursor cells of the other three major cell groups not yet expressing the phenotypic
markers used for their detection. Alternatively, the virus infection may cause a suppression of
normal cell markers (Jennings et al., 1985), which then escape proper classification. Whatever
the explanations are, the null cell population deserves more attention considering its major
contribution to the virus persistence in PBL of tolerant cattle.
In conclusion, the establishment of the identities of the infected cells in persistently BVDV-
infected calves will now allow a further elucidation of the mechanisms governing establishment
and maintenance of virus persistence, including the possible effects on the specialized functions
of the immune cells, whether of a direct nature due to virus infection per se or indirectly by
upsetting normal functional circuits, as the shifts in relative frequencies (Table 4) may indicate
(Woodruff & Woodruff, 1975).

The authors thank Dr W. C. Davis (WashingtonState University, Pullman, Wash., U.S.A.) and Dr A. J. Teale
(ILRAD, Nairobi, Kenya) for the generousgift of MAbs. The provisionof BVDV-freeMDBK cellsand the access
to animals made possible by Dr A. Meyling (The National Veterinary Laboratory, Copenhagen, Denmark) are
greatly appreciated. Thanks are also due to Helle Kurstein and Irene Kosokowskyfor typing the manuscript.
Financial support was receivedby H. BielefeldtOhmann from the Danish Agricultural and VeterinaryResearch
Council (grant no. SJVF 13-3681).
 B VD V in blood o f persistently infected cattle 1981
REFERENCES
BALDWIN, C. L., TEALE, A. J., NAESSENS, J. G., GODDEERIS, B. M., M^cHUGH, N. D. & MORRISON, W. I. (1986).
Characterization of a subset of bovine T lymphocytes that express BoT4 by monoclonal antibodies and
function: similarity to lymphocytes defined by human T4 and murine L3T4. Journal of Immunology 136, 4385-
4391.
BIELEFELDT OHMANN, H. (1981). Bovine viral diarrhoea virus infections. II. Pathogenic studies of spontaneous and
experimental cases. Pathomorphological, immunocytochemical and serological investigations. Ph.D. Thesis, The
Royal Veterinary and Agricultural University, Copenhagen.
BIELEFELDT OHMANN, n. (1983). Pathogenesis of bovine viral diarrhoea-mucosal disease: distribution and
significance of BVDV antigen in diseased calves. Research in Veterinary Science 34, 5-10.
mELEFELDr Oh'MANN,H. (1987). Double-immuno-labelling systems for phenotyping of immune cells harbouring
bovine viral diarrhoea virus. Journal of Histochemistry and Cytochemistry (in press).
BIELEFELDTOHMANN,H. & BABIUK,L. A. (1986a). Alteration of alveolar macrophage functions by in vivo treatment
with recombinant interferons alpha-1 and -gamma. Proceedings of the 1st International VeterinaryImmunology
Symposium, Guelph, Ontario, Canada, July 1-4, p. 71.
BIELEFELDTOHMANN,H. & BABIUK,L. A. (1986b). Viral infections in domestic animals as models for studies of viral
immunology and pathogenesis. Journal of General Virology 67, 1-25.
BIELEFELDTOHMANN,H. & BLOCH, B. (1982). Electron microscopic studies of bovine viral diarrhoea virus in tissues
of diseased calves and in cell cultures. Archives of Virology 71, 57-74.
BIELEFELDT OHMANN,H., FILION, L. G. & BABIUK,L. A. (1983). Bovine monocytes and macrophages: an accessory
role in suppressor-cell generation by ConA and in lectin-induced proliferation. Immunology 50, 189 197.
mELEFELDT OHMANN, H., GILCHRIST, J. E. & BAmUK, L. A. (1984). Effect of recombinant DNA-produced bovine
interferon alpha (BolFN-alpha) on the interaction between bovine alveolar macrophages and bovine
Downloaded from www.microbiologyresearch.org by

herpesvirus type 1. Journal of General Virology 65, 1487-1495.

BIELEFELDTOHMANN,H., DAVIS,W. C. & BABIUK,L. A. (1985). Functional and phenotypic characteristics of bovine
natural cytotoxic cells. Immunobiology 169, 503-519.
BIELEFELDT OHMANN, H., DAVIS, W. C. & BABIUK, L. A. (1986). Surface antigen expression by bovine alveolar
On: Tue, 06 Sep 2016 21:46:32

macrophages: functional correlation and influence of interferons in vivo and in vitro. Immunobiology 171, 125-
142.
IP: 120.188.92.144

BOLIN, S. R., McCLURKIN, A. W., CUTLIP, R. C. & CORIA, M. F. (1985). Severe clinical disease induced in cattle
persistently infected with noncytopathic bovine viral diarrhoea virus. American Journalof Veterinary,Research
46, 573-576.
BROWNLIE, J., CLARKE,M. C. & HOWARD, C. J. (1984). Experimental production of fatal mucosal disease in cattle.
Veterinary. Record 114, 535 536.
BROWNLIE, J., CLARKE,M. C., HOWARD, C. J. & POCOCK, D. H. (1986). Mucosal disease - a combination of precise
immunotolerance and viral superinfection. Proceedings of the 1st International Veterinary Immunology
Symposium, Guelph, Ontario, Canada, July 1-4, p. 105.
DAVIS,W. C., MARUSIC,S., LEWlN, H. A., SPLITTER, G. A., PERRYMAN,L. E., McGUIRE,T. C. & GORHAM,J. R. (1987). The
development and analysis of species specific and cross reactive monoclonal antibodies to leukocyte
differentiation antigens and antigens of the major histocompatibility complex for use in the study of the
immune system in cattle and other species. Veterinary"Immunology and lmmunopathology (in press).
ELLIS, J. A., M~,cHUGH,N. D., DAVIS,W. C. & MORRISON,W. I. (1986). Bovine mononuclear phagocyte differentiation
antigens identified by monoclonal antibodies. Proceedings of the 1st International Veterinary Immunology
Symposium, Guelph, Ontario, Canada, July 1-4, p. 57.
GARDINER, A. C., NETTLETON, P. F. & BARLOW, R. M. (1983). Virology and immunology of a spontaneous and
experimental mucosal disease-like syndrome in sheep recovered from clinical border disease. Journal of
Comparative Pathology 93, 463-469.
GENDELMAN,H. E., NARAYAN, O., MOLINEAUX,S., CLEMENTS,J. E. & GHOTBI, Z. (1985). Slow, persistent replication of
lentiviruses: role of tissue macrophages and macrophage precursors in bone marrow. Proceedings of the
National Academy of Sciences, U.S.A. 82, 7086-7090.
HORZINEK, M. C. (1981). Non-arthropod-borne Togaviruses. London: Academic Press.
JACOBSON,S. & McFARLAND,H. F. (1982). Measles virus persistence in human lymphocytes: a role for virus-induced
interferon. Journal of General Virology 63, 351-357.
JENNINGS, S. R., RICE, P. L., KLOSZEWSKI,E. D., ANDERSON,R. W., THOMPSON,D. L. & TEVETHIA,S. S. (1985). Effect of
Herpes Simplex virus types 1 and 2 on surface expression of class I major histocompatibility complex antigens
on infected cells. Journal of Virology 56, 757-766.
KOSKI, I. R., POPLACK,D. G. & BLAESE,R. M. (1976). A non-specific esterase stain for the identification of monocytes
and macrophages. In In Vitro Methods in Cell-Mediated and Tumor Immunology, p. 359. Edited by B. R.
Bloom & J. R. David. New York: Academic Press.
LARSSON,B. (1986). Increased suppressor activity of F gamma + ceils in cattle persistently infected with bovine viral
diarrhea virus (BVDV). Proceedings of the 1st International Veterinary Immunology Symposium, Guelph,
Ontario, Canada, July 1-4, p. 111.
LIESS, B., FREY, H.-R., ORBAN, S. & HAFEZ, S. M. (1983). Bovine virus-diarrhoea (BVD)-"Mucosal Disease":
persistente BVD-Feld-virus Infektionen bei serologisch selektierten Rindern. Deutsche tieri~rztliche
Wochenschrift 90, 261-266.
 1982 n . B I E L E F E L D T O H M A N N , L. R O N S H O L T A N D B. B L O C H

LITTLEJOHNS, I. R. & WALKER, K. H. (1985). Aetiology and pathogenesis of mu~osal disease of cattle: current
concepts, observations and speculation. Australian Veterinary Journal 62, 101-103.
McCLURKIN, A. W., LITTLEDIKE,E. T., CUTLIP, R. C., FRANK, G. H., CORIA, M. F. & BOLIN, S. R. (1984). Production of
cattle immunotolerant to bovine viral diarrhea virus. Canadian Journal of Comparative Medicine 48, 156-161.
MATTHEWS,J. H. & VORNDAM,A. V. (1982). Interferon-mediated persistent infection of Saint Louis encephalitis virus
in a reptilian cell line. Journal of General Virology 61, 177-186.
MORRISON, W. t. (1986). Monoclonal antibodies against bovine macrophage-populations. Proceedings of the 1st
International Veterinary Immunology Symposium, Guelph, Ontario, Canada, July 1-4, p. 3.
MORRISON, W. I., GODDEERIS, B. M., TEALE, A. J., BALDWIN, C. L., BENSAID, A. & ELLIS, L (1986). Cell-mediated
i m m u n e responses of cattle to Theileria parva. Immunology Today 7, 211-216.
NARAYAN,O., SHEFFER, D., GRIFFIN, D. E., CLEMENTS,J. & HESS, J. (1984). Lack of neutralizing antibodies to caprine
arthritis-encephalitis lentivirus in persistently infected goats can be overcome by immunization with
inactivated Mycobacterium tuberculosis. Journal of Virology 49, 349-355.
RICHMAN, D. D., CLEVELAND, P. H., REDFIELD, D. C., OXMAN, M. N. & WAHL, G. M. (1984). Rapid viral diagnosis.
Journal of Infectious Diseases 149, 298-310.
ROEDER, P. L. & DREW, T. W. (1984). Macosal disease of cattle : a late sequel to fetal infection. VeterinaryRecord 114,
309-313.
ST6BER, M. (1984). Neuere Erkenntnisse fiber das BVD-syndrom des Rindes: Erreger, Immunit/itsgeschenen,
Verlauf and Verbreitung, Bekampfung. Praktischer Tierarzt iv, 88-98.
TERFSTRA, C. (1981). Border disease: virus persistence, antibody response and transmission studies. Research in
Veterinary Science 30, 185-191.
VON RHEINBABEN,F., STITZ, L. & RO'I'r, R. (1985). Influence of interferon on persistent infection caused by Borna
disease virus in vitro. Journal of General Virology 66, 2777-2780.
Downloaded from www.microbiologyresearch.org by

WESTAWAY,E. G., BRINTON, M. A., GAIDAMOVICH,S. YA., HORZINEK, M. C., IGARASHI,A., K.$~,.RI~.INEN,L., LVOV, D. K.,
PORTERFIELD, J. S., RUSSELL, P. K. & TRENT, D. W. (1985). Togaviridae. Intervirology 24, 125-139.
WESTaURY, H. A., NAP'rHINE, D. V. & STRAUBE, E. (1979). Border disease: persistent infection with the virus.
Veterinary Record 104, 406-409.
On: Tue, 06 Sep 2016 21:46:32

WOODRUFF, J. F. & WOODRUFF, J. J. (1975). The effect of viral infections on the function of the i m m u n e system. In
Viral Immunology and Immunopathology, pp. 393-418. Edited by A. L. Notkins. London: Academic Press.
IP: 120.188.92.144

(Received 18 November 1986)