C. Dearden I, W. AI-Nakib 2, K. Andries 3, R. Woestenborghs 3, and D. A. J. Tyrrell 1
1MRC Common Cold Unit, Salisbury, England 2 Department of Microbiology, Faculty of Medicine, University of Kuwait, Safat, Kuwait
3 Department of Virology, Janssen Research Foundation, Beerse, Belgium
Accepted August 22, 1989
Summary. Viruses were isolated from nasal washings of volunteers receiving experimental therapy for rhinovirus type 9 infection with intranasal sprays of a new synthetic antiviral R61837. On a screening test nine subjects yielded drug sensitive virus and four resistant virus. In four others the virus was sensitive at first but became resistant later, while in one the reverse occurred. Evidence is given that at least some of the resistant viruses were present in the respiratory tract and were not selected during virus isolation. Of six viruses studied in detail, five had a low degree of resistance and one was highly resistant. The degree of resistance of the five was similar for an anfiviral chalcone, dichloroflavan and disoxaril. The sixth was different in that the resistance to disoxaril was relatively less than to the other drugs.. The significance of these results is discussed--these are the first experiments in man to show the selection of drug resistant rhinovirus.
Introduction
In recent years we have studied several antivirals which exert their effect by direct interaction with the capsid proteins of rhinoviruses, namely a dichlo- roflavan (DCF), a chalcone, and R61837. The first two were tested in volunteers as oral preparations and nasal sprays, but had no beneficial effect even though they were given repeatedly from a day before the virus drugs were given [3, 4, 9, 10]. However R61837 was found to suppress colds induced by RV9 I-2]. In the first experiment the drug was given for four days but appeared to suppress symptoms only while it was given; the experiment was therefore repeated and drug was given for six days, until just before the volunteers left the Unit, and symptoms and signs were prevented almost completely during this administra- tion. This paper reports studies of materials collected during the first experiment.
It has been known for some time that it is relatively easy to recover viruses resistant to DCF and chalcone by treating virus pools in vitro or propagating
72 C. Dearden et al.
virus in the presence o f drug [1, 8, 131. We therefore wondered whether viruses resistant to R61837 might be found also, and more important ly , whether they might emerge during t rea tment o f volunteers. We chose to examine this by testing specimens f rom volunteers who took par t in the first trial o f the drug; these were of par t icular interest in that they were collected not only during t rea tment but also for three days after nasal sprays o f the drug had been stopped.
Viruses were cul tured and tested in order to de termine whether dur ing this experimental t r ea tment they had become resistant against R61837 and related drugs.
M a t e r i a l s a n d m e t h o d s
Volunteers
The experiments were conducted with the prior approval of the Harrow District Ethical Committee. The general methods of isolation and clinical evaluation have been described elsewhere [6] and the trials have also been reported [2]. In brief, after a quarantine period, volunteers received regular intranasal sprays of drug or placebo. More details are given in Table 1.
Tissue cultures
A sensitive strain of cells (Ohio HeLa) was propagated and subcultured weekly. Tubes were seeded at 2 x 105 cells/ml in BME containing 10% newborn calf serum and incubated at 37°C at an angle of 15 °. Once the cells were 90% confluent the growth medium was removed and replaced with 1 ml maintenance medium (BME, 4.4% sodium bicarbonate, 2% FCS, 0.13% tryptose phosphate broth, 30 mM magnesium chloride and antibiotics) per tube. For some experiments this medium contained Janssen R61837 at 0.1 gg/ml. Once inoculated (0.2ml specimen per tube) the tubes were rolled at 33 °C for eight days. The medium was changed on days 1 and 6 after infection and the cells observed for rhinovirus specific cytopathic effect on days 6 and 8 after infection. Positive cultures were frozen at - 70 °C and thawed three times and the tissue culture fluid harvested and stored at - 70 °C for further investigation.
Virus" isolation
Nasal washings were collected in saline, mixed with an equal volume of broth, divided into aliquots and stored at - 7 0 °C until studied.
Samples were inoculated into roller tube cultures of Ohio HeLa [12] which were rolled at 33 °C and examined by low power microscopy. The presence of viruses was recognized by the appearance of a typical cytopathic effect.
Drug preparation
R61837 (3-methoxy-6-[4-(3-methylphenyl)-l-piperazinyl]pyridazine) was obtained from Janssen Research Foundation as a solution of 2.5mg/ml in 10% hydroxy-propyl 13 eyclodextrin and was then further diluted in maintenance medium. DCF (4-6 dichloro- flavan) from Wellcome Research, the chalcone Ro 09-01410 (4-ethoxy-2-hydroxy-4,6'- dimethoxychalcone) from Nippon Roche and disoxaril from Sterling-Winthrop were dis- solved in dimethyl sulphoxide at 1 mg per ml and diluted as required in culture medium.
Tests of minimum inhibitory concentration (MIC) Pools of virus were prepared and titrated for infectivity in roller tube cultures. Serial dilutions of drug were prepared in maintenance medium. The medium was added to the pairs of
Drug resistant rhinoviruses from the nose of experimentally treated volunteers 73
tubes and then all were inoculated with 50 TCDs0 of virus in 0.2 mt of medium. Cultures were rolled at 33 °C and read when the tubes without drug showed extensive CPE, usually six to eight days after infection. This method was chosen because the viruses used did not produce a regular cytopathic effect in microtitre plates.
Results
The patterns of virus excretion
We first reviewed the pattern of results seen when viruses were isolated from treated volunteers. In attempting to isolate viruses we removed the culture medium (with drug derived from the inoculum) 1 1/2 hours after inoculation. Each attempt was performed in duplicate in two tubes containing plain MM and in another two tubes to which drug was added at a final concentration of 0.1 gg/ml. This concentration was chosen because it was found to completely inhibit virus replication in tubes inoculated with washings from volunteers who had been given placebo or with virus used as inoculum or with a laboratory- passaged strain of RV9. As a first approximation we could say that the growth of virus in both pairs of tubes indicated the presence of a drug resistant virus-- in the first pair only a drug sensitive virus.
Table 1 outlines the plan of the trial and shows when drug and virus were administered and nasal washes were collected. All volunteers given placebo produced drug sensitive virus only, while those given drug often yielded drug resistant virus. Tables 2 summarises the results obtained. Virus was recovered from a rather higher proportion of volunteers given placebo (83%) than those given drug (68%). Six individuals given drug developed colds. However, those given drug continued to shed virus after the drug was withdrawn (Table 2) and indeed cleared virus later than those given placebo. Several patterns were ob- served (Table 3). In nine cases (volunteers 1-9) all isolates were drug sensitive and in four (14-17) they were all drug resistant. In four others (10-13) the drug sensitive virus found at first was followed by drug resistant virus--after treat- ment ceased but in one case (18) the reverse occurred; there was one instance (19) of an apparently drug dependent virus.
Thus the general picture was that drug resistant virus often appeared in the treated group. However, this conclusion needed to be examined further. It could be that the viruses became resistant and were selected during the virus isolation process, while being propagated in the test tube. However, no resistant viruses were recovered in the initial screen when washings were inoculated into cultures containing drug. We nevertheless repeated some isolations trying to eliminate the drug. Specimens were extracted with chloroform but virus isolation attempts were unsuccessful--presumably traces of solvent impaired the sensitivity of the cells to small amounts of virus. We therefore inoculated tubes with washings, rolled them for 60 min, washed them with PBS and then continued incubation; we thus removed from the culture most of the drug which might influence the first and subsequent progeny virus. Furthermore we examined washings taken at the end of the trial, when no drug had been given for over three days and
'Fab
le 1
. D
ay t
o d
ay p
lan
of
the
tria
l ,,-
..1
Day
of
tria
l
Qua
rant
ine
peri
od
0 1
2 3
4 5
6 7
8
Ent
er i
sola
tion
0.
4 m
l dr
ug
0.4
ml
drug
0.
4 m
l dr
ug
0,4
ml
dru
g
0.4
ml
dru
g
or
plac
ebo
or p
lace
bo
or
plac
ebo
or
plac
ebo
or
plac
ebo
Blo
od f
or
at
at
at
at
at
bioc
hem
istr
y 8
a.m
, 8
a.m
. 8
a.m
, 8
a.m
. 8
a.m
. an
d ha
emat
olog
y 11
a.m
. 11
a.m
. 11
a.m
. 11
a.m
. 2
p.m
. 2
p.m
. 2
p,m
. 2
p.m
. B
lood
for
5
p,m
. 5
p.m
. 5
p.m
. 5
p.m
. bi
oche
mis
try
8 p.
m.
8 p,
m.
8 p,
m,
8 p.
m,
and
haem
a-
10.3
0 p.
m,
10.3
0 p.
m.
10.3
0 p,
m.
10.3
0 p.
m.
tolo
gy
8.10
a.m
, N
asal
was
h V
irus
N
asal
N
asal
N
asal
N
asal
ch
alle
nge
was
h w
ash
was
h w
ash
12 n
oo
n
En
d
isol
atio
n
Nas
al
was
h
Dru
g w
as a
dmin
iste
red
at 2
.5 m
g/m
l in
cyc
lode
xtri
n
Tab
le 2
. F
req
uen
cy o
f vi
rus
shed
ding
Tre
atm
ent
gro
up
V
irus
iso
late
d fr
om
N
o. i
sola
ted
on
day
N
o.
of
cold
s
5 6
7 8
9
Pla
cebo
15
/18
(83%
) 3
8 10
4
2 4
(22%
) D
rug
19
/28
(68%
) 3
6 10
12
12
6
(21%
)
t"--
O
¢-q
O
0
o
0 (D
(3
Drug resistant rhinoviruses from the nose of experimental ly treated volunteers 75
when chemical analysis showed tha t drug concentra t ions in the washings were
very low (Table 4). Virus was still recovered tha t grew in the presence o f drug.
However, growing up virus in the absence o f drug might allow it to regain
sensitivity. Indeed was passed five resistant viruses serially three times in tissue
culture wi thout drug and observed small increases in sensitivity; tha t of viruses
10, 12, 14, and 18 increased two-fold and of 11 increased three-fold. F r o m six volunteers virus was isolated (on days 8 or 9) in the absence o f
drug and also in the presence o f 0.1 gg/ml to prevent reversion. The MICs of
all isolates were measured and are summarized in Fig. 1. These results should
give us upper and lower limits for the drug sensitivity o f virus recovered f rom
the nose. In some cases both results are the same and there is a clear cut difference f rom those found with viruses obta ined f rom volunteers given pla-
cebo. The concentra t ions o f drug in nasal washings are recorded in Table 4. This
shows tha t the concentra t ions recovered were high but very variable during drug adminis t ra t ions and declined steadily between the sixth and ninth day
when virus was mos t often recovered f rom treated volunteers. In four instances (10, 11, 14, and 17) resistant viruses were cult ivated f rom specimens which conta ined no detectable drug. One of the two volunteers (17) f rom w h o m
Drug resistant rhinoviruses from the nose of experimentally treated volunteers 77
v
~o
,,=, G0
Z g o_
>12-8-
2.5-
2-0"
1"5"
1.0-
0-5' [i, [
I I l I
lb 1"~ 1'2 1'4 'i'7 lh I VOLUNTEER TREATED WITH .21 VOLUNTEERS
ACT,VE R6 ,83 , , T R E A ~ D W , ~ PLACEBO
Fig. 1. Degree of resistance of viruses isolated in the presence and absence of drug from volunteers given drug or placebo. * Virus grows to toxic drug levels. MIC of virus isolated
• with or • without R61837 in media
resistant virus was cultured when full concentrations of drug were present turned out to have a highly resistant virus when cultured after drug had been cleared (see below and Fig. 1). On the other hand, another such volunteer (18) later shed drug sensitive virus; this was possibly connected with the lower concen- trations of drug during treatment or with selection of drug resistant virus by exposure to the drug in the earlier nasal washings. By contrast sensitive virus was recovered from two volunteers (12 and 13) on days 5 and 6 when drug concentrations were high.
We therefore concluded that drug resistant virus was present in the nose and apparently propagating there. Since some of the volunteers with colds shed sensitive as well as resistant virus the resistant virus may not have caused their disease. But since only resistant virus was recovered from volunteer 17 it is likely that it caused the cold which was observed.
Resistance to other drugs
It is of interest to know whether the type of resistance is similar to that evoked by exposure to other drugs in vitro. It is known that RV2 selected for resistance to chalcone may be resistant to DCF [8] so we looked for cross resistance in the viruses obtained from the volunteers. The MIC measurements were done to reduce experimental variation as much as possible. In each experiment the
78 C. Dearden et al.
Table 5. Ratio of MIC of four drugs for viruses isolated from 6 volunteers treated with active Janssen R61837 to mean MIC of viruses isolated from 7 volunteers treated with
Mean MIC of viruses ~< 0.1 0.032 0.049 3.04 from placebo volunteers (gg/ml- 1)
MIC inoculum virus 0.1 0.032 0.049 3.04 (gg/ml)
* Still positive at toxic drug levels
viruses f rom treated and control volunteers were tested together using a single master set of drug dilutions, a single set of tissue cultures and a single set of virus dilutions kept at - 7 0 °C. The results shown in Table 5 are the means of two independent experiments which gave almost identical results.
Table 5 shows clearly that the resistant viruses f rom the volunteers given R61837 were all resistant to the other drug while the viruses f rom volunteers given placebo were uniformly sensitive. However, the results were not exactly the same. Some viruses showed a low degree of resistance to all drugs (14 and 18) others were somewhat more resistant (10,11, 12) but one (17) was highly resistant.
Table 5 also shows the results of MIC tests with disoxaril [7]. This molecule was included because its reaction with RV 14 has been studied by X-ray crys- tal lography of crystals o f virus soaked in the drug [ 11 ]. This shows that disoxaril and related molecules are all inserted into a deep region of the capsid near the '"canyon" and drug resistant mutants show substi tutions in amino acids in that region [G. Diana, pers. comm.] . It is therefore likely that R61837 and the other drugs are producing their antiviral effects by hydrophobic interactions with the corresponding region of RV9 and other rhinoviruses. However, Table 5 shows
Drug resistant rhino'~iruses from the nose of experimentally treated volunteers 79
that the virus most resistant to R61837, and other drugs, exhibited relatively little increase in resistance to disoxaril. Thus within the isolates studied there are differences in the specificity as well as the degree of sensitivity to these drugs.
Discussion
Rhinoviruses resistant to the new generation of virus-inactivating drugs can be recovered relatively easily in vitro as mentioned earlier. The general picture seen in this study was that drug resistant virus often appeared in treated vol- unteers.
However this conclusion must be examined critically. It could be that re- sistant virus was selected by exposure to the drug, even before inoculation into the test tube. R61837 is known to rapidly inactivate 70 to 80% of RV9 particles on contact. Although free drug can be extracted with lipid solvents such as chloroform, drug bound to the virus cannot be removed using this procedure [5]. Attempts to isolate residual virus after interaction with R61837 are therefore likely to yield the less sensitive viruses. Indeed one can select less sensitive virus by simple exposure of an RV9 pool in vitro to low concentrations (0.01 gg/ml) of drug before and during adsorption [K. Andries, unpublished]. Therefore, neither chloroform extraction of the samples nor washing of the cultures with PBS after the adsorption step can completely prevent the selection of resistance in vitro rather than in vivo. However in four instances resistant viruses were cultivated from specimens which contained drug concentrations below 0.002 lag/ ml and therefore we conclude that these appeared in the respiratory tract of volunteers treated with R61837, and multiplied there.
We conceive of the rhinoviruses as being like all RNA viruses, potentially unstable and variable genetically. The population in the respiratory tract is no doubt maintained relatively constant by the need to replicate efficiently in the respiratory epithelium and to be shed and infect successfully. However the introduction of the drug into the airway represents a major perturbing factor and it is interesting to see how quickly the virus may adapt to its arrival and removal.
When RV9 or RV2 are adapting to exposure to chalcone it has been noted that the first drug resistant virus to emerge may grow poorly in tissue culture, though on later passages it may grow well and be pathogenic [S. Abdelmajid, unpublished]. In the present studies the viruses recovered usually grew rather less well than usual in tissue culture but one (17) apparently caused disease.
A series of volunteer experiments has been done with this compound [2] and we report studies on viruses from one in which there was no clinical benefit. The results given here indicate that treatment delayed the host's clearance of the virus and generated resistant virus some of which nevertheless reverted to sensitivity. In many instances the degree of resistance was not extreme and may not be great enough to prevent the drug being effective. For instance, in a further trial drug administration was continued for longer and colds were
80 C. Dearden et al.
suppressed [2]. We have since found that colds are suppressed even when drug administrat ion is delayed until the day after virus exposure (unpublished). However the possible development of drug resistance will have to be allowed for when planning clinical trials. After all, drug resistant organisms are com- monly found after the clinical use of effective antibacterial drugs.
Finally, study of the MIC indicates that the site and mode of action of R61837 are similar to but not identical with that of the Sterl ing-Winthrop series o f drugs. This p h e n o m e n o n deserves further study at the molecular level.
Acknowledgements
We wish to thank the volunteers for their co-operation, Dr. P. G. Higgins for trial infor- mation and Dr. G. I. Barrow for clinical data and Dr. J. Heykanto for support in the analysis of the nasal washes for drug.
References
1. Ahmad ALM, Dowsett AB, Tyrrell DAJ (1987) Studies of rhinovirus resistant to an antiviral chalcone. Antiviral Res 8:27-39
2. A1-Nakib W, Higgins PG, Barrow GI, Tyrrell DAJ, Andries K, Vanden Bussche G, Taylor N, Janssen PAJ (1989) Suppression of colds in human volunteers challenged with rhinovirus by a new synthetic drug (R61837). Antimicrob Chemother 33: 522- 525
3. A1-Nakib W, Higgins PG, Barrow GI, Tyrrell DAJ, Lenox-Smith I, Ishitsuka H (1987) Intranasal chalcone Ro 09-0410 as prophylaxis against rhinovirus infection in vol- unteers. Antimicrob Chemother 20:887-892
4. A1-Nakib W, Willman J, Higgins PG, Tyrrell DAJ, Shepherd WM, Freestone DS (1987) Failure of intranasally administered 4',6-dichloroflavan to protect against rhi- novirus infection in man. Arch Virol 92:255-260
5. Andries K, Dewindt B, De Brabander M, Stokbroekx R, Janssen PAJ (1988) In vitro activity of R61837, a new antirhinovirus compound. Arch Virol 101 : 155-167
6. Beare AS, Reed SE (1977) The study of antiviral compounds in volunteers. In: Oxford JS (ed) Chemoprophylaxis and virus infections of the respiratory tract, vol 2. CRC Press, Cleveland, pp 27-55
7. Diana G, Otto MJ, McKinlay MA (1985) Inhibition of picornavirus uncoating as antiviral agents. Pharmacol Ther 29:287-297
8. Ninomaya Y, Ohsawa C, Aoyama M, Umeda I, Suhara Y, Ishitsuka H (1984) Antivirus agent Ro 09-0410 binds to rhinovirus specifically and stabilizes the virus conformation. Virology 134:269-276
9. Phitlpotts R J, Higgins PG, Willman JS, Tyrrell DAJ, Lenox-Smith I (1984) Evaluation of the antirhinovirus chalcone Ro 09-0415 given orally to volunteers. J Antimicrob Chemother 14:403-409
10. Phillpotts RJ, Wallace J, Tyrrell DAJ, Freestone DS, Shepherd WM (1985) Failure of oral 4'-6-dichloroflavan to protect against rhinovirus infection in man. Arch Virol 75: 115-121
11. Smith T J, Kremer H J, Luo M, Vriend G, Arnold E, Kamer G, Rossmann MG, McKinlay MA, Diana GD, Otto MJ (1986) The site of attachment in human rhinovirus 14 for antiviral agents that inhibit uncoating. Science 233:1286-1293
Drug resistant rhinoviruses from the nose of experimentally treated volunteers 81
12. Stott EJ, Tyrrell DAJ (1968) Some improved techniques for the study fo rhinoviruses using HeLa cells. Arch Ges Virusforsch 23:236-244
13. Tisdale M, Selway JWT (1984) Effect of dichloroflavan (BW683C) on the stability and uncoating of rhinovirus type 1 B. J Antimicrob Chemother 14 [Suppl A]: 97-105
Authors' address: Dr. D. A. J. Tyrrell, MRC Common Cold Unit, Coombe Road, Salisbury, Wiltshire SP2 8BW, England.
