J.
gen. Virol. (1983), 64, 1819-1822.
Printed in
Great Britain
1819
Key words: IFN (HulFN)/ lymphocytes/inducer
An Interferon-fl-like or Interferon-inducing Protein Released by Mitogenstimulated Human Leukocytes
By J. V A N D A M M E , A. B I L L I A U , * M. D E L E Y AND P. D E S O M E R
Department o f Human Biology, The Rega Institute, University o f Leuven,
Minderbroedersstraat 10, B-3000 Leuven, Belgium
(Accepted 11 M a y 1983)
SUMMARY
Human peripheral blood leukocytes were treated with concanavalin A (Con A) to
produce interferon 3' (HuIFN-7). On gel filtration this interferon eluted as a protein
with a molecular weight of 45 000. In addition to this, the culture supernatant contained
an interferon-like protein of apparent molecular weight 22000 (22K factor). The
antiviral activity of this protein was neutralizable by a highly specific antibody to
HuIFN-fl. Yet, the 22K factor differed from classical HuIFN-fl in several
characteristics: lack of activity on certain homologous and heterologous cells which are
sensitive to HuIFN-fl; lack of affinity for zinc-chelate and Con A-Sepharose columns;
failure to bind to an anti-HuIFN-fl antibody column. Moreover, a specific antiserum
raised against the 22K factor did not neutralize HuIFN-fl. Two alternative explanations
of these findings are proposed: (i) the 22K factor is an interferon whose molecular
structure resembles that of the known HuIFN-fl but it is not identical to it, or (ii) the
22K factor is not an interferon but a protein that can induce the production of H u I F N fl in certain lines of fibroblastoid cells.
One of the lymphokines produced by antigen- or mitogen-stimulated lymphocytes is an
interferon called HuIFN- 7 (for reviews, see Stewart, 1979; Epstein, 1979). Although its physicochemical and biological properties are different from those of other interferons, its distinctiveness is essentially based on its reactivity in neutralization tests with specific antisera (Havell et
al., 1975; De Ley et al., 1981). Cells induced to synthesize interferon often produce a mixture of
different types, and we found that peripheral blood lcukocytes induced with concanavalin A
(Con A) released, in addition to HuIFN-7, an antiviral factor of molecular weight 22 000 (22K
factor), serologically characterizable as HuIFN-fl-like (De Ley et al., 1980; Van D a m m e et al.,
1981). However, in the present study we show that this 22K factor can be distinguished from
HuIFN-fl by several biological and physicochemical properties.
Suspension cultures of peripheral blood leukocytes obtained from human burly coats were
stimulated by Con A (Van Damme et aL, 1981). The crude interferon obtained was concentrated
and partially purified by adsorption to silicic acid and dcsorption in 50~o ethylene glycol. It was
then fractionated by gel filtration, the antiviral activity (as titrated on human diploid cells)
eluting in two peaks with apparent molecular weights of approx. 45000 (45K component) and
approx. 22000 (22K component, henceforth called 22K factor).
The 45K component had the essential characteristics of HuIFN-~ : no serological relationship
to HuIFN-~ and -fl, neutralization by a homologous specific antiserum, sensitivity to acid (pH 2)
and no activity on heterologous cells (De Lcy et al., 1980; Van D a m m e et al., 1981). The 22K
component, on the other hand, was neutralizable by a specific anti-HulFN-13 antiserum but not
by anti-HuIFN-~ or -7 antisera (Table 1). It was protein in nature, its antiviral effect was
directed against viruses of different taxonomic groups but was restricted to the species of origin
of the cells. Establishment of the antiviral state required intact cellular m R N A synthesis.
Finally, in a single-cycle R N A virus infection system, treatment with the 22K factor inhibited
viral R N A synthesis, concordant with the mechanism of action of known intefferons on R N A
viruses.
Downloaded from www.microbiologyresearch.org by
0022-1317/83/0000-5680 $02.0001983
SGM
IP: 88.99.165.207
On: Wed, 12 Jul 2017 20:51:23
Short communication
1820
T a b l e 1. Serological characterization of the 22K component of Con A-induced interferon
Antiserum added*
Specificity
Control (no antiserum)
Anti-HuIFN-ctt
Anti-HulFN-fl:~
Serum no. 1
Residual antiviral activity (loglo U/ml) of
loglo Dilution
- 4
-2
- 3
Serum no. 2
Serum no. 3
Serum no. 4
Serum no. 5
Anti-HuIFN-y§
Anti-22KlI
-4
- 3
-4
-1
-2
-2
- 2
-2
HuIFN<t
3-6
2-3
HuIFN-fl
4.8
4.6
HuIFN-?
3.0
2.8
22K
3.3
3-2
3.7
4-0
4-1
3-7
3-7
3-4
3.7
2-0
2-8
4.1
< 2-0
2.7
<2.0
2-4
2.3
4-8
4.9
2-6
2.6
3.0
2-6
2.9
2-3
2-5
2.7
< 1.5
2-9
<1.0
3.5
3-6
< 1.0
3.2
< 1.0
1.4
<1.0
<l.0
1.3
3.4
< 1.0
* Constant concentration of antibody added to each dilution of interferon assay (2 h at 37 °C).
t Serum provided by Dr B. Dalton (Wistar Institute, Philadelphia, Pa., U.S.A.). Origin: sheep immunized with
purified HuIFN-~.
Origins of anti-HulFN-fl sera: 1, goat immunized with fibroblast interferon of sp. act. 106 U/rag; 2, as 1, sp.
act. l09 U/mg; 3, rabbit immunized with fibroblast interferon, serum provided by Dr J. Vil6ek, New York
University, N.Y., U.S.A. ; 4, sheep immunized with fibroblast interferon of sp. act. 106 U/mg, serum obtained
from Dr B. Dalton; 5, rabbit immunized with HulFN-fl from E. coliexpressing a plasmid-borne HulFN-fl gene
(Dr H. Schellekens, TNO, Rijswijk, The Netherlands).
§ Serum prepared from a rabbit immunized with 45K component of human immune interferon (De Ley et al.,
1981).
]] Serum prepared from a rabbit immunized with 22K component of human immune interferon.
T a b l e 2. Antiviral activity of the 22K factor on homologous and heterologous cells in comparison with
known interferon types
Antiviral activityl"
¢
Cell type*
IFN-ct IFN-fl
Human diploid fibrobtasts (E6SM), low passage level
100
100
Human diploid fibroblasts (E6SM), high passage level
320
45
Human trisomic-21 cells (LiR)
1070 2075
Human trisomic-21 cells (RoL)
1065 1040
Human amnion cells (WISH)
63
49
Human epidermal carcinoma cells (HEp-2)
71
26
Human osteosarcoma cells (MG-63)
29
21
Monkey kidney cells (Vero)
40
40
Porcine kidney cells (PK)
400
10
Primary calf kidney cell culture
320
40
Primary rabbit kidney cell culture
0.3
A
IFN-),
100
25
860
329
60
28
68
2-5
< 1-0
<0.5
<0.6
Interferon• producing
22K
capacityt:~
100
100
<1
<8
299
5
<2
61
0-3
<0-5
0.4
64
100
8
5
286
<1
<0-1
* Origin of cell lines has been described by De Ley & Billiau (1982).
t Figures are means of three separate determinations, expressed as percentages of the interferon titres on human
diploid fibroblasts of low passage level; all interferon preparations had a s p . act. /> 106 U/mg.
:~Cells were tested for interferon production using the superinduction method with polyriboinosinic, ribocytidilic acid (50 ~tg/ml), cycloheximide (10 ~tg/ml) and actinomycin D (1 Ixg/ml) as described by Van Damme & Billiau
(1981).
H o w e v e r , s o m e o t h e r c h a r a c t e r i s t i c s o f t h e 2 2 K f a c t o r d i s t i n g u i s h e d it f r o m k n o w n HulFN-13:
it w a s s t a b l e in a c i d o r S D S - c o n t a i n i n g buffers, b u t u n l i k e H u l F N - f l it h a d n o affinity for
i m m o b i l i z e d z i n c - c h e l a t e or C o n A. W h e r e a s classical H u l F N - f l is a c t i v e o n cells o f v a r i o u s n o n h u m a n species, t h e 2 2 K f a c t o r w a s f o u n d to b e strictly species-specific, t h e r e b y r e s e m b l i n g
H u l F N - ? ( T a b l e 2). I n a d d i t i o n , w h e n t e s t e d o n d i f f e r e n t h u m a n cell t y p e s , it s h o w e d a
r e m a r k a b l e cell specificity. I t w a s i n a c t i v e in H E p - 2 a n d W I S H cells w h i c h a r e s e n s i t i v e to
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 20:51:23
Short communication
1821
classical HuIFN-fl, as well as to HuIFN-~ and -y. It also failed to act in certain disomic and
trisomic cells which were highly sensitive to HuIFN-fl.
In view of these data we investigated more closely why the antiviral effect of the 22K factor is
neutralizable by anti-HuIFN-fl. The possibility was considered that our antisera against
HuIFN-fl also contained antibodies against various other proteins, including perhaps the 22K
protein. Sera from different sources (Table 1) were therefore used for the neutralization
reactions. All sera caused neutralization; most significantly, a serum prepared against HuIFN-fl
produced by Escherichia coli, expressing a plasmid-borne HuIFN-fl gene, also had neutralizing
activity for the 22K factor, indicating that the cross-reactivity was not due to antibodies other
than those directed against HuIFN-fl.
We also examined the affinity of the anti-HuIFN-fl antibody for the 22K protein. Neutralization reactions were carried out with a fixed concentration of antibody in each of the
interferon dilutions before addition to the assay cells. Lowering of the concentration to limit
amounts did not reveal consistent differences in the kinetics of interaction of the 22K factor and
classical HuIFN-fl with the HuIFN-fl antibody (see Table t). However, 22K failed to adsorb to a
column of anti-HuIFN-fl antibody fixed on Sepharose beads, although this column did adsorb
nearly 100% of classical HuIFN-fl and released this interferon on elution with acid. Furthermore, an antiserum prepared against the 22K factor by immunizing a rabbit was quite active in
neutralizing the biological activity of the 22K factor, but was inactive against classical HuIFNfl. The latter two observations suggested that neutralization of the biological activity of the 22K
factor by anti-HuIFN-fl might result from indirect interference of the antibody in the assay
system.
There are several possible explanations for our data. Firstly, the 22K factor may be a novel
HuIFN, different from known HuIFN-~, HulFN-fl and HulFN-y. Specifically, it may be a
previously unknown subtype of HulFN-fl. This idea is supported by evidence for the existence
of multiple genes for HulFN-fl as demonstrated by Sehgal &Sagar (1980). However, according
to Tavernier et al. (1981) there exists only one HuIFN-fl gene, thereby eliminating the possibility
of pleomorphism in primary structure. Also, the 22K protein might share only a very few
epitopes with HulFN-B. This does not necessarily imply stretches of identical amino acid
sequence; areas with similar steric configuration would theoretically be sufficient to explain its
relationship to HulFN-fl. A precedent for this is a protein described by Weissenbach et al.
(1980) as HulFN-flz. This protein is obtained by translation in oocytes of a subclass of mRNA
isolated from induced fibroblasts. In immunoprecipitation reactions this protein seems to be
related to HulFN-fl, yet it is still not clear whether its biological activity is also serologically
related to HulFN-fl.
Secondly, the 22K factor and classical HulFN-B might have the same primary structure but
differ in glycosylation. This might explain different cell- and species-specificities as well as
different affinities for adsorbents, but would less easily explain the lack of binding of the 22K
factor to anti-HulFN-fl antibody, and the lack of cross-neutralization in reactions with anti-22K
serum.
Thirdly, it might be a component of HulFN-y, since Yip et al. (1982) have demonstrated that
mild treatment of HulFN-y (58K) with SDS, while destroying most of the biological activity,
converts the residual material to two antiviral proteins of lower molecular weight: 20K and 25K.
However, we consider this unlikely since neither derivative was reported to be neutralizable by
anti-HulFN-p. Also, these materials have affinity for Con A while the 22K factor does not bind
to Con A.
A very different possibility is that the 22K factor is not an interferon but an interferoninducing protein. In this case its antiviral activity in bioassays would result from production, by
the assay cells, of HulFN-fl. Inclusion of antibody to HulFN-fl in the assay medium would
neutralize the antiviral activity. This hypothesis has the advantage of being compatible with all
the physicochemical and serological data presented here. Some further support for it is available
from a comparison of the sensitivity of different human cell lines to the 22K factor and their
capacity to produce HulFN-fl. Diploid cells and MG-63 cells (van Damme & Billiau, 1981) are
known to be good producers of HulFN-fl, and, as shown in Table 2, are also very sensitive to the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 20:51:23
1822
Short communication
22K factor. Also, the 22K-sensitive trisomic RoL cells were good producers, while the 22Kinsensitive W I S H and HEp-2 Cells were low producers. A n exception to this parallelism were the
trisomic LiR cells, which were relatively good producers of H u l F N - f l but were poorly sensitive
to the 22K factor.
Attempts to obtain direct evidence for the inducer hypothesis by demonstrating the
production of H u l F N - f l in diploid cell cultures after treatment with the 22K factor were always
unsuccessful: supernatants and cell lysates of such cultures were always devoid of detectable
amounts of HulFN-/L Thus, if our hypothesis is correct the amounts of interferon induced by the
22K factor would be small, but still sufficient to make the cells virus-resistant. Similar
experiments with MG-63 cells yielded positive results, provided cycloheximide was included in
the medium. This lead is currently being pursued in collaboration with a laboratory (that of D r J.
Content, Pasteur Institute, Brussels) experienced in detecting H u l F N - f l m R N A in cells.
This study was supported by the Cancer Research Foundation of the Belgian A.S.L.K. (General Savings and
Retirement Fund) and by the 'Geconcerteerde Onderzoeksacties'. Marc De Ley is Bevoegdverklaard Navorser of
the Belgian NFWO (National Fund for Scientific Research). The authors thank their colleagues C. Vermylen,
M. Peetermans, H. Claeys and L. Muyle from the Blood Transfusion Services of Leuven and Antwerp, for
providing buffy coats. The excellent technical assistance of R. Conings, J.P. Lenaerts, W. Put and I. Ronsse, as
well as editorial help of C. Callebaut are gratefully acknowledged.
REFERENCES
DELEY,M.&mLLIAU,A. (1982). Responsivenessof human ceils trisomic for chromosome 21 to the antiviral action
of human immune interferon. Antiviral Research 2, 97-102.
DE LEY, M., VAN DAMME, J., CLAEYS, H., WEENING, H., HEINE, J., BILLIAU, A., VERMYLEN, C. & DE SOMER, P. (1980).
Interferon induced in human leukocytes by mitogens: production, partial purification and characterization.
European Journal of Immunology 10, 877-883.
DE LEY, M., VAN DAMME, J., BILLIAU, A. & DE SOMER, P. (1981). The preparation of antibodies directed against human
immune interferon. Journal of Virological Methods 3, 149-153.
E~TEIN,L. B. (1979). The comparative biologyof immune and classical interferons. In Biology of the Lymphokines,
pp. 443-514. Edited by S. Cohen, E. Pick & J. Oppenheim. New York: Academic Press.
HAVELL, E. A., BERMAN, B., OGBURN, C. A., BERG, K., PAUCKER, K. & VILCEK, J. (1975). Two antigenically distinct
species of human interferon. Proceedings of the National Academy of Sciences, U.S.A. 72, 2185-2187.
SEHGAL,V. B. &SAGAR,A. n. (1980). Heterogeneity of poly(I), poly(C)-inducedhuman fibroblast interferon mRNA
species. Nature, London 288, 95-97.
STEWART,W. E., U (1979). The Interferon System, p. 421. Wien & New York: Springer-Verlag.
TAVERNIER,J., DERYNCK,R. & FIERS,W. (1981). Evidence for a unique human fibroblast (IFN-]~I) chromosomal
gene, devoid of intervening sequences. Nucleic Acids Research 9, 461-471.
VANDAMME,J. &mLLIAU,a. ( 1981). Large-scaleproduction of human fibroblast interferon. Methods in Enzymology
78, 101-119.
VAN DAMME, J., DE LEY, M., CLAEYS, H., BILLIAU, A., VERMYLEN, C. & DE SOMER, P. (1981). Interferon induced in
human leukocytesby concanavalin A: isolation and characterization of y- and p-type components. European
Journal of Immunclogy 11, 937-942.
WEISSENBACH, J., CHERNAJOVSKY, Y., ZEEVI, M., SHULMAN, L., SOREQ, H., NIR, U,, WALLACH, D., PERRICAUDET, M.,
TIOLLAIS,P. & REVEL,M. (1980). Two interferon mRNAs in human fibroblasts: in vitro translation and
Escherichia coli cloning studies. Proceedings of the National Academy of Sciences, U.S.A. 77, 7152-7156.
YIP,. Y, K., BARROWCLOUGH, B. S., URBAN, C. & VILCEK, J. (1982). Molecular weight of human gamma interferon is
similar to that of other human interferons. Science 215, 411-413.
(Received 10 March 1983)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 12 Jul 2017 20:51:23