Download Structure-Function Analysis of Mouse Interferon Alpha Species

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Genomic imprinting wikipedia , lookup

Ribosomally synthesized and post-translationally modified peptides wikipedia , lookup

Community fingerprinting wikipedia , lookup

Secreted frizzled-related protein 1 wikipedia , lookup

Protein–protein interaction wikipedia , lookup

Western blot wikipedia , lookup

Vectors in gene therapy wikipedia , lookup

Biochemical cascade wikipedia , lookup

Signal transduction wikipedia , lookup

Protein wikipedia , lookup

Magnesium transporter wikipedia , lookup

Gene therapy of the human retina wikipedia , lookup

Gene nomenclature wikipedia , lookup

Metabolism wikipedia , lookup

Gene wikipedia , lookup

Expression vector wikipedia , lookup

Biosynthesis wikipedia , lookup

Gene expression wikipedia , lookup

Genetic code wikipedia , lookup

Gene regulatory network wikipedia , lookup

Gene expression profiling wikipedia , lookup

Two-hybrid screening wikipedia , lookup

Amino acid synthesis wikipedia , lookup

Protein structure prediction wikipedia , lookup

Point mutation wikipedia , lookup

Endogenous retrovirus wikipedia , lookup

Silencer (genetics) wikipedia , lookup

Biochemistry wikipedia , lookup

Proteolysis wikipedia , lookup

Artificial gene synthesis wikipedia , lookup

Transcript
J. gen. Virol. (1988), 69, 67-75. Printedin Great Britain
67
Key words: interJeron (MulFN-a)/hybrid gene/gene structure
Structure-Function Analysis of Mouse Interferon Alpha Species:
MulFN-aI0, a Subspecies with Low Antiviral Activity
By J. T R A P M A N , * M. V A N H E U V E L , P. D E J O N G E , ~ I. J. B O S V E L D ,
P. K L A A S S E N AND E. C. Z W A R T H O F F
Department o f Pathology, Erasmus University, P.O. Box 1738, 3000 D R Rotterdam,
The Netherlands
(Accepted 24 September 1987)
SUMMARY
A mouse interferon alpha gene (MuIFN<tl0) was isolated from a BALB/c cosmid
genomic library. The gene was located on a 1.8 kb HindIII fragment and a 5.1 kb EcoRI
fragment. The coding region and parts of the 5' and 3' non-coding regions were
sequenced. The results showed that the MuIFN-cd0 gene encoded a protein of 167
amino acids. Like most other MuIFN-ct species it contained a putative N-glycosylation
site at amino acid positions 78 to 80. It also possessed cysteine residues at positions 1,
29, 86, 99 and 129. In the signal peptide, in addition to cysteine 21, which is present in
all MuIFN-ct species sequenced so far, a cysteine was found at position 22. At the
amino acid level MuIFN-cd0 showed strong homology to MuIFN-ctl (only 15 out of 167
amino acids were different). The MuIFN-cd0 gene was transiently expressed in
monkey COS cells under the direction of the simian virus 40 early promoter. The
protein product secreted by COS cells was equally active on mouse (L929) and hamster
(CHO) cells. However, as compared to MuIFN-ctl and MulFN-ct4 the specific activity
on mouse cells of the protein was 10- to 100-fold lower. To find out which region of its
structure was responsible for this low activity, hybrids of the genes encoding MuIFNctl0 and MuIFN-ctl were constructed using the two common XmnI sites which
correspond to positions between amino acids 67 and 68 and 123 and 124, respectively.
The data showed that hybrid constructs which were MuIFN-~l-like from amino acid
68 or MuIFN-cd0-1ike from position 124 to the C terminus possessed high antiviral
activity. Other hybrid constructs were hardly active at all. This implied that the amino
acid 68 to 123 region was mainly responsible for the low antiviral activity of MuIFN~10. In this part of the molecule MulFN-ctl and MulFN-ctl0 differed in only five amino
acids. A serine at position 110 and a valine at 85 were unique to MuIFN-ctl0 as
compared to all known MulFN-ct and human IFN-¢t subspecies.
INTRODUCTION
Three antigenically distinct types of interferons (IFNs) can be recognized. These are
commonly known as IFN-~, IFN-/~ and IFN-7 (for a recent review, see Weissmann & Weber,
1986). IFN-~s are encoded by a multigene family. In the mouse system, nine different complete
genes have so far been isolated and characterized. Eight of these [~1, ~2, c~4,c~5,~6T, c~6P, ~A (~7)
and ~9] encode biologically active proteins (Shaw et al., 1983; Daugherty et al., 1984; Zwarthoff
et al., 1985; Kelley & Pitha, 1985 a; Self & DeMaeyer-Guignard, 1986; Kelley et al., 1986) and
the other is a pseudo-gene (LeRoscouet et al., 1985). Results so far show that all mouse (Mu)
IFN-~ genes are clustered on chromosome 4 (Kelley et al., 1983; Lovett et al., 1984; Dandoy et
al., 1984, 1985; Van der Korput et al., 1985).
The mature MuIFN-~ species are 166 or 167 amino acids long. An exception is MuIFN-~4,
which possesses a deletion of five amino acids corresponding to amino acids 103 to 107 in the
t Present address: Department of Microbiology, Technical University, Delft, The Netherlands.
000-7914 © 1988 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 08 May 2017 23:20:06
68
s. TRAPMAN AND OTHERS
other proteins (Zwarthoff et al., 1985; Kelley & Pitha, 1985 b). The mutual structural homology
of the eight different subspecies varies from 749/0 to 88~o. Most MuIFN-ct species [excluding
MuIFN-ct6T and MulFN-~A (ct7)] contain an N-glycosylation site (Asn-Ala-Thr) at positions 78
to 80 of the mature protein. Although highly homologous, MuIFN-c(s show different antiviral
properties. Their specific activity as measured on mouse cells varies about tenfold (Van Heuvel
et al., 1986, 1987). Moreover, there are large differences in antiviral activity on hamster cells
between the various species.
The availability of a family of very homologous genes of which the protein products show
different biological activities provides the opportunity to perform a detailed structure-function
analysis of these proteins. During the course of our study on this subject we isolated a M u I F N
gene (MuIFN-ctl 0) which was closely related to MuIFN-ctl, but showed a tenfold lower specific
activity on mouse (and hamster) cells. Here we describe the characterization of this I F N species
and of hybrid proteins derived from MuIFN-~l and MuIFN-~10 by recombination of the
natural genes using common restriction enzyme sites.
Preliminary data have been published elsewhere (Trapman et al., 1986). In that abstract the
gene described as MuIFN-~tI0 here was tentatively named MuIFN-c(9. Because of the recent
isolation of another gene which was also called MuIFN-ct9 (Self & DeMaeyer-Guignard, 1986)
the nomenclature has been changed.
METHODS
A cosmid mouse genomic library prepared from WEHI cells was the kind gift of Dr A. de Klein, Department of
Genetics, Erasmus University. The MulFN-ctl0 gene was detected under standard stringent hybridization
conditions using MuIFN-ctl as a probe. Preparation of genomic and plasmid DNA, restriction enzyme digestions,
ligations, nick translation and Southern blotting were essentially as described by Maniatis et al. (1982).
Sequencing was done by the dideoxy chain termination method (Sanger et al., 1977). The appropriate fragments
were cloned in M13mp18/19 (Messing et al, 1981).
Transfection of COS cells, PAGE of 35S-radiolabelled supernatant proteins and calculation of the specific
activity of MulFN-cts were as described previously (Van Heuvel et al., 1986, 1987).
Interferon titrations were by the cytopathic effect reduction assay on mouse L929 cells or Chinese hamster ovary
(CHO) cells using vesicular stomatitis virus as a challenge. G-002-904-511was used as an International Standard
on L929 cells. MuIFN-cd was used as a standard on CHO cells (Van Heuvel et al., 1986).
RESULTS
Isolation o f the M u l F N - ~ I O gene
A BALB/c mouse genomic library was screened with a MuIFN-ctl gene fragment as a probe.
One of the positive clones was further characterized by restriction enzyme mapping. The
MuIFN-~ gene could be localized on a 5.1 kb E c o R I fragment. This fragment was subcloned and
analysed in more detail, its restriction map is depicted in Fig. 1 (a). Fine mapping showed that
the gene was situated on a 1.8 kb HindIII fragment and a 1.2 kb H i n d I I I - P s t I fragment.
To ensure that we had isolated a naturally occurring MuIFN-~ gene, rather than a
recombinant gene arising from two different genes by homologous recombination during
propagation of the cosmid library, we hybridized mouse genomic D N A with the 1-2 kb and
0.6 kb H i n d I I I - P s t I fragments (Fig. 1 b and c). Hybridization with the 1.2 kb probe showed a
large series of bands both in the E c o R I and HindIII digests, representing the MuIFN-~
multigene family. Among the most prominent bands were a 5.1 kb E c o R I and a 1.8 kb HindIII
fragment, corresponding in size to the respective restriction enzyme fragments of the isolated
clone. When the 0-6 kb flanking fragment was used as a probe, a single band was observed (also
5.1 and 1-8 kb in size in both digests). From these results we concluded that we had isolated a
natural MuIFN-ct gene (MuIFN-~10).
Structure o f the MuIFN-otlO gene and protein
The coding region and parts of the 5' and 3' flanking regions of MuIFN-~10 were sequenced
(Fig. 2). The results indicated that it was a functional MuIFN-~t gene. The presence of an open
reading frame of 570 bp was obvious. Furthermore, a T A T T T A A box, which is a characteristic
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 08 May 2017 23:20:06
Properties of MulFN-~IO
69
(a)
0
EcoRI
(b)
1
2
(c)
1
2
PvulI
1
23-6-9.6 m
6.6~
2
--5"1
4.3--
HindlII
3
- XmnI
5'
3"
~
10
4
2"3 I
2.0 u
i NeoI
XmnI
XmnI
--1.8
PstI
NcoI
HindlII
5
EcoRI
Fig. 1. (a) Physical map of the genomic fragment on which the MulFN-~I 0 gene is situated. Distances
are in kbp. (b) Southern blot hybridization of mouse genomic DNA with the 1.2 kbp HindlII-PstI
MulFN-~I0 probe. Lane 1, HindlII digest; lane 2, EcoRI digest. (c) Southern blot hybridization of
mouse genomic DNA with the 0.6 kbp PstI-HindlII MulFN-cd0 flanking probe. Lane 1, HindlII
digest; lane 2, EcoRI digest. Markers in kbp are HindlII fragments of phage 2.
of IFN-~ genes, was found about 100 bp upstream from the A T G start codon, while 32 bp
downstream from the (modified) T A T A box an A G sequence was found, which is thought to
represent the transcriptional start of IFN-~ m R N A s . In the small stretch of 3' non-coding
nucleotides sequenced, no polyadenylation signal could be detected.
Fig. 3 (a) shows the amino acid sequence of the signal peptide of the MulFN-~10 Protein as
deduced from the nucleotide sequence; Fig. 3 (b) compares the primary structure of the mature
MulFN-~10 protein with the sequence of other MulFN-~ proteins. In the signal peptide
MulFN-~10 diverged from all other known sequences at one position. In addition to the Cys
residue on $21, MulFN-~10 contained a Cys at $22, where normally Ser is found. The mature
protein was composed of 167 amino acids. It possessed the five cysteine residues observed in all
MulFN-~ species at positions 1, 29, 86, 99 and 139. It also contained the putative Nglycosylation site (Asn-Ala-Thr) at 78 to 80 found in M u l F N - ~ species with the exception of
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 08 May 2017 23:20:06
70
J. T R A P M A N A N D O T H E R S
-125
-100
-50
~AAAATA~AA~EIAGT~TTTTT~TCC~ATTTAAGA~ACATT~ACCCA~GAT~TCTTC~GAGAACCTA~A~G~AAG~AT~AgGAC~AAACA~CCCA~AA~A~A~
1
CAGCAI~GGfAA£M~CAEC
50
~[g gC~ AGG E~C TG~ gc~ ~]E zig M G gI~ EIg BEg GIG A~G AGE IAC TGG CCA ACE Igl IGT
tO0
ETA GGA IGT GAC CTT CCT CAG ACT CAT AAC CTC AGG AAC AAG AGA GCC ITA ACC CTC CTG GTA AAA ATG AGG AGA CIC
150
200
ICE EEl CIC TCC TGC CIG AAG GAC EGG AAG GAC TIT GGA TTT CCC CAG GCA AAG GIG GAT GCC CAG CAG ATC CAG GAG
250
GCI CAA GEE ATC CCI GTC CTG AGI GAG CTG ACE CAG CAG AIC CIG AAC ATE TIC ACA TEA AAG GAC TCA TCT GCI GCT
300
350
IGG AAI GCA ACE CTC ETA GAC TEA GTC TGC AAT GAC CTC CAC gAG CAG CTC AAT GAg gig CAA GGC Tgf CTG ATG CAG
400
A50
GAG GTG GGG GIG CAG GAA CTI ICE CIG ACE CAA GAA GAC ICC CIO CTG GCI GIG AGG AAA TAC TTC CAC AGG ATE ACT
5OO
GIG ITC C]g AGA GAG AAG AAA CAC AGE CCC TGT GEE IGG GAG GIG GTC AGA GCA GAA ATE TGG AGA GCC CTG TEl ICE
550
600
TEA GCC AAC TIE CTG GCA AGA CIG AGI GAG AAG AAG GAG TGA GICCTGAGACAAAGIAGAGAGGAICTCCAGGATTAGGACACTGCACC
699
650
TCICTGTCCAGATTCTACCAICTCAAAAAIAICATACAAITTTGCATIAAITGAAGCAAGITGCCIAGATCTTCTGCAG
FiB. 2. Nucleotide sequence o f the M u I F N - ~ I O gene. T A T A box, A T ( } start codon, T O T codon
encoding the first a m i n o acid in the mature protein and T G A stop codon are underlined.
(a)
St
~10
S10
$20
MARLCAFLMVLAVMSY-WPTCCLG
~1
• : ...................
G(2
........
S,.
VM.I . . . . . .
c(4
.........
m5
...........
I.VM...Y.SA.S..
P.L . . . . . . .
m6T
..................
c(6P
.............
c(9
...PF . . . . . .
otA
SI.S..
.........
S..
S.. S . .
L.......
S..
V.I .... S..S..
T.L ......
S..S..
(b)
10
1
• 10
20
30
40
50
• I
...................
~2
....
H.Y ........
a4
....
H.Y..G
~5
........................................
E..G
~6T
.......
E...TLK..KEK
~6P
........................................
a9
~A
Q ....................
KV.AQ ....
......
............
E .......
PF . . . . . . .
V.EE ....
K ..........
Q .....
70
P ..............
....
K .......
LE...N
....
K ....
K1 . . . .
E..G
AQ . . . . . . . . . . . . . . . . . . . .
.......
...................
R...E
.......
.............
110
........
120
....
L .....
L..T
........
D .....
TL .................
TL .................
Q .....
140
V .............
150
D.S ....
160
~CNDLHQQLN~L~GCLMQEvGvQELSLTQEDSLLAvRKYFHR~FLREKKHSPCA~EVvRAE~WRALSSSANLLARL~EK~E
~I
F .................
Q .....
FP . . . . .
A .............
Y .................
V .........
F ............
T ....
Q .....
PP . . . . .
A .............
Y .................
V .......
• 4
F ...........
KA.V..-
....
~5
F..EV
.......
KA.V..Q
.....
~6T
f ....
Y .......
~6P
f
~A
A ............
V .................
T .......
N .......
130
T..L
V ....................
E ........................
Q ................
100
RD . . . .
L..RD
~2
~9
80
KK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
LE...N
IQ ....................
...................
90
~10
60
CDLPQ~HNLRNKRALTLL~KMRRL~PL~CCKDRKDFGFPQAKVDAQQIQEAQA~P~L~EL~Q~LN~F~SKD~AAWNA~LLD~
......
FRG...L
F ...........
.PP ...........
A..V.Q.RL..PP
SP . . . . . . . . . . . . . . . . . . .
....
Q.EI.A.P
...........
..........
L..MK..P
......
PP . . . . .
Q..MK
Y .....
Y..K
....
L .....
I...V
Y .................
V ..............
L ..........
KA.V ........
T .......
Y .................
V.G..R.E.V.,.P
.......
V .......
V .........
E..
KEE-
V.G..R.E.-
V ........
K .....
........
Y .................
V .......
V ........
T .......
Y .................
V...M
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 08 May 2017 23:20:06
E..
V .......
] .........................
....
....
T ........
K .......
N.DEE..
E..
T
Properties of MulFN-c¢IO
71
MulFN-~6T and MulFN-aA (a7). MulFN-al0 showed the highest homology to MulFN-~I
(only 15 out of 167 amino acids are different) and the lowest to MulFN-~4 (36 differences).
MulFN-~I0 contained unique amino acids at four positions, conserved in other MulFN-~
proteins: positions 41 (Ala instead of Glu), 85 (Val instead of Phe), 110 (Ser instead of Pro) and
148 (Ile instead of Val).
Antiviral properties of the MulFN-~IO protein
To obtain information on the antiviral properties of the MulFN-~10 protein, the gene was
inserted into the eukaryotic expression vector pSV328A (Van Heuvel et al., 1986) which
contains the origin of replication and early promoter of simian virus 40 (SV40) and the
polyadenylation signal of the rabbit fl-globin gene. pSV328A was digested with PstI (partially)
and HindlII. The 1.2 kb HindlII-PstI fragment of MulFN-~10 was cloned into the partially
digested pSV328A, resulting in pSV~10HA. Next, the NcoI fragment of pSVccl0HA was
replaced by the NcoI fragment of pSV~I, thus removing the MulFN-~10 promoter.
Alternatively, the MstlI fragment of pSVcOOHA was isolated and exchanged with the MstlI
fragment of pSV~I. The latter approach was possible because in the coding region of the mature
proteins the first amino acids of MulFN-~I and -~10 were identical. In this part MulFN-~I and
-~10 differed only in the signal peptide sequence (see Fig. 3).
Both constructs were transiently expressed in monkey COS-I cells and the supernatant was
monitored for antiviral activity and protein production. Fig. 4 shows the results of SDS-PAGE
of MulFN-cd0 (NcoI) secreted by COS cells as compared to control IFNs. Table 1 summarizes
the corresponding antiviral activity. The protein pattern illustrates that comparable amounts of
different proteins (MulFN-~I, -c¢4 and -~10) were present; however, the amount of antiviral
activity measured in the different samples varied considerably. The Cys residue in the signal
peptide on $22 had no effect on the production of the protein [compare MulFN-~10 (NcoI) and
MulFN-~10 (MstlI) in Table 1]. As shown earlier (Van Heuvel et al., 1986) MulFN-c~4 possesses
the highest specific activity, while the activity of MulFN-~10 is more than 100-fold less and at
least 10-fold lower than that of MulFN-~I. Using an independent isolate of the MulFN-~I0
gene from a different library, identical values were obtained (data not shown). From these
results, MulFN-al0 was the mouse IFN subspecies with the lowest specific antiviral activity on
mouse cells. The antiviral activity on hamster (CHO) cells was relatively high and identical to
that on mouse cells (see Table 1).
Antiviral properties of MulFN-alO~I hybrid proteins
To obtain a more detailed insight into the region of MulFN-~10 that is responsible for its low
biological activity we constructed a series of hybrids between the highly homologous MulFN• 10 and MulFN-~I genes using the two common XmnI sites which correspond to amino acid
positions 67 to 68 and 123 to 124, respectively. The hybrid genes were transiently expressed in
COS cells and supernatants were analysed for protein composition and antiviral activity (Fig. 5,
Table 2). Varying amounts of the different hybrid proteins were present in the cell culture
medium. MulFN-~I~10 (Xb) and MulFN-~10~I (Xa) (Fig. 5, lanes 5 and 7) were present in
amounts comparable to the parental proteins, no MulFN-~I ~10 (Xa) and a decreased amount of
MulFN-~I0~I (Xb) were seen (Fig. 5, lanes 4 and 8). From these findings we concluded that
there is a relationship between high activity and the presence of MulFN-~I sequences in the
middle part (from amino acid 68 to 123) of the protein molecule (Table 2). If MulFN-~10
sequences are present in this region no IFN or an IFN with a much lower specific activity is
detected. This strongly suggests that amino acids which differ between MulFN-~I and -~10 in
this part of the protein are responsible for the low antiviral activity of MulFN-cO0.
Fig. 3. (a) Amino acid sequenceof the signal peptide of the MulFN-~10 protein as deduced from the
nucleotide sequence compared to the signal peptides of other MulFN-~s. (b) Amino acid sequence of
the mature MulFN-~I0 protein as compared to other MulFN-a subspecies.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 08 May 2017 23:20:06
J. TRAPMAN AND OTHERS
72
l
2
3
4
5
--97
--69
m46
--30
IFN-~--
--18
Fig. 4. Fluorograph of an SDS-polyacrylamide gel containing 35S-labelled proteins secreted by COS
cells which had been transiently transfected with IFN expression plasmids. Lane 1, MuIFN-~I ; lane 2,
MulFN-~4; lane 3, MulFN-~10 (NcoI); lane 4, control COS supernatant; lane 5, molecular weight
markers (Mr x 10-3).
T a b l e 1. Antiviral activity of MulFN-a subspecies on mouse (L929) and hamster (CHO) cells
MulFN
Antiviral activity on L929 cells
(IU/ml)
Antiviral activity on CHO cells
(U/ml)
al
cc4
al0 (NcoI)
cd0 (MstlI)
3200
32000
240
240
3200
48
320
320
T a b l e 2. Antiviral activity of MulFN-cG~IO hybrid proteins on mouse (L929) cells
MuIFN
cd
~1~10
alcd0
~10
cd0~l
cd0~l
(Xa)
(Xb)
(Xa)
(Xb)
Antiviral activity
(IU/ml)
Specific activity
(IU/mg protein)
3200
<1
1600
240
3200
2
1-5 x 107
ND*
0.8 × 107
1-5 × 106
1.5 × 107
3.1 x 104
* ND, Not done.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 08 May 2017 23:20:06
Properties of MulFN-ulO
1
2
3
4
5
6
73
7
8
9
IFN-~--
Fig. 5. Fluorograph of an SDS-polyacrylamide gel containing 35S-labelled proteins secreted by COS
cells transfected with MulFN-c¢I~10 hybrid expression plasmids. Lanes 1 and 9, molecular weight
markers (Mr x 10-3); lane 2, control COS supernatant; lane 3, MulFN-cd ; lane 4, MulFN-~¢I~10 (Xa);
lane 5, MulFN-~lcd0 (Xb); lane 6, MulFN-~10; lane 7, MulFN-ccI0~I (Xa), lane 8, MulFN-cO0c¢l
(Xb). Xa and Xb indicate XmnI sites in the corresponding MulFN-c¢ genes used for the construction of
hybrid genes (see also Table 2).
DISCUSSION
In this study we describe the characterization of the protein product of the MulFN-cd0 gene.
Direct comparison of the primary structure of MulFN-cd 0 with that of other MulFN-~ species
(Fig. 3) revealed a considerable degree of homology (91 9/00to 78 ~). MuIFN-~10 differs from all
other MuIFN-~ species evaluated so far by its much lower antiviral activity on mouse cells. The
at least 10-fold difference in antiviral activity between MuIFN-cd0 and MulFN-cd combined
with the high degree of mutual sequence homology provided us with a tool for a more detailed
structure-function analysis. Hybrids were constructed between the two genes using common
restriction enzyme sites and the protein products were analysed. Previously we have used a
similar approach to characterize the regions that are involved in the high antiviral activity of
MulFN-~I on hamster cells and that of MulFN-c~4 on mouse cells (Van Heuvel et al., 1987).
These data indicated a role of N-terminal sequences in high activity of MuIFN-c~ species on
hamster cells and the importance of the C-terminal part of the MuIFN-a4 for its high activity on
mouse cells. Our data here show that a third region (amino acids 68 to 123) is important for the
low antiviral activity of MuIFN-gl0 on mouse cells. In this part of the molecule MuIFN-~I and
MuIFN-~ 10 differ in only five amino acids, at positions 85, 103, 109, 1 l 0 and 116. Of these, the
Val residue at 85 and the Ser residue at 110 are unique to MuIFN-~10 as compared to all other
MuIFN-~ species. All MuIFN-~s possess a Phe and a Pro residue at positions 85 and 110,
respectively. It is conceivable that the aberrant composition of MuIFN-~10 at one of these
highly conserved positions is of importance to its diminished antiviral properties. The absence
of a Pro residue at 110, which can lead to structural changes of the protein, is especially striking.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 08 May 2017 23:20:06
74
J. T R A P M A N A N D O T H E R S
Because of the large differences in antiviral activity between hybrids containing the middle part
of gl0 and the parental al0 protein, these hybrids are not well suited for a more detailed
structure-function analysis of MuIFN-gl0 than that done so far. The low activity of these
hybrids is most probably explained by additional structural changes of the protein and not by a
direct reduction of the affinity of the binding domain(s) for the IFN receptor. Site-directed
mutations of the various positions discussed above are needed to extend our observations and to
confirm our hypothesis of the importance of Pro 110 (and Phe 85) to the antiviral activity of
MuIFN-~ species.
The human (Hu) IFN-~ multigene family is composed of at least 14 different members
(Weissman & Weber, 1986). Comparison of the consensus HuIFN-~ amino acid sequence
(Weissman & Weber, 1986) with that of MulFN-~s shows a mutual homology of around 70%.
The most striking structural differences between HuIFN-~s and MuIFN-~s are Noglycosylation
of MuIFN-~s (N-glycosylation site at positions 78 to 80) and the presence of a Cys residue at
amino acid position 86 in MuIFN-~s, which is absent in HuIFN-gs with the exception of
HuIFN-~tl (D). Cys residues at positions 1, 29, 99 and 139 are found in all MuIFN-ct and
HuIFN-~ species examined. Similarly, Pro residues are seen at positions 4, 26, 39, 110 and 138 in
almost all IFN-~s from both species [MuIFN-~s contain an additional Pro at 55, which is also
present in HuIFN<t2 (A)]. Like MuIFN-~10, HuIFN-al6 lacks Pro 110. Unfortunately, data
about the antiviral properties of HuIFN-ctl6 or hybrids derived from it, which would be
interesting for comparison to MuIFN-~10, are not available.
We are indebted to Mrs M. Hanegraaff for typing the manuscript and to Mrs P. Delfos for photography. This
study was supported by a grant from the D u t c h Cancer Foundation K W F .
REFERENCES
DANDOY, F., KELLEY, K. A., DEMAEYER-GUIGNARD,J., DEMAEYER,E. & PITHA, P. M. (1984). Linkage analysis of the
murine interferon-a locus on chromosome 4. Journal of Experimental Medicine 160, 294-302.
DANDOY, F., DEMAEYER, E., BONHOMME, F., GUENET, J. L. & DEMAEYER-GUIGNARD, J. (1985). Segregation of
restriction fragment length polymorphism in an interspecies cross of laboratory and wild mice indicates a
tight linkage of the murine IFN-fl gene to the murine IFN<t genes. Journal of Virology 56, 216-220.
DAUGHERTY,B., MARTIN-ZANCA,D., KELDER, B., COLLIER,K., SEAMANS,T. C., HOTTA,K. & PESTKA,S. (1984). Isolation
and bacterial expression of a murine alpha leucocyte interferon gene. Journal of Interferon Research 4,
635-643.
KELLEY, K. A. & PITHA, P. M. (1985a). Characterization of a mouse interferon a gene locus. II. Differential
expression of a-interferon genes. Nucleic Acids Research 13, 825-839.
KELLEY, K. A. & PITHA, P. M. (1985b). Characterization of a mouse interferon gene locus. I. Isolation of a cluster of
four interferon genes. Nucleic Acids Research 13, 805-823.
KELLEY, K. A., KOZAK, C. A., DANDOY, F., SOR, F., SKUP, D., WINDASS, J. D., DEMAEYER-GUIGNARD,J., PITHA, P. M. &
DEMAEYER, E. (1983). M a p p i n g of murine interferon-a genes to chromosome 4. Gem 26, 181-188.
KELLEY, K. A., RAY, N. B. K., KELLUM, M. & PITHA, P. M. (1986). Synthesis of fusion and mature murine alpha
interferons in Escheriohia coll. Gene 45, 317-325.
LEROSCOUET,D., VODJDANI, G., LEMAIGZE-DUBREUIL,Y., TOVEY, M. G., LATTA,M. & DOLY, J. (1985). Structure of a
murine alpha interferon pseudogene with a repetitive R-type sequence in the 3"-flanking region. Molecular
and Cellular Biology 5, 1343-1348.
LOVETT, M., COX, D. R., YEE, D., BOLL, w., WEISSMANN,C., EPSTEIN, C. J. & EPSTEIN, L. B. (1984). The chromosomal
location of mouse interferon a genes. EMBO Journal 3, 1643-1646.
MANIATIS, T., FRITSCH, E. F. & SAMBROOK,J. (1982). Molecular Cloning: A Laboratory Manual. New York: Cold
Spring Harbor Laboratory.
MESSING, J., CREA, R. & SEEBURG, P. H. (1981). A system for shotgun D N A sequencing. Nucleic Acids Research 9,
309-321.
SANGER, F., NICKLEN, S. & COULSON,R. (1977). D N A sequencing with chain-terminating inhibitors. Proceedings of
the National Academy of Sciences, U.S.A. 74, 5463-5467.
SEIF, I. & DEMAEYER-GUIGNARD, J. (1986). Structure and expression of a new murine interferon-alpha gene:
MuIFN-a9. Gene 43, 111-121.
SHAW, G. D., BOLL, W., TAIRA, H., MANTEL, N., LENGYEL, P. & WEISSMANN,C. (1983). Structure and expression of
cloned murine IFN-ct genes. Nucleic Acids Research 11, 555-573.
TRAPMAN,J., VAN HEUVEL, M., ZWARTHOFF,E. C., BOSVELD,I. J. & KLAASSEN,P. (1986). Properties of modified mouse
interferon alpha species. Journal of Cellular Biochemistry, supplement 10c, 246.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 08 May 2017 23:20:06
Properties o f MulFN-otlO
75
VAN DER KORPUT, A. G. M., HILKENS, J., KROEZEN, V., ZWARTHOFF, E. C. & TRAPMAN, J. (1985). M o u s e i n t e r f e r o n
alpha and beta genes are linked at the centromere proximal region of chromosome 4. Journal of General
Virology 66, 493-502.
VAN HEUVEL, M., BOSVELD, L J., MOOREN, A. T. A., TRAPMAN, J. & ZWARTHOFF, E. C. (1986). P r o p e r t i e s o f n a t u r a l a n d
hybrid murine alpha interferons. Journal of General Virology 67, 2215-2222.
VAN HEUVEL, M., BOSVELD, I. J., KLAASSEN, P., ZWARTHOFF, E. C. & TRAPMAN, J. (1987). Structure-function a n a l y s i s of
murine interferon-alpha: antiviral properties of novel hybrid interferons. Journal of Interferon Research (in
press).
WEISSMANN,C. & WEBER, H. (1986). The interferon genes. In Progress in Nucleic Acid Research and Molecular
Biology, vol. 33, pp.251-300. Edited by W. E. Cohn & K. Moldave. New York: Academic Press.
ZWARTHOFF, E. C., MOOREN, A. T. A. & TRAPMAN, J. (1985). Organization, structure and expression of murine
interferon alpha genes. Nucleic Acids Research 13, 791-804.
(Received 15 June 1987)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Mon, 08 May 2017 23:20:06