Pathogenicity and transmissibility of the
1918 Spanish influenza pandemic virus
Terrence Tumpey
Influenza Division
Centers for Disease Control and Prevention
Influenza A HA and NA Subtypes in Nature
H1
H2
H3
H4
H5
H6
H7
H8
H9
H10
H11
H12
H13
H14
H15
H16
N1
N2
N3
N4
N5
N6
N7
N8
N9
Timeline of Emergence of
Influenza A Viruses in Humans
Avian
Influenza
Asian
flu
Spanish
Influenza
H2
H1
1918
H9
H5
Hong
Kong
flu
H3
H1
1957 1968 1977
(A/WS/33)
H7
H5
1997
2006
1918 ‘Spanish’ Influenza Pandemic
• Total deaths in 1918-1919 estimated to be 20-50
million
• U.S. Deaths = 550,000-675,000
• Flu deaths in Philadelphia in October 1918 =
10,959.
• U.S. Military deaths to flu = 43,000 (out of
~100,000 U.S. Troop casualties in W.W.I.)
U.S. Life Expectancy 1900-1960
70
65
Age
60
55
50
45
40
35
1900 1906 1912 1918 1924 1930 1936 1942 1948 1954 1960
Date
1918 Spanish Influenza
Iowa State College
Reconstruction of the 1918 influenza virus
. .. .
.
Lung specimens
archived since 1918
Gene sequencing
Gene reconstruction
Drs. Palese,
Garcia-Sastre,
& Basler
Virus rescue
Infectious 1918 virus in
BSL-3 enhanced lab
CDC Atlanta
1918 autopsy cases
• Case 1: 21 y.o., PVT, Ft. Jackson, SC,
died after 6 day course on 26 Sept. 1918
A/South Carolina/1/18
• Case 2: 30 y.o., PVT, Camp Upton, NY,
died after 3 day course on 26 Sept. 1918
A/New York/1/18
1918 lung block
1918 autopsy case 3
Johan Hultin
in 1997 at the same gravesite
Johan Hultin as a young man
in 1951 at the Brevig gravesite
46 years later
Attempt to grow live 1918 virus in 1951
Frozen cadaver lung tissue
Jeffery Taubenberger and Ann Reid
X
FAILED
1918 viral gene
sequencing
Third 1918 Case: Alaska
• Case 3: ~30 y.o. Inuit female from Teller Mission,
Alaska, died in <5 days in Nov. 1918; Exhumation and
lung biopsy in Aug. 1997.
A/Brevig Mission/1/18
Use the 1918 virus as a model for
pandemic influenza
Main Objectives
1. Identify properties that are responsible for the
extraordinary virulence of the 1918 influenza
virus
2. Identify genetic determinants responsible for the
transmissibility of this pandemic virus
The hemagglutinin (HA) and neuraminidase (NA) are
the major viral surface proteins that play an important
role in virulence
NA
HA
PA
PB2
PB1
HA
NP
NA
M
NS
Lipid
bilayer
1918 HA and NA genes enhanced the virulence of a
contemporary H1N1 subtype virus
Mean lung titers
Log10 EID50/ml
8
1918 HA/NA:Tx/91
rescued Texas/36/91 (Tx/91)
Wild-type Tx/91
8/8 dead
6
4
2
0/8 dead
0
0
2
6
Days after infection
Tumpey et al. 2004 PNAS 101:3166
Kabasa et al. 2004 Nature 431:703
8
BALB/c mouse lung pathology at
4 days following infection
A/Texas/36/91 (H1N1)
1918 HA/NA:Tx/91
Reverse-genetics system for generation of
influenza viruses from plasmids
PB1
PB2
Transfection
PB1
PB2
PA
HA
NA
NP
M
NS
293T/MDCK Cells
PA
NP
4 protein-expression plasmids
for viral polymerase and
NP proteins expressed from
pol I vectors.
8 plasmids expressing
viral RNAs expressed
from polI vectors.
Recombinant influenza virus
Fodor E, et. al. (1999) J. Virol. 73: 9679-9682.
Reconstructed 1918 pandemic virus
Negative stain EM of
1918 influenza virus
EM by Cynthia Goldsmith, Infectious Disease Pathology Activity, CDC
Thin section EM of MDCK cell pellets infected with
the 8-gene 1918 influenza virus
EM by Cynthia Goldsmith, Infectious Disease Pathology Activity, CDC
1918 recombinant viruses generated
using reverse genetics
Virus*
Growth in MDCK cells
(PFU/ml)
1918
9.0 X 10
1918:Tx/91 HA (7:1)
3.0 X 10
1918 HA/NA/M/NP/NS:Tx/91 P’s (5:3)
7.0 X 10
Contemporary H1N1 (/Tx/91)
2.3 x 10
7
7
7
7
* The identity of the 1918 and Tx/91 influenza virus genes was confirmed by RT-PCR and
sequence analysis.
1918 hemagglutinin (HA) is
essential for lethality in mice
% Mouse survival
100
80
Tx/91
60
Tx HA:1918 (7:1)
1918 (1)
1918 (5:3) Tx/91
1918 (2)
40
1918 (1)
1918 (2)
20
0
0
2
4
6
8
10
Days after infection
12
14
1918 HA and P genes are essential for maximal
replication in mouse lungs
Mean lung titers
log10 EID50/ml
10
8
*
*
Tx/91
Tx HA:1918
6
1918 5:3 Tx/91
4
1918 (1)
2
1918 (2)
0
Day 4 after infection
H5N1 versus 1918 virus in BALB/c mice
Virus
Subtype
Lung Titers
(EID 50/ml)
(log10)
Tx/36/91
H1N1
3.7
Not lethal
1918
H1N1
7.1
3.5
A/Vietnam/1203/04
H5N1
6.3
2.2
A/Thailand/16/04
H5N1
7.7
1.7
* Expressed as the log10 PFU required to give 1 LD50
LD50*
Influenza transmission
•
Classical experimentation by Andrewes
and Glover (1941) determined that
human influenza virus may transmit
from infected ferret to uninfected ferret.
•
The molecular basis of influenza virus
transmission are not well understood.
• The identification of molecular determinants of influenza
virus transmission may provide a framework for the future
identification of influenza viruses with pandemic potential.
Ferret Model
• Naturally susceptible to
influenza virus infection
• Distribution of sialic acid
receptors in the respiratory
tract is similar to humans
• Exhibit similar symptoms to
influenza virus infection as
humans • Fever
• Lethargy
• Nasal discharge
• Sneezing
Influenza Virus
Transmission in Ferrets
Inoculated
Inoculated
Inoculated
Contact
Inoculated
Inoculated
Contact
Inoculated
Inoculated
Inoculated
Inoculated
Contact
Ferret Model of Respiratory Droplet Transmission
Nasal wash virus titers
Log10 EID50/ml
Inoculated
Contact ferrets
8
6
Human H3N2
4
Log10 EID50/ml
2
8
1
1
5
3
3
7
5
6
Avian H5N1
4
(HK/486/97)
2
1
3
5
1
3
5
Days post inoculation/contact
7
9
Respiratory droplet transmission of
avian H1N1 viruses
Nasal Wash Titers: A/Duck/NY/15024/96 (H1N1)
Virus titer (log10 EID50/mL)
7
Inoculated
Contact ferrets*
6
5
4
3
2
1
1
7
5
3
Days Post-Inoculation
* Influenza sero-neg at day 0 and 18 p.c.
1
7
9
5
3
Days Post-Contact ferrets
11
Respiratory droplet transmission of
avian H1N1 viruses
Nasal Wash Titers: A/duck/Alberta/35/76 (H1N1)
Inoculated
Virus titer (log10 EID50/mL)
8
Contact ferrets*
7
6
5
4
3
2
1
1
3
5
7
9
Days Post-Inoculation
* Influenza sero-neg at day 0 and 18 p.c.
1
3
5
7
9
Days Post-Contact ferrets
11
1918 virus transmission experiment
Six ferrets for respiratory droplet transmission
Inoculated
Dose: 106 PFU of
1918 virus i.n.
Contacts
Untreated
Slide adjacent
cages together
24 hrs later
Monitor disease signs and collect nasal washes
daily from inoculated and contact ferrets
1918 virus inoculated ferrets
1526
1621
1500
1559
Weight change (gms)
1400
1300
1/3 survived
1200
1100
1000
900
800
-2
-1
0
1
2
3
4
5
6
7
8
9
10
Days after infection
11
12
13
14 15
16
Pathogenesis of 1918 virus in ferrets –
1918 virus spread to naïve contacts
Clinical Signs
Inoculated
Contacts
Lethality
66%
33%
Virus in nasal wash
3/3+
3/3+
Max Temp Change (%)
+ 5.1
+ 3.9
Sneezing
Yes
Yes
Respiratory droplet transmission of
human H1N1 viruses
Log10 EID50/ml
Inoculated
Contact ferrets
8
6
4
1918
2
†
1
3
5
7
†
†
9
1
3
5
7
9
11
Log10 EID50/ml
8
6
Texas/36/91
4
2
1
3
5
7
Days Post-Inoculation
1
3
5
7
9
Days Post-Contact
11
Does receptor binding specificity of influenza viruses influence
transmission of H1N1 viruses in mammals?
Distribution of sialic acids/receptor preference
• Human influenza viruses prefer αlpha 2,6 linkages
• Avian influenza viruses prefer αlpha 2,3 linkages
Human
Sia(2-6)Gal
Avian
Sia(2-3)Gal
Influenza virus receptors in the human airway
Upper respiratory tract
α2,6 sialic acid
α2,3 sialic acid
+++
+/-
Lower respiratory tract
α2,6 sialic acid
α2,3 sialic acid
+++
+++
Single amino acid substitutions in the 1918
HA changes the receptor binding specificity
1918 HA
viral strain
1918 SC
1918 NY
avian
77
D
D
D
138
A
A
A
amino acid position
186
190
P
D
P
D
P
E
194
L
L
L
225
D
G
G
Properties of rescued 1918 viruses
Enzymatically modified
chicken red blood cells (CRBCs)
Amino acid
position in HA
Infectivity
Titer
Virus
190
225
(PFU/ml)
SC 18
D
D
4.8 x 107
NY 18
D
G
5.0 x 107
AV 18
E
G
5.0 x 107
E
G
2.2 x 107
Dk/Alb (wt)
Presence or Absence of
Hemagglutination
α2,6
CRBCs
α2,3
CRBCs
Untreat.
CRBCs
+
+
-
+
+
+
+
+
+
+
Two amino acid substitutions in the 1918 HA abolishes
transmissibility of the pandemic virus
Nasal Wash Titers: AV18 virus
Virus titer (log10 EID50/mL)
Inoculated
Contact ferrets*
8
7
6
5
4
3
2
† †
1
1
3
5
7
9
Days post-inoculation
* Influenza sero-neg at day 0 and 18 p.c.
1
3
5
7
9
Days Post-contact
11
Transmissibility of the 1918 NY virus
Virus titer (log10 EID50/mL)
Inoculated
Contact ferrets*
8
7
6
5
4
3
2
†
1
1
3
5
7
9
1
Days Post-Inoculation
* 2/3 seroconversion to NY 18 virus on day 18 p.c.
3
5
7
9
Days Post-Contact ferrets
11
Influenza pathogenesis and transmission
of H1N1 viruses in ferrets - summary
Spread to Contacts
Virus
Binding
preference
Mortality
Virus
Seroreplication conversion
Texas/36/91
α 2,6
Not lethal
Yes
Yes
Duck/Alb/76
α 2,3
Not lethal
No
No
1918 (Human HA)
α 2,6
66%
Yes
Yes
1918 (avian HA)
α 2,3
66%
No
No
33%
1/3
2/3
1918 (NY HA)
α 2,6/α 2,3
Summary
•
The 1918 HA and P genes are essential for maximal virus
replication and optimal virulence.
•
The parental 1918 (SC18) virus and the mutant 1918 viruses are
virulent in ferrets.
•
Two amino acid mutations that cause a switch from the human
α2,6 to the avian α2,3 SA receptor binding preference resulted in
a virus incapable of respiratory droplet transmission between
ferrets, but that maintained its lethality and replication
efficiency in the upper respiratory tract.
•
Poor transmission of a 1918 virus with dual α2,6/α2,3 specificity
suggests that a predominant α2,6 SA binding preference is
essential for optimal transmission of this pandemic virus.
Acknowledgements
Centers for Disease
Control and Prevention
Influenza Division/IVPB
Taronna Maines
Neal van Hoeven
Claudia Pappas
Cynthia Goldsmith
Mount Sinai School of
Medicine
University of Washington
School of Medicine
Armed Forces
Institute of Pathology
USDA/Southeast Poultry
Research Laboratory
The Scripps Research
Institute
NIH Grants; 5R01 AI0506919-02 and
AI058113-01