THE REGULAR MOSQUITO IRIDESCENT VIRUS (RMIV) IN THE

BLACK SALTMARSH MOSQUITO, AEDES TAENIORHYNCHUS

(WIEDEMANN) : PRODUCTION, PURIFICATION,

TRANSOVARIAL TRANSMISSION, SITE OF ENTRY,

DEVELOPMENT OF INFECTIONS, AND

POLYPEPTIDE COMPOSITION

BY
STEPHEN CARSON HEMBREE

A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT? OF THE
REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA
1974

Dedicated to Jane, my wife

ACKNOWLEDGMENTS

I wish to express gratitude to ray supervisory com-
mittee: Dr. F. S. B Ian ton, Dr. R. E. Lowe, Dr. S. G. Zam,
Dr. J. L. Nation, and Dr. F. W. Zettler.

Appreciation is also expressed to Dr. D. L. Wiedhaas,
Director, and to the staff of the insects Affecting Man
and Animals Research Laboratory, U. S. Department of Agri-
culture, for their support of this project. I wish
especially to thank Mr. D. W. Anthony, for assistance with
electron microscopy, and Mrs. A. L. Cameron, Mr. H. R.
Ford, and Mr. E. Green for assistance in many ways.

111

:■:. i

TABLE OF CONTENTS -

PcK

ACKNOWLEDGMENTS ■ i

LIST OF TABLES vi;

LIST OF FIGURES

ABSTRACT

INTRODUCTION

METHODS AND MATERIALS 21

Production of Virus 21

Routine Virus production procedure ..... 21
Determination of Optimum Quantity

of RMIV Inoculum 25

Determination of the Optimum Larval

Age for Exposure to RMIV 26

Purification of RMIV 27

Quantification of RMIV Production 31

Attenuation of Infect ivity of RMIV

During Purification 32

Transmission of RMIV 33

Per Cent of Larvae Infected with

RMIV by Transovarial Transmission 33

Transovarial Transmission Through

Isolated Females 34

Sex Specificity of Transovarial

Transmission of RMIV 3 6

Transmission of RMIV by

Integument Puncture 3 7

TABLE OF CONTENTS (Continued)

Page

Production of Antiserum to RMIV and

to A. taeniorhynchus Antigens 38

Purification of Gamma Globulin 40

Conjugation of Anti-RMIV Antibody with

Fluoresceine Isothiocyanate (FITC) 41

Pilot project on Autoradiography of
3H-Methyl Thymidine Uptake and Dis-
tribution in Larval A. taeniorhynchus 46

Production of 3H-Methyl Thymidine

Labeled RMIV 51

Site of Entry of RMIV into Host Tissue 52

Light Microscope Studies 52

Autoradiography 52

Fluorescent antibody 53

Electron Microscope Studies 53

Development of RMIV Infections 55

Autoradiography 55

Fluorescent Antibody 56

Overt infections 56

Covert infections 57

Number and Molecular Weights of RMIV

Structural Polypeptides 58

Isolation of RMIV Structural Polypeptides

by Hydroxy lapatite Chromatography 60

RESULTS AND DISCUSSION 62

Production of RMIV 62

Determination of Optimum Quantity

of RMIV Inoculum 62

TABLE OF CONTENTS (Continued)

Page

Determination of Optimum Larval

Age for Exposure to RMIV 65

Quantification of RMIV Production 65

Attenuation of Infectivity of

RMIV During Purification 68

Transmission of RMIV 70

Per Cent of Larvae Infected with

RMIV by Transovarial Transmission 70

Transmission of RMIV Through

Isolated Females 72

Sex Specificity of Transovarial

Transmission 73

Total of Overt and Covert

Transmission 75

Transmission of RMIV by Integument

Puncture 73

Pilot Project on Autoradiography of

^H-Methyl Thymidine Uptake and

Distribution in Larval A. taeniorhvnchus .... 79

Site of Entry of RMIV Into Host Tissues 82

Light Microscope Studies 82

Autoradiography 82

Fluorescent antibody 86

Electron Microscope Studies 86

Development of RMIV Infections 97

Autoradiography 97

Fluorescent Antibody 100

Overt infections 100

Covert infections 107

TABLE OF CONTENTS (Continued)

Page

Number and Molecular Weights of

RMIV Structural polypeptides 110

Separation of RMIV Structural

Polypeptides by Hydroxylapatite

Chromatography 114

SUMMARY 115

BIBLIOGRAPHY H9

BIOGRAPHICAL SKETCH 12 5

LIST OF TABLES

Table page

1 Net Virus Production Resulting from

24 Hr Exposure of Day Old A. taeniorhynchus

Larvae to RMIV 63

2 RMIV Production Resulting from 24 Hr
Exposure of A. taeniorhynchus Larvae
of Different Ages to 50 LEQ of

Inoculum 65

3 Attenuation of Infectivity Resulting from
Purification of RMIV Through the Stage

Indicated 69

4 Transovarial Transmission of RMIV by
Reciprocal Mating of Exposed and Unex-
posed Male and Female A. taeniorhynchus ... 74

5 Total of Overt and Covert Transmission
of RMIV to Larvae Exposed at Different

Ages 77

6 Proteins Used as Markers for the Estimation
of the Molecular Weights of RMIV Struc-
tural Polypeptides ;Q1

7 Relative Migration Fractions and Molecular

Weights of RMIV Structural Polypeptides . . . 112

LIST OF FIGURES

Figure Page

1 Per Cent Transmission of RMIV in

A. taeniorhvnchus Larvae Exposed to

Increasing Amounts of Inoculum 54

2 per Cent of RMIV Infection Among Progeny
of Mosquitoes Exposed to RMIV as Larvae

of Different Ages -^

Malpighian Tubule Cell with Labeled
Nucleus

SO

Anterior Midgut Region of Larva Showing
Residual Deposits of Tritiated Thymidine
Peripheral to Caudal Extremity of Foregut
Invagination into Midgut 23

RMIV Infected Fatbody Cell in A.

taeniorhynchus Larva with Graination

Over Cytoplasmic Deposits of Tritiated

Thymidine Labeled Viral DNA 85

Midgut Epithelium Cell Containing

Scattered RMIV Particles 88

Midgut Epithelium Cell with Disarranged

Microvilli and Containing Scattered RMIV

Particles 89

RMIV Particles pressed Against Peritrophic

Membrane in Midgut of A. taeniorhynchus ... 90

RMIV Particles pressed Against Peritrophic

Membrane in Midgut of A. taeniorhynchus_ ... 91

LIST OF FIGURES (Continued)

Figure page

10 RMIV particle Between peri trophic

Membrane and Midgut Epithelium 92

11 Virus Like Particle Apparently Within

a Microvillus 93

12 Small Pocket of RMIV in Non-Epithelial
Cell in Midgut Wall of A. taeniorhynchus

Larva 95

13 Methyl Green Staining of RMIV Infected

Cells 99

14 Fluorescent Antibody Staining of Epidermal
Cells, Imaginal Bud Tissues and Fatbody

Tissue Infected with RMIV 102

15 Fluorescent Antibody Staining of RMIV

Infected Tracheal Epithelium Cell 103

16 Fluorescent Antibody Staining of Fatbody
Surrounding Testis 105

Abstract of Dissertation Presented to the Graduate Council

of the University of Florida in Partial Fulfillment of the

Requirements for the Degree of Doctor of Philosophy

THE REGULAR MOSQUITO IRIDESCENT VIRUS (RMIV) IN THE

BLACK SALTMARSH MOSQUITO, AEDES TAENIORHYNCHUS

(WIEDEMANN) : PRODUCTION, PURIFICATION,

TRANSOVARIAL TRANSMISSION, SITE OF ENTRY,

DEVELOPMENT OF INFECTIONS, AND

POLYPEPTIDE COMPOSITION

By

Stephen Carson Hembree

March, 1974

Chairman: Franklin S. Blanton

Major Department: Entomology and Nematology

A system permitting the weekly production by 1 person
of gram quantities of the regular mosquito iridescent virus
(RMIV) in the black saltmarsh mosquito, Aedes taeniorhynchus
(Wiedemann) , was described. The efficiency of the system re-
sulted from exposure of larvae at their most susceptible age
to an optimum quantity of RMIV inoculum and from the use of
a virus purification procedure designed to minimize waste.
Transmission rates of about 10 per cent were routinely
achieved. An average of 58 ug of virus was produced per in-
fected larva, constituting 9.35 per cent of the dry weight of
the average infected larva. Attenuation of infectivity oc-
curred throughout the virus purification procedure, no
major part of it being attributable to a single cause.

Transovarial transmission of the virus increased with
the age of larvae at. exposure through day 5 of larval
life, but decreased when exposure was made just before
pupation. All of the larvae produced by 9 isolated,
transovarially transmitting females were infected. Trans-
mitting females produced fewer average progeny than non-
transmitting females. Males did not transmit virus to
previously unexposed females. Total transmission (i.e.,
transmission resulting in overt disease plus transmission
indicated only by infections among progeny) decreased with
the age of larvae at exposure and was greatest for larvae
in the second instar when exposed.

A probable site of entry of RMIV into host tissue was
found by eiectronmicroscopy in midgut epithelium at the
level of the foregut invagination. Infections were found
in fatbody, epidermis, imaginal buds and tracheal epithelium
by autoradiography of 3H-methyl thymidine treated larvae
84 hr after the initiation of exposure to the virus. In-
fections were found in fatbody by fluorescent antibody
staining 48 hr after initiation of exposure. Infection had
appeared in epidermis and in imaginal buds by 72 hr after
initiation of exposure, and in tracheal epithelium by 96
hr after initiation of exposure. Infections resulting in

transovarial transmission could not be detected by fluorescent
antibody staining.

Twenty-five polypeptides were found in RMIV by poly-
acrylamide gel disc electrophoresis of sodium dodecyl
sulfate — 2-mercaptoethanol disrupted virus. The molecular
weights of these were estimated by polyacrylamide gel disc
electrophoresis and ranged from 23,500 to 331,000. Fourteen
per cent of the genome of the virus could code for its
structural polypeptides, assuming no redundancy and that
no 5S ribosomal RNA or transfer RNA were coded for.

INTRODUCTION

The relationships between viruses and arthropods are
of great interest and importance to man. Viruses are
etiologic agents of many diseases of man, of his domestic
animals, and of his crops. Arthropods are vectors for many
of these and serve also as reservoirs for some diseases
of both man and animals (Chamberlain, 1968) and of plants
(Maramorosch, 1968) . This necessitates an especially inti-
mate relationship between the disease agents and their vectors
Arthropods themselves are susceptible to other viral diseases
that are not known to infect either plants or other animals.
Most of these have been found affecting insects (Smith,
1967) . These diseases are of particular interest in consi-
deration of current efforts to find non-chemical means
of reducing some arthropod populations. Although the
pathogenicity of the viral diseases of arthropods has long
been recognized, the realization that many arboviruses and
arthropod-transmitted plant viruses are pathogenic to
their vectors is relatively recent (Chamberlain, 1968;
Maramorosch, 1968) . The relationships between viruses and

arthropods are now known to form a symbiotic spectrum
ranging from mutual benefit at one extreme to population
decimating lethal parasitism at the othe~ (Maramorosch
and Jensen, 1963) .

The viruses of insects can be divided into two groups
on the basis of whether the virions are enveloped in
protein crystals (occluded) or whether they are not (non-
occluded) (Smith, 1967, p. 8) . Of the non-occluded insect
viruses, the iridescent viruses have attracted the atten-
tion of virologists for several reasons. They are the
largest symmetrical viruses known (Smith, 1967, p. 84);
they are produced in great relative quantity, composing up
to 25 per cent of the dry weight of their host (Williams
and Smith, 1957) ; they are probably the easiest of all
viruses to isolate and purify (Smith, 1967, p. 78) ; and
they are DNA containing viruses which replicate in the
cytoplasm of host cells (Wildy, 1971) .

The International Committee on Nomenclature of Viruses
(ICNV) established at the International Congress for
Microbiology in 1966 placed the iridescent viruses in the
genus Iridovirus (Wildy, 1971) . The members of this genus
were defined as large (130 - 180 nm) , non-occluded,
icosahedral, DNA containing viruses which replicate in

the cytoplasm of the host cell. The Tipula irridescent virus
(TIV) , isolated from the leather jacket , Tipula paludosa
(Meigen) (Diptera) , was designated as the type species.
Other members of the genus were Chilo iridescent virus
(CIV), from the rice stem borer, Chile suppressalis Walker
(Lepidoptera) , and Ser ices this iridescent virus (SIV) ,
isolated from the pruinose scarab, Sericesthis pruinosa
(Dalman) (Coleoptera) . The literature of these three viruses
recently has been well reviewed by Smith (1967) and Bellett
(1968) . The mosquito iridescent virus (MIV) , first described
from the black saltmarsh mosquito, Aedes taenlorhynchus
(Wiedemann), by Clark et al. (1965), was listed as a probable
member of the genus. The uncertainty shown at that time
was due to a lack of information of the physical and chemi-
cal characteristics of this virus which subsequently
became available (Matta, 1970; Matta and Lowe, 1970) . At
the present time, no species names have been approved for
iridoviruses by the ICNV, and an informal nomenclature
that identifies the viruses by use of the name of the host
has gained acceptance in the current literature and has
been used in this dissertation.

The most conspicuous feature of the iridoviruses that
have been recognized by the ICNV is their characteristic

iridescence when viewed by reflected light. Visible color
is evident both in heavily infected host tissue and when
the purified virus is pelleted by ultracentr if ugation .
This iridescence has been explained as Bragg reflection of
visible light from randomly oriented microcrystals of the
virus, the formation of which is made possible by the ex-
treme regularity of size and shape of the virions (Williams
and Smith, 1957; Smith and Williams, 1958; Klug et al . ,
1959) . This feature of the genus Iridovirus is shared with
a number of other insect viruses which could not be included
in the genus because too little was known of their charac-
teristics. The extent to which the phenomenon of iridescence
is correlated with the physical, biological and biochemical
characteristics which define the genus Iridovirus is un-
certain. It is conceivable that other viruses might be
found that iridesce but differ from the iridoviruses as
they have been defined. Several icosahedral cytoplasmic
deoxyriboviruses are known which do not form into the
paracrystalline arrays necessary for iridescence (Stoltz,
1971). Nevertheless, the iridescent viruses of insects
are generally considered to be closely related (Bellett,
1968; Bellett and Inman, 1967; Cunningham and Tinsley,
1968; Glitz et al . , 1968; Wildy, 1971). Tinsley and Kelly

(1970) suggested an interim nomenclature system for this

group of viruses. Their published list of 12 viruses

should now contain two additional ones recently reported

from a chaoborid, Corethrella brake leyi , in Louisiana

(Chapman et al. , 1971) and from a grass grub, Costelytra

zealandica, in New Zealand (Kalmakoff et al., 1972).

Iridescent viruses have been reported infecting several

species of Aedes mosquitoes and one of Psorophora from widely

separated localities:

Aedes taeniorhynchus (Wiedemann)

Florida (Clark et al . , 1965; Hall and Lowe, 1971)
Louisiana (Chapman et al . , 19G6)

Aedes annulipes (Meigen)

Czechoslovakia (Weiser, 1965)

Aedes cantans (Meigen)

Great Britain (Tinsley et al . , 1971)
Czechoslovakia (Weiser, 1965)

Aedes fulvus pallens (Ross)

Louisiana (Chapman et al . , 1966)

Aedes vexans (Meigen)

Louisiana (Chapman et al . , 1966)

Aedes dorsalis (Meigen)

Nevada (Chapman et al . , 1966)

Aedes detritus (Haliday)

Great Britain (Service, 1968)
Tunisia (Vago et al. , 1969)
France (Hasan et al. , 1970)

Aedes stirr.ulans (Walker)

Connecticut (Anderson, 1970)

Psorophora ferox ( Hurab o 1 t )

Louisiana (Chapman et al . , 196^)

The relationship of these isolates is not completely
known. Cross infectivity studies were conducted among
the known hosts of mosquito iridescent viruses in Louisiana
(Woodard and Chapman, 1963) , and in all tests the isolates
were infective for species other than their own host.
However, there was consistently less than 1 per cent in-
fectivity, compared to 15 per cent to 18 per cent infectivity
in the original hosts. These data suggest the viruses from
different species were not identical. Aedes sollicitans
(Walker) is frequently found in natural breeding sites with
A. taeniorhynchus, but has never been found infected with
RMIV in nature, even in the presence of infected A. taenior-
hynchus . However, it can be infected artificially in the
laboratory with about the same infectivity rate as the
natural host. The iridescent virus from A. stimulans is
apparently distinct from other MIV, since it is considerably
smaller (135 nm vs. 180 nm) and is lethal to the larvae in
the late third rather than the fourth stadium (Anderson,
1970) . Isolates of MIV from A. detritus, collected in
Tunisia and Southern France, were compared morphologically

and serologically and were considered to be identical
(Hasan, eb al. , 1970). The iridescent virus from A. cantans
in Great Britain showed a relationship to that from
A. taeniorhynchus by the tube precipitin test but showed
no relationship to TIV, SIV, CIV, or to the iridescent
virus from Wiseana cervinata (Lepidoptera) in New Zealand
(Tinsley et al. , 1971). in addition, the iridescent virus
from A. taeniorhynchus was unrelated to TIV or SIV when
tested by complement fixation (Cunningham and Tinsley,
1968) .

Of the iridescent viruses reported from mosquitoes,
only the one from A. taeniorhynchus in Florida (Clark et al . ,
1965) has been well characterized. Two strains of this
virus were reported from Louisiana. One exhibited a yellow
to green, and sometimes pink, iridescence and the other a
blue iridescence (Woodard and Chapman, 1968) . The latter
has been collected only from Louisiana, and it has been
suggested that this isolate is not as prevalent as the
former (Hall and Lowe, 1971) . The blue isolate was
designated the turquoise mosquito iridescent virus (TMIV) ,
and the orange isolate, the regular mosquito iridescent
virus (RMIV) (Matta and Lowe, 1970) . The infectivity of
the 2 isolates was shown to be about the same (Woodard

8

and Chapman, 1968) . Slightly more DNA was present in
RMIV than in TMIV (Faust et al. , 1968; V7agner et al. ,
1973) and neither contained RNA (Faust et al . , 1968) .
RMIV and TMIV have been further compared physically,
chemically, and serologically (Lowe et al. , 1970, Hall and
Lowe, 1972) . Measurement of the virions of both types,
in tissue sections and after negative staining of purified
virus, indicated that RMIV was larger than TMIV. Also,
RMIV had a greater sedimentation rate in sucrose density
gradients and greater density in csci gradients. No
antigenic differences could be detected between the strains
at the level of sensitivity of agar gel diffusion tests.
Both strains infected the same host tissues and caused
similar cytopa thology. Tubular structures of unknown origin,
slightly smaller in diameter than the virus they accom-
panied, were found in tissues infected with both strains
(Lowe et al . , 1970; Hall and Anthony, 1971). A recent
study attributed several biophysical differences between
the strains to their relative sizes (Wagner et al. , 1973).
Additionally, their electrophoret ic mobilities were dif-
ferent, but their isoelectric points were similar. Dif-
ferences were shown in the percentages of protein, DNA,
and lipid, but structural differences other than size were
not detected.

Several physicochemical characteristics of RMIV have
been reported by Matt a (1970) . Spectrophotometry analysis
showed that the ratios of ultraviolet absorption to the
concentration of virus suspension were constant at a wave
length of 700 nm, and an extinction coefficient (E^qq)
of 10.8 was determined to permit easy quantification of
purified virus suspensions. The sedimentation coefficient
(S20'w) was 4458/ and the density was calculated to be
1.354 g/cm . The particle weight, calculated from the
diameter (180 nm) and density of the particle, was 4.129
x 10 g or 2.486 x 10 daltons. The DNA content of
15.97 _ 0.2 9%, determined by diphenylamine with a known
quantity of virus, was larger than the 11.7 per cent re-
ported previously by Faust et al. (1968) and accounted for
a DNA molecular weight of 397 x 106 daltons. Wagner et al .
(1973) reported a content of 16.9 ± 0.1% DNA, and, using
this figure and the particle weight, they calculated the
molecular weight of the DNA to be 464 x 106 daltons. Both
of these figures for the molecular weight of DNA are more
than twice as large as those of any other known non-
iridescent virus (Wagner et al. , 1973). The amino acid
composition was found to be very similar to that of SIV
and TIV. Wagner et al. (1973) also found lipid in RMIV,

10

but the possible structural role of this lipid, like that
found in TIV (Glitz et al . , 1968), remains unresolved.
Some of these data were used to differentiate RMIV from
other iridescent viruses (Lowe et al. , 1970).

The structure of RMIV has been studied by high resolu-
tion electronmicroscopy as one representative of icosahedral
cytoplasmic deoxyriboviruses by Stoltz (1971). He reported
that the electron dense nucleoid was surrounded by a
trilaminar membrane, morphologically resembling a unit
membrane, that closely followed the contour of the nucleoid
and was thought to be a part of it.

Surrounding the nucleoid and the inner membrane was an
outer membrane, closely appressed to the latter in intact
virions. The contact line of the 2 membranes was electron
dense and twice as thick as a single membrane lamella.
In sectioned material there appeared to be a layer of
morphological subunits on the surface of the outer membrane,
but they could not be resolved by negative staining of
intact virions. The rigidity of these external structural
subunits was thought to be responsible for the icosahedral
shape of the virus. The outei membrane and morphological
subunits thereon were defined as the shell of the virus,
because they tended to maintain their characteristic angular

11

configuration in disrupted particles. The inner membrane
appeared flexible and could maintain the contour of either
the nucleoid or the shell. The inner membrane remained
attached to the nucleoid in disrupted particles more fre-
quently than to the shell. In a note appended to his paper,
Stoltz (1971) proclaimed that morphological subunits had
been detected by negative staining on the surface of
collapsed virions and that these were organized into
triangular shaped "trisymmetrons " (terminology of Wrigley,
1969, 1970) which were hexagonally arranged. Thus, the
structure of the shell of RMIV resembled that of SIV
(Wrigley, 1969) and TIV (Wrigley, 1970; Stoltz, 1971) .

Clark et_al. (1965) demonstrated that per os trans-
mission of MIV was possible, and they also reported that
the infectivity of the virus apparently was destroyed by
either drying or putrefaction. Woodard and Chapman (1968)
routinely transmitted the virus in the laboratory by
per os exposures for up to 68 serial passages with a mean
transmission rate of 16 per cent. Their studies also
indicated that various exposure methods influenced the
success of transmission of the virus. Larvae exposed to
RMIV for 24 hr during the first through the third, but not
the fourth, stadium demonstrated infection before pupation.
Optimal transmission appeared to occur when the larvae

12

were exposed during the second stadium. Transmission
increased with the duration of exposure up to 48 hr , but
additional exposure caused only a slight increase in per
cent transmission. When late first and early second
instar larvae were exposed for 24 hr to various numbers
of macerated, infected larvae, there was a sharp increase
in the transmission rate as the quantity of inoculum in-
creased but also a sharp increase in larval mortality.
However, transmission was quantitated on the number of
surviving larvae rather than the number of exposed larvae,
and the figures given may be deceptive. Also, the net
number of infected larvae produced (total number produced
minus number of infected larvae used in the inoculum) from
the test with the most inoculum was only about 1/3 of
that produced from a group exposed to 1/10 of the same
amount of inoculum.

Linley and Nielsen (1968a) conducted a series of
carefully described experiments on the transmission of
RMIV in the laboratory. The percentage of transmission
in these experiments was determined on the basis of the
number of larvae exposed, and exposure was not done in
the condition of extreme crowding that resulted in high
larval mortality for Woodard and Chapman (1968) . However,

13

even with the modifications, their results generally-
substantiated those of Woodard and Chapman (1968).

Matta and Lowe (1970) reported 2 separate infectivity
tests with known numbers of first stadium A. taeniorhynchus
larvae exposed to known amounts of inoculum and in which
the number of survivors was determined. Transmission rates
were 7.73 per cent and 15.6 per cent when calculated on
the basis of number of survivors, but only 2.67 per cent
and 5.95 per cent, respectively, when calculated on the
basis of number of larvae exposed, and larval mortality
was greater than 60 per cent in both tests. The increase
in virus produced (number of infected larvae produced,
minus number of infected larvae used as inoculum, divided
by number of infected larvae used as inoculum) was 167.5
per cent and 197.5 per cent in the respective experiments.
It was stated that the infectivity of the virus was not
significantly reduced by treatment with ether, but it
was not stated whether the test was performed as pre-
viously described by Andrewes and Horstmann (1949) .

Experiments relative to the transmission of RMIV in
nature were described by Linley and Nielsen (1968b) .
These experiments demonstrated that infection could be
acquired per os by healthy larvae exposed to virus from

14

either macerated or unmacerated larvae -under field con-
ditions. Virus from macerated hosts retained its in-
fectivity in 10 per cent sea water for 3 days in the
laboratory, and then slowly attenuated over a period of
30 days. There was less retention of infectivity in
artesian water, and the virus was uninfective after 2
days on damp sod.

Transovarial transmission of RMIV was first observed
and substantiated by Woodard and Chapman (1968) . Infected
larvae (never more than 2 per cent) were always found among
larvae reared from the eggs of survivors of experiments on
the infectivity of RMIV. Sterilization of all containers,
instruments and media did not reduce the level of trans-
mission. When late third or early fourth instar larvae
were exposed to virus, they showed no gross symptoms of
infection before or after becoming adults. However, in 3
experiments the larvae developing from eggs deposited by
these adults showed an average infection of almost 30 per
cent. Linley and Nielsen (1968a) excluded the possibility
of transovum transmission by surface sterilization of
the eggs. Additional support for a transovarial mechanism
was provided when Hall and Anthony (1971) found the virus
in both larval and adult ovaries.

15

All the progeny of 5 isolated, transmitting females
observed by Linley and Nielsen (1968a) and 3 observed by
Hall and Anthony (1971) were infected, suggesting an
all or none mechanism of transovarial transmission. The
5 transovarially transmitting females produced an average
of only 7.4 larvae while non-transmitting females of the
same generation produced an average of 26.7 larvae (Linley
and Nielsen, 1968a) . This indicates the proportion of
transovarially transmitting females in a population is not
necessarily equal to the proportion of infected larvae
produced by that population. Evidence from 19 separate
tests indicated that virus infections were not carried
covertly from the second to the third generation (Linley
and Nielsen, 1968a) .

Both Woodard and Chapman (1968) and Linley and
Nielsen (1968a) suggested that RMIV could be transmitted
to only a susceptible proportion of an exposed population.
They made no attempt to test that hypothesis nor did they
propose an explanation of why only a proportion of a
population would be susceptible. Woodard and Chapman
(1968) believed that the susceptible proportion of a
population remained fairly constant regardless of when the
larvae were exposed. This hypothesis was based on the

16

untenable assumption that the proportion of covertly in-
fected adults in a population was equal to the proportion
of infected larvae they produced.

Linley and Nielsen (1968b) proposed a tentative
account of the natural transmission of RMIV in A.
taeniorhynchus . They stated,

Transovarial transmission produces in-
fected larvae which die in the fourth
instar. These then provide possibly the
only, but certainly by far the most im-
portant, source of new infection, which
is acquired per os by the healthy larvae
when they feed on the diseased cadavers
prior to pupation. This in turn leads
to the presence of infected adults which
complete the cycle by depositing infected
eggs (p. 24) .

Few observations have been made on the gross pathology
accompanying the development of infections with RMIV.
Chapman et al . (1960) observed that iridescence appeared
first in the thorax of late third or early fourth instar
larvae and then spread throughout the abdomen during the
fourth larval stadium. Most of the patently infected
larvae died before pupation. Matta and Lowe (1970) main-
tained that iridescence first appeared in the lateral
portions of one of the first four abdominal segments.
Linley and Nielsen (1968a) noted that larvae infected
transovarially appeared to move, feed and develop normally

17

until shortly before death, when they became sluggish and
ceased to feed. After death, the abdomen often contracted,
and some of the larvae remained suspended from the surface
of the water and were often cannibalized.

Clark et al . (1965) first studied the histopathology
of RMIV infections and stated that the virus appeared to
develop in cell cytoplasm of adipose tissue. Matta and
Lowe (1970), using darkfield microscopy and a staining
technique previously described by Matta and Lowe (1969),
observed the virus in fatbody and imaginal discs and noted
that the destruction of these tissues was usually complete
in patently infected larvae. In rare instances larvae sur-
vived, and in the pupae the fatbody had not been completely
destroyed. Using the electron microscope, Hall and
Anthony (1971) found the virus in tracheal epithelium and
epidermal tissues, in addition to fatbody and imaginal
discs, and to a lesser extent in hemocytes, foregut, nerve,
muscle, and in both larval and adult ovaries. Disruption
of metabolic processes in the larval fatbody was thought
to be the cause of death. Virus was never found in midgut
or hindgut epithelium or in malpighian tubules.

In attempts to locate the site of entry of RMIV into
host tissue Hall and Anthony (1971) found virus particles

in the esophageal region of the foregut epithelium after
larvae were exposed to suspensions of the virus. it is
unlikely that the virus passed through the cuticular lining
in this region of the gut, and it was not stated how long
the larvae had been exposed to the virus suspension.
Stoltz and Summers (1971) presented the hypothesis that
virus entered the epithelium in the extreme anterior region
of the midgut. This hypothesis was based, in part, on tne
observation that intact virus particles were not found more
than 100 p. distal to the foregut invagination into the
midgut. The peritrophic membrane has been considered an
effective barrier to intact virions, but both Hall and
Anthony (1971) and Stoltz and Summers (1971) expressed the
possibility that subviral components of nucleic acid or.
nucleoprotein might pass through the membrane.

The present dissertation further elucidates the re-
lationships between RMIV and A. taeniorhynchus . Techniques
for mass production and purification of large amounts of
RMIV in minimal amounts of time were essential to several
of these efforts. Tinsley and Harrap (1972) stated at
the Second International Congress for Virology that one of
the main difficulties in research with MIV was its low
virulence and the resulting lack of sufficient quantities

19

of the virus obtained from laboratory hosts. Therefore,
priority was placed on the development of methods to
produce large (gram) quantities of purified RMIV. Stan-
dardized management of the host was regarded as essential
to the mass production of the virus. Several host manage-
ment methods involving different feeding regimens and the
use of different rearing containers were attempted. Only
the most successful of these has been described. Experi-
ments were conducted to determine the optimum quantity of
inoculum necessary and the optimum age at which to expose
larvae to assure maximum net virus production. The produc-
tion of RMIV and the effect of infection with the virus on
the weight of the host were quantitated. The attenuation
of infectivity of RMIV resulting from purification was
studied to determine if it could be attributed predominantly
to any particular one of the procedures in purification.
Several experiments were conducted to elucidate the quanti-
tative aspects of vertical transmission within a host popu-
lation and to determine if it was possible for RMIV to be
transmitted by males to previously unexposed females.
Attempts were made to transmit RMIV by integument puncture,
as described by Clark (1966) , to determine if resistance
to entry of virus into host tissues was responsible for

20

failures to achieve high rates of transmission by per os
exposures. The site of entry of RMIV into host tissue
was sought by fluorescent antibody and autoradiographic
observations with the light microscope and by examination
of ultrathin sections of exposed first instar larvae with
the electron microscope. The development of overt in-
fections with RMIV in larvae exposed in the second larval
stadium was studied by fluorescent antibody staining and
by autoradiography. The development of covert infections
with RMIV in larvae, pupae and adults, exposed as late
instar larvae, was studied by fluorescent antibody staining,
The number of structural polypeptides composing RMIV was
determined and their molecular weights estimated by
polyacrylamide gel disc electrophoresis. Attempts were
made to isolate these polypeptides by hydroxylapat ite
chromatography.

METHODS AND MATERIALS

Production of Virus

Routine Virus production
procedures

The initial source of RMIV used in these studies
was live, infected, fourth stadium larvae of A. taenior-
hynchus mosquitoes collected from a small saltmarsh on
the southwest side of Atsena otie Key, an island near
the town of Cedar Key, Florida. All mosquito eggs used
to produce larvae for routine virus production were acquired
from the mosquito colony of the U. S. Department of Agri-
culture Insects Affecting Man and Animals Research Labora-
tory, at Gainesville, Florida. The colony procedures used
for mosquito production are being described for publica-
tion (Ford and Green, in prep.) .

Eggs were obtained from the colony on damp spnagnum moss
in pans covered with aluminum foil. They could be stored
at room temperature for up to 3 months without unacceptable
loss of vigor. Periodic examinations of larvae produced
in the colony provided assurance that there were no latent

21

22

vxrus infections present. Tap water was used to rinse
the eggs from the moss into a round enamel pan, using
a screen to collect the moss and debris. The water was
swirled rapidly to accumulate the eggs in the center of
the pan, and they were then pipetted into graduated conical
bottom centrifuge tubes, 1 cm3 per tube. Each tube of
eggs was poured into an enamel hatching pan containing 2
liters of 0.05 M Nad made with distilled water. The eggs
were again accumulated by swirling, and an infusion of
200 mg of live brewer's yeast was added to each pan. The
eggs were allowed to hatch for 1 hr after which the un-
hitched eggs were removed. One cm of eggs provided about
20,000 larvae in each pan. These were allowed to develop
for 24 hr in the yeast infusion.

The progression of mosquito life was measured by days,
or 24 hr periods, beginning at hatching, regardless of the
time of day this occurred. No effort was made to coordinate
any treatment with moulting or ecdysis, but it is pertinent
that passage through each of the first 3 stadia required
about 24 hr v/ith adequate food and space (Nayar, 1967).
All water temperatures were maintained at 27°C throughout
rearing.

Twenty-f our-hr-old larvae were poured onto an organdy

23

cloth screen, washed with tap water, and placed in a small
volume of 0.05 M saline. Eight oz waxed-paper cups were
used as containers to expose the larvae to virus. Twenty-
five hundred larvae were counted into 1 cup containing
200 ml of 0.05 M saline, and this cup was then used as a
visual standard to pipette an estimated 2500 larvae into
each exposure container to be used. Inoculum was prepared
by triturating fresh, RMIV infected fourth stadium larvae
in a Ten Broeck type tissue grinder until their head cap-
sules were completely destroyed. It was quantitated as
larval equivalents (LEQ) , 1 LEQ being the virus from 1
patently infected larva, whether freshly triturated or at
some stage in virus purification. The inoculum was filtered
through an organdy screen before being pipetted into the
exposure containers. Routine exposures were for 24 hr
without additional food.

Rearing containers, prepared several hours before
needed, were white plastic trays,1 56.7 x 44.7 x 8.0 cm,
containing 7.5 liters of 0.05 M saline made with table

■"■panel Control Corp., Detroit, Michigan.

24

salt and tap water. After exposure, the larvae and
inocula were poured into the rearing trays. One gram
of high protein, low fat content hog food supplement1
(ground fine enough to pass through a screen of 50 meshes
per inch) was added as a thoroughly wetted slurry to
each tray. The larvae were fed an additional 2 grams
of food 48 and 72 hr after being placed in the rearing
trays. With this schedule, no aeration was necessary,
and no fouling of the water occurred. Occasionally, a
few pupae appeared late in the fifth day of larval life,
but pupation usually began late in the sixth day after the
eggs hatched.

The collection of infected larvae began when the first
pupae were observed in the trays. The contents of the
trays were poured through a screen of 30 meshes per inch
to retain the larvae. The trays were washed immediately
with cold tap water and a scrub cloth to remove all old
infusion and detritus. Soap was never used, and no attempt
was made to sterilize the trays, although they were allowed
to dry thoroughly before being used again. The larvae

-•-Southeast Hog Supplement (40% protein) , Purina,
St. Louis, Missouri.

25

were placed .in a black photographic tray containing ice
water to immobilize them. The infected larvae, which
were conspicuously iridescent against the black back-
ground, were removed with a bulbed pipette and held in
cold 0.05 M Tris-HCl buffer, pH 7.0.

Determination of Optimum
Quantity of RMIV Inoculum

Eggs were hatched and larvae allowed to develop for
24 hr as described above. Six experimental groups of
larvae, each with replicates of 2500 larvae, were exposed
for 24 hr to 5, 10, 25, 50, 75, and 100 LEQ of inoculum
per group, respectively. Controls were exposed to an equal
number of triturated uninfected larvae. Groups exposed
to an inoculum of only 5 LEQ were given a small amount of
food. At the end of the exposure period, all groups were
examined for larval mortality, and the larvae were trans-
ferred to rearing trays and maintained as described above.
Feeding of groups that experienced mortality during
exposure was reduced to avoid contaminating the rearing
containers with excessive food. Infected larvae from these
treatments were harvested and counted late in the fourth
stadium. Only larvae showing visible signs of RMIV in-
fection were considered infected. Net production of virus,

26

quantitated as LEQ, was determined. A qualitative judge-
ment of mortality was made, since only a low level of
mortality could be tolerated in any virus production system
to be used routinely. The experiment was repeated 3
times in successive weeks to be certain the system was
replicable.

Determination of the Optimum
Larval Age for Exposure to RMIV

Six experimental groups, composed of 2 replicates of

2500 larvae each, were exposed for 24 hr to 50 LEQ of

RMIV inoculum during days 1 through 6 of larval life,

respectively. Exposure of group 1 was made immediately

after hatching, and exposure of group 2 was made during the

second day of larval life as described for routine virus

production. Larvae of groups 3 through 6 were placed in

rearing trays when 24 hr old, fed and held by routine

methods until the respective groups were old enough to

expose. After exposure, all larvae were carefully rinsed

with tap water to remove external inoculum and returned

to rearing trays. Groups that had been exposed were

screened and infected larvae collected and counted at 12

hr intervals beginning early in the fourth day of larval

life and continuing through day 7, when pupation was almost

27

complete. Net production of virus, quantitated as LEQ,
was determined from 2 experiments conducted at an interval
of 1 month.

Purification of RMIV

All virus purification procedures were performed at
0° - 4°C in 0.05 M Tris - HC1 buffer, pH 7.0. Virus
infected larvae were triturated in a Ten Broeck type tissue
grinder until their head capsules were completely destroyed.
The triturant was washed through organdy cloth into a
beaker, and the volume was adjusted to 50 ml per 1000 larvae.
The beaker was covered and the contents stirred with a
magnetic stirrer for 12 - 24 hr to further fragment the
tissues. The triturant was then placed in a separatory
funnel and brought to a 20 per cent (v/v) mixture with
cold ethyl ether. The mixture was shaken vigorously for
5 min, placed in an explosion proof refrigerator, and
allowed to separate for 6 hr . The buffer layer was removed
and re-extracted with ether. The first ether layer was
washed with the original volume of buffer and allowed
to separate an additional 6 hr . These extractions were
effective in removing much of the debris from the virus
suspension. The final buffer layers were combined in a

28

beaker and placed in an ice bath. The residual ether

was

removed by stirring the mixture gently (to avoid foaming)
for 12 - 24 hr with a magnetic stirrer under a ventilated
hood, while passing a gentle stream of air over the top
of the virus suspension. Following ether extraction,
the virus was pelleted by centrif ugation in 50 ml round
bottom centrifuge tubes at 10,000 g for 40 min, in an
International Equipment Company (IEC) PR-6 refrigerated
centrifuge equipped with a high speed accessory attachment
The supernatant buffer was discarded, 10 ml of new buffer
were added to each tube, and the pellets were allowed to
soften overnight before being resuspended by vortexing.
After resuspension, the volume was adjusted to 80 ml per
1000 LEQ being processed, and the virus was subjected to
4 cycles of differential centrif ugation . The low speed
cycle was 15 min at 1000 g, and the high speed cycle was
40 min at 10,000 g. The low speed pellets of the first
3 cycles were scavenged by resuspending them to 25 per
cent of their original volume, combining them, and
recentrifuging them for 20 min at 1500 g. The virus in
the supernatant buffer was pelleted by a high speed run.
All high speed pellets were allowed to soften in buffer
for at least 6 hr before being resuspended by vortexing.

2 9

Virus in the high speed pellets from the scavenging runs
was returned to the main bulk of virus before the succeed-
ing low speed run. The final low speed pellet was dis-
carded.

The concentration of resuspended virus from the final
high speed run was estimated spectrophotometrically and
adjusted to approximately 20 mg/ml with a Beckman DU-2
spectrophotometer, and the absorption curve at 700 nm
(Matta, 1970) . Sucrose density gradients were prepared by
successively layering 2 . 5 ml aliquots of 55 per cent
(w/v) , 44 per cent, 30 per cent, and 15 per cent sucrose
in 13 ml, 1.4 x 9.7 cm centrifuge tubes. The gradients
were allowed to form naturally in cold for at least 12 hr
before use, and gradients not used within 48 hr of prepara-
tion were discarded. The virus suspension was diluted
1:1 with 10 per cent sucrose, and 2 ml of this material
were layered on the gradients and centrifuged at 20,000 g
for 20 min in an IEC B-60 ultracentrif uge equipped with a
type SB-283 head. After centrif ugat ion, the virus was
seen as a distinctive band about two-thirds of the way
down the tube, with a diffuse layer of top component slightly
above it. Initially, a small pellet of iridescent material
was seen on the bottom of the tubes after sucrose gradient

30

centr if ugaticn, but later results showed that proper care
in disrupting and resuspending the final high speed
pellet from differential centr if ugation prevented its forma-
tion. The virus band, but not the top component, was
removed with a needle and syringe. The virus suspension
was diluted 1:1 with distilled water and pelleted in a
Beckman Model L preparative ultracentrif uge equipped with
an SW-39 head. The pellets were resuspended and washed
twice with distilled water to remove the residual sucrose.
The final pellets were resuspended in a minimum volume of
distilled water, quantitated spectrophotometrically , ad-
justed to a concentration of 15 mg/ml, and lyophylized
with a Virtis lyophylizer. The lyophylized virus (15 mg/
tube) was stored at -30°C.

Virus purified by this procedure reacted vigorously
with its homologous antiserum formed in rabbits, but not
with normal rabbit serum, in tube precipitin tests. It
did not react with antiserum against acetone extracted
A. taeniorhynchus tissue in tube precipitin tests, by agar
gel diffusion tests, or by immunoelectrophoresis in agarose.
It appeared homogeneous in the electron microscope.
Matta (1970) found that virus purified by a similar but
less rigorous method produced a single sedimentation boun-
dary in an analytical ultracentr ifuge equipped with

31

Schlieren optics .

Quantification of RMIV
Production

Fifteen groups of 2500 larvae each were exposed to
an inoculum of 50 LEQ per group and reared to the advanced
fourth stadium as described for routine virus production.
Infected larvae were collected and counted, and the total
number was divided into 2 equal groups. A third group of
larvae, showing no overt symptoms of RMIV infection and
equal in number to 1/2 the total number of infected larvae,
was also collected. The virus from one group of the in-
fected larvae was purified by methods described above and
quantitated spectrophotometr ically . The other group of
infected larvae and the group of apparently uninfected
larvae were lyophylized with a Virtis lyophylizer until
weight loss stopped. The weight of both groups was deter-
mined on a Mettler H-5 balance. From these data the
following results were determined: (1) the per cent of
transmission acquired by routine virus production pro-
cedures, (2) the quantity of purified virus produced per
infected larva, (3) the weights of infected and uninfected
larvae from the same rearing set, and (4) the per cent of
larval dry weight attributed to the virus.

32

Attenuation of Infectivity
of RMIV During Purification

Transmission studies with purified, lyophylized virus
indicated that there was lower transmission with this
material than when other preparation were used. Experi-
ments were conducted to determine if the loss of infectivity
occurred at any particular stage of the virus purification
procedure .

Six experimental classes were established, each com-
posed of 5 groups of 200 larvae. Twenty-f our-hr-old larvae
were exposed for 24 hr to virus which had been purified
through various stages of purification as follows: (1)
fresh, triturated infected larvae; (2) buffer extracted
virus; (3) ether extracted virus; (4) virus subjected to
4 cycles of differential centr ifugation ; (5) virus which
had been subjected to sucrose density gradient centri-
fugation; and (6) virus which had been purified and
lyophylized .

Larvae in classes 1 and 2 were exposed to 5 LEQ of
inoculum per group. Larvae in classes 3 and 4 were exposed
to 10 LEQ per group. Larvae in classes 5 and 6 were
exposed to 20 LEQ per group. The use of increasing amounts
of inoculum was necessary to assure infections in all
classes, because preliminary tests showed that exposure of

33

classes I and 2 to the quantities of inoculum necessary
to produce infections in classes 5 and 6 resulted in high
mortality during exposure. A small amount of food was
provided larvae in classes 3 through 6 during exposure,
and for each group 200 larvae were reared in rectangular,
white enamel pans containing 1 liter of 0.05 M Nacl.
The larvae were examined for RMIV infection in the late
fourth stadium, and per cent transmission was determined
for each class. The experiment was replicated each week
for 4 weeks.

Transmission of RMIV

Per Cent of Larvae Infected
with RMIV by Transovarial
Transmission

Experiments were conducted to determine the per cent
of infected Fl generation larvae produced by adults that
had survived exposure to RMIV at different times during
their larval life. The surviving adults from the 6 experi-
mental groups in the experiment "Determination of the Optimum
Age for Exposure to RMIV" were maintained in an adult
rearing room at 80°C and 70 per cent relative humidity.
Cotton pads soaked in 5 per cent sucrose were continually
available as a food source, and shaved guinea pigs provided

34

blood meals on consecutive days. Five days after emergence,
16 oz cups, containing loosely packed, damp sphagnum moss
were offered as an oviposition medium. The moss was
left in the cages for 5 days, then covered with aluminum
foil, and the collected eggs were allowed to develop an
additional 10 days. The eggs then were hatched and reared
for 24 hr as described for routine virus production. Two
trays of 2500 larvae from each group were reared to the
advanced fourth stadium. They were examined for RMIV
infection, and the per cent transmission in each group
was determined. The experiment was conducted twice in
successive months.

Transovarial Transmission
Through Isolated Females

Larvae were reared to 4 days of age by procedures

described above, at which time groups of 2500 were exposed

to 50 LEQ of inoculum for 24 hr . The larvae were rinsed

thoroughly after exposure to remove inoculum, placed in a

rearing tray and checked periodically for evidence of

overtly infected larvae until pupation was completed. The

pupae were placed in a cage to emerge, and the adults

were fed as described above. Five days after emergence

was completed, 100 blooded females were taken from the

group and isolated in 35 ml glass vials containing damp
sphagnum moss as an oviposit ion medium. Boiled raisins
were placed on the screened tops of the vials for food and
as a source of liquid. The females were left in the vials
for 1 week, then removed. The vial tops were taped to
prevent desiccation of the eggs, and the eggs were allowed
to develop an additional 2 weeks. The moss from each
vial was then placed on a screen and washed. Egg batches
from vials in which oviposition occurred were placed indivi-
dually in rectangular, white enamel pans with 1 liter of
0.05 M NaCl. Larvae hatching in the pans were fed and
allowed to develop to the advanced fourth stadium, counted
and examined for RMIV infection. Determined from these
data were: (1) the per cent of isolated blooded females
that oviposited, (2) the per cent of the egg batches pro-
duced which hatched, (3) the per cent of the egg batches
which hatched that contained infected larvae, (4) the per
cent of infected larvae from each egg batch producing
infected larvae, and (5) the number of larvae produced
by transmitting and by non-transmitting females.

36

Sex Specificity of Transovarial
Transmission of RMIV

Four rearing trays of 2500 A. taeniorhynchus larvae
each were allowed to develop to 5 days of age. The larvae
in 2 of the trays were exposed for 24 hr in 16 oz cups
to 100 LEQ of fresh inoculum per cup of 2500 larvae.
After exposure, the larvae were rinsed thoroughly and re-
turned to rearing trays. The 2 untreated trays were main-
tained on a regular rearing schedule. Pupae from all 4
trays were collected and put into cages for emergence.
At 6 hr intervals during a 48 hr emergence period, the
adults were immobilized in a cold room and separated both
according to sex, and whether or not they had been exposed
to virus. When emergence was complete the adults in
each cage were again immobilized and divided into 2 equal
groups and recombined to result in the following experi-
mental classes:

Class 1: Unexposed Males X Unexposed Females
Class 2: Unexposed Males X Exposed Females
Class 3: Exposed Males X Unexposed Females
Class 4: Exposed Males X Exposed Females
The adults were fed as previously described. Eggs
were again collected by the standard method and allowed

37

to develop for 10 days after oviposit ion. They were then
hatched, and 2 groups of 2500 larvae from each class were
allowed to develop to advanced fourth stadium, when they
were examined for evidence of RMIV infection. The per
cent of transovarial transmission of RMIV in each class
was determined.

Transmission of RMIV by
Integument puncture

An attempt was made to transmit RMIV by integument
puncture to determine if rates of transmission higher than
those acquired per os could be achieved. Late third instar
larvae (early in their fourth day of larval life) were
rinsed thoroughly with tap water and placed in a suspension
of 1 mg/ml of purified RMIV in a petri dish placed in a
tray of shaved ice. The depth of the virus suspension
was sufficient to cover the larvae, and when they were
immobilized, they were pinched with a pair of fine tipped
forceps, prepared as described by Clark (1966), and trans-
ferred to a covered tray of 0.15 M Nacl. After 24 hr ,
the survivors were transferred to a tray of clear 0.05
M NaCl. To extend their development time, they were
maintained at 24°C rather than 27°C, as in routine rearing,
and only a small amount of food was provided. Beginning

48 hr after inoculation, the larvae were checked at 12 hr
intervals for infection with RMIV -

Production of Antiserum to RMIV and
to A. taeniorhynchus Antigens

Antibodies to both RMIV and A. taeniorhynchus antigens
were produced in rabbits. Baseline normal sera were
acquired by cardiac puncture of nonimmunized rabbits.
Formation of antibody to RMIV was stimulated by 4 weekly
injections of virus into each rabbit used. Five mg of
purified RMIV was suspended in 1 ml of normal saline,
emulsified with 1 mg of Freund's complete adjuvant, and
1 ml of this emulsion was injected intramuscularly into
each hip. Weekly bleedings (35-40 ml) by cardiac puncture
began 1 wk after the fourth injection. Injection of
antigen vithout adjuvant was continued after the fourth
week, with the injections being given immediately after
blood was withdrawn. The blood was allowed to clot in 50
ml centrifuge tubes for 1 hr at room temperature. The
clots were then ringed with a sterile applicator stick and
allowed to retract for 24 hr at a temperature of 4°C. The

Difco Laboratories, Detroit, Michigan.

3 9

tubes v/ere centrifuged at 5,000 g for 20 min, and the serum
was collected with sterile pipetts and stored without
glycerine at -30°C. The presence of antibody to RMIV
was determined by the tube precipitin test.

Powdered tissue of A. taeniorhynchus was prepared
by triturating healthy fourth stadium larvae in a Ten
Broeck type tissue grinder, containing acetone, in a dry
ice-acetone bath. The tissue was extracted with acetone
and the powdered tissue collected by filtration at -30°C.
The powdered tissue was air dried, and water soluble
proteins to be used as antigens were solubilized with
phosphate buffered saline, pH 7.0, by stirring overnight
at 4°C. Insoluble matter was removed by centrifugation
at 15,000 g for 30 min and discarded. The antigen was
sterilized by passing it through an 0.45 u Millipore
filter. The concentration of protein was determined by
the Lowry method (Lowry et al . , 1951) , using bovine serum
albumin fraction V1 as a standard. Antibody to this
antigen was produced in a rabbit by injecting 10 mg of
antigen suspended in Fruend's complete adjuvant weekly
for 4 weeks. Only 1 collection of 35 ml of blood

iSigma chemical Company, St. Louis, Missouri.

40

was made by cardian puncture before the rabbit expired.
The antiserum was prepared and stored as described
above. Presence of antibody was determined by the tube
precipitin test.

Purification of Gamma Globulin

Methods used to purify gamma globulin from anti-
RMIV antiserum were partially those of Goldman (1968) and
partially those of Kawamura (1969) . All procedures were
performed at 0° - 4°C . Antiserum was diluted 1:1 with
sodium phosphate buffered saline, pH 7.0. Saturated enzyme
grade ammonium sulfate, (NH4)2S04, was adjusted to pH
7.0 with 0.1 N NaOH. The globulins were precipitated twice
with half-saturated (NH4) 2S04 and the euglobulins were
precipitated twice with 1/3 saturated (NH4)2S04- The
precipitates were resolubilized in distilled water to the
original undiluted serum volume, except for the final
precipitates, which were dissolved in 1/2 the original
serum volume. The residual (NH4) 2S04 was removed by
dialysis against 0.15 M NaCl for 24 hr with stirring and
frequent changes of saline. The globulins were then
dialized against 0.005 M phosphate buffer, pH 8.0. The

-Sigma chemical Company, St. Louis, Missouri.

protein concentration was determined by the Lowry method
(Lowry et al . , 1951) and adjusted to 40 mg/ml by
pressure filtration in an Am icon pressure filtration

1 9

device equipped with a PM-10 membrane. A DEAE cellulose
column was prepared with a dry weight of DEAE cellulose
equal to 20 times the actual quantity of proteins to be
separated. A column with a height : diameter ratio of
more than 10:1 was used. The column was equilibrated and
eluted with 0.005 M phosphate buffer, pH 8.0, in which
both alpha and beta globulins, but not gamma globulin,
adsorbed to the column. The column was eluted with a flow
rate of 10 ml/hr. The gamma globulin from the column was
quantitated spectrophotometrically, using an extinction
coefficient of 1.24, and the concentration was adjusted
to 20 mg/ml. The gamma globulin was stored at 4°C for
a short time before further processing.

Conjugation of Anti-RMIV Antibody with
Fluoresceine Isothiocyanate (FITC)

Conjugation was performed by the method of Clark and
Shepard (1963) as described by Goldman (1968) . The

iAmicon Corporation, Lexington, Massachusetts

meg/g.

2

Sigma Chemical Company, St. Louis, Missouri. 0.89

42

concentration of the antibody solution was adjusted to
20 mg/ml and the pH was adjusted to 9.5 with 0.5 M
carbonate-bicarbonate buffer, pH 9.5, containing 1 mg

FITC per mi of immune globulin G (IgG) solution.

2
A G-25 Sphadex desalting column was prepared and

equilibrated with 0.0175 M phosphate buffer, pH 6.3. The

solution of conjugated antibody and free dye was added to

the column, and the column was washed with the equilibrating

buffer. The conjugated antibody was eluted after the

void volume, while the free dye was retained on the column.

The volume of the labeled antibody was adjusted to the

pre-labeling IgG volume with an Amicon pressure filtration

device equipped with a PM-10 membrane. The labeled IgG

was then dialyzed overnight against a large volume of

0.0175 M phosphate buffer, pH 6.3.

At this point the antibody solution consisted of

molecules of unlabeled antibody and antibody molecules to

which were conjugated 1 or more FITC molecules.

^-Baltimore Biological Laboratory, Baltimore, Maryland,

2

Sigma Chemical Company, St. Louis, Missouri.

43

Kawamura (1969) stated that the unlabeled antibody should
be removed, because it would compete with labeled anti-
body for binding sites. Also, antibody molecules with
more than 2 FITC molecules attached should be removed,
because they had a high net negative charge and tended
to combine non-specif ically with tissue. To this end, the
labeled antibody solution was subjected to chromatography
on a DEAE cellulose column equilibrated with 0.0175 M
phosphate buffer, pH 6.3. In this buffer unlabeled anti-
body would not absorb to the column, and it could be
eluted from the column with the equilibrating buffer.
The labeled antibody was eluted from the column, according
to the number of FITC molecules attached, by eluting the
column with a flow rate of 10 ml/hr of 0.0175 M phosphate
buffer, pH 6.3, containing successively 0.1, 0.2, and
0.3 M NaCl .

The fluoresceine : protein (F : P) ratio of the
conjugate was determined by the method of Holborow and
Johnson (1967), using the formula:

Concentration of FITC (ug/ml) =

Absorption4Q2 ~ Absorption-^g x Dilution

2

0.2

44

Protein concentration was then determined by the Lowry
method (Lowry et al. , 1951), and the F : P ratio was
determined by the formula:

F : P = 0.41 x uq FITC/ml
mg/ml protein

Fractions with F : P ratios of 0.36 to 1.79 were
pooled, with a resulting composite F : P ratio of 0.76.
The labeled antibody was dialyzed exhaustively against
phosphate buffered saline, pH 7.2. The concentration was
adjusted to 8.33 mg/ml, and the conjugate was lyophylized
in lots of 10 mg and 15 mg/tube and stored at -30°C.

For evaluation, the labeled antibody was reconstituted
to its pre-lyophylization volume with distilled water,
diluted with phosphate buffered saline, pH 7.2, to a con-
centration of 3 mg/ml, and clarified by centrif ugation at
15,000 g for 10 min. Dilutions of 2 mg/ml, 1 mg/ml, 0.5
mg/ml and 0.25 mg/ml were prepared to determine the
optimum concentration to use for staining. Unexposed and
known infected larvae as test specimens were embedded
together in Paraplast, the blocks were sectioned at a

■'-Scientific products, Evanston, Illinois.

45

thickness of 8 u, and the sections were deparaf f inized in
tertiary butanol at 40°C. The specimens were hydrated
through an ethanol series to distilled water and washed
in phosphate buffered saline before being incubated in a
humidified chamber with dilutions of the conjugated
antibody. Following exposure to the antibody, they were
given 4-5 min washes in phosphate buffered saline and
covers lips were mounted over the specimens with a
media of 1 per cent Tris /~Tris (hydroxymethyl) amino-
menthane_7 in 9 parts glycerine and 1 part water, with
the pH adjusted to 9.9 with 0.1 N NaOH (Mrenova and
Albrecht, 1966) . The specimens were examined and photo-
graphed with a Zeiss Ultrophot II microscope, using dark
field illumination. The light source was an Osram HBO-200
Mercury vapor lamp. The primary filter used was the
Zeiss UG-l/3mm, and the secondary filter admitted all
wave lengths of light greater than 410 nm. Dilutions of
the labeled antibody as low as 0.5 mg/ml were found
to stain foci of virus infections, and subsequent studies
utilized a concentration of 1 mg/ml. The specificity of
the stain was indicated by its failure to stain sections
of uninfected larvae mounted alternately with sections
of infected larvae, on the same slide. The staining

46

reaction of the labeled antibody was inhibited by treating
sections of infected larvae with unlabeled anti-RMIV anti-
serum prior to treatment with the labeled antibody. Treat-
ing sections of infected larvae with antiserum to A.
taeniorhynchus did not inhibit the staining reaction.

Pilot Project on Autoradiography of 3H-Methyl

Thymidine Uptake and Distribution in

Larval A. taeniorhynchus

No literature was available on certain parameters
essential to an autoradiographic study of the development
of RMIV infection in A. taeniorhynchus larvae. A pilot
project was conducted with uninfected larvae to determine:
(1) the level of radioactivity necessary in an aqueous
medium to assure uptake of 3H-methyl thymidine into larvae
in quantities detectable autoradiographically, (2) the
length of time larvae could be exposed to this activity
without destroying the resolution of the technique by
translocation of the tritiated methyl group by metabolism,
and (3) the optimum length of time to expose a nuclear
track emulsion to sections containing the tritiated
thymidine, before developing them.

Dilutions of 3H-methyl thymidine,1 with a specific

International Chemical and Nuclear Corporation, Irvine,
California .

4 7

activity of 17.4 Curies/mill imole (Cj/mM) and a concentra-
tion of 1 millicurie/raM (mCi/mM) , were made with 0.05 m
Nad to provide solutions of 2 . 5 microcuries/Ml (uCi/ml) ,
5uCi/ml, 10 uCi/ml, and 25 uCi/ml. Ten ml of each con-
centration were placed in each of 4 aluminum weighing cups
and a small amount of brewer's yeast infusion added. Ten
late third stadium larvae of A. taeniorhynchus were placed
in each cup. At intervals of 30 min, 1 hr, 2 hr, and 4 hr
after initiation of exposure, the larvae in 1 of the 4
cups of each concentration were removed, rinsed thoroughly,
and placed for 30 min in 3 oz cups containing a yeast
infusion. The post-exposure feeding was intended to flush
unabsorbed label from the gut to minimize the nonspecific
distribution of the label by the microtome knife during
sectioning of the specimens. Following exposure to label,
the larvae were washed thoroughly in a dilute solution of
Tween-80, rinsed in distilled water, fixed in Carnoy's
fixative for 4 hr at room temperature, dehydrated through
absolute ethanoi, cleared in tertiary butanol and embedded,
two larvae per block, in paraplast. Serial sections of the
larvae from the 16 experimental groups were cut at a
thickness of 8 p on a Leitz microtome and placed on
gelatin cooled slides. The slides were divided into 5

48

sets, each set containing the serial sections of two
larvae from each of the experimental groups. The slides
were deparaf f inized in xylene, hydrated through descending
concentrations of ethanol to distilled water and allowed
to dry. All subsequent dipping, packaging and developing
operations were performed in total darkness. The slides
were emulsion coated by dipping them in Kodak NTB-2 nuclear
track emulsion that had been melted and held in a constant
temperature water bath at 45°C. Control blank slides, to
determine background graination, and unlabeled sections,
to check for possible nonspecific tissue activity, were
also coated, and all slides were allowed to dry for 2 hr
in a near vertical position. After drying, the slides were
sealed in slide boxes containing a drying agent, wrapped
in several layers of aluminum foil, and placed in a
refrigerator at 4°C . At intervals of 2 , 3, 5, 7 and 10
weeks after dipping, 1 set of the slides was developed
in Kodak Dektol for 1.5 min at 18°C, dipped for 30 sec
into Kodak Indicator Stopbath, fixed for 10 min in Kodak
Rapid Fix (diluted 1:1), and washed in running water for
15 min (Rogers, 1967) . After rinsing in distilled water,

Eastman Kodak Company, Rochester, New York.

49

the slides were stained in methyl green-puronine (Pease,
1953) as recommended by Thurston and Joftes (1963) . After
staining, the slides were dehydrated in tertiary butanol,
cleared in tolune and covers lips were mounted with a
synthetic resinous mounting medium.

When all sets of the autoradiograms had been processed,
they were examined at 500 x (oil) magnification with an
AO-Spencer microscope. Background was rated on the con-
trol slides and also on each experimental slide, by examin-
ing areas of the slide distant from sections, as light
( 5 grains/100 u2) , moderate (5-10 grains/100 u2) or heavy
( 10 grains/100 u2) . The quality of the stain reaction
was noted, and the control sections which contained no
radioisotope were examined for evidence of nonspecific
activity. The following tissues were examined for evidence
of Label uptake: epidermis, fatbody, foregut epithelium,
midgut epithelium, gastric caecae, imaginal discs, hindgut
epithelium, malpighian tubules, tracheal epithelium and
gonads. Casual observation was made of incorporation into
nerve and muscle tissue. Labeling over the midgut contents
was observed to see if the gut had been swept clean by
post exposure feeding. Incorporation of label into the
cells of each tissue, as indicated by supranuclear

50

graination, was rated semiquantitatively by code as

follows :

Frequency of labeled cells (nuclei)

1 = <1/10 of cells labeled

2 = 1/10 - 1/2 of cells labeled

3 = >l/2 of cells labeled
Intensity of activity in labeled cells (nuclei)

1 ( = ^25 grains per nucleus

2 = 25-50 grains per nucleus

3 = 50-100 grains per nucleus

4 = /"100 grains per nucleus

Only the effect of the various exposure concentrations
and durations on the labeling of the midgut epithelium
cells was considered in deciding the concentration of label
to use in subsequent experiments. These cells were ob-
served as standards, because (a) they were uniformly large,
(b) the nuclei occupied less than 1/2 the area of the cells
in cross-sections, making it easy to discern successful
localization of the isotope within the nucleus, and (c)
because they had good staining properties.

51

■5

Production of ->H-Methyl Thymidine
Labeled RMIV

A. taeniorhynchus larvae were hatched and reared for
24 hr, and 2000 of them were exposed to RMIV as described
for routine virus production. The larvae were reared
to advanced third stadium and then transferred to a pan
containing a liter of 0.05 M NaCl, 10 mCi of 3H-methyl
thymidine (specific activity of 66 Ci/mM) , and a small
amount of food. The larvae were exposed to the tritiated
thymidine for 24 hr . At the end of exposure 128 con-
spicuously infected fourth stadium larvae were harvested,
as were 10 larvae which displayed no evidence of infection.
The uninfected larvae and four of the infected larvae were
prepared for autoradiography as described above. The
virus in the remaining 124 infected larvae was purified by
methods previously described and quantitated spectro-
photometrically . Five mg of labeled virus were recovered
and diluted to 50 mg . An 0.2 ml sample gave 1200 disin-
tegrations per min (dpm) above background when counted in
a Packard Tri-carb Liquid Scintillation Counter, indicat-
ing a total radioactivity of 0.14 uCi for the purified
virus. The virus was used immediately in autoradiographic
study of the site of penetration of RMIV into host tissue.

52

Site of Entry of RMIV into
Host Tissue

Light Microscope Studies

Autoradiography. — Eggs of A. taeniorhynchus were
hatched and the larvae allowed to develop for 24 hr by
standard procedures. Two thousand larvae were exposed to
5 mg of RVIV, labeled with 3H-methyl thymidine as
described before, with a total radioactivity of 0.14
uCi. At intervals of 1/4, 1/2, 1, 2, 4, 8, and 12 hr,
125 of the larvae were withdrawn from the exposure medium,
rinsed thoroughly with tap water, and fixed in Carnoy's
fixative for 4 hr at room temperature. The specimens
were processed histologically, dipped in Kodak NTB-2
nuclear track emulsion, developed after 2 weeks' exposure
and stained as previously described. They were examined
at magnifications of 500x (oil) and lOOOx (oil) in an
attempt to detect a site of entry of the labeled virus
through the gut wall.

Exposed specimens that were not processed for auto-
radiography were reared to the advanced fourth stadium,
when they were examined to determine the per cent infection
with RMIV.

53

Fluorescent antibody. — Eggs of A. taeniorhynchus
were hatched, the larvae were allowed to develop for 24
hr, and 2500 larvae were exposed to an inoculum of 50 LEQ
of RMIV as described for routine virus production. At
intervals of 1/4, 1/2, 1, 2, 4, 8, 12, and 24 hr after the
initiation of exposure to virus, samples of 125 larvae
were removed from the exposure container. They were rinsed
thoroughly, fixed in Carnoy's fixative for 4 hr at room
temperature, dehydrated through absolute ethanol, cleared
in tertiary butanol, and embedded in Paraplast. Sections,
processed as before, were incubated for 45 min with
1 mg/ml conjugated anti-RMIV antibody and examined as
for other studies for evidence of a specific site of entry
of RMIV into host tissue.

Larvae not collected for processing were allowed to
develop to the advanced fourth stadium when they were
examined to determine per cent infection with RMIV.

Electron Microscope Studies

Two hundred freshly hatched A. taeniorhynchus larvae
were placed in a 10 cm diameter petri dish of distilled
water containing an inoculum of 10 freshly triturated
RMIV infected fourth stadium larvae. Samples of exposed
larvae were removed from the exposure medium at intervals

54

of 5, 10, 15, 20, and 30 rain, and also at 1 1/2, 2, 3,
4, and 5 hr after the initiation of exposure. Head
capsules and air tubes were excised in glutaraldehyde ,
and the larvae were fixed for 4 hr at room temperature
in 4 per cent glutaraldehyde in 0.1 M cacodylate buffer,
pH 7.2. They were washed in cacodylate buffer over night,
post fixed for 1 hr in 1 per cent osmium tetroxide in
cacodylate buffer, dehydrated through ascending concen-
trations of ethanol , cleared in propylene oxide, and
embedded in epon-araldite (Mollenhauer , 1964) . Silver to
light gold sections, approximately 60 nm thick, were cut
on a Sorvall porter -Blum MT-2 ultramicrotome with glass
knives, and the sections were taken up on carbon or
Formvar coated grids. The sections were stained with
saturated uranyl acetate and lead citrate (Venable and
Coggeshall, 1965) and examined and photographed with a
Hitachi 125 E electron microscope using accelerating
voltages of 50 or 75 KV. Thick sections were also cut
and examined by phase contract microscopy as an aid to
orientation.

5 5

Development of RMIV Infections

Autoradiography

Eggs of A. taeniorhynchus were hatched, and the larvae
were allowed to develop for 48 hr in the hatching container,
by which time they had reached the second larval stadium.
Two groups, each with 2500 larvae, were exposed for 24 hr
to 100 LEQ of RMIV inoculum and then transferred to rearing
containers. Twelve hours after the termination of virus
exposure, and at 24 hr intervals thereafter, 125 of the
exposed larvae were removed from the rearing container and
placed for 2 hr into 125 ml of 0.05 M Nad containing 20 uCi/ml
of ^H-methyl thymidine with a specific activity of 13 Ci/mM.
A small amount of food was provided to encourage the intake
and passage of material through the gut. At the termination
of exposure to tritiated thymidine, the larvae were
thoroughly rinsed and placed in a suspension of food for
30 min. The larvae were processed histologically, and
autoradiography and staining were performed as has been
described. Exposure of this emulsion was for a period of
10 days. The activity of the exposure medium was increased
to 20 uCi/ml, despite the results of the pilot project,
and the emulsion exposure time was decreased in an attempt

56

to shorten the experiment. Autoradiograms were examined
at magnifications' 500x (oil) and lOOOx (oil) to determine
the tissues in which infection could be detected by this
technique and to determine how soon after the exposure
infections in these tissues could be detected.

Larvae not collected for study were reared to the
advanced fourth stadium and examined to determine the
per cent infection with RMIV .

Fluorescent Antibody

Overt infections . --Eggs of A. taeniorhynchus were
hatched, the larvae were reared for 24 hr , and 2500 of the
larvae were exposed to 50 LEQ of RMIV inoculum as described
for routine virus production. At the termination of ex-
posure, the larvae were transferred to a rearing container.
At that time and at 24 hr intervals thereafter, samples
of 125 larvae were collected, processed histologically,
and stained with fluorescent antibody, 1 mg/ml , as described
above.

The exposed specimens were examined to determine in
which tissues infection could be detected and at what
time interval after initiation of exposure infections
could be detected by this method.

57

Larvae not collected for this study were reared to
the advanced fourth stadium and examined to determine the
per cent of infection with RMIV.

Covert infections . --Specimens for this study were
collected from the exposed mosquitoes in the experiment,
"Sex Specificity of Transovarial Transmission of RMIV. "
Larvae were exposed to 100 LEQ of RMIV inoculum per group
of 2500 larvae during the fifth day of larval life.
Specimens for this study were collected as larvae at the
end of the sixth day of larval life (48 hr after exposure
was initiated) , as pupae at the end of the eighth day of
life, just prior to eclosion (72 hr after exposure was
initiated) , and as adult females prior to bleeding (120
hr after exposure was initiated) . Known transovarially
transmitting females, surviving the experiment "Transmission
of RMIV through Isolated Females," were also examined.
The specimens were processed histologically, stained with
fluorescent antibody to RMIV (1 mg protein/ml) and examined
as described above to determine which tissues were infected
and at what time interval after the initiation of exposure
infection could be detected, with special attention paid
to reproductive tissues. The amount of transovarial trans-
mission occurring through the experimental group, exclusive

58

of the known infected females, was reflected in the trans-
mission through class 4 of the experiment "Sex Specificity
of Transovarial Transmission of RMIV. "

Number and Molecular Weights of RMIV
Structural Polypeptides

Analytical sodium dodecyl sulfate (SDS) — polyacrylamide
gel disc electrophoresis was performed on SDS — 2-
mercaptoethanol disrupted samples of highly purified RMIV
by the methods of Maizel (1971) . Purified RMIV was sub-
jected to sucrose gradient centrifugation 1 or 2 additional
times, washed 3 times with distilled water to remove
residual sucrose, and quantitated spectrophotometrically
(Matta, 1970) . The concentration was adjusted to 10 mg
(protein) /ml in distilled water. The virus was structurally
disrupted and the proteins denatured to polypeptides by
heating the virus to 100°C for 1 min in 1 per cent SDS1
and 0.1 per cent 2-mercaptoethanol . 2 Samples of 100 ug,
200 ug, 300 ug and 400 ug of protein in 250 ul volume of
12 per cent sucrose were overlaid onto 10 x 0.6 cm resolving

1Fisher Scientific Company, Fairlawn, New Jersey.
^Matheson, Coleman and Bell, Cincinnati, Ohio.

59

gels of 10 per cent or 13 per cent acrylamide with
2.5 x 0.6 cm stacking gels of 3 per cent acrylamide. Both
resolving gels and stacking gels contained 0.1 per cent
SDS. Bromphenol blue was added to the samples as a
tracking dye. Simultaneously, 50 ug samples of proteins
of known molecular weight, also treated with SDS and 2-
mercaptoethanol, were run in identical gels. A current of
2mA per tube was applied until the tracking dye had migrated
0.5 cm into the resolving gel, and then 4mA per tube were
applied until the tracking dye approached the bottom of the
resolving gels. The gels were cut at the dye front when re-
moved from their tubes. They were fixed and stained in 0.2 per
cent Coomassie Brilliant Blue R 250 in 50 per cent methanol
and 7 per cent acetic acid for 12 hr and destained in repeated
changes of 5 per cent methanol in 7 per cent acetic acid.
The gels were stored in stoppered test tubes containing 7
per cent acetic acid. The stained bands of viral poly-
peptides were counted, and the relative migration of both

•^Sigma Chemical Company, St. Louis, Missouri.

2

Schwarz/Mann, Orangeburg, New York.

Sigma Chemical Company, St. Louis, Missouri.

60

viral polypeptides and polypeptides of known molecular
weight was determined in relationship to the dye front.
Curves were drawn, relating to relative migration of the
standard polypeptides in both 10 per cent and 13 per cent
resolving gels, to the Log10 of their molecular weights.
The molecular weights of the viral polypeptides were then
estimated from these curves.

Isolation of RMIV Structural Polypept ides
by Hydroxylapatite Chromatography

Hydroxylapatite chromatography techniques described
by Moss and Rosemblum (1972) were utilized with RMIV.
Hydroxylapatite (Bio-Gel HT) 1 was washed repeatedly with

0.01 M sodium phosphate, pH 6.4, 0.1 per cent (W/V) SDS ,

2

and 1.0 mM dithiothreitol (DTT) . Columns, 2.5 x 10 CM,

were poured over a layer of fine grade Sephardex G-502
or over filter paper discs. Twenty to 40 mg samples of
RMIV, purified as described above, were made to 5 ml in
1 per cent SDS, 0.1 per cent 2-mercaptoethanol and placed
in a boiling water bath for 2 min. The denatured samples

1Bio-Rod Laboratories, Richmond, California.

2

Sigma Chemical Company, St. Louis, Missouri.

61

were then diluted to a concentration of 1 mg/rnl with 0.01
M sodium phosphate, pH 6.4, 0.1 per cent SDS and placed on
the columns. They were washed into the columns with 2
column volumes of the 0.01 M phosphate-SDS-DTT solution and
eluted with a linear gradient formed by a 2-chambered
gradient former, containing 200 ml of 0.1 M sodium phosphate,
pH 6.4, 0.1 per cent SDS, 1 mM DTT in the starting chamber and
200 ml of 0.7M sodium phosphate, pH 6.4, 0.1 per cent SDS, 1
mM DTT in the trailing chamber. A flow rate of about
5 ml/hr was maintained by gravity. Three ml fractions were
collected in a refrigerated Buchler fraction collector.
The O.D28q of the fractions was determined in a Beckman
DU-2 spectrophotometer, and curves were drawn relating
fraction number to O.D-280

RESULTS AND DISCUSSION

Production of RMIV

Determination of Optimum
Quantity of RMIV Inoculum

For purposes of this dissertation, the optimum
quantity of RMIV inoculum for use in the routine produc-
tion of RMIV was defined as that quantity which resulted
in the maximum net production of virus without causing
significant larval mortality during exposure. The combined
results of 3 experiments designed to determine that
quantity are shown in Table 1. An inoculum of 50 LEQ per
exposure cup of 2500 larvae was found to be the optimum
inoculum for use in the virus production system described
and was used in routine virus production.

A graph of these data (Figure 1) indicates that the
per cent transmission of RMIV increased in proportion to
the quantity of inoculum used until inocula in excess of
50 LEQ/2500 larvae were used. Above this point, high
larval mortality during exposure obscured the relationship
between per cent transmission and quantity of inoculum.

62

63

Table 1

Net Virus Production Resulting from 24 Hr Exposure

of Day Old A. taeniorhynchus

Larvae to RMIV

Inoculum (LEQ) Per 2500 Larvae
5 10 25 50 75 100

Gross Number of Infected

Larvae Per 15,000 Exposed 234 317 758 1174 863 131

Per Cent Transmission 1.56 2.1 5.1 7.2 5.7 0.8

Net Virus Production

(LEQ) Per 15,000 Exposed 219 287 863 927 638 0

Mortality During Very

Exposure Low low Low Low High High

64

10

20

30

40

50

60

70

80

90

100

Inoculum (LEQ) per 2500 larvae

Fig. 1. Per Cent Transmission of RMIV in A. taeniorhynchus
Larvae Exposed to Increasing Amounts of Inoculum.

65

Mortality during exposure in the control groups
paralleled that in the experimental groups, indicating
that the high mortality in experimental groups exposed
to 75 and 100 LEQ of inoculum was not due to a toxic
effect of the virus, as experienced by Bellett and Mercer
(1964) in studies on the multiplication of SIV in cell
cultures .

No infection was observed among the control larvae.

Determination of Optimum
Larval Age for Exposure
to RMIV

The combined results of two experiments designed to

determine the effect of the age of A. taeniorhynchus larvae

at exposure to RMIV on the net production of RMIV are

shown in Table 2. Maximum net production of RMIV resulted

from exposures initiated when the larvae were 24 hr old.

These results corroborate those of Woodard and Chapman

(1968) .

Quantification of RMIV
Production

Thirty-eight hundred infected, advanced fourth stadium

larvae were collected from 37,500 larvae exposed to RMIV

and maintained as described for routine virus production.

66

Table 2

RMIV Production Resulting from 24 Hr Exposure

of A. taeniorhynchus Larvae of Different

Ages to 50 LEQ of Inoculum

Day of Larval Life Exposure
Was Effected
1 2 3 4 5 6

Gross Number of Infected

Larvae per 10,000 Exposed 1033 1084 702 329 0 0

Per Cent Transmission 10.33 10.84 7.02 3.20 0 0

Net Virus Production
(LEQ) per 10,000 Larvae
Exposed 833 884 502 129 0 0

67

This represented a transmission rate of 10.13 per cent.
Nineteen hundred infected larvae and an equal number of
uninfected larvae from the same experiment had dry weights
of 1.19698 grams and 1.39290 grams, respectively. This
represented a mean dry weight of 630 ug and 733 ug for
infected and uninfected larvae, respectively. From the
additional 1900 infected larvae harvested, 112 mg of RMIV
were purified. This was a recovery of 58.9 ug of purified
virus per infected larva and showed that 9.35 per cent of
the dry weight of an average infected, advanced fourth
stadium larva was composed of virus.

The host production system and the virus purification
procedures described here resulted in a maximum production
by 1 person of 940 mg of purified virus in 1 week and have
approached that quantity several other times. Therefore,
it is possible for 1 person, with an adequate mosquito
colony, setting 80 trays (200,000) of exposed larvae per
week, to conveniently produce and purify 1 net gram of
virus per week. Virus produced in 1 week can be purified
the subsequent week while a new generation of infected
hosts are produced. Although this quantity of virus may
not be adequate for extensive biological control field
testing, it certainly is adequate for numerous laboratory
studies .

68

Attenuation of I nf activity
of RMIV Duri nq Purification

An attenuation of the infectivity of RMIV during
purification is apparent from the data shown in Table 3.
Almost 70 times as many infected larvae per LEQ of inoculum
were produced by freshly triturated inoculum as by an
inoculum of purified virus that had been lyophylized. This
attenuation may have resulted from both quantitative loss
of virus during purification and from some damage to virus
particles during processing that reduced the ability to
enter a host cell or the ability to establish active infec-
tions. The proportion of attenuation attributable to any
one of these causes cannot be determined from the data.
The greatest attenuation appeared to occur following
ether extraction and sucrose gradient centrifugation. This
was surprising, since RMIV was shown to be insensitive to
ether by Matta and Lowe (1970) and because sucrose density
gradient centrifugation is commonly used in virus purifi-
cation because it is a relatively mild technique. The
greater relative loss of infectivity during these stages of
purification was possibly a reflection either of the use of
increased quantities of inoculum in those experimental groups
or a reflection of a non-linear relationship between trans-
mission and the quantity of inoculum used. Despite the

69

C

•h x:

P tn
O

3
CD

<u x:

Pi E-t

>1>

p H 4->

•H s (0

> Pi o

•H -H

4J IH -C

u o c

4-) (T3

IW (d 4J

O Oto
-H

C >U d)

O -H X!

•H S-i 4->

P 3

ftJ ^

3

C E

d> O

-P M

P M-l

di o

> J

rtf u

ft 3

0) B 3

P CD O

U U O

<U 3

IH fl H

c o

H M M-i
CM O

P

c c

d> I o

u tn -h

C CD

s-i rd en

CD S-l -H

ft Eh E

CD

> T3

S-< M CD

nJ d) en

j ft o
CM

CD d> w

P (J

U 3 O

<D 13 o

4-1 O O

c u -

l-i CU rf

3

u

o o

c o

H O

M— I <tf

O
S-l

>i d)

-U ft
•H
P

c a

3 ^

a ~1

CD

tn

rd C

p o

tn -h

-p

0) (T3

> u

•H -H

cn m

en -h

0) S-i

S-l 3

tn ft
o

u m

ft o

c
o

•H

p
(0

tn

3
U-l
■H

S-l
-P

c

CD
u

p
c

0) c

•H o

TS -H

rd P

S-i rd

O tn
3

CD M-l

cn -h

O M

S-l P

u c

3 CD

CO U

XI

a
o
>i

70

ambiguity of the data, it was apparent that the attenuation
detected was not caused by any single step in purification,
but occurred throughout the whole procedure.

Transmission of RMIV

Per Cent of Larvae Infected
with RMIV by Transovarial
Transmission

A relationship was found between the age of the parent
generation when exposed to RMIV and the frequency of in-
fections transmitted transovarially . As shown in Figure 2,
little or no transovarial transmission into progeny occurred
when the parent generation was exposed to virus during early
larval life. The amount of transovarial transmission in-
creased as the age of the parent generation at exposure
increased, up to a point, and then decreased. These data
suggest that, when sufficient time for development of
infections in larvae is available, overt infections with
RMIV result. No definite explanation can be given for
the decrease in transovarial transmission through parents
exposed during the sixth day of larval life. It is possible
that some host physiological change associated with pupa-
tion interrupts some infections before reproductive tissues
become infected. This possibility will be discussed further
in later sections.

71

(3.54%)

3 -

2 -

1 -

(1.99%)

12 3 4 5 6
Day of larval life during which the parent genera-
tion was exposed to RMIV

Fig. 2. per Cent of RMIV Infection Among Progeny of Mos-
quitoes Exposed to RMIV as Larvae of Different Ages

Data from 2 experiments were combined, and a total
of 10,000 larvae were in each experimental group.

72

Transmission of RMIV
Through Isolated Females

Eighty-three of 100 isolated, bl ooded females, exposed

to RMIV during the fifth day of larval life, oviposited.

Of these, eggs of 67 of the females (81%) hatched, and 9

groups (13.4%) contained infected larvae. Initially,

some of the larvae in 4 of the infected groups displayed

no gross infection early in the fourth stadium. This

suggested that there were exceptions to the apparent all-or-

none mechanism of transovarial transmission observed by

Woodard and Chapman (1968) and by Hall and Anthony (1971) .

However, the uninfected larvae were isolated, and without

exception they developed overt infections and died before

pupating. The 9 groups of infected larvae consisted of

from 13 to 104 larvae, with a mean of 63.3 per group,

compared to the uninfected groups, which ranged from 1 to

156 larvae with a mean of 94.2 per group. From the total

of 6,053 larval progeny observed from the isolated females,

only 591 (9.8%) were infected, although 13.4 per cent (9

of 67) of the females were known to have been infected.

These data suggested that the per cent of infection among

larvae produced in transovarial transmission experiments

was not an accurate reflection of the per cent of infection

73

occurring among the females that produced them. In the
experiment described, the per cent of known infected
females among the isolated females exceeded the per cent
of overt infections developing among the progeny of the
experimental group by a factor of 1.37 (13.4% i- 9.8% = 1.37)
Although this factor was acquired from limited data, it
will be used in subsequent discussions to estimate the
per cent of infected (transmitting) females in transovarial
transmission experiments from the per cent of infection
among the progeny of the experimental group.

Sex Specificity of
Transovarial Transmission

Males of A. taeniorhvnchus apparently play no part in
the transovarial transmission of RMIV (Table 4) . When males
were exposed to RMIV as 4 day-old larvae and mated to unex-
posed females, none of 5,000 progeny were infected. Also,
per cent of infection among progeny of exposed females was
not diminished by mating them to unexposed males rather than
to exposed males .

As will be discussed in later sections, overt infec-
tions appeared as often among male larvae as among female
larvae when mixed groups were exposed in early instars.
Since RMIV is a cytoplasmic virus, the apparent inability

74

Table 4

Transovarial Transmission of RMIV by Reciprocal

Mating of Exposed and Unexposed Male and

Female A. taeniorhynchus

Per Cent Infection Among Progeny
Mating (5,000 Larvae Per Class)

Unexposed Males x Unexposed Females o

Unexposed Males x Exposed Females 4.5

Exposed Males x Unexposed Females 0

Exposed Males x Exposed Females 4.8

75

of males to transmit the virus may be related to the very
small quantity of cytoplasm occurring in sperm.

Total of Overt and
Covert Transmission

The total amount of RMIV infection acquired by a

group of exposed larvae was defined as the sum of overt

infections (identified by the characteristic iridescence

of infected larvae and pupae) and covert infections. Overt

infection with RMIV were easy to quantitate, and since

they are fatal almost without exception, they play no

part in transovarial transmission. Little is known of

covert infections with RMIV, since at present the only

evidence for their existence is the transovarial transmission

resulting from a portion of them. The proportion of covert

infections which do not result in transmission is unknown.

Also, it is not known whether covert, infections can result

in transovarial transmission into only 1 of numerous

ovipositions by a single female. Nevertheless, an attempt

was made to determine the total transmission of RMIV to

groups of larvae exposed to the virus at different ages.

This was done to test the hypothesis (Woodard and Chapman,

1968) that about the same per cent of infections occurred

whether the larvae were exposed when young (resulting in

76

overt infections) or when older (resulting in transovarial
transmission) .

The following assumptions were necessary: (1) males
play no significant part in vertical transmission of the
virus, although they become infected with the same fre-
quency as females; (2) the per cent of covertly infected
females is about 1.37 times the per cent of infected
larvae among the progeny of the exposed generation; and (3)
the proportion of covertly infected females which transmit
the virus through their first egg batch is not influenced
by the age of the larvae when infection is acquired.
Support for the first 2 assumptions has been offered. The
latter is not unreasonable, considering that overt infections
are manifested 3 or 4 days after initiation of exposures
in larvae, and that adults are usually a week to 10 days
old before they oviposit. Therefore, infections should
have had time to develop regardless of when the larvae
acquired infection. Overt infections occurring in the
experiment "Determination of Optimum Larval Age for
Exposure to RMIV" were related to covert infections, re-
flected by transovarial transmission, in the experiment
"Per cent of Larvae Infected with RMIV by Transovarial
Transmission," because the ova for the latter experiment

77

Table 5

Total of Overt and Covert Transmission

of RMIV to Larvae Exposed at

Different Ages

Per Cent Total

Day During Which per Cent Overt Covert Trans-

Larvae were Exposed Transmission Transmission* mission**

1 10.33

2 10.84

3 7.02

4 3.29

5 0

6 0

0

10.33

0.23

11.07

0.62

7.64

3.07

6.36

4.85

4.85

2.73

2.73

♦Covert transmission was calculated as 1.37 times per cent
of infection among progeny of the exposed generation.

**These data represent observation of 10,000 larvae per
class .

78

were produced by the survivors of the former. Both experi-
ments were conducted twice at an interval of 1 month, and
combined data are given in Table 5. The^e data suggest
that older larvae are less susceptible to infection with
RMIV than younger larvae and that larvae are more sus-
ceptible to RMIV during the second day of larval life than
at any other time.

Transmission of RMIV
by Integument Puncture

Attempts to transmit RMIV to A. taeniorhynchus larvae

by integument puncture, as described by Clark (1966), were

unsuccessful. Clark reported up to 70 per cent transmission

of various microbial mosquito pathogens by this method.

One hundred and seventy-five of 269 third instar larvae

survived the exposure and were fourth instar larvae or

pupae 5 days after exposure. Pupation was delayed by

a poor diet and by maintaining the larvae at 24°C rather

than at 27°C. These conditions attempted to prolong

development and may have interfered with the development

of infection. Also, there is no assurance that virus ever

entered the host, so the data are inconclusive.

79

Pilot. Project on Autoradiography of ''K -Methyl

Thymidine Uptake and Distribution in

Larval A. taeniorhynchus

Autoradiography detected uptake of 3H-methyl thymidine
into the nuclei of cells of third instar A. taeniorhynchus
larvae in the following tissues : epidermis, fatbody,
foregut epithelium, midgut epithelium, gastric caecae,
imaginal buds, hindgut epithelium, malpighian tubules,
tracheal epithelium and the gonads. Uptake was greatest
in the malpighian tubules, hindgut epithelium, and gastric
caecae, and only slightly less in midgut epithelium,
maginal buds, and fatbody. Less uptake was detected in
epidermis, foregut epithelium, tracheal epithelium and gonadal
tissue. Uptake into nerve and muscle cells was rarely
detected. Exposure to 2 . 5 uCi/ml, for up to 4 hr , resulted
in labeling too sparse to be useful, but exposure to
25 uCi/ml for even 1/2 hr over labeled many tissues. A
2 hr treatment of 10 uCi/ml labeled the cells of most
tissues consistently and adequately. On the basis of the
response of midgut epithelium (used as a representative
tissue) this treatment was selected as ideal for a study
of the development of RMIV infections in host tissue. In
Figure 3 is shown the sharply circumscribed location of

80

Fig. 3. Malpighian Tubule Cell with Labeled Nucleus
(lOOOx) .
(N - Nucleus)

81

graination over a labeled nucleus. The specific activity
of the labeled thymidine used in this pilot study was
17 Ci/mM and exposures of 2 weeks were adequate for
specimens exposed to the accepted treatment of 10 uCi/ml
for 2 hr. Grains over the cells of active tissues
increased in number as duration of exposure of the emulsion
increased, but background graination also increased rapidly
and reduced the resolution of the technique.

Examination of control materials indicated that back-
ground graination intrinsic to the emulsion was low enough
not to reduce the resolution of the technique. A small
amount of very fine graination occurred over the anterior
thoracic fatbody of all specimens, whether exposed to
label or not, and could not be explained. The staining
reaction of methyl green-pryonine was very good in unin-
fected tissue. DNA in the nuclei stained a pale blue,
and the graination was easy to detect. FN A stained a pale
pink in the cytoplasm and a bright pink in the nucleoles,
which was very helpful in detecting the nuclei of small
cells .

Feeding larvae to flush unabsorbed tritiated thymidine
from the gut lumen after exposure was very helpful. In
the few specimens that did not feed during the flushing

82

period, graination was very heavy over the gut contents.
It was observed that some of the radioactive thymidine
was dispersed from the gut into surrounding areas by the
microtome blade during sectioning. An interesting
phenomenon was observed in larvae that had fed actively
during the flushing period. As shown in Figure 4, a pocket
of tritiated thymidine remained in the otherwise cleanly
flushed gut. The residual pocket was ring-shaped and
located in the anterior part of the midgut at the level of
the caudal extremity of the foregut invagination into the
midgut. This possibly represented an "eddy" in the flow
of material through the gut, and its possible significance
is discussed later with studies on the site of entry of
RMIV into host tissue.

Site of Entry of RMIV
Into Host Tissue

Light Microscope Studies

Autoradiography . — The site of entry of RMIV into host
tissue could not be detected on autoradiograms of serial
sections of 700 early instar larvae of A. taeniorhynchus
exposed to purified RMIV that had been labeled with
3H-methyl thymidine. Three and one-tenth per cent (30/973)

33

Fig. 4. Anterior Midgut Region of Larva Showing Residual
Deposits of Tritiated Thymidine Peripheral to
Caudal Extremity of Foregut Invagination into
Midgut (lOOx) .
(I - Foregut invagination; C - Gastric caecae)

84

of the survivors of exposure to the purified, labeled
virus developed infections with RMIV, when they were
reared to the fourth stadium. That was about the same
per cent transmission as achieved with purified, unlabeled
virus in the study "Attenuation of Infectivity of RMIV
During Purification," suggesting that labeling did not
reduce the infectivity of the virus. Small foci of grains
were seen scattered in the emulsion over the midgut
contents of larvae that had been exposed to the labeled
virus, indicating that labeled virus was consumed and
that it could be detected autoradiographical ly. However,
these foci of grains were rare. Only 0.14 uCi of activity
were detectable in the entire virus inoculum (5 mg) by
liquid scintillation counting. Autoradiograms of sections
of 4 of the infected larvae in which the inoculum was
produced were examined. These revealed that label
incorporation into foci of virus, identified in the
cytoplasm of infected cells by their staining reaction with
methyl green, was much less than into the nuclei of the
infected cells (Figure 5) . The evidence suggests that the
virus inoculum was not heavily labeled. This could have
occurred if most of the viral DNA had been produced before
the infected larvae were exposed to tritiated thymidine.

85

Fig. 5. RMIV Infected Fatbody Cell in A. taeniorhynchus
Larva with Graination Over Cytoplasmic Deposits
of Tritiated Thymidine Labeled Viral DNA (lOOOx)

86

If this is true,, it might account for failure to observe
exit of the virus from the gut lumen into host tissue by
autoradiography.

Fluorescent antibody . --The site of entry of RMIV into
host tissue was not detected by examination of fluorescent
antibody stained serial sections from 800 early instar
larvae exposed to an inoculum of triturated, infected
hosts. Infection with RMIV developed in 16.8 per cent (191/
1131) of the survivors of the exposure, when they were
reared to the fourth stadium. Studies to be discussed
below suggested that RMIV entered host tissue as isolated
virions which did not become aggregated in cells of the mid-
gut epithelium. Neither these virions nor the fluorescence
from them by attached labeled antibody could have been
detected with the light microscope because their size is
below the resolution of the equipment used.

Electron Microscope Studies

A probable site of entry of RMIV into host tissue was
found by examination of tissues with the electron micro-
scope. Numerous virus particles with the size and appearance
of RMIV were found in midgut epithelium of first instar
A. taeniorhynchus larvae killed and fixed 5 hr after

initiation of exposure to the virus. The infected
epithelial cell shown in Figures 6 and. 7 is at the level
of the foregut invagination into the midgut. This was de-
termined by the relative positions of the infected cell,
the peritrophic membrane and the cuticular lining of the
foregut invagination. The larvae examined were from a
colony known to be free of RMIV infection prior to exposure.

A possible mechanism of entry of RMIV into host tissue
is as follows. Some of the virus particles entering the
midgut are swept into the eddys forming a ring around the
caudal extremity of the foregut invagination. (See Figure
4 and related text.) Some of these virus particles are
pressed against the peritrophic membrane (Figures 8 and 9) ,
emerging between the foregut invagination and the midgut
wall (Figures 6 and 7) . where perforations occur in the
peritrophic membrane due to faulty construction or to
trauma, virus particles invade the space between the
peritrophic membrane and the midgut epithelium. Virus
particles and other objects of equal or larger size were
seen between the peritrophic membrane and the midgut wall
(Figures 8, 10 and 11) . virus particles enter the midgut
cells, perhaps passing into and down the microvilli
(Figures 10 and 11) , or perhaps crossing the cell membrane
in an area where the microvilli are disarrayed or damaged

88

■ . '%FS>Z

W:

j. %

Fig. 6. Midgut Epithelium Cell Containing Scattered RMIV
Particles (7500x) .

(V - Virus; I - Foregut invagination; M - Microvilli;
P - Peritrophic membrane)

89

Fig. 7. Midgut Epithelium Cell with Disarranged Microvilli
and Containing Scattered RMIV Particles (15,000x)
(V - Virus; M - Microvilli; p - Peritropnic' ' '
membrane; I - Foregut invagination)

90

Fig. 8. RMIV Particles Pressed Against peritrophic

Membrane in Midgut of A. taeniorhynchus (9C00x)
(V - Virus; p - Peritrophic membrane; M -
Microvilli)

91

Fig. 9. RMIV particles pressed Against Peritrophic
Membrane in Midgut of A. taeniorhynchus
(33,000x) ,
(V - Virus; P - peritrophic membrane)

92

Fig. 10. RMIV Particle Between Peritrophic Membrane and
Midgut Epithelium (30,000x) .
(V - Virus; p - peritrophic membrane; M -
Microvilli)

93

Fig. 11. Virus Like particle Apparently Within a Micro-
villus (30,000x) .
(VLP - Virus like particle; M - Microvilli)

94

(Figure 7) . The large numbers of virions in the infected cell
in Figure 6 suggests the possibility that the virus may repli-
cate in some midgut cells. It is also possible that the virus
passes to adjacent cells (Figure 12) before entering and being
distributed by the nemo lymph. Only selected epithelial cells
may be receptive, and the availability of receptive cells
would constitute a barrier to entry in addition t<_ the
peritrophic membrane. The section of an infected midgut
cell seen in Figure 6 contains over 100 virus particles,
while no virus particles can be seen in any of the adjacent
epithelial cells. Also, the microvilli of the cell are
disarranged (Figure 7) , and the cell is not vacuolated, as
the adjacent cells are. It is possible these observations
reflect only the level at which the cells were sectioned.
However, disarrangement of the microvilli may reflect damage
to the cell surface permitting easier entry by the virus.
The absence of vacuolation in the infected cell may indicate
either that it is different in age than the other cells or
that metabolic changes resulting from invasion by the
virus have occurred. It is a possibility that the age of
a cell may influence its susceptibility.

Some characteristics of the transmission of RMIV in
the laboratory were enigmatic. Per cent transmission

95

• - ■

•

*

if " '4*

fc

Fig. 12. Small pocket of RMIV in Non-Epithelial Cell in

Midgut Wall of A. taeniorhynchus Larva (14,000x)
(V - Virus)

96

appeared to be linearly related to quantity of inoculum,
as shown by the rising side of the curve in Figure 1.
However, this was true only to a point, beyond which quan-
titation of the relationship was impossible, since larval
mortality during exposure, probably due to anoxia,
became almost total in extremely heavy inocula. Although
it has not been established that there is a finite limit
to per cent transmission in any given population, no one
has achieved high levels of transmission, regardless of
the quantity of inoculum used or the conditions of ex-
posure .

Two of the possible mechanisms that may limit trans-
mission of RMIV are the barriers to entry mentioned above
and metabolic inhibition of replication by virus which
gains entry. The biochemistry of the process by which
RMIV is replicated is not known, although the recent find-
ing of RNA polymerase activity in SIV and in CIV (Kelly
and Tinsley, 1973) represents a beginning. Therefore,
it is impossible to speculate on mechanisms by which trans-
mission is controlled by host reactions limiting replica-
tion. If it were possible to achieve high levels of
transmission by injection of the virus into larvae, this
would suggest that control of transmission was effected

97

by limiting entry of the virus into the host. However,
it has not been possible to do this. The failure of
the injection experiment is of no significance because the
procedures used may not have assured entry of the virus
into the host. Also, it is possible that the virus must
pass through the gut epithelium as a prerequisite to the
establishment of infection. Why high levels of trans-
mission of RMIV have not been possible remains unresolved.

Development of RMIV Infections

Autoradiography

Infections with RMIV were detected by autoradiography
in 16 of 100 late fourth stadium larvae that were killed
and fixed 108 hr after initiation of exposure to the virus.
Infections were also detected in 20 of 100 early fourth
stadium larvae that were killed and fixed 84 hr after
initiation of exposure to the virus. It was possible to
visually detect infections in 348 of 2000 (17.4%) late
fourth stadium larvae exposed with the above experimental
larvae. Infections with RMIV were not detected autoradio-
graphically in any of 2 groups of 100 larvae each, killed
and fixed as second and third instar larvae 36 and 60 hr
after initiation of exposure to the virus. Infected cells

98

were easily recognized by the presence of heavy graination

in the emulsion over the cytoplasm, signalling the

3

incorporation of H-methyl thymidine into viral DNA

(Figure 5) . Graination over the cytoplasm of uninfected
cells was only slightly higher than background graination

(Figure 3) . Infected cells were also recognized by the
histochemical reaction between viral DNA in the cytoplasm
and the methyl green stain used. Stained foci of viral
DNA appeared as conspicuous green to greenish purple
bodies in the cytoplasm of infected cells (Figure 13) .
Infected cells were found among the following tissues:
fatbody, imaginal buds, and epidermis. Fatbody cells,
which appeared to be the preferred host of the virus, were
conspicuously enlarged when infected.

An attempt to shorten the time of exposure of the
emulsion, by exposing larvae to greater activities of

H-methyl thymidine than those indicated as optimal by
the pilot project on autoradiography, resulted in over-
labeling of the tissue and reduced the resolution of the
method. Infections in cells with small cytoplasm volume
were difficult to detect. The stained foci of virus in
epidermal cells were about the same size as the nuclei of
these cells, and they could not always be distinguished by

99

Fig. 13. Methyl Green Staining of RMIV Infected Cells
(lOOx) .
(F - Fatbody)

100

the presence of the pink staining nucleolus. use of a
lighter stain specific for the nucleus only, such as
nuclear fast red, would have enhanced the resolution of
the technique.

Infected tissues were distributed throughout the bodies
of infected larvae by the time infection could be detected
either autoradiographically or histochemically . No point
source of origin of the infection was evident. A remarkable
progression of disease apparently occurred late in the third
stadium or early in the fourth stadium. Within a period
of 24 hr , the infection progressed from being inapparent
by the methods used to being conspicuously present and
distributed through the diseased larvae.

Fluorescent Antibody

Overt infections. — Infections with RMIV were detected
by fluorescent antibody staining of viral antigens in 12
of 100 fourth stadium larvae, killed and fixed 96 hr after
initiation of exposure to the virus. Infections were also
detected in 14 of 100 third stadium larvae, and 1 of 100
second stadium larvae, that had been killed and fixed 72
and 48 hr after the initiation of exposure, respectively.
No evidence of infection was found in 100 second instar

101

larvae, killed and fixed immediately after exposure.
Overt infections were observed in 253 of 2000 (12.7%)
advanced fourth stadium larvae examined visually 120 hr
after initiation of their exposure with the experimental
larvae .

Infected cells were recognized by green fluorescence
emanating from labeled antibody combined with foci of
viral antigen in the cytoplasm. No other fluorescence
was present in any location or configuration that could
have been confused with the stained antigen. However, a
green natural fluorescence was observed lining the entire
enteric track, in the oviduct and surrounding the ova of
gravid females, and in the cytoplasm of cells forming the
malpighian tubules. The latter was caused by small,
discrete spherules that were consistent in size, while
stained foci of viral antigen were larger and very ir-
regular in shape.

Infection was detected in cells of fatbody, imaginal
buds, epidermis, and tracheal epithelium tissues (Figures
14 and 15) . Infection was found only in fatbody cells
in the infected second instar larva. These occurred singly
or in pairs on the medial surface of the fatbody in almost
every abdominal segment, but were scattered and infrequent
in the thorax. The infected cells were about double normal

102

Fig. 14. Fluorescent Antibody Staining of Epidermal Cells,
Imaginal Bud Tissues and Fatbody Tissue Infected
with RMIV (250x) .
(E - Epidermis; B - Imaginal bud; F - Fatbody)

103

Fig. 15. Fluorescent Antibody Staining of RMIV Infected
Tracheal Epithelium Cell (lOOOx) .
(N - Nucleus; T - Tracheal taenidiura)

104

size, and the foci of antigen were small and clearly de-
fined. Both epidermis and fatbody were infected in third
instar larvae, and occasionally a few cells of the
imaginal buds had small, well-circumscribed foci of
stained antigen. Again, foci occurred in the fatbody of
each segment of the abdomen, and were now also common in
the thorax. The number of infected cells per abdominal
segment focus was fairly constant within, but varied between,
hosts. In some hosts only a few of the cells in each
abdominal segment were infected; in others as many as 1/2
of all cells observed were infected. Epidermal cells were
never observed to be infected except when they were adja-
cent to infected fatbody. Infected fatbody cells in third
instar larvae were markedly enlarged and many were filled
with the stained foci of viral antigen. Infections were
massive in all infected fourth stadium larvae examined.
In some larvae virtually every fatbody cell was infected ,
including those in the head capsule. Both epidermal cells
not adjacent to infected fatbody, and tracheal epithelium
cells were infected. Infection in the fatbody surrounding
the testes was heavy (Figure 16) , but no antigen was found
within either the testes or the ovaries. No evidence of
infection was seen in epithelium of foregut, midgut, or

105

Fig. 16. Fluorescent Antibody Staining of Fatbody Sur-
rounding Testis (250x) .
(F - Fatbody; G - Testis)

106

hindgut or in nerve tissue. Infection in muscle tissue
may have occurred, but probably could not have been seen
because of the bright, silvery autof luorescence of that
tissue. Some fluorescence was seen in a few oenocytes ,
but it appeared as small foci. There were several foci
in each cell, and these probably represented phagocytized
fragments of disrupted, infected cells.

The pattern of distribution of infected cells in the
earliest infection observed suggested that infection was
initiated in isolated cells throughout the body rather
than at one location. This would be expected if virus
penetrating the gut epithelium was distributed randomly
by the hemolymph. If large numbers of infectious virus
particles had penetrated a localized portion of midgut,
a generalized infection, rather than a few isolated infected
cells, would probably have resulted. It appears likely
that a few infectious particles penetrate the gut, enter
the hemolymph, are distributed randomly, and infect cells
throughout the body. The virus replicates and accumulates
in these cells which finally rupture. This releases the
large number of infectious particles that initiate the
generalized infection that rapidly kills the host.

107

Covert infections. — RMIV infections were not detected
by fluorescent antibody staining among 100 advanced fourth
instar larvae or among 100 unblooded female adults, killed
and fixed 48 and 96 hr, respectively, after initiation
of exposure to the virus. Also, infections were not de-
tected in serial sections of 3 adult female mosquitoes that
were known to have transmitted the virus transovar ially .
Only 1 of 100 pupae, killed and fixed 72 hr after the
initiation of exposure, was infected, and this infection
appeared abnormal. In sections of the infected pupa, infected
fatbody cells were in groups of 3-10 cells per focus.
These foci were scattered and infrequent in occurrence,
rather than symmetrically located as in overtly infected
larvae .

Failure to find evidence of infections among the
larvae and unblooded females was surprising. Transovarially
acquired infections occurred among 4.8 per cent of the
progeny of the group of mosquitoes from which specimens
examined were collected. This indicated that infections
probably were present in about 6.5 per cent (4.8% x 1.37)
of that group--yet, only 1 of 300 examined presented any
evidence of infection when examined by fluorescent antibody
staining. (As shown in the study entitled "Transovar ial

108

Transmission of RMIV Through Isolated Females," per cent
of infection among parents exposed as late instar larvae
exceeds that observed among their progeny by a factor of
about 1.37.) Adequate time passed between exposure and
sampling for infections to have become detectable, since
infections can be detected as early as 48 hr after
initiation of exposure. This was shown in second instar
larvae, which have much less fatbody than the specimens
studied here. The method used has detected foci of in-
fection as small as 1 cell. Therefore, it appears likely
that infections by which transovarial transmission is
effected are not massive and generalized as is characteristic
for overt infections that result in death of the host.

An hypothetical explanation of the above observations
follows. It incorporates the fact (Figure 2) that less
transovarial transmission occurred through mosquitoes
exposed during the sixth day of larval life than through
those exposed during the fifth day of larval life. When
late instar larvae feed on the cadavers of infected larvae,
infectious virus particles enter the tissues of some of
them. The rate at which infections develops in these
larvae is proportional to the number of infectious particles
acquired. The extent to which infections progress is

109

proportional both to the number of infectious particles
that replicate and to time. Host changes associated with
impending pupation stop or sharply reduce the rate of viral
replication, halting the progression of the infection.
When large numbers of infectious particles enter larvae
early enough, infections may progress to such an extent
that infected cells can still be found after pupation.
When fewer infectious particles are acquired during
exposure or when less time is available before pupation
interrupts viral replication, infections do not progress
to a point at which they can be detected by fluorescent
antibody staining. However, the initial quantity of
virus entering the host or that produced by subsequent
replication will have been sufficient for infectious
particles to have been distributed to reproductive tissues
in some females. Receptiveness of reproductive tissues
to virus invasion might be a factor working to assure
infection of all oocytes in infected females. Also, virus
might replicate, either in reproductive tissues or else-
where, at a level not detectable by the methods used,
with the result that all the oocytes become infected.

110

Number and Molecular Weights of
RMIV Structural Polypeptides

proteins present in purified RMIV were resolved into
25 polypeptides by SDS-polyacrylamide gel disc electrophoresis
Table 6 is a list of proteins of known monomer number and
molecular v/eight, which were electrophoresed at the same
time as the viral polypeptides, to serve as markers for
estimating the molecular weights of the RMIV polypeptides.
The relative migration fractions (Rf 's) of RMIV polypeptides
in both 13 per cent and 10 per cent gels and their es-
timated molecular weights are shown in Table 7. Markers in
the molecular weight range of the smaller viral poly-
peptides passed through 10 per cent gels with the dye
front, as did the smaller viral polypeptides, and their
Rf ' s could not be determined in 10 per cent gels. Therefore,
molecular weights of small viral polypeptides were deter-
mined from a curve drawn with the Rf ' s of small markers
from 13 per cent gels. The RF of thyroglobulin, molecular
weight 335,000, was not included in drawing the curve*
The larger markers and viral polypeptides penetrated the
13 per cent gels such a short distance that errors in
measuring their migration distances were very large.
Therefore, molecular weights of large viral polypeptides
were determined from a curve drawn with the Rf 's of large

Ill

Table 6

Proteins Used as Markers for the Estimation
of the Molecular Weights of RMIV
Structural Polypeptides

Log1Q

Monomer

Monomer

Relative

M

igration

Molecular

Molecular

Fract

ion

protein

Weight

Weight

13% Gels
0.702

10% Gels

Cytochrome C

12,400

4.0934

Myoglobin

17,500

4.2430

0.660

Apoferritin

24,000

4.3802

0.590

0.594

Chymotrypsinogen

25,000

4.3979

0.562

Aldolase

40,000

4.6021

0.312

0.486

Ovalbumin

45,000

4.6532

0.305

0.467

Human Igc

(heavy chains)

50,000

4.6990

0.229

0.385

Bovine Serum

Albumin

67,000

4.8261

0.179

0.301

Thyroglobulin

335,000

5.5250

0.019

0.053

112

Table 7

Relative Migration Fractions and Molecular
Weights of RMIV Structural Polypeptides

Poly-

Relative Migrat

ion Fraction

Molecular

Weight

peptide

13% Gel

10% Gel

13% Gel

10% Gel

1

0.044

331,000

2

0.068

290,000

3

0.080

271,000

4

0.097

246,000

5

0.112

224,000

6

0.146

181,000

7

0.076

0.155

80,700

174,000

8

0.093

0.186

77,200

146,000

9

0.098

0.191

75,800

141,000

10

0.123

0.214

72,700

124,000

11

0.132

0.252

69,200

100,000

12

0.164

0.306

64,100

73,000

13

0.216

0.369

55,200

51,000

14

0.221

0.379

54,500

48,800

15

' 0.225

0.408

53,900

29,600

16

0.275

0.471

47,100

22,200

17

0.309

0.519

42,900

20,900

18

0.319

0.529

41,800

17,400

19

0.345

0.563

39,000

16,300

20

0.358

38,000

21

0.365

37,100

22

0.414

32,600

23

0.473

27,800

24

0.483

27,300

25

0.539

23,500

113

markers from 10 per cent gels.

As can be seen in Table 7, the molecular weights of
RMIV polypeptides determined in both 13 per cent and 10
per cent gels differed greatly. This difference was
minimal (about 8%) for viral polypeptide 13. To determine
the sum of the molecular weights, the molecular weights
of polypeptides 1-13 from 10 per cent gels, rather than 13
per cent gels, were taken as probably more nearly accurate.
Likewise, the molecular weights of polypeptides 14-25 from
13 per cent gels, rather than 10 per cent gels, were taken
as probably more nearly accurate. The total of these
molecular weights was 2,817,000. The molecular weight of
the DNA in RMIV has been shown to be about 400 x 106
(Matta, 1970; Wagner et al . , 1973). Assuming there is no
redundancy in the DNA molecule, and that it does not code
for 5S ribosomal RNA or transfer RNA, it should be able
to code for about 20 x 10 daltons of protein. Thus, only
about 14 per cent of the genome of RMIV would be able to
code for its structural polypeptides. This figure is less
than 1/2 of the 30 per cent of the genome of SIV and CIV
required to code for their structural polypeptides (Kelly
and Tinsley, 1972) .

114

Separation of RMIV Structural Polypeptides
By Hydroxylapatite Chromatography

The structural polypeptides of RMIV could not be
separated by hydroxylapatite chromatography at pH 6.4,
Elution of the columns used was quantitative, but all
fractions eluted as one peak.

SUMMARY

A system for the production of RMIV in A. taeniorhynchus
larvae that yielded a maximum weekly production of 940 mg
of purified virus was described. Several features of the
production system accounted for its success. Larvae were
exposed to an experimentally determined optimum quantity
of inoculum (50 LEQ per 2500 larvae exposed) . The larvae
were exposed to inoculum during the second day of larval
life, their age of greatest susceptibility. Methods were
used to expose and rear larvae that caused essentially no
mortality. This permitted standardization of procedures,
and as a result, the virus production system is adaptable
to mass production of the virus. Additionally, virus
purification procedures were used that minimized waste of
virus during purification. The production of virus from
infected larvae was quantitated, with an average of 58.9
ug of purified virus per larva. This constituted 9.35 per
cent of the host's dry weight. Attenuation of viral in-
fectivity occurred throughout the purification procedure
and could not be attributed to any single step in the

115

116

purification procedure.

Transovarial transmission of RMIV was found to in-
crease as the age of the parent generation at per os ex-
posure to virus increased, up to a point, but decreased
if parents were exposed during the 24 hr immediately
preceding pupation. Transovarial transmission studies
with isolated females indicated that infected (transmitting)
females produced fewer larvae (avg. = 63.3) than uninfected
females (avg. = 94.2) . Quantitation of this difference
in fecundity provided a factor with which the per cent of
covert infections among adults, exposed in the laboratory,
was estimated from the per cent of infection among their
progeny. With a means of estimating the per cent of covert
infections available, total transmission was estimated
when larvae of different ages were exposed to inoculum.
This decreased as the age of larvae at exposure increased.

Males were shown to play no part in transovarial
transmission of RMIV.

Preliminary studies of the autoradiographic detection
of tritiated thymidine uptake in A. taeniorhynchus indi-
cated that a 2 hr exposure of larvae (1 larva/ml) to a
solution of 10 uCi/ml of 3H-methyl thymidine (Sp. Act.
17 Ci/raM) was adequate to label most larval cells. An

117

autoradiographic study of developing RMIV infections showed
that infected cells could be detected in fatbody, imaginal
buds, epidermis and tracheal epithelium 34 hr after the
initiation of exposure. This was the first attempt to
study the development of RMIV infections by autoradiography,
and the method displayed great potential.

A probable site of entry of RMIV into host tissue was
detected by electronmicroscopy . For the first time, intact
virus particles were seen between the peritrophic membrane
and the midgut wall of second instar larvae that had been
exposed to virus for 5 hr preceding fixation. Those seen
were at the level of the foregut invagination into the
midgut, and large numbers of virus particles were also
seen in midgut epithelium. Although these particles were
not packed in paracrystalline arrays, the number of parti-
cles present provided the first indication that RMIV may
replicate in midgut cells.

An effective fluorescent antibody stain for RMIV
was produced and used to study the development of overt
infections with the virus. Viral antigen was detected in
larvae as early as 48 hr after the initiation of exposure.
Infection occurred in epidermis, fatbody, imaginal buds,
and tracheal epithelium. However, covert infections

118

among known transovarial transmitters of RMIV could not
be detected with fluorescent antibody staining. This
observation plus the decrease in transovarial transmission
through parents (exposed for 24 hr prior to pupation)
suggested that virus replication in pupae and adults, and
perhaps in late fourth stadium larvae, was arrested or
retarded .

The number of structural polypeptides in RMIV was
determined, and their molecular weights were estimated by
polyacrylamide gel disc electrophoresis. it was calculated
that a minimum of 14 per cent of the genome of RMIV could
code for its structural polypeptides. This information is
available for only 2 other iridescent viruses. When similar
information becomes available for other iridescent viruses,
it will be valuable in assessing the relationships among
iridoviruses and used to facilitate identification.

BIBLIOGRAPHY

Anderson, J. F. 1970. An iridescent virus infecting the
mosquito Aedes stimulans. J. Invertebrate Pathol.,
15, 219-224.

Andrewes, C. H., and Horstmann, D. M. 1949. The suscepti-
bility of viruses to ethyl ether. J. General Microbiol.
3_, 290-297.

Bellett, A. J. D. 1968. The iridescent virus group.
Adv. Virus Res., _13, 225-246.

Bellett, A. J. D., and Inman, R. B. 1967. Some properties
of deoxyribonucleic acid preparations from Chilo,
Sericesthis , and Tipula iridescent viruses. J. Mol .
Biol., 25_, 425-432.

Bellett, A. J. D., and Mercer, E. H. 1964. The multiplica-
tion of Sericesthis iridescent virus in cell cultures
from Antheraea eucalypti Scott. I. Qualitative
experiments. Virology, 2_4_, 645-653.

Chamberlain, R. W. 1968. Arboviruses, the arthropod-borne
animal viruses. Curr. Topics Microbiol. Immunol.,
42, 38-58.

Chapman, H. C, Clark, T. B., Anthony, D. W. , and Glenn,

F. E., Jr. 1971. An iridescent virus from larvae of
Corethrella brakeleyi (Diptera: Chaeboridae) in
Louisiana. j. Invertebrate pathol., 18_, 284-286.

Chapman, H. C, Clark, T. B., Woodard, D. B., and Kellen,
W. R., 1966. Additional mosquito hosts of the
mosquito iridescent virus. J. invertebrate pathol.,
8, 545-546.

Clark, H. F., and Shepard, C. C. 1963. A dialysis technique
for preparing fluorescent antibody. Virology, 20,
642-644.

119

120

Clark, T. B. 1966. A method for the inoculation of

mosquito larvae (viral and bacterial pothogens) , J.
Invertebrate Pathol., 8, 425-426.

Clark, T. B., Kellen, W. R., and Lum, P. T. 1965. A

mosquito iridescent virus (MIV) from Aedes taeniorhynchus
(Wiedemann). J. Invertebrate Pathol., 1_, 519-521.

Cunningham, J. C, and Tinsley, T. W. 1968. A serological
comparison of some iridescent non-occluded insect
viruses. J. General Virol., 3_, I-8-

Faust, R. M., Dougherty, E. M., and Adams, J. R. 1968.
Nucleic acid in the blue-green and orange mosquito
iridescent viruses (MIV) isolated from larvae of
Aedes taeniorhynchus. J. Invertebrate Pathol . , 10,
160.

Glitz, D. G., Hills, G. J., and Rivers, C. F. 1968. A
comparison of the Tipula and Sericesthis iridescent
viruses. J. General Virol., .3, 209-220.

Goldman, M. 1968. "Fluorescent Antibody Methods . "
Academic press, New York.

Hall, D. W., and Anthony, D. W. 1971. pathology of a
mosquito iridescent virus (MIV) infecting Aedes
taeniorhynchus . J. Invertebrate Pathol., 18,
61-69.

Hall, D. W., and Lowe, R. E. 1971. A new distribution
record for the mosquito iridescent virus (MIV) .
Mosquito News, 31, 448-449.

Hall, D. W., and Lowe, R. E. 1972. physical and serological
comparison of "R " and "T " strains of mosquito
iridescent virus from Aedes taeniorhynchus. J.
Invertebrate Pathol., _19, 317-324.

Hasan, S., Croizier, G., Vago, C, and Duthoit, J. J. 1970.
Infection a virus irisant dans une population naturelle
d 'Aedes detritus Haliday en France. Ann. Zool. Ecol.
Anim., 2, 295-299.

121

Holborow, E. J., and Johnson, G. D. 1967. Immunofluorescence.
Chapt . 16, pp. 571-596 in "Handbook of Experimental
Immunology." D. M. Weir, ed . Blackwell, Oxford,
England .

Kalmakoff, J., Moore, S., and Pottinger, R. p. 1972.

An iridescent virus from the grass grub Costelytra

jzealandica: Serological study. J. Invertebrate
Pathol., 20_, 70-76.

Kawamura, A., Jr. 1969. Editor, "Fluorescent Antibody
Techniques and Their Applications," University park
Press, Baltimore.

Kelly, D. C, and Tinsley, T. W. 1973. Ribonucleic acid
polymerase activity associated with particles of
iridescent virus types 2 and 6. J. Invertebrate
Pathol., 22., 199-202.

Klug, A., Franklin, R. E., and Humph ye ys -Owen , S. p. F.
1959. The crystal structure of Tipula iridescent
virus as determined by Bragg reflection of visible
light. Biochim. Biophys . Acta, 3_2, 203-219.

Linley, J. R., and Nielsen, H. T. 1968a. Transmission of
a mosquito iridescent virus in Aedes taeniorhynchus .

I. Laboratory experiments. J. Invertebrate Pathol.,
12, 7-16.

Linley, J. R., and Nielsen, H. T. 1968b. Transmission

of a mosquito iridescent virus in Aedes taeniorhynchus.

II. Experiments related to transmission in nature.
J. Invertebrate Pathol., 1_2_, 17-24.

Lowe, R. E., Hall, D. W., and Matta, J. F. 197 0. Com-
parison of the mosquito iridescent viruses (MIV) with
other insect iridescent viruses. Proc. 4th Int. Colloq.
Insect Pathol., 1970, College Park, Maryland, pp.
163-170.

Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall,
R. J. 1951. Protein measurement with the folin
phenol reagent. J. Biol. Chem., 193, 265-275.

Maramorosch, K. 1968. plant pathogenic viruses in insects.
Curr. Topics Microbiol. Immunol., 42, 94-107.

122

Maramorosch, K. , and Jensen, D. D. 1963. Harmful and
beneficial effects of plant viruses in insects.
Ann. Rev. Microbiol., 17_, 495-530.

Maizel, J. v., Jr. 1971. Polyacrylamide gel electrophoresis
of viral proteins. Chapt . 5, pp. 180-244 in "Methods
in Virology," Vol. 5, K. Mamorosch and H. Koprowski ,
eds . , Academic press, New York.

Matta, j. f. 1970. The characteristics of a mosquito

iridescent virus. II. physicochemical characteriza-
tion. J. invertebrate Pathol., 1_6, 157-164.

Matta, J. F., and Lowe, R. E. 1969. A differential stain-
ing technique for a mosquito iridescent virus.
J. Invertebrate Pothol . , 13_, 457-458.

Matta, J. F., and Lowe, R. E. 1970. The characterization
of a mosquito iridescent virus (MIV) . I. Biological
characteristics, infectivity, and pathology.
J. Invertebrate pathol . , 1_6, 38-41.

Mollenhauer, H. H. 1964. Plastic embedding mixtures for
use in electronmicroscopy . Stain Technol . , 39,
111-114.

Moss, B., and Rosenblum, E. N. 1972. Hydroxylapat ite
chromatography of protein-sodium dodecyl sulfate
complexes. J. Biol. Chem., 247, 5194-5198.

Mrenova, M., and Albrecht, P. 1966. Stabilization of

fluorescence in preparates treated by the fluorescent
antibody technique. Nature, 212, 1256-1257.

Nayar , J. K. 1967. The pupation rhythm in Aedes

taeniorhynchus (Diptera: Culicidae) . II. Ontogenetic
timing, rate of development, and endogenous diurnal
rhythm of pupation. Ann. Ent. Soc. Am., 60,
946-971.

Pease, A. G. E. 1953. "Histochemistry: Theoretical and
Applied." Little, Brown and Co., Boston.

Rogers, A. W. 1967. "Techniques of Autoradiography."
Elserier, Amsterdam.

123

Service, M. W. 1968. The ecology of the immature stages
of Aedes detritis (Diptera: Culicidae) . J. Appl .
Ecol. , 5, 613-630.

Smith, K. M. 1967. "Insect Virology." Academic press,
New York.

Smith, K. M., and Williams, R. C. 1958. Insect viruses
and their structure. Endeavor, 17, 12-21.

Stoltz, D. B. 1971. The structure of icosahedral cytoplas-
mic deoxyriboviruses . J. Ultrastruct. Res., 37 ,
219-239.

Stoltz, D. B., and Summers, M. D. 1971. pathway of in-
fection of mosquito iridescent virus. I. Preliminary
observations on the fate of ingested virus. J. Virol.,
8, 900-909.

Thurston, J. M. , and Joftes, D. L. 1963. Stains com-
patible with dipping radioautography . Stain Technol.,
38, 231-235.

Tinsley, T. W. , and Harrap, K. A. 1972. "Moving frontiers
in invertebrate virology, " Monographs in Virology,
Vol. 6, Krager, Basel, Switzerland.

Tinsley, T. W. , and Kelly, D. C. 1970. An interim
nomenclature system for the iridescent group of
insect viruses. J. Invertebrate Pathol., 16, 470-472.

Tinsley, T. W. , Robertson, J. S., Rivers, C. F., and

Service, M. W. 1971. An iridescent virus of Aedes
cantans in Great Britain. J. Invertebrate Pathol.,
18, 427-428.

Vago, C, Rioux, J. A., Duthoit, J. L . , and Dedet, J. p.
1969. Infection spontanee a virus irisant dans une
population d 'Aedes detritus (Hal. 1833) des environs
de Tunis. Ann. Parasit. Hum. et Comp., 4_4, 667-676.

Venable, J. H., and Coggeshall, R. 1965. A simplified
lead citrate stain for use in electron microscopy.
J. Cell Biol., 25, 407-408.

124

Wagner, G. W., Paschke, J. D., Campbell, W. R., and Webb,
S. R. 1973. Biochemical and Biophysical properties
of two strains of mosquito iridescent virus. Virology,
52, 72-80.

Weiser, J. 1965. A new virus infection of mosquito
larvae. WHO Bull., 33.' 586-588.

Wildy, P. 1971. "Classification and Nomenclature of
Viruses," Monographs in Virology, Vol. 3, Krager,
Basel, Switzerland.

Williams, R. C, and Smith, K. M. 1957. A crystallizable
insect virus. Nature, 17 9, 119-120.

Woodard, D. B., and Chapman, H. C. 1968. Laboratory
studies with the mosquito iridescent virus (MIV) .
J. Invertebrate Pathol., 1JL, 296-301.

Wrigley, D. G. 1969. An electron microscope study of the
structure of Ser icesthis iridescent virus. J. General
Virol., 5_, 123-134.

Wrigley, D. G. 1970. An electron microscope study of the
structure of Tipula iridescent virus. J. General
Virol., 6, 169-173.

BIOGRAPHICAL SKETCH

Stephen Carson Hembree was born September 20, 1938,
in Murphy, Cherokee County, North Carolina. His parents,
Noah Webster Hembree and Susie Ferguson Hembree, were
descendants of pioneer farming families in western North
Carolina. He graduated from high school in Murphy, North
Carolina in 1956 and entered North Carolina State Univer-
sity at Raleigh, North Carolina, where he studied civil
engineering for 2 years. fie then entered the Army and served
as an enlisted man for 3 years in the 101st Airborne Division.
In 1961 he returned to North Carolina State University hwere
he studied fishery biology and zoology and graduated with
honors from the pre-medical curriculum in 1964. He entered
graduate school at his alma mater in 1964 and completed a
Master of Science degree in zoology, with a major in
parasitology, in 1966. He re-entered the Army as a com-
missioned officer in 1966 and has since served at the
406th Medical Laboratory in Japan, at the Walter Reed
Army Institute of Research in Washington, D. C, and with
the U. S. Army Special Forces - waiter Reed Army Institute

125

126

of Research Field Epidemiology Research Team in South
Vietnam. While in South Vietnam, he was awarded 2 Bronze
Star Medals while conducting epidemiological research
among the Montagnard tribes along the Cambodian and Laotian
borders of South Vietnam. From September, 1970, until the
present time he has worked toward the degree of Doctor of
Philosophy at the University of Florida, Gainesville, Florida,
under Army sponsorship.

Stephen Carson Hembree is married to the former Jane
Van Horn and is the father of two sons, Michael and Sean.

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.

zL^

^J^\

-^Z^~^

ranklin S. Blanton, Chairman
Professor of Entomology

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.

Ronald E. Lowe, Co-Chairman
Associate Professor of Entomology

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.

/James l. Nation
Professor of Entomology

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.

Stephen G. Zam

Associate professor of Zoology

I certify that I have read this study and that in my
opinion it conforms to acceptable /standards of scholarly
presentation and is fully adequate,', —in scope and quality,
as a dissertation for the degree/pf Doctor of Philosophy.

Francis W. Zettler

Associate Professor of Plant Pathology

This dissertation was submitted to the Graduate Faculty of
the College of Agriculture and to the Graduate Council, and
was accepted as partial fulfillment of the requirements for
the degree of Doctor of Philosophy.

March, 1974

Dean, College of Agriculture

Dean, Graduate School