_ in
2.:

55.1.
a .

A. dads. v , r .
. . . ,4 J. 1
am an: :3 Hr .
-1. , .

‘ ,

m»

a...
2

.4...”

fibdv

? 3...... 5 5
. . 3‘25
3 #4.?!3.

_ 1%

i=1 .
9.44 2. Inuit:
.I 931131 IL;
(girl. .. 47: 511%.

$.05
J1. Diniyll.
. SI!!! ’1: .I.rfl
SvuIlt I1!» . I. .f. :1
1:37.13‘ 21L: . V I v}

 

 

 

 

 

 

. It:
Iv}!
. |$1tvl ..l|...‘

 

..I\tv:. al.hAtu.u
sln‘...u 111} a . .9-

...II It. a: Ali-“P9 July/91‘? I1
3|”! . .\J;:f .~:f.
If... it“: 1 -
MQI‘3L0LIII, V”. ‘ .tv

 

THES$

LIBRARY
Michigan State
University

This is to certify that the

thesis entitled

CLONING AND CHARACTERIZATION OF THE CHITIN SYNTHASE
GENE FROM THE OOMYCETE PATHOGEN PYTHIUM INSIDIOSUM

presented by

John D. Perpich

has been accepted towards fulfillment
of the requirements for

M. S . degree inClinical Laboratory

Sciences

/
flit/IV“

Vfiv

Major professor

Date [Z/lZ/af

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

PLACE IN RETURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

 

| DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDue.p65-p.15

ABSTRACT

CLONING AND CHARACTERIZATION OF THE CHITIN SYNTHASE GENE
FROM THE OOMYCETE PATHOGEN nyhium insidiosum

By

John D. Perpich

Pythium insidiosum is an organism in the kingdom Stramenopila class
Oomycetes that causes disease in mammals including humans. Current
therapeutic strategies, including anti-fungal drug treatment, have been largely
ineffective. Chitin synthase genes (CHS) have been found in the true Fungi and
in the Oomycetes and may be a potential target for drug development. The
objective of this study was to clone and characterize the CHSZ gene from P.
insidiosum (PiCHSZ). Degenerate PCR primers (Bowen 1992) were used to
amplify a conserved 600 bp region of PiCHSZ. The deduced amino acid
sequence of PiCHSZ and homologous sequences from true Fungi and other
Oomycetes were used to create a phylogenetic tree. The Oomycetes CHS
grouped together away from the true Fungi, which suggests that CHS from the
Oomycetes represent a new family of OHS genes. A biased genomic DNA
library was screened using the 600 bp fragment of PiCHSZ as a probe but failed
to yield a clone containing PiCHSZ. A second product co-amplified with the 600
bp fragment of PiCHSZ did not share any significant sequence homology with
any known CHS gene or any sequences in the GenBank database. However,
the presence of the PiCHSZ gene in the genome was supported by Southern Blot

analysis.

ACKNOWLEDGMENTS

I would like to acknowledge the people whose support has made this
thesis possible. My major professor, Leonel Mendoza, for his unwavering
support of this research project through all of its successes and failures and for
always taking the time to listen. Frances Trail who generously provided her
laboratory space and resources for a portion of this project and for serving on my
guidance committee. John Gerlach for providing additional guidance. Douglas
Estry for serving on my guidance committee. Roger Herr for playing a role in

training me in the laboratory and for very helpful advice along the way.

TABLE OF CONTENTS

 

 

 

 

 

 

 

 

LIST OF FIGURES ........................................................................ vii
INTRODUCTION ........................................................................... 1
1.1 History of Infections Caused by Pythium insidiosum ......................... 1
1.2 Life Cycle of Pythium insidiosum .................................................. 2
1.3 Pythiosis insidiosi in Humans and Animals ..................................... 5
Clinicaflesentgtion .............................................................. 6
Diagnosis ............................................................................ 7
Treatment ........................................................................... 9
1.4 Chitin and Chitin Synthase ......................................................... 14
Chitin: Polv N-acetvlgluccgamine ............................................. 14
Chitin §vntha_se Genes (QHS) ................................................ 15
Inhibition of CHS Enms .................................................... 16
§_tu_dy of CHS from the Oomycetes .......................................... 18
QHS thlogenv ................................................................... 20
1.5 Objectives .............................................................................. 21
MATERIALS AND METHODS ......................................................... 24
2.1 DNA Extraction ........................................................................ 24
2.2 Polymerase Chain Reaction (PCR) .............................................. 25
2.3 Southern Hybridization .............................................................. 26
2.4 Construction of a Biased Genomic DNA Library .............................. 28

2.5 Screening of the P. insidiosum Biased Genomic DNA Library
(Plaque Lifts) ............................................................................... 29

2.6 PCR Screening of the Putative Positive Clones Isolated by Plaque

Lifts ............................................................................................ 30
2.7 DNA Isolation from Putative Positive Lambda Phage Clones ............ 30
2.8 Isolation of the Insert DNA from the Lambda Phage DNA ................. 32
2.9 Sub-cloning of the Insert DNA into the pZErOTM Plasmid Vector... 32

2.10 Isolation of Plasmid DNA from Bacterial Clones Harboring pZErOTM.. 32

2.11 Nucleotide Sequencing ............................................................ 33
2.12 Engineering a Specific PiCHSZ Probe ......................................... 34
2.13 Southern Blot Using the Purified 500 bp PiCHSZ Probe .................. 34
2.14 Screening Putative Positive Clones from the Biased Library by Dot

Blot Hybridization .......................................................................... 35
2.15 Screening a Putative Positive Clone Obtained by Dot Blot

Hybridization ................................................................................ 35
RESULTS ................................................................................... 37
3.1 Amplification of a 600 bp Fragment of PiCH82 ............................... 37
3.2 Southern Hybridization Using Genomic DNA from P. insidiosum ........ 37
3.3 Constructing a Biased Genomic DNA Library ................................ 37
3.4 Screening the Library ............................................................... 41

3.5 Purification and Sub-cloning of the DNA Insert from a Putative Positive
Phage Clone ............................................................................... 45

3.6 Sequencing the Putative PiCHSZ Molecule and the 600 bp Probe..... 49
3.7 Southern Blot Using Each Probe Separately ................................. 49
3.8 Dot Blot of Nine Clones from the Primary Screening of the Library ..... 53
3.9 Screening the Entire Library with the 500 bp PiCHSZ Probe .............. 56
DISCUSSION .............................................................................. 58

FUTURE DIRECTION .................................................................... 64

BIBLIOGRAPHY ........................................................................... 65

vi

LIST OF FIGURES

Figure 1: CH8 Phylogenetic Tree ......................................................... 22
Figure 2: Chitin Synthase PCR ............................................................ 38
Figure 3: Southern blot analysis ........................................................... 39
Figure 4: Screening the Mixture of Genomic DNA Fragments Prior to

Cloning ........................................................................................... 40
Figure 5: Primary screening of the biased library ..................................... 42
Figure 6: Secondary screening of the biased library ................................. 43
Figure 7: Tertiary screening of the biased library ..................................... 44

Figure 8: Not I Digestion and Blot of the DNA insert from a Putative Positive
Clone ............................................................................................. 46

Figure 9: PCR Screening of Plasmid Clones carrying either the 4 kb, 7.5 kb
or 11 kb insert from the Not I digestion (see figure 8) of the lambda phage
clone ............................................................................................. 48

Figure 10: Sequence of Both Products Obtained by PCR with the CH8
Primers .......................................................................................... 50

Figure 11: Sequences of one of the Bowen primers and primer binding
regions of PiCHSZ and the unknown sequence ..................................... 52

Figure 12: Southern blot using 3 different probes and 3 identical blots ........ 54

Figure 13: Dot blot of nine putative positive clones from the primary
screening of the biased library ............................................................ 57

vii

1. Introduction
1.1 History of Infections Caused by Pythium insidiosum

Pythium insidiosum is a microorganism classified in the kingdom
Stramenopila, phylum Pseudofungi, class Oomycetes, order Pythiales and family
Pythiaceae. P. insidiosum is the causative agent of a disease termed pythiosis
insidiosi, which is characterized by the formation of cutaneous granulomas in
infected animals. The history of pythiosis is reviewed by Mendoza (1998). The
first documented cases of equine cutaneous granulomas linked to pythiosis
occurred during the 19th century in India. Smith in 1884, Fish in 1895 and Drouin
in 1896, however, first suggested its fungal-like etiology. But they did not isolate
the etiologic organism. The organism was first isolated by de Haan and
Hoogkamer in 1901 who were working with equines with cutaneous granulomas
in Indonesia. They named the disease hyphomycosis destruens equi. The
identity of the organism remained obscure and this uncertainty was compounded
by the fact that pythiosis insidiosi was clinically confused with other equine
cutaneous diseases, specifically, a skin condition caused by species of the
genus Habmnema. During the 20th century Bridges and Emmons isolated an
organism from Texan horses with cutaneous granulomas. They believed it was
a fungus similar to Mortierella. Austwick and Copland induced the formation of
biflagellate zoospores upon transfer to an aqueous medium containing rotten
maize silage (Austwick 1974). Based on their findings they suggested the genus
Pythium was a more appropriate classification for this organism. lchitani and

Amemiya induced oogonia with aplerotic spores and suggested the species

Pythium gracile. Later deCock et al., working with several isolates from different
geographic locations, concluded that this agent was a single species and
Pythium insidiosum was introduced (de Cook 1987). Today, the disease remains
an important cause of economic loss for horse farmers, a significant killer of
humans in tropical and subtropical third world countries like Thailand and an

important cause of death in dogs and cats.

1.2 Life Cycle of Pythium insidiosum

Organisms of the genus Pythium are important plant pathogens and some
species are also pathogens of fishes. However, P. insidiosum is the only
species of the genus that causes disease in mammals. Pythium insidiosum has
a worldwide distribution in temperate, sub tropical and tropical areas.

The life cycle of P. insidiosum in nature was originally studied by Miller
(1983) and involves the growth of hyphal elements on plant tissue, the formation
of sporangia followed by the release of motile biflagellate zoospores. Miller was
able to show that the zoospores of P. insidiosum swam toward the hairs of all
mammals tested and even suggested that they swam toward the follicle of the
hair (1983). In this experiment the zoospores changed from a random path to a
continuous line to the hairs. The same experiment was performed using leaves,
stems and roots of the Australian water lily (Nymphaea gigantean), the water
hyacinth (Eichhomia crassipes) and salvinia weed (Salvinia molesta). Of all
plants tested, the zoospores showed affinity for the Australian water lilies and

displayed a similar change of direction as previously described. Miller proposed

that P. insidiosum grows as hyphae on water lilies where it forms sporangia and
zoospores. The zoospores are released and swim in the swamp until
encountering either another plant host, in which case the cycle is repeated, or
the tissue of an injured animal where they germinate and causes pythiosis.

The model proposed by Miller, although plausible, does not address some
important issues such as the mechanisms for the colonization and germination of
the zoospores in an animal host. Mendoza et al. studied the phenomenon of
chemotaxis in P. insidiosum and proposed a more detailed model of its life cycle
(1993). Using a scanning electron microscope, they not only confirmed the
chemotactic behavior of the zoospores, but they also showed that, upon
reaching the tissue, the zoospores lost their flagella, encysted and then secreted
an amorphous substance over their surface. The encysted zoospores, using the
amorphous substance, attached themselves to the new host and then emitted a
germ tube. Mendoza et al. also found that the germ tubes were always oriented
toward the infected tissue. This observation was in agreement with similar
findings in other Pythium species and in other Protoctistans, which also secreted
a similar amorphous material (Mendoza 1993). Mendoza et al. proposed that
chemotactic signals, produced by the plant and the animal tissues, direct the
zoospores to swim toward the new host, secrete the amorphous substance and
orient their germ tubes toward the tissue. The zoospores in their control
experiment, which contained no plant or animal tissue, swam in a random
manner, finally encysted, produced a germ tube but did not secrete the

amorphous substance. Mendoza et al. proposed that secretion of the

amorphous substance aids the zoospores in attachment to the host's tissue and,
therefore, could be an important virulence factor. Finally, they asserted that the
zoospores are more likely the infectious propagule. Based on their findings they
proposed a model similar to the one proposed previously by Miller. They
suggested that P. insidiosum first colonizes a plant, grows in hyphal form and
then differentiates into sporangia with zoospores. The zoospores, attracted by
chemotactic signals, swim toward a new plant or an injured animal, encyst,
secrete an amorphous substance, attach to the host tissue and produce a germ
tube that penetrates the host’s tissue. If the new infected host is a plant, the life
cycle repeats, however, if the new host is an animal, the zoospores germinate
causing pythiosis.

The models proposed by these two groups of investigators are both
essentially the same on the most basic level, however, the Mendoza et al. model
illustrates the life cycle at higher resolution. Their model shows an association
between the production of the amorphous substance and the presence of animal
tissue. This association suggests that the amorphous substance is involved in
the virulence of P. insidiosum, however, the study by Mendoza et al. fails to show
the molecular events involved in the attachment and penetration of the infected
tissues. Furthermore, to prove that this substance is truly involved in the
virulence of P. insidiosum, chemical identification of the amorphous substance
followed by mutation of the genes involved in its synthesis would be required. To

achieve this goal, an animal or plant model of the infection would be of great

value to assess the virulence of P. insidiosum in isolates that express or lack this

substance.

1.3 Pythiosis insidiosi in Humans and Animals

Infections caused by P. insidiosum occur in temperate, subtropical and
tropical areas of the world with cases being reported in Argentina, Australia,
Brazil, Colombia, Costa Rica, Haiti, India, Indonesia, Japan, New Guinea, New
Zealand, Thailand and the United States (Mendoza 1998). Horses, dogs, cats
and humans are all susceptible to the infection, which usually occurs after
exposure to stagnant water containing zoospores of P. insidiosum. However,
some reported cases of animal and human pythiosis without exposure to
stagnant water suggest that infection can be induced by other propagules of P.
insidiosum (Mendoza 1998).

Three main forms of human pythiosis have been described:
1) ophthalmic which includes keratitis and corneal ulcer, 2) subcutaneous
lcutaneous and 3) systmeiclvascular (T hianprasit 1996). Predisposing factors
such as or-thalassemia, B-thalassemia, hemoglobin E and paroxysmal nocturnal
hemoglobinuria have all been incriminated in reported cases of systemic and
cutaneous human pythiosis in Thailand (T hianprasit 1996). Patients infected
with P. insidiosum did not receive deferoxamine therapy, so some Thai clinicians
believed that iron overload might contribute to the mechanism of infection
(Wanachiwanawin 1993). However, based on data from experimental rabbit

pythiosis and reports of human pythiosis in Australia, Haiti, New Zealand and the

USA, the infection does not appear to be associated with impaired immune
function (Miller 1983). Thus, the high incidence of Thalassemia in Thailand
could better explain why human pythiosis is always associated with this disorder.
Based on reports of pythiosis in animals and humans, it has been shown that
pythiosis occurs in apparently healthy hosts (T riscott 1993, Shenep 1998,
Thitithanyanont 1 998).

Clinical Presem

The cutaneous form of the disease is commonly observed in the lower
parts of the limbs but can occur in virtually any anatomical area of body that has
been exposed to contaminated swamp water in the endemic areas. This form of
the disease is common in horses, however, there have been at least two cases
of human subcutaneous pythiosis in Thailand affecting the extremities
(T hianprasit 1996). The clinical manifestation of the disease in horses is
characterized by tumor like lesions with necrotic tissue and yellow-white coral like
masses known as “kunkers” (Mendoza 1986). These granulomatous cutaneous
lesions are 5mm to 500mm in diameter and have a characteristic
serosanguineous discharge. Intense pruritus of the infected areas is also
characteristic of the disease in horses (Mendoza 1998).

Patients with systemic arterial pythiosis usually develop progressive
gangrene and pain in the extremities. Clinically, progressive ascending arteritis
with thrombosis and aneurysms of the main arteries is observed (T hianprasit
1996). After exposure and infection due to P. insidiosum, the duration of the

disease varies from 1-6 months before diagnosis or death (Thianprasit 1996).

The ophthalmic form of the disease rarely involves patients with
hemoglobinopathies. This form occurs in otherwise healthy people usually after
handling infected plants. Corneal ulcers and keratitis are typical clinical findings
and both are indistinguishable from eye infections caused by fungi. Patients
often end up with keratoplasty, evisceration or enucleation (lmwidthaya 1994).

Diagnosis

Pythiosis is difficult to diagnose based on clinical features alone.
Cutaneous forms of the disease are often confused with infections caused by
zygomycetes such as Basidiobolus ranarum and Conidiobolus coronatus.
Microscopic examination of clinical material from the infected areas is the most
common means of diagnosing pythiosis. The coral-like kunkers in equine
pythiosis are the ideal clinical material for laboratory testing. The masses should
be cut into small pieces and placed in 10% KOH and then examined using light
microscopy (Mendoza 1998). Humans with pythiosis rarely develop kunkers, so
tissue biopsies, pus or tissue debris should be used for microscopic examination.
The presence of coenocytic hyphae from infected areas 4.0-9.0um in diameter
with branches at about 90° angles is indicative of pythiosis. The production of
biflagellate zoospores is an important diagnostic tool and can be performed
within 24 hours. Once biflagellate zoospores are induced, one can rule out the
confusing possibility of an infection caused by a zygomycete.

Histological examination of tissue sections from infected areas can also
be used to diagnose infections caused by P. insidiosum. Coenocytic hyphae

with branches at 90-degree angles surrounded by eosinophils is a typical

histopathological feature of pythiosis. However, these tissue sections can easily
be mistaken for those found during infections caused by organisms in the fungal
phylum Zygomycota, specifically B. ranarum and C. comnatus, which also cause
cutaneous granulomas in horses. Thus, other methodologies have been
developed to differentiate these etiologic agents.

Serological tests have proven to be the most reliable way to diagnose
infections caused by P. insidiosum. The enzyme-linked immunosorbent assay
(ELISA) has been used to diagnose pythiosis in horses. The test proved to be
specific and sensitive in dogs, humans and horses with active pythiosis
(Mendoza 1998). This test employs dilute culture filtrate antigen (CFA) and anti-
horse lgG or lgM. Some horses in endemic regions without any other signs of
disease were found to be positive in this test, suggesting that sub-clinical
infections may occur (Mendoza 1998).

Immunohistochemical assays including immunofluorescence (Mendoza
1987) and immunoperoxidase (Brown 1988, Triscott 1993) tests have been used
to identify P. insidiosum hyphae in tissue sections. The immunofluorescence
test is specific for P. insidiosum in infected cats, dogs and humans, but uniform
fluorescence among tissue sections of the “kunkers” was detected in equine
pythiosis (Mendoza 1998). The immunoperoxidase test was able to distinguish
P. insidiosum in tissue sections from the troublesome zygomycetes B. ranarum
and C. coronatus (Triscott 1993).

The complement fixation test was also used in the past; however, this test

detects only 82% of horses infected with P. insidiosum. This test is also prone to

false positives, specifically, in sera from cured patients or from apparently
healthy horses (Miller 1982). These facts combined with the difficulty in
performing the test have precluded its use as a diagnostic tool.

Finally, several laboratories have used the immunodiffusion test (ID) with
success. The ID test is by far the best serological test for diagnosis. Its success
is due to its economical features and also its simplicity to perform and interpret.
Although it is insensitive, its specificity is remarkable. The test uses
concentrated CFAs from P. insidiosum (Mendoza 1986). This test did not show
precipitin bands using sera from horses recovering from infection after two
months of successful treatment. Further, the test showed no cross-reaction
using sera from patients with conidiobolomycosis, habronemiasis or horses with
excessive granulomatous tissue (Mendoza 1986). This test is also useful in
monitoring response to treatment and may be applied to cats, cattle, dogs,
horses and humans with pythiosis (Mendoza 1998).

In summary, the diagnosis of pythiosis can be accomplished using direct
microscopic examination of clinical samples from infected sites followed by a
confirmatory ID test as well as other serological tools. In well-equipped
laboratories the histochemical assays can provide a reliable alternative to the
immunodiffusion and ELISA tests. In all cases care should be taken to avoid
misdiagnosis of cutaneous zygomycosis.

Treatment

Surgical removal of lesions caused by P. insidiosum has been the main

treatment in horses and humans. This procedure is not practical since most

lesions occur on the limbs and this area, at least in horses, possesses a delicate
anatomical structure. Furthermore, it is very difficult to remove all of the infected
tissues and reoccurrence of the disease is common (McMullan 1977, Alfaro
1990). Amputation to treat pythiosis has been attempted in humans with the
vascular form (Thianprasit 1996). Although some patients responded to this
form of treatment, many did not. Patients that do not respond to amputation
often die of disseminated abdominal pythiosis (Mendoza 1998). Nonetheless,
surgery is one of the options available to treat pythiosis in horses and humans.
A combination of surgery with other forms of treatment has been attempted. For
example, two horses with subcutaneous pythiosis were treated by surgical
removal of the granulomatous masses on their legs. The sites were then treated
with a neodymium:yttrium-aluminum-gamet laser and the horses were given
trimethoprim-sulfa-methoxazole and phenylbutazone for 2 weeks after surgury
(Sedrish 1997). One year later both horses were disease free and had minimal
scarring.

In summary, surgery has limited use in cases of systemic pythiosis and
amputation is often required. Reoccurrence of the disease is common following
surgical treatment of cutaneous forms due to incomplete remoVal of all of the
infected tissue. However, surgery followed by drug therapy or laser
cauterization, has been used successfully to treat cutaneous forms of the
disease.

The most significant advance in the treatment of pythiosis was the

development of a curative vaccine in 1981. Miller first used an

10

immunotherapuetic vaccine consisting of sonicated hyphal antigens to cure
horses with pythiosis (1981). Miller achieved a 53% cure rate after vaccination
alone and a 33% cure rate using the vaccine after surgery (n=40: 30 horses
vaccine only, 10 horses surgery followed by vaccine). Miller suggested that the
failure of the vaccine might be due to the age of some of the lesions or the age
of the vaccine preparation. Later, Mendoza developed a new version of the
vaccine made of precipitated proteins from CFAs (1986). This vaccine gave
results similar to one Miller tested in 1982 but it had a longer shelf life and fewer
side effects. Mendoza et al. noted that horses with chronic pythiosis (lesions >2
months old) did not respond to the vaccine. Overall, horses with chronic
pythiosis all died in spite of receiving the vaccine, whereas, horses with acute
pythiosis were all cured after vaccination (Mendoza 1992). Horses that were
cured by immunotherapy all developed a local inflammatory response at the site
of the injection. Horses that were not cured by the vaccine did not respond at
the site of the injection. These findings strongly suggest that the success of
immunotherapy could be directly related to the immune status of the horse
(Mendoza 1998).

In addition to its usefulness in horses, there is one documented case
where the vaccine was successfully used to cure human pythiosis and at least
seven new human cases from Thailand (Thitithanyanont 1998, Mendoza
unpublished data, personal communication). The first case treated with the
vaccine was a boy from Thailand with chronic disseminated arterial pythiosis.

The vaccine was given as a last resort. He progressed over a 4-week period

11

until all symptoms of the disease virtually disappeared. Two years after the
vaccination, the boy was still in good condition and free of pythiosis. The other
seven cases also responded in the same fashion. The vaccine does not appear
to provide any lasting immunity to the disease.

Based on histopathological and serological findings in cases of vaccinated
and cured humans and animals, a potential mechanism of action for the curative
properties of the vaccine has been proposed (Thitithanyanont 1998). It was
found that, in cured cases, the classic eosinophilic inflammatory reaction
gradually changed to a mononuclear response with macrophages and possibly
cytotoxic T lymphocytes (CTLs). This switching of the response indicated this is
an example of a possible immunomodulation from a Th2 (eosinophils and lgE) to
a Th1 response (mononuclear cells). While the immunotherapuetic vaccine
remains an exciting new treatment option, further investigation on the
mechanism of its curative properties should be pursued before its
commercialization. Although the vaccine is of value in acute cases, it is of little
use in chronic equine pythiosis presumably because horses become anergic and
unable to respond to the stimulation provided by the immunogens in the
vacccine.

Drug therapy is another therapeutic option that has been used over the
years to treat pythiosis in animals and humans. Many traditional anti-fungal
compounds target ergosterols, however, this compound is not present in the cell
membrane of P. insidiosum. Interestingly, some investigators have reported

success rates up to 80% using amphotericin B, a drug that targets ergosterols, to

12

treat pythiosis in horses (McMullan 1977). Strikingly, since that report, other
reports have indicated that Amphotericin B is ineffective for treating pythiosis
(lmwidthaya 1994, Thianprasit 1996). In addition, amphotericin B has no effect
on P. insidiosum in vitro and it is highly toxic to the host. Potassium iodide is
another compound that has been used to treat equine pythiosis with
contradictory results. Although some reports indicate that potassium iodide or
sodium iodide is effective against the pathogen (Gonzales 1979, Hutchins 1972),
other reports indicate these drugs are not effective (Mendoza 1998, Murray
1978). In addition to their low cure rate these drugs are toxic the host.

In human pythiosis, amphotericin B was ineffective in curing systemic
pythiosis, the most common type (Wanachiwanawin 1993, Thianprasit 1996,
Sathapatayavongs 1989). A combination therapy using terbinafine and
itraconazole was effective in treating one case of human ophthalmic pythiosis
(Shenep 1998). In two cases of subcutaneous pythiosis a combination of
amphotericin B and 5-flurocytosine was effective (Triscott 1993). Additionally,
some studies indicate that a saturated solution of potassium iodide can cure
subcutaneous human pythiosis (Thianprasit 1996, Mendoza 1998). However,
other reports have suggested that potassium iodide is ineffective (Mendoza
1998). A combination of amphotericin B, flucytosine and hydrocortisone was
successful in treating two cases of subcutaneous human pythiosis in Australia
(T riscott 1993). Although drug therapy has proven to be feasable in some cases
of subcutaneous pythiosis, it is ineffective in cases of systemic human pythiosis.

Unfortunately, many clinicians resort to amputation above the infected site for

13

the treatment of systemic pythiosis (Thianprasit 1996). In summary, human
subcutaneous and ophthalmic pythiosis have been treated successfully with
drugs in a few cases, however, drug therapy is not effective in treating systemic
or vascular pythiosis.

Drug therapy for pythiosis applies anti-fungal compounds to a disease that
is not caused by a fungus, consequently, the drugs give contradictory results.
This is most likely because the targeted compounds of the drugs are not present
in P. insidiosum. Therefore, new drugs with a broader spectrum of action should
be developed in order to treat non-fungal diseases like pythiosis. New targets
common to fungal and fungal-like organisms should be explored. Some
potential targets for this type of therapeutic strategy are the cell wall components

of P. insidiosum including the glucan and chitin synthase enzyme systems.

1.4 Chitin and Chitin Synthase

_C_h_it_in: Polv N-acetvlglucosamine

Chitin is an unbranched polysaccharide widely distributed in the Kingdom
Fungi. It has also been found in animals (insects and crustacea), plants (algae)
and protoctista. In fungi and fungal-like microorganisms chitin is part of the cell
wall and gives strength and rigidity to the cell. Chitin is composed of repeating
units of N-acetyl-D-glucosamine (GIcNAc). A single strand of chitin is estimated
to be between 1800 and 2300 residues long (Ruiz-Herrera 1992). Individual
chitin strands associate with other strands by hydrogen bonding and non-bonded

interactions between pyranose rings in arrays of sheets or chains (Ruiz-Herrera

14

1992). Chitin is classified as alpha, beta or gamma according to how the strands
interact with each other. Like DNA strands, chitin strands have polarity, which is
based on their reducing and non-reducing ends. Alpha chitin is the only form
identified to date in true Fungi.

Chitin Synthase Genes (CH8)

The cell wall of most fungi is composed, in part, of the polysaccharides
chitin and glucan, which provide rigidity and strength to the cell. It has been
known for quite some time that Oomycetes possess cellulose and glucan in their
cell walls. More recently, however, it was reported that some Oomycetes also
contain chitin. There is evidence that chitin and glucan are structurally linked in
the cell wall, a fact that may be important in planning therapeutic strategies (El-
Sherbeini 1995, Kollar 1995). More importantly, the enzymes responsible for the
synthesis of these two polymers, glucan and chitin synthases, are inhibited by
the echinocandins and nikkomycins, respectively. This finding suggests that the
cell wall constituents might be good targets for the development of new anti-
fungal drugs.

Several unique genes or CHS genes encode the chitin syntheses. Much
of what is known about CHS comes from studies of organisms in the Kingdom
Fungi. True Fungi typically have multiple isoforms of chitin synthase, each with a
different role in the cell. In yeast, such as Saccharomyces cerevisiae, at least
three isoforms of CHS have been found (Bulawa 1993). CHS1 and CHSZ are
both activated by protease treatment, whereas, CHS3 is not (Bulawa 1993).
CHS1 and CHSZ together produce only 10% of cellular chitin while CHS3 is

15

believed to make up the remaining 90%. Using mutagenic studies, it was found
that CHS1 is required for normal budding while CHSZ is required for normal
morphology, septation and cell separation (Bulawa 1993). CHS3 mutants are
resistant to Calcofluor and KIuyveromyces Iactis killer toxin. They lack alkali-
insoluable glucan, and are, in some cases, temperature sensitive for growth
(Bulawa 1993). In filamentous fungi, like Asperyillus nidulans, four CHS genes
have been described so far. These CHS genes are involved in conidia formation
and growth at the apical portion of the hyphae (Borgia 1996, Motoyama 1997).
In general, these isozymes have redundancy and multiple CHS genes must be
disrupted in order to perturb the growth of the fungus.

Inhim of CHS Enzymes

The problems associated with current anti-fungal drugs including limited
effectiveness, toxicity and cost have prompted interest in finding new drugs with
targets more specific to the pathogens. The cell wall has proven to be a very
effective target in prokaryotes and is now being investigated as a potential target
in fungal and para-fungal pathogens.

The echinocandins, inhibitors of glucan synthase, are currently in clinical
trials. They proved to be effective in murine models infected With Candida sp.,
Aspergillus fumigatus and Histoplasma capsulatum (Powles 1998). An
echinocandin class anti-fungal drug was fungicidal to Candida albicans in an in
vitro study (Gordee 1988). This compound inhibited glucose incorporation into [3-
glucan in about 75-95% of treated fungal cells. A semi-synthetic pneumocandin

was also effective in vivo against the fungus Pneumocystis carinii in a rat model

16

of the infection (Powles 1998). Echinocandins are limited by their low oral
absorption, so they need to be applied parenterally until perfected for oral use.
The anti-fungal drugs that target glucan synthase could also be applied to
patients infected with P. insidiosum, which also contains glucan.

Another class of compounds, the nikkomycins, have been explored as
potential chemotherapeutic agents. Although the nikkomycins showed
effectiveness in vitro against some of the pathogenic dimorphic fungi (Hector
1990), they are no longer in clinical trials (Georgopapadakou 2001). Efficiency of
transport into fungal cells and susceptibility to degradation may account for their
limited clinical use (El-Sherbeini 1995). In spite of this, the current nikkomycin
anti-microbials may find application in dual therapy with echinocandins. A
combination of a nikkomycin and an echinocandin was effective in vitro against
Asperyillus fumigatus, Coccidioides immitus (Stevens 2000) and Candida
albicans (Hector 1986). One study indicates that fungi may compensate for
inhibition of synthesis of one polymer by up regulation of another polymer
(Stevens 2000). In another study, it was found that the mycelial phase of
Paracoccidioides brasiliensis synthesized more a-glucan in response to
treatment with papulacandin B, an inhibitor of B-glucan synthesis (Hector 1990).
These studies suggest that dual inhibition of glucan and chitin synthase may
enhance the therapeutic value of drugs that target the cell wall components.

Although current chitin synthase inhibitors have not proven effective in
treating fungal infections, it seems that the short half-life and delivery of the drug

to the fungal target might explain the problems. If these pharmacological

17

problems could be solved, chitin synthase inhibitors may find application alone or
in concert with other cell wall inhibitors. Therefore, chitin synthase remains an
unexploited target. Characterization and comparison of the cell wall components
of fungal-like organisms to true fungi may be important for the evaluation of new
anti-fungal drugs for the treatments of infections caused by fungal-like organisms
such as P. insidiosum.

Study of CHS from the Oomycetes

The Oomycetes, including P. insidiosum, are a group of organisms in the
kingdom Stramenopila that are morphologically, but not phylogenetically, related
to organisms in the kingdom Fungi. These organisms are protists near the
plants and the green algae (Kwon-Chung 1994). Due to their morphological
features, the Oomycetes were at one time considered to be members of the
kingdom Fungi. Before phylogenetic analysis, the Oomycetes and the true fungi
were separated into two groups based on their cell wall polymers. The
Oomycetes were placed into the cellulose-glucan group and the fungi into the
chitin-glucan group (Bartnicki-garcia 1968).

During the middle of the 20th century, several investigators explored the
possibility of chitin in some Oomycetes (Thomas 1942, Frey 1950, Cooper 1967,
Lin 1970, Cherif 1992, Bulone 1992, Mort-Bontemps 1997). Thomas in 1942
first proposed the idea of chitin and cellulose in the Oomycete Pythium
debaryanum. However, Frey (Frey 1950), who found only cellulose in this
Oomycete, could not replicate his work. Using X-ray diffraction and working with

Apodachlya sp., Lin and Aronson revealed the presence of chitin in this

18

Oomycete (1970). Using a wheat germ agglutinin-ovomucoid-gold complex
(WGA), a lectin that binds to GlcNAc, Cherif et al. found chitin in Pythium
ultimum (1992). However, it is well known that the WGA assay is not entirely
specific for chitin and that it has weak affinity for N-acetylneuraminic acid, N-
acetylgalactosamine and Man-B-(1-4)-GlcNAc-B-(1-4)-GlcNAc-B-N-Asn (Bulawa
1993). Later, Bulone et al. were able to demonstrate the presence of chitin in
the Oomycete Sapmlegnia monoica using several assays. They demonstrated
chitin in the cell wall using X-ray diffraction, electron diffraction and infrared
spectroscopy. They obtained a residue hydrolysable by chitinase, an enzyme
that degrades chitin. They also detected a ‘chitin synthase activity’ from
membrane fractions of the mycelium of this oomycete and showed that the
activity was enhanced after treatment with trypsin, a protease that activates
chitin synthase in vitro. In addition, the activity was decreased by addition of
polyoxin D, a competitive inhibitor of chitin synthase. They also detected chitin
synthase activity in regenerated protoplasts. These investigators noted that
although chitin synthase was present in S. monoica, the levels of chitin were low
relative to most fungi.

Later, Mort-Bontemps at al. (1997) demonstrated the existence of chitin
synthase genes in the Oomycete Saprolegnia monoica for the first time. This
was the first chitin synthase gene ever cloned from an Oomycete. This group
used a pair of degenerate primers, designed by Bowen et al. (1992) to amplify a
conserved 600 bp region from fungal chitin synthase genes, to amplify

homologous regions in three different oomycetes: S. monoica, Phytophthora

19

capsici and Achlya ambisexualis. They cloned and characterized the entire
CHSZ gene from S. monoica. In addition, they also detected a chitin synthase
activity from the mycelium of S. monoica and demonstrated the expression of
chitin synthase transcripts from mycelium and regenerated protoplasts using RT-
PCR. As part of this work, Mort-Bontemps et al. constructed a phylogenetic tree
using the deduced amino acid sequences of CHSZ from S. monoica, and the
fungi Rhizopus oligosporus, Candida albicans, Aspergillus nidulans,
Saccharomyces cerevisiae and Neurospora crassa. They suggested that the
fungi and Oomycetes, “despite having a divergent evolution, have evolved with
conserved chitin synthase systems" (Mort-Bontemps 1997). However, their
analysis was limited by the fact that only one Oomycete was used for
comparison. In addition, their results clearly show that CH82 from S. monoica
segregates away from fungal CH8 genes, which formed a compact group.

C_H§£hvlggemr

Later, Perpich et al. (1999), using the same set of degenerate primers,
amplified a 600 bp fragment of the CHSZ gene from the Oomycete P.
insidiosum. They used the deduced amino acid sequences of the conserved
600 bp region from four Oomycetes and several fungi, rather than the entire
CHS molecule, to construct a phylogenetic tree. Their phylogenetic analysis
(figure 1) showed that the Oomycetes grouped together away from the fungal
CHS. In addition, the Oomycetes, like the fungi, appear to be further divided into
clades composed of CHS1 and CHSZ classes. These results suggested that

CHS from the Oomycetes represents a new family of CHS genes distinct from

20

the fungal CHS. This finding also agrees with the current taxonomical placement

of the Oomycetes in a separate kingdom from the true fungi.

1.5 Objectives
The general objective of the current study was to establish the presence
of chitin synthase genes in P. insidiosum. The original specific goals of this
study were:

- Cloning and sequence characterization of CHSZ from P. insidiosum
(PiCHSZ).

- Determine the number of CHS isozymes present in P. insidiosum.

- Phylogenetic analysis using the deduced amino acid sequence of PiCHSZ.

21

Figure 1

CH8 Phylogenetic Tree. The tree was constructed using deduced amino acid
sequences from several fungal chitin synthase genes and known Oomycete
sequences: Aa=Achlya ambisexualis, Pc=Phytophthora capsici,
Sm=8aprolegnia monoica and Pi=Pythium insidiosum. The fungi grouped into
clades as CHS1, CH82, CHS3 or other isoforms of CH8. The Oomcyetes
(bottom of the tree) grouped together away from the true fungi, which suggests
that the Oomycetes CH8 might be a new family of CH8 genes. The phylogram
was generated using PAUP (Phylogenetic Analysis Using Parsimony) with 500

bootstrap replicates.

22

Figure 1

100

 

CiCHSl
79

 

MgCHSl
r— BdCHSl

62
9O 8 1

 

 

 

 

 

 

94 —— HcCHSl

 

 

99

 

 

 

 

 

WdCHSZ
PzCHSZ
— HcCHS3

 

 

 

79

 

 

100 —— PbCHSZ

 

 

PeCHSZ

 

100 98

 

 

 

—— FpCHSZ
SsCHSZ
—— EjCH83

 

 

52
68

 

 

100

 

 

—— HcCHS2

 

 

 

NcCHS3

SmCHSl

 

84

 

 

‘00 PcCHSl

 

 

 

 

AaCHSl

 

 

SmCHSZ

 

PiCHSZ

23

2. Materials and Methods
2.1 DNA Extraction

Pythium insidiosum (ATCC #200269) was subcultured in Sabouraud broth
for 5 days at 37°C while shaking at approximately 150 rpm. The culture was
killed by the addition of merthiolate (0.05% final concentration). The hyphae
were concentrated using a vacuum pump and filter flask system. The hyphal
mat was washed with sterile water and placed into a pre-cooled mortar. The mat
was then ground under liquid nitrogen using a pestle and reduced to a fine
powder. Microfuge tubes were filled half way up the conical portion with the
powder and 600uL of lysis buffer (5 mM Tris-HCI (pH 7.2), 5 mM EDTA (pH 8.0),
0.9% SDS, 0.01% 2-mercaptoethanol) was added. The mixture was
homogenized using a syringe and a 20 G 1 ‘/2 needle and then centrifuged at
maximum speed in a microcentrifuge. After centrifugation, 250p.L of the
supernatant was placed into a fresh tube and 250pL of lysis buffer was added.
Proteinase K (200uglmL final concentration) was added and the mixture was
incubated at 55°C for 1 hour. Following incubation, 500uL of Phenol:
Chloroforrn: isoamyl alcohol (25:24:1) was added and the tube mixed until an
emulsion formed. The mixture was centrifuged for 15 minutes at maximum
speed at room temperature. Following centrifugation, 400p.L of the supernatant
was placed into a fresh tube and 10p.L of 3M sodium acetate followed by 2
volumes of ice cold 100% ethanol were added. The tube was inverted gently
several times. After an overnight incubation at -20°C, the tube was centrifuged

at maximum speed at room temperature for 5 minutes. One mL of 70% ethanol

24

was added to the resulting nucleic acid pellet. The tube was centrifuged at
maximum speed at room temperature for 2 minutes. The 70% ethanol was
removed from the tube and the pellet was air dried for 10 minutes and then
resuspended in 50-100pL of sterile water depending on the downstream
application of the DNA.

The RNA was removed by the addition of 30 units of DNase free RNase1
(Promega, Madison, WI) and buffer to a 1X working concentration and the
mixture was incubated for 1 hour at 37°C. Following this incubation, the mixture
was extracted with an equal volume of phenol: chloroform: isoamyl alcohol
(25:24:1), precipitated and washed following the technique above for the nucleic

acid isolation.

2.2 Polymerase Chain Reaction (PCR)

The Polymerase Chain Reaction was prepared on ice in a total reaction
volume of 25pL containing the following (final concentration): 12.25uL H20, 1X
Geneamp® PCR Buffer ll (50mM KCI and 10mM Tris-HCI (pH 8.3, Applied
Biosystems, Foster City, CA), 2 mM MgCl2, 1.25 mM each dNTP, 20pmol each
primer, 500ng DNA template (P. insidiosum genomic DNA) and 1.25 U Taq
polymerase (Applied Biosystems, Foster City, CA). Negative control reactions
containing the same reagents, except the DNA template, were included in all
experiments. The PCR thermal cycling conditions were as follows: hot start for

10 minutes at 95°C, 40 cycles as follows: 1 minute at 94°C, 2 minutes at 50°C

25

and 3 minutes at 72°C and an extension file: 1 minute at 94°C, 2 minutes at
50°C and 10 minutes at 72°C.

Amplification products were analyzed by running portions of each reaction
on a 1.4% agarose gel at 135 volts for 45 minutes to 1 hour. Ethidium bromide
(0.5ug/mL) was added to all agarose gels. DNA was isolated from the agarose
by using a protocol from the S.N.A.P.TM mini-prep kit (lnvitrogen®, Carlsbad,
CA). This purified PCR product was cloned into the TOPO-TA vector (lnvitrogen
®, Carlsbad, CA) and used for sequencing or labeled radioactively and used as a

probe for Southern hybridization experiments.

2.3 Southern Hybridization

Genomic DNA from P. insidiosum was digested with Bam H l (Promega,
Madison, WI) as follows: 40pg DNA, 6 mM Tris-HCI (pH 7.5), 100mM NaCl,
6mM MgClz and 1mM DTT and 40 units of Bam H I (buffer and enzyme from
Promega, Madison, WI) in a total reaction volume of 60pL. The mixture was
incubated at 37°C for 3 hours and heat inactivated by incubating at 65 °C for 15
minutes.

The reaction was loaded into 2 lanes of a 0.8% agarose gel along with
undigested DNA, unlabeled probe and a DNA size marker as controls. The gel
was run at 75 volts for approximately 4 hours.

The DNA was transferred to a nylon membrane using the capillary
method. For this, the gel was first treated with 0.25 N HCI in a shaking bath at

room temperature for 30 minutes. The gel was then placed into a shaking bath

26

containing 1.5 M NaCl and 0.5 N NaOH for 30 minutes and then transferred to a
fresh bath of the same solution and the incubation was repeated. Finally, the gel
was placed in a bath containing 0.5M Tris-HCU1.5M NaCI (pH 8.0) and incubated
for 15 minutes. Once the gel treatment was complete, a capillary transfer was
set up using 3MM paper filters, 20X 880 (3M NaCl and 0.3M sodium citrate pH
7.0) as the transfer buffer and a positively charged nylon membrane (BioRad,
Hercules, CA). After overnight transfer, the membrane containing the DNA was
rinsed briefly in 10X 88C, to remove debris, and then briefly blotted. The DNA
was cross-linked to the membrane by exposure to 50mJoules of UV energy in
the Gene linker UV chamber (BioRad, Hercules, CA).

Pre-hybridization of the membrane was carried out at 42°C for 2 hours in
50% Forrnamide, 5X Denhardt’s solution (0.1% ficoll Type 400, 0.1 %
polyvinylpyrrolidone and 0.1% BSA), 0.25% 808, 3X SSPE (3M NaCl, 0.02 M
EDTA and 0.2M NaHzPO4 pH 7.4) and salmon sperm DNA (0.1mg/mL).

The 600 bp PCR amplicon was radioactively labeled with 32F using
random priming with hexamers (Gibco, Rockville, MD). Unincorporated
nucleotides were removed from the labeling reaction using a push column
(Stratagene, La Jolla, CA). The probe was denatured by heating in a boiling
water bath for 5 minutes and immediately transferred to ice. The labeled
denatured probe was added to the pre-hybridization mixture, containing the
membrane, and incubated at 42°C for 24 hours. The hybridization buffer was
decanted from the hybridization bottle and 100mL of low stringency wash buffer

(1X 88C and 0.5% SDS) was added. Washing at low stringency proceeded for

27

30 minutes at room temperature with one change of solution. After the second
low stringency wash solution, 100mL of high stringency wash buffer (0.5x SSC
and 0.5% SDS) was added and washing proceeded at 68°C for 30 minutes with
one change of solution. The membrane was removed from the hybridization
bottle, wrapped in plastic wrap and placed on an X-ray film (Eastman Kodak,
New Haven, CT) at -70°C. The film was developed a various times depending

upon the signal strength using an automated film processor (Eastman Kodak,

New Haven, CT).

2.4 Construction of a Biased Genomic DNA Library

The same conditions as previously described in section 2.3 (Southern
Hybridization) were repeated and a small portion of the agarose containing DNA
fragments between 6-12 kb was physically excised from the agarose gel. The
DNA was purified using glassmilk (Bio101 Carlsbad, CA).

A portion of the purified DNAs was run on an agarose gel, blotted and
hybridized with the radioactively labeled CH8 PCR product as described in
section 2.3 (Southern Hybridization). The purified 6-12 kb mixture of DNA
fragments was cloned into the Lambda FIX II vector (Stratagene La Jolla, CA)
according to the manufacturer’s protocol. The cloning reaction was scaled down
proportionally to accommodate 10pg of DNA rather than the suggested 50ug.
The resulting biased library was amplified and purified according to the

manufacturer’s protocol.

28

2.5 Screening of the P. insidiosum Biased Genomic DNA Library (Plaque Lifts)
Primary screening of the library was performed by plating 50,000 plaques
on a 150mm plate. Five microliters of a 11100 dilution (50,000 pfu) of the
amplified library were plated by mixing with E. coli cells as described in the
Lambda FIX ll protocol. Plates were incubated at 37°C for 8 hours until the
plaques were each about 1mm in diameter. The plates were then chilled for 2
hours at 4°C. Plaque lifts were done by placing circular nylon membrane over
the plates for 2 minutes for the first membrane and 4 minutes for the second
replica. The membranes were transferred to a solution of 1.5 M NaCl and 0.5 N
NaOH for 2 minutes at room temperature. They were then placed into a solution
containing 1.5M NaCl and 0.5M Tris-HCI (pH 8.0) for 5 minutes and finally a
solution containing 0.2M Tris-HCI (pH 7.5) and 2X SSC for 1 minute. The
membranes were briefly blotted and cross-linked by exposing to 120mJouIes of
UV energy using a Gene Linker (BioRad, Hercules, CA). The membranes were
washed in 5X SSC, 0.5% SDS and 1mM EDTA for 1 hour at 42°C with shaking.
Pre-hybridization and hybridization steps were performed as described
above for Southern Hybridization (section 2.3). Washing was done in the same
manner as the Southern Hybridization except for the low stringency and high
stringency wash buffers, which contained 1X SSC and 0.1% SDS and 0.2X SSC
and 0.1% 808, respectively. Low stringency washes were performed at room
temperature and high stringency washes were performed at 42°C. Membranes
were placed on an X-ray film and incubated at -70°C. Putative positive clones

were picked up from the original phage/E. coli plates and placed in 1 mL SM

29

buffer (100 mM NaCl, 8mM M9804, 50mM Tris-HCI (pH 7.5), 0.01% gelatin) and
20uL of chloroform.

For secondary screening, a portion of eluted phage was used to produce
well-separated plaques on a 150mm agar plate. This plate was screened in the
same manner as for the primary screening except high stringency washing steps
were performed at 52°C. Single putative positive clones were picked up from the
plates and placed in 0.5 mL of SM buffer with 10 pL of chloroform. A tertiary
screening was performed using the phage eluent from the secondary screening.
This was done in the same manner as the secondary screening, except the high

stringency washing, which was performed at 62°C.

2.6 PCR Screening of the Putative Positive Clones Isolated by Plaque Lifts
Five microliter portions of SM buffer containing putative positive clones
were used as template for PCR as described in section 2.2 above. A single
clone that was positive by both PCR and plaque lifting was selected for DNA

isolation and further analysis.

2.7 DNA Isolation from a Putative Positive Lambda Phage Clone
A putative positive clone was plated to 50,000 plaques on a 150mm plate
and incubated for 8 hours. The plate was overlaid with 10mL SM buffer and
incubated at 4 °C overnight with shaking. The SM buffer containing the lambda
phage was collected from the plate and 500 uL of chloroform was added. The

mixture was incubated for 15 minutes at room temperature and then centrifuged

30

at 500 X g for 10 minutes. The supernatant, containing the phage, was
transferred to a fresh tube and 30 pL of chloroform was added. DNase (3 pg,
Boehringer Manheim, Indianapolis, IN) and Rnase (3.6 pg, Promega, Madison,
WI) were added to the tubes which contained about 10 mL of SM with the phage.
The tubes were incubated at 37 °C for 15 minutes. An equal volume of a
solution containing 20% polyethylene glycol (PEG 8000) and 2M NaCl in lambda
diluent (10mM Tris-HCL (pH 7.5) and 10mM M9804) was added and this mixture
was vortexed gently and incubated in ice water for 1hour to precipitate the
lambda phage. The phage were recovered by centrifugation at 4°C and 10,000
X g for 10 minutes. The supernatant was removed and discarded. The pellet
containing the phage was resuspended in 0.5 mL of TE buffer (pH 8.0) by
vortexing and 5p.L of 10% SDS was added. The mixture was incubated for 5
minutes at 68°C. Ten microliters of 5M NaCl was added and the lambda DNA
was extracted once by mixing with an equal volume (500pL) of
phenolzchloroform and once by mixing with an equal volume of chloroform alone.
The aqueous phase was transferred to a fresh tube between extractions.
Following the last organic extraction, an equal volume of isopropanol was added
and the tube was mixed and stored at -80°C for 15 minutes. The precipitated
DNA was recovered by centrifugation at 12,000 X g at 4°C for 15 minutes. The
pellet was washed with 70% ethanol and resuspended in water. The results of
the DNA isolation were assessed by running a small portion of the DNA sample

on a 0.8% agarose gel.

31

2.8 Isolation of the Insert DNA from the Lambda Phage DNA

Purified DNA from a putative positive clone was treated with Not I
(Promega, Madison, WI). Four identical Not I digests were set up as follows: 1 pg
DNA from the positive clone, 6mM Tris-HCI (pH 7.9), 0.15M NaCI, 6mM MgCIz
1mM DTT, 0.1mglmL BSA and 5 units of Not I in a total volume of 20 pL. The
digestions were incubated at 37°C for 1, 2, 3 and 4 hours. Each digestion was
heat inactivated by incubating at 65°C for 15 minutes.

Following inactivation, the reactions were loaded into a 0.7% agarose gel
and run at 75 volts for 4 hours and 30 minutes. The gel was blotted and the 600
bp CH8 PCR product was used as a probe in a Southern hybridization
experiment as described in section 2.3. The restriction enzyme digestion and
electrophoresis were repeated and bands were excised from the gel and purified
using a capture column from the S.N.A.P. mini-prep kit for purifying DNA from

agarose gels (lnvitrogen®, Carlsbad, CA).

2.9 Sub—cloning the Insert DNA into the pZErOTM Plasmid Vector
The purified insert DNA from the lambda phage screening was sub-cloned
into the pZErOTM-Z vector (lnvitrogen® Carlsbad, CA), which uses Kanamycin
selection. The insert DNA from the lambda phage clone was recombined with

pZErOm-Z vector according to the manufacturer’s protocol.

2.10 Isolation of Plasmid DNA from Bacterial Clones Harboring pZErOm-Z

Bacterial clones were grown for 24 hours in LB broth supplemented with

32

Kanamycin. Plasmid DNA was isolated using the S.N.A.P. mini-prep kit
(lnvitrogen® Carlsbad, CA) and quantified by absorbance at 260 nm before

sequencing.

2.11 Nucleotide Sequencing

Sequencing was performed using the Sanger dideoxy chain termination
method, big dye terminator chemistry (Applied Biosystems, Foster City, CA) and
the Applied Biosystems ABI Prism 310 automated DNA sequence analyzer. Dye
terminator sequencing reactions were set up as follows: 500ng of plasmid DNA,
4pL TRRM (Template Ready Reaction Mixture, Applied Biosystems, Foster City,
CA), 3.2 pmol of primer (lnvitrogen, Carlsbad, CA) and water to a final volume of
10pL. The Reactions were cycled 25 times as follows: 10 seconds at 96°C, 5
seconds at 50°C and 4 minutes at 60°C. The reactions were incubated at 4°C
until the cleanup step.

Before loading on to the ABI Prism 310 Genetic Analyzer, a clean-up
procedure was used to eliminate unincorporated Dye Terminators from the
reaction mixture. This was performed by mixing the entire 10pL sequencing
reaction with 32pL of 95% ethanol and 6pL of sterile water. This mixture was
vortexed briefly and incubated at room temperature for 15 minutes. The mixture
was then centrifuged at maximum speed for 30 minutes at room temperature in a
microcentrifuge. The supernatant was removed and discarded and the DNA
pellet was dried for 5 minutes in a vacuum centrifuge. The pellet was

resuspended in 18pL of TSR (Template Suppression Reagent, Applied

33

Biosystems, Foster City, CA). The tube was vortexed, briefly centrifuged,
incubated at 95°C for 2 minutes and immediately chilled on ice for 10 minutes.
The tube was vortexed and briefly centrifuged before being loaded on to the ABI
Prism 310 Genetic Analyzer. Primer walking was employed for sequencing the
insert by successive rounds of gathering sequence data and generating new

primers.

2.12 Engineering a Specific PiCH82 Probe
The following primer set (hereafter referred to as CH8 nested) was used
in a PCR reaction as described in section 2.2 to generate a 500 bp amplicon of
PiCH82: 5’-AGG AAC ACT CAC AGG CAT CGC AAC-3' and 5'-TGG GTT GAA
GTA GTT GGG CTG CTC-3'. The PCR products were cloned and sequenced

as described in sections 2.2 and 2.11, respectively.

2.13 Southern Blot Using the Purified 500 bp PiCH82 Probe

A Southern hybridization experiment was performed as described in
section 2.3 using the following three probes: 1) the 600 bp PCR product of
PiCH82 obtained using the degenerate primers (Bowen 1992), 2) the purified
600 bp unknown PCR product and 3) the purified CH82 500 bp PCR product
obtained using the CH82 nested primers. The following control samples were
loaded into a single agarose gel in triplicate: all three unlabeled probes, Not I
digestion (prepared as described in section 2.8) of a pZErOm-Z plasmid vector

containing a 7.5 kb fragment of the insert DNA from the putative positive lambda

phage clone, Bam H l digestion of P. insidiosum genomic DNA (prepared as
described in section 2.3) and a DNA marker (Promega, Madison, WI). Following
electrophoresis and capillary transfer, the nylon membrane was divided into 3
equal parts each containing a copy of all the above samples. Each membrane
was reacted with one of the three probes. The methods for this Southern

hybridization experiment were the same as those described in section 2.3.

2.14 Screening Putative Positive Clones from the Biased Library by Oct Blot
Hybridization

Aliquots of nine positive phage clones from the primary screening,
dissolved in SM buffer, were spotted on a nylon membrane in triplicate. The
membrane was allowed to dry, divided into three and treated as described in
section 2.5. The membranes were briefly blotted and the DNA was cross-linked
using 120mJouIes of UV energy (Biorad, Carlsbad, CA). Each of the
membranes was screened either with the 600 bp mixture probe, the 500 bp
CH82 probe or the 600 bp unknown probe in a Southern hybridization
experiment. The pre-hybridization, hybridization and washing steps were

performed as described in section 2.3.

2.15 Screening a Putative Positive Clone Identified by Dot Blot Hybridization
PCR reactions were prepared using phage clones in SM buffer as
template. A positive control reaction was prepared using genomic DNA from P.

insidiosum as template. The reactions were set up using the same formulation

35

as described in section 2.2 but with 5 pl of each phage eluent as template. The
reactions were cycled as described in section 2.2. The results were assessed by

running 10 pl of each reaction on a 2% agarose gel for 45 minutes at 135 volts.

36

3. Results
3.1 Amplification of a 600 bp Fragment of PiCH82
PCR amplification of P. insidiosum genomic DNA using the primers
designed by Bowen et al. (1992) yields CH8 amplicons of approximately 600 bp.
Figure 2 shows a single band at approximately 600 bp in all reactions containing
genomic DNA from P. insidiosum and no reaction in a negative control

containing no DNA template.

3.2 Southern Hybridization Using Genomic DNA from P. insidiosum

Figure 3 shows hybridization to undigested genomic DNA and unlabeled
600 bp probe. A strong band at approximately 8 to 9 kb is present in each of the
two lanes containing the Bam H I digestion. A faint band at approximately 11 kb
was also observed in the 8am H I digestion in some blots exposed to film

overnight. A slight cross-reaction is observed with the DNA used as a marker.

3.3 Constructing a Biased Genomic DNA Library
The biased library of DNA fragments was screened for PiCH82 by
Southern blot using the 600 bp PiCH82 PCR amplicon as a probe prior to
cloning. A single band is observed in the lane containing the library of fragments
that matches the 8 kb band in the Barn H I digestion (Figure 4). The probe
hybridized to the undigested DNA control and the unlabeled 600 bp probe

control.

37

750 bp
500 bp

250 bp

 

Figure 2

Chitin Synthase PCR. Amplicons obtained using genomic DNA from P.
insidiosum as template and Bowen et al. (1992) CH8 primers. The figure shows
a single 600 bp product. Lane 1- DNA marker, lanes 2, 3 and 4- identiml PCR
reactions with P. insidiosum genomic DNA as template and lane 5- Negative

control reaction (no DNA-reagent blank).

38

lOkb 1,
6kb

Figure 3
Southern blot analysis. The figure shows a single distinct band at

approximately 8 kb. Lane 1- 1 kb DNA marker (Promega Madison, WI), lanes 2
and 3- ldentiwl Bam H l digestions of P. insidiosum genomic DNA and lane 4-

undigested genomic DNA from P. insidiosum.

39

12345

IOkb
6kb

600 bp

 

Figure 4

Screening the Mixture of Genomic DNA Fragments Prior to Cloning.
Screening the biased library of fragments by Southern hybridization using the
600 bp fragment of PiCH82 as a probe prior to cloning into Lambda FIX ll. Lane
1- 1 kb DNA marker (Promega, Madison, WI) lane 2- Barn H I digestion of P.
insidiosum genomic DNA, lane 3- biased library of fragments used for cloning
into Lambda FIX II, lane 4» undigested genomic DNA from P. insidiosum and

lane 5 unlabeled 600 bp PiCH82 probe.

40

3.4 Screening the Library

Primary screening was done by making five plates with 50,000 plaques
each. Two replicas of each plate were made using nylon membranes. The
membranes were used in hybridization reactions with the 600 bp PCR amplicon
as a probe. The resulting X-ray films were aligned with the plates and plaques
that were positive on both membranes were selected. Figure 5 shows an X-ray
film from one of the primary screening plates. Plaques were excised from the
agar plates and the phage were eluted into buffer. The phage eluent was used
to make two additional plates for secondary screening. Each of these plates was
screened with two membranes and the 600 bp PiCH82 probe. Figure 6 shows
the results of the secondary screening. Several plaques were positive on both
membranes and two well-separated plaques were selected for further screening.
These plaques were excised from the plates and the phage were eluted into SM
buffer.

A tertiary screening was done to select phage clone. Two separate
clones were selected from the secondary screening and each was grown on an
agar plate with approximately 50 plaques on each plate, which gave adequate
separation of the plaques. The results of the tertiary screening can be seen in
Figure 7. All of the plaques on one of the plates were positive except one.
Several plaques were selected and placed individually into buffer.

PCR was performed using the intact phage clones from the tertiary screening as
template to ensure that the clones contained PiCH82 as an insert. Several

clones that were positive by plaque screening were selected. Additionally,

41

 

Figure 5

Primary ecreenlng of the blased library. Nunbemd spots represent putative
positive plaques that reacted with the 600 bp PiCH82 probe on both
merrbranes.

42

 

Figure 6

Secondary screening of the biased library. Plaques on these plates were
larger and better separated than the plaques observed during the primary
screening. Seven ofthe plaques hybridized to the probe on both membranes in

the same relative position.

43

 

Figure 7
Tertiary screening of the biased library. The plaques were well separated on

the agar plates and 44 of 45 plaques hybridized to the 600 bp probe and were

deemed putative positive clones.

several clones that were negative by plaque screening were also selected as
negative controls. Agreement between the PCR screening and plaque screening

methods was noted for all clones tested (data not shown).

3.5 Purification and Sub-Cloning of the DNA Insert from a Putative Positive
Phage Clone

The insert of the recombinant lambda FIX ll vector was isolated as an
intact cassette by treating with Not I. DNA was extracted from a positive clone
and digested with Not I. The DNA was treated for one, two, three and four
hours, respectively. Five bands were observed after treating with Not I for 1
hour. The gel was transferred to a membrane and hybridized to the 600 bp
probe. Two of the bands hybridized with the 600 bp probe. One band at 11 kb
was slowly degraded over the 4 hours of digestion and began to disappear over
time on the blot (Figure 8b) and on the gel (Figure 8a). A second band at
approximately 7.5 kb increased in intensity on the blot and the gel. A third band
at 4 kb was more intense after 4 hours of digestion (on the gel) but it did not bind
to the probe.

PCR using the 11kb, 7.5kb and 4kb fragments from the Not I digestion,
cloned separately into plasmids, as templates with the CH8 primers revealed
600 bp amplicons in reactions containing the 11kb and 7.5kb Not I fragments
(Figure 9). No amplification of a 600 bp product was observed in the reaction
containing the 4kb fragment as template. A positive control reaction prepared

using genomic DNA from P. insidiosum as template revealed an expected

45

Figure 8

Notl Digestion and Blot of the DNA insert from a Putative Positive Clone.
Not I digestion of DNA from the lambda phage clone after 1, 2, 3 and 4 hours I
and the corresponding blot using the 600 bp PCR amplicon as a probe.

A) The gel before transfer. Lane 8- unlabeled 600 bp probe (Note that the
amount of DNA loaded is lower than the limit of detection using ethidium
bromide), lane 7- 1kb DNA marker (Promega, Madison, WI), lane 6- 1 hour, lane
5— 2 hours, lane 4- 3 hours, lane 3- 4 hours and lane 2- undigested DNA from the
lambda phage clone, lane 1- 23 kb DNA marker (BioRad, Carlsbad, CA). The
relevant MW of selected bands in the lanes containing DNA markers is indicated.
B) The corresponding blot using the 600 bp PCR amplicon as a probe. Lanes
are labeled the same as in figure 9B. The unlabeled 600 bp PiCH82 probe is

visible at the bottom.

46

Figure8
A. 12345678

23 kb

9kb

 

B. 12345678
. .- 23kb

11 kb
O... 75 kb
-
O 600bp

47

 

Figure 9

PCR Screening of Plasmid Clones carrying either the 4 kb, 7.5 kb or 11 kb
Insert from the Notl digestion (see figure 8) of the lambda phage clone.
Bowen et al. (1992) CH8 primers were used. Lane 1- 11 kb fragment, lane 2-
7.5 kb fragment, lane3- 4 kb fragment, lane 4_ positive control (P. insidiosum
genomic DNA used as template), lane 5- negative control (no DNA template)

and lane 6- 1kb DNA marker.

48

product of approximately 600 bp. A negative control reaction containing no DNA

template was also prepared and showed no amplification.

3.6 Sequencing the Putative PiCH82 Molecule and the 600 bp Probe

PCR was used to amplify a 600 bp product from plasmid clones of the 7.5
kb fragment of the original lambda phage clone insert DNA. The 600 bp product
was cloned and sequenced. The sequence of this product did not match the
PiCH82 sequence obtained by Perpich et al. (1999). This unknown sequence
(Figure 10) was one bp shorter than PiCH82 and did not share significant
homology with any sequence in the GenBank database. The unknown
sequence matched the sequence of the Bowen CH8 PCR primers, however, it
had different residues at the degenerate positions when compared to PiCH82
(Figure 11).

The 600 bp probe was cloned into a plasmid vector and sequenced. The
probe was found to be a mixture of the 600 bp PiCH82 sequence obtained by
Perpich et al. (1999) and a 600 bp fragment of the unknown sequence obtained

from the biased lambda phage library.

3.7 Southern Blot Using Each Probe Separately
A Southern blot experiment was performed using each product, obtained by PCR
with genomic DNA from P. insidiosum, as a probe. Three probes were prepared.

First, the same probe used to screen the library was selected. This probe is a

49

Figure 10

Sequence of Both Products Obtained by PCR with the CH8 Primers.
Sequence comparison of the two products obtained by PCR using the
degenerate CH8 primers designed by Bowen et al (1992). Primer binding sites
are from residues 1 to 27 and from 553 to 579 (on PiCH82). PiCH82 is the 600
bp conserved region of the chitin synthase class 2 gene from P. insidiosum. The
unknown fragment is the lower sequence (Unknown), which was co-amplified
with PiCH82 using the primers designed by Bowen et al (1992). Sequence

dissimilarities are shown with an asterisk.

5O

Figure 10

CCCCCCCCCCCCCCCCCC

ofio omw on 0*» owe I“... an OH...

thQ‘dflfluo DB‘Uhfidduo UOBBHDBHGB mafiflduufluo UBDBDBOOBO UBUOQ‘UQ‘B 0084081000 QatfihOOBtfi

 

 

COOC*§O£.I IOCIIICII¢ CCCCCCCICI lCCICC¢ICC ICICCICIOC ICICICOCI. II IIIIIIIIIIIIIIIIII

 

 

 

 

«Micah a

go
o~
«macaw a

«mundm o

 

51

A
C

G C C A C
1) 5' CTG AAG CTT ACT ATG TAT AAT GAG GAT 3’
2) 5’ CTG AAG CTT ACG ATG TAC AAC GAA GAT 3'
3) 5' CTG AAG CTT ACT ATG TAT AAT GAG GAT 3’

Figure 11

Sequences of one of the Bowen primers and primer binding regions of
PiCH82 and the unknown sequence. 1) Bowen et al. degenerate CH8 primer,
2) Primer binding region of 600 bp unknown sequence and 3) Primer binding
region of 600 bp PiCH82. Positions with multiple residues are degenerate
areas. Differences between PiCH82 and the unknown sequence at the

degenerate positions are shown in bold.

52

mixture of the unknown sequence and PiCH82. Second, the 600 bp unknown
sequence alone. This probe was obtained by PCR using the cloned 7.5 kb
fragment of genomic DNA on a plasmid as template. Third, a 500 bp portion of
PiCH82 was used. This probe was obtained by PCR using a set of nested
primers designed to amplify only PiCH82 but not the unknown sequence from
genomic DNA of P. insidiosum. Purity of all of the probes was confirmed by
sequencing.

The mixture probe reacted with all three unlabeled probes and the 7.5 kb
insert DNA from the lambda phage clone. In addition, this probe hybridized to
the same 8 kb band observed in the Barn H l digestion of previous Southern blot
experiments. It also hybridized weakly to an 11 kb fragment in the Barn H I
digestion. The probe composed only of the unknown sequence recognizes itself,
the unlabeled mixture probe and the 7.5 kb insert DNA from the lambda phage
clone. It bound to a band larger than 8 kb in the Barn H l digestion. Finally, the
500 bp PiCH82 probe bound to the mixture probe and to itself. It did not react
with the 7.5 kb insert from the biased library. This probe hybridized to a band at
the 8 kb position but did not display a band at the 11 kb position in the Barn H l

digestion.
3.8 Dot Blot of Nine Clones from the Primary Screening of the Library

Nine of the strongest positive clones from the primary screening of the

biased library were selected for this experiment. Each clone was spotted once

53

Figure 12

Southern blot using 3 different probes and 3 identical blots. All samples
were loaded in triplicate and reacted with a different probe. The samples for the
mixture probe: lane 1- unlabeled mixture probe, lane 2- unlabeled unknown
probe, lane 3- unlabeled 500 bp PiCH82 probe, lane 4- 7.5 kb insert from
lambda phage clone (digested from a plasmid), lane 5- 1 kb marker (Promega,
Madison, WI) and lane 6- Sam H I digest of P. insidiosum genomic DNA. The
samples for the unknown probe: lane 1- unlabeled mixture probe, lane 2-
unlabeled unknown probe, lane 3- unlabeled 500 bp PiCH82, lane 4- Bam H I
digestion, lane 5- 1 kb marker and lane 6- 7.5 kb insert. The samples for the 500
bp PiCH82 probe: lane 1- Bam H l digest, lane 2- 1 kb marker, lane 3- 7.5 kb
insert DNA, lane 4- unlabeled 500 bp PiCH82 probe, lane 5- unlabeled unknown

probe and lane 6- unlabeled mixture probe.

54

Figure 12

Mixture Probe 600 bp Unknown 500 hp PiCH82
12345 6 -123 456 -1234 56

R

 

55

each on three small pieces of nylon membrane. The membranes were reacted
with all three probes used in the experiment in section 3.8 (Figure 13). The
probe composed of the unknown sequence reacted with all nine clones. The
mixture probe also reacted with all nine clones. The 500 bp PiCH82 probe
reacted with clone six only.

PCR screening of clone six and several other clones that were negative in
the dot blot using the CH8 nested primers revealed an expected 500 bp
amplicon for clones five, six and seven. Pythium insidiosum genomic DNA was
used as a positive control in a separate PCR. Clone six showed the strongest
band at 500 bp, which matched the size of the product from genomic DNA, while
clones five and seven showed weaker bands at the same molecular mass.

A primary screening performed using the phage eluent of clone six
revealed no phage clones that reacted with the 500 bp probe. These results

were obtained despite screening 250,000 plaques.

3.9 Screening the Entire Library with the 500 bp PiCH82 Probe
The entire biased library was screened with the purified 500 bp PiCH82
probe. After screening five plates and a total of 250,000 clones, no putative

positive plaques were identified.

56

 

§ O .
123456789
J

Figure 13

Oct blot of nine putative positive clones from the primary screening of the
biased library. Sanples were spotted in triplicate on three different
membranes. Each membranewasmactedwithoneofthethreeprobesusedfor
the Soulhem blot experirmnt (Figure 12): 600 bp mixture probe (top membrane),
600 bp unknown probe (middle rnerrbrane) and 500 bp PiCH82 (bottom

membrane).

57

4. Discussion

The primary goal of this study was to establish the presence of CH8
genes in P. insidiosum and to isolate and sequence one of the genes. A biased
genomic DNA library was constructed and screened for the chitin synthase class
2 gene from P. insidiosum (PiCH82). A putative positive lambda phage clone
was isolated in the tertiary screening of the biased library (Figure 7) and DNA
was extracted. The insert DNA was excised from the vector by treating with Not
I. The results of this treatment (Figure 8 a and b) show five distinct bands. The
vector arms are 9 kb and 20 kb and are present at approximately those
positions. The remaining bands are approximately 11 kb, 7.5 kb and 4 kb. The
11 kb band disappears over time while the 7.5 kb and 4 kb bands increase in
intensity over time. This suggests that the insert DNA from the putative positive
lambda phage clone is 11 kb in size and contains a Not I recognition site which
produces bands that are 7.5 kb and 4 kb in size (Figure 8b). The 11 kb and 7.5
kb bands was recognized by the probe while the 4 kb band was not. Therefore,
the probe recognition site is present within the 7.5 kb portion of the 11 kb insert
DNA. These conclusions are supported by PCR (Figure 9) where the expected
600 bp amplicon was observed using the 7.5 kb and 11 kb fragments as
template, while the 4 kb fragment did not yield the 600 bp band.

The nucleotide sequencing of the 600 bp PCR product showed that the 11
kb fragment did not carry the PiCH82 gene. Instead, the 600 bp product was
found to contain primer recognition sites for the Bowen et al. CH8 primers but

was not homologous to any known CH8 (Figures 10 and 11).

58

A Southern blot experiment was prepared using a purified 500 bp
fragment of PiCH82, the unknown probe and the mixture of the two (Figure 12).
The 500 bp PiCH82 probe reacted with a band at approximately 8 kb that
appears to be at the same MW as observed in the original Southern blot. The
600 bp unknown probe, however, reacted with a band at a higher molecular
mass, nearly 11 kb. This finding suggests that the strong 8 kb band observed in
the original Southern blot (figure 3) is a fragment that contains the PiCH82 gene
and that the 11 kb band is a fragment that contains the unknown sequence.
Therefore, since the 11 kb fragment was isolated from the lambda phage library,
it was possible that the 8 kb fragment could be contained in another putative
positive clone from the library. Since the primary screening of the library (Figure
4) revealed several putative positive clones, one of them could contain the
PiCH82 gene.

The results of the dot blot experiment (figure 12) show that clone six could
contain PiCH82. PCR screening, using the primers known to amplify a pure 500
bp product of PiCH82, was done using this clone and several clones negative in
the dot blot (data not shown). Clone six reacted strongly, two other putative
negative clones were weakly positive and the clone containing the 11 kb
fragment was negative. These results, while seemingly inconclusive, could be
explained by diffusion across the plate of clone six to clones in other areas. The
phage eluent containing clone six was plated and plaque lift screening was done.
This method failed to isolate a putative positive phage clone.

The purified 500 bp PiCH82 probe was used to screen the entire library.

59

No putative positive clones were obtained despite screening 250,000 plaques.
Thus, this library seems to contain several copies of clones with the 11 kb insert
but none containing the 8 kb PiCH82 gene.

An unexpected result of this research was the co-amplification by PCR,
using the Bowen et al. CH8 primers, of a segment of DNA approximately 600 bp
in length that shows no homology to any known chitin synthase genes. The
primers used for this PCR experiment are well characterized and have been
widely used (Bowen 1992, Motoyama 1994, Mort-Bontemps 1997, Nino-Vega
1998). Bowen et al. (1992) designed the primers to amplify conserved 600 bp
regions of chitin synthase genes from fungi. Since their design in 1992, several
investigators have used them to amplify fungal and Oomycete CH8. None of
these investigators reported anomalous amplification of non-CH8 DNA
segments. Perpich et al. (1999) used these primers to amplify a 600 bp
fragment of CH82 from P. insidiosum and did not report any non-CH8
amplicons. The PCR conditions in this study were not different from those
reported in that study. Yet, these data show two PCR products using the
degenerate CH8 primers. Due to the fact that the PCR artifact was only 1 base
pair smaller than the intended product it was not detected after agarose gel
electrophoresis and, consequently, was co-purified with the 600 bp fragment of
PiCH82. The artifact was detected later as a minor contaminant of the CH8
probe preparation.

Artifacts during PCR amplification do occur. Some investigators have

reported the formation of hybrid molecules during PCR amplification. This was

60

usually reported when heterogeneous mixtures of rDNA templates were used for
PCR. Hybrid products composed of a stretch of one sequence fused to a
different but homologous sequence were found (Amann 1995). This example,
however, does not explain why Perpich et al. (1999) did not obtain contaminated
amplicons under conditions similar to those used in this study. Despite
sequencing over 30 clones of the 600 bp PCR product, those investigators
obtained only the 600 bp fragment of the PiCH82 gene. This discrepancy could
be explained by a manufacturer’s error in the synthesis of the primers used for
this research. Both experiments were done in the same laboratory with similar
reagents and conditions. The primary difference was the lot of primers used for
amplification. This study used new primers synthesized by a different company.
Therefore, aberrant products from the synthesis of the primers may account for
the differential results. However, examination of Figure 11, which shows a
comparison of the sequences from the Bowen et al. primers and the two PCR
products obtained in this study, suggests that the primers were synthesized
correctly and the anomalous amplification of the artifact may be due to the
degeneracies in the primers. This raises the possibility that the original primers
used by Perpich et al. (1999) were somehow biased for amplification of PiCH82.
This idea is supported by the fact that the authors obtained only one CH8
isozyme, which is in contrast to all other reports were mulitiple isozymes of CH8
were identified using the same primer set. For example, Mort-Bontemps et al.
(1997) reported two PCR products, homologous to CH81 and CH82 from

Saccharomyces cerevisiae, using the Bowen et al. primers to amplify genomic

61

DNA from the Oomycete Saprolegnia monoica. In addition, most fungal chitin
synthase systems studied to date contain two or more isozymes. Finally, it is
possible that the artifact obtained in this study is actually an ancient CH8
isozyme that has become a pseudogene. However, no other reports of such
pseudogenes have been discovered so far using the Bowen et al. primers.

The problems isolating PiCH82 could be due to the vector used for
cloning in phage lambda. Lambda FIX ll preferentially packages lambda DNAs
with inserts ranging from 9 to 23 kb. The size of the Barn H I fragment of
PiCH82 in this study is approximately 8 to 9 kb, which is at the low end of this
range. Although an 8 kb insert would Iigate into FIX II more efficiently than an 11
kb insert, such a construct would not package with nearly the efficiency of a
lambda DNA carrying an 11 kb insert since packaging efficiency is largely
dependent upon the size of the DNA in the head of the phage. This is consistent
with the fact that all of the putative positive clones obtained in the primary
screening hybridized to the purified unknown probe. Presumably all of these are
equivalent clones containing the 11 kb Bam H I fragment. However, this was an
unexpected finding since other investigators have cloned fragments as low as 6
kb using the same vector and cloning kit (Mendoza, personal communication).

In conclusion, the problems encountered during this research seem to be
technical rather than biological in nature. Chitin synthase genes have been
identified in several fungi and Oomycetes (Mort-Bontemps 1997, Bowen 1992,
Motoyama 1994, Au-Young 1990). These genes were identified by PCR using

the same primers that were used in this research. The entire CH82 from 8.

62

monoica was isolated and characterized (Mort-Bontemps 1997). This gene is
expressed during hyphal growth and in regenerated protoplasts. Furthermore,
the amplification of conserved regions of CH8 genes containing open reading
frames in a total of four Oomycetes including P. insidiosum suggests that these
genes are not pseudogenes. Rather, they are likely important to the Oomycetes
at some stage in their life cycles. In addition, the taxonomic classification
proposed by Bartnicki-Garcia which placed the Oomycetes into the cellulose-
glucan group and the true fungi into the chitin-glucan group is no longer
appropriate in light of the fact that the Oomycetes clearly contain chitin and chitin
synthase enzymes in their cell walls. Further, it appears that the presence of
CH8 in the Oomycetes is widespread.

This research confirmed the existence of a CH82 gene in P. insidiosum
and established that CH8 genes from the Oomycetes are phylogenetically
distinct from fungal CH8 genes. Further work is needed in order to characterize
the entire PiCH82 gene. The results of this study will undoubtedly be important
for future studies on the cloning and characterization of CH8 genes in P.

insidiosum.

63

5. Future Direction

As soon as PiCH82 has been isolated from P. insidiosum, expression
experiments should be performed to determine the role of chitin in this organism.
If P. insidiosum utilizes chitin during hyphal growth as 8. monoica does, CH8
could be an excellent target for drug design. Although development of drugs that
target chitin synthase is in its infancy, chitin synthase remains an attractive target
for drug design in this group of pathogens. In addition, the recent development
of glucan synthase inhibitors could also be investigated as a chemotherapeutic

option in the infections caused by P. insidiosum.

BIBLIOGRAPHY

Alfaro AA, Mendoza L. 1990. Four cases of equine bone lesions caused by
Pythium insidiosum. Equine Veterinary Journal. 22(4): 295-97.

Amann RI, Ludwig W, Schleifer K. 1995. Phylogenetic identification and in situ
detection of individual microbial cells without cultivation. Microbiological Reviews.
Mar, 59(1): 143-69.

Austwick PKC Copland JW. 1974. Swamp cancer. Nature (London). 250: 84.

Au-Young J, Robbins PW. 1990. Isolation of a chitin synthase gene (CH81) from
Candida albicans by expression in Sacchammyces cerevisiae. Mol Microbiol.
Feb; 4(2) 197-207.

Bartnicki-Garcia S. 1968. Cell wall chemistry, morphogenesis and taxonomy of
fungi. Ann. Rev. Microbiol. 22: 87-108.

Borgia PT, lartchouk N, Riggle PJ, Winter KR, Koltin Y, Bulawa CE. 1996. The
chsB gene of Aspergillus nidulans is necessary for normal hyphal growth and
development. Fungal Genet Biol. Sep; 20(3): 193-203.

Bowen AR, Chen-Wu JL, Momany M, Young R, Szaniszlo PJ, Robbins PW.
1992. Classification of fungal chitin synthases. Proc. Natl. Acad. Sci. USA. 89:
519-23.

Brown CC, Roberts ED. 1988. Intestinal pythiosis in a horse. Aust. Vet. J. 65: 88-
9.

Bulawa CE. 1993. Genetics and molecular biology of chitin synthesis in fungi.
Ann. Rev. Microbiol. 43: 505-34.

Bulone V, Chanzy H, Gay L, Girard V, Fevre M. 1992. CharacteriZation of chitin
and chitin synthase from the cellulosic cell wall fungus Sapmlegnia monoica.
Experimental Mycology. 16: 8-21.

Cherif M, Benhamou N, Belanger RR. 1992. Occurrence of cellulose and chitin in
the hyphal walls of Pythium ultimum: a comparative study with other plant
pathogenic fungi. Can. J. Microbiol. 39: 213-22.

Cooper BA, Aronson JM. 1967. Cell wall structure of Pythium debaryanum.
Mycologia. 59: 658-70.

65

Davila T, San-Blas G, San-Bias F. 1986. Effect of papulacandin B on glucan
synthesis in Paracoccidioides brasiliensis. J. Med. Vet. Mycol. 24: 193-202.

de Cock WAW, Mendoza L, et al. 1987. Pythium insidiosum sp. nov. the
etiological agent of pythiosis. J. Clin. Microbiol. 25: 344-9.

El-Sherbeini M, Clemas JA. 1995. Nikkomycin Z supersensitivity of an
echinocandin-resistant mutant of Saccharomyces cerevisiae. Antimicrob. Agents
Chemother. 39: 200-207.

Frey R. 1950. Chitin and zellulose in pilzzellwanden. Ber. Schweiz. Bot. Ges. 60:
199-230. ‘
Georgopapadakou NH. 2001. Update on antifungals targeted to the cell wall:
focus on beta-1,3-glucan synthase inhibitors. Expert Opin lnvestig Drugs. Feb;
10(2): 269-80.

Gonzalez HE, Threebilcock P et al. 1979. Tratamiento de la ficomicosis equine
subcutanea empleando yoduro de potasio. Revista Colombiana de Agriculture.
14: 1 15-21 .

Gordee RS, Zeckner DJ, Howard LC, Alborn WE, Debono M, et al. 1988. Anti-
candida activity and toxicology of LY121019, a novel semisynthetic polypeptide
antifungal antibiotic. Ann. NY. Acad. Sci. 544: 294-301.

Hector RF, Braun PC. 1986. Synergistic action of nikkomycins X and Z with
papulacandin B on whole cells and regenerating protoplasts of Candida albicans.
Antimicrob. Agents Chemother. 29: 389-94.

Hector RF, Zimmer BC, Pappagianis D. 1990. Evaluation of nikkomycins X and Z
in murine models of coccidioidomycosis, histoplasmosis and blastomycosis.
Antimicrob. Agents Chemother. 34(4): 587-93.

Hutchins DR, Johnston K. 1972. Phycomycosis in the horse. Aust. Vet. J. 48:
269-78.

lchitani T, Amemiya J. 1980. Pythium gracile isolated from the foci of granular
dermatitis in the horse (Equus cabal/us). Trans. Mycol. Soc. Japan. 21: 263-5.

lmwidthaya P. 1994. Human pythiosis in Thailand. Postgrad Med J. 70: 558-60.

Kollar RE, Petrakova E, Aswell G, Robbins PW, Cabib E. 1995. Architecture of
the yeast cell wall. J. Biol. Chem. 270: 1170-78.

Kwon-Chung KJ. 1994. Phylogenetic spectrum of fungi that are pathogenic to
humans. Clinical Infectious Diseases. 19(suppl 1): 81-7“.

66

Lin CC, Aronson JM. 1970. Chitin and cellulose in the cell walls of the oomycete
Apodachlya sp. Arch. Microbiol. 72: 1 1 1-1 14.

Matsumoto Y, Matsuda 8, Tegoshi T. 1989. Yeast glucan in the cyst wall of
Pneumocystis carinii. J. Protozool. 36: 215-225.

McMullan WC, Joyce JR, et al. 1977. Amphotericin B for the treatment of
localized subcutaneous phycomycosis in the horse. J. Am. Vet. Med. Assoc. 170:
1293-8.

Mendoza L, Chapter 33. Pythium insidiosum. In: Topley & Wilson’s Microbiology
and Microbial Infections. Medical Mycology. vol. 4 9 " edition. L Ajello and RJ Hay
eds. Arnold, London. Sydney, Auckland. Pp. 617-630, 1998.

Mendoza L, Alfaro A. 1986. Equine pythiosis in Costa Rica: Report of 39 cases.
Mycopathologia. 94: 123—29.

Mendoza L, Hernandez F, Ajello L. 1993. Life cycle of the human and animal
oomycete pathogen Pythium insidiosum. J. Clin. Microbiol. 31(11): 2967-73.

Mendoza L, Kaufman L, Standard P. 1987. Antigenic relationship between the
animal and human pathogen Pythium insidiosum and nonpathogenic Pythium
spp. J. Clin. Microbiol. 25: 2159-62.

Mendoza L, Kaufman L, Standard P. 1986. Immunodiffusion test for diagnosing
and monitoring pythiosis in horses. J. Clin. Microbiol. 23: 813-16.

Mendoza L, Villalobos J, Calleja C, Solis A. 1992. Evaluation of two vaccines for
the treatment of pythiosis insidiosi in horses. Mycopathologia. 119: 89-95.

Miller RI. 1981. Treatment of equine phycomycosis by immunotherapy and
surgery. Aust. Vet. J. 57: 377-82.

Miller RI. 1983. Investigations into the biology of three phycomycotic agents
pathogenic for horses in Australia. Mycopathologia. 81: 23-8.

Miller RI, Campbell RS. 1982. Immunological studies on equine phycomycosis.
Aust. Vet. J. 58: 227-31.

Miller RI, Campbell RS. 1983. Experimental pythiosis in rabbits. Sabouraudia. 21:
331-41.

Mort-Bontemps M, Gay L, Fevre M. 1997. CH82, a chitin synthase gene from the
oomycete Saprolegnia monoica. Microbiology. 143: 2009-2020.

67

Motoyama T, Kojima N, Horiuchi H, Ohta A, Takagi M. 1994. Isolation of a chitin
synthase gene (ChsC) of Aspergillus nidulans. Biosci Biotechnol Biochem. Dec;
58 (12) 2254-2257.

Motoyama T, Fujiwara M, Kojima N, Horiuchi H, Ohta A, Takagi M. 1997. The
Aspergillus nidulans genes chsA end chsD encode chitin syntheses which have
redundant functions in conidia formation [corrected and republished article
originally appeared in Mol Gen Genet 1996 Jun 251(4): 442-50. Mol Gen Genet.
Jan 27; 253(4): 520-28.

Murray DR, Ladds P, et al. 1978. Metastatic phycomycosis in a horse. J. Am.
Vet. Med. Assoc. 172: 834-6.

Nino—Vega GA, Burrrnen E, Gooday G, San-Bias G, Gow N. 1998. Molecular
cloning and sequencing of a chitin synthase gene (CH82) of Paracoccidioides
brasiliensis. Yeast. 14: 181-187.

Perfect JR, Wright KA, Hector RF. 1991. Synergistic interactions of nikkomycin
and cilofungin against diverse fungi, see ref. 132a. 369-79.

Perpich JD, Herr RA, Gerlach JA, Mendoza L. Chitin synthase class 2 (CH82)
gene from the oomycete pathogen Pythium insidiosum. 99‘" General Meeting of
the American Society for Microbiology, May 1999, Chicago, IL.

Powles MA, Liberator P, Anderson J, Kerkhanis Y, Dropinski JF, Bouffard FA,
Balkovec SM, Fugioke H, Aikawa M, McFadden D, Schmatz D. 1998. Efficacy of
MK-991 (L-743, 872), e semisynthetic pneumocandin, in murine models of
Pneumocystis carinii. Antimicrob Agents Chemother. 42(8): 1985-9.

Ruiz-Herrera J. Chitin and Chitosan. In: Fungal cell wall: structure, synthesis and
assembly. Boca Raton: CRC press. 1992.

Sathapatayavongs B, Leelachaikul P, Prachaktam R, Atichertekam V,
Sriphojanart S, Trairatvorekul P, Jirasirithem S, Nontasut S, Eurvilaichit O, Flegel
T. 1989. Human pythiosis associated with thalassemia hemoglobinopathy
syndrome. J Infect Dis. Feb; 159(2): 274-80.

Sedrish SA, Moore RM, Valdes-Vasquez MA, Haynes PF, Vicek T. 1997.
Adjunctive use of a neodymium: yttrium-aluminum-gamet laser for treatment of
pythiosis granulomas in two horses. JAVMA. 211(4): 464-5.

Shenep JL, English BK, Kaufman L, Pearson T, Thompson JW, Kaufman RA,
Frisch G, Rinaldi MG. 1998. Successful medical therapy for deeply invasive facial
infection due to Pythium insidiosum in a child. Clin. Infect. Diseases. 27: 1388-
93.

68

Stevens DA. 2000. Drug interaction studies of e glucan synthase inhibitor (LY
303366) and a chitin synthase inhibitor (nikkomycin Z) for inhibition and killing of
fungal pathogens. Antimicrob. Agents Chemother. 44(9): 2547-48.

Thianprasit M, Chaiprasert A, lmwidthaya P. 1996. Human pythiosis. Curr. Top.
in Med. Mycol. 7(1): 43-54.

Thitithanyanont A, Mendoza L, Chuansumrit A, Pracharktam R, Laothametas J,
Sathapatayavongs B, Lolekha S, Ajello L. 1998. Use of an immunotherapeutic
vaccine to treat a life-threatening human arteritic infection caused by Pythium
insidiosum. Clin Inf. Diseases. 27: 1394-400.

Thomas RC. 1942. Composition of fungus hyphae Ill. The pythiaceae. Ohio J.
Sci. 42: 60-2.

Triscott JA, Weedon D, Cabana E. 1993. Human subcutaneous pythiosis. J.
Subcutaneous Pathology. 20: 267-71.

Wanachiwanawin W, Thianprasit M, Fucharoen S, et al. 1993. Fetal arteritis due

to Pythium insidiosum infection in patients with thalassemia. Trans. Roy. Soc.
Trop. Med. Hyg. 87: 296-8.

69