|
|
||||||||
1 From the Vanderbilt University Medical School, Nashville, Tennessee; 2 Mitotix, Inc., Cambridge, Massachusetts; and 3 Baylor College of Medicine, Houston, Texas.
| Abstract |
|---|
|
|
|---|
METHODS. Two strains, SC5314 and VE175, were examined. Standardized inocula were placed on the débrided corneal surface of one eye in Dutch belted rabbits and covered with a contact lens. A temporary tarsorrhaphy was opened after 24 hours with removal of the contact lens. Six days later, corneas were photographed and animals killed. Corneas were bisected with one half for quantitative isolate recovery and the other for stromal penetration by hyphae.
RESULTS. Strain SC5314 was significantly more virulent. The mean hyphal penetration into the cornea was 24.4% ± 8.5% of the corneal thickness, and in three of six corneas hyphae penetrated through the entire cornea. In contrast, for VE175, the mean hyphal penetration was 2.6% ± 1.2%. The difference between these two strains was statistically significant (P = 0.0297). Hyphae did not penetrate into the deep layers of the cornea in any of the six rabbits infected with VE175. The grading of clinical disease was consistent with histology, in that strain SC5314 caused more severe infection than VE175 and the difference was statistically significant (P = 0.0048). There was no difference in isolate recovery.
CONCLUSIONS. Wild-type strains of C. albicans can differ significantly in virulence as measured by depth of fungal invasion into corneas and clinical evaluation of infection. Further characterization of the intrinsic genetic differences between such strains may help identify factors responsible for fungal virulence.
| Introduction |
|---|
|
|
|---|
| Materials and Methods |
|---|
|
|
|---|
Inoculum
For these experiments, both strains were cultured overnight in
yeast extract-peptone-dextrose broth at 30°C on a rotary shaker at
200 rpm. We used 25 ml of media in a 50-ml sterile polypropylene tube.
Just before inoculation the cultures were centrifuged at
2500g for 3 minutes, the supernatant was poured off, and the
tubes were inverted and allowed to drain. The yeast pellet was used to
inoculate the rabbits corneas.
Animals
Dutch belted rabbits of either sex weighing 1 to 2 kg were used in
these experiments. All animals were treated in accordance with the ARVO
Statement for the Use of Animals in Ophthalmic and Vision
Research.
Model of Invasive Keratitis
Dutch belted rabbits were anesthetized with intramuscular ketamine
and xylazine.32
Corneal anesthesia was obtained with
topical 0.5% proparacaine hydrochloride. The nictitating membrane was
removed by sharp dissection. A 7-ml disc of epithelium was marked by
gentle pressure with a corneal trephine and completely removed by
atraumatic scraping with a 15 BardParker blade. An inoculum of 100
µl, containing approximately 2 x 108
blastoconidia, was placed on the denuded cornea and covered with a
flexible contact lens made from etafilcon A (Acuvue; Johnson and
Johnson Vision Products, Jacksonville, FL). A tarsorrhaphy prevented
extrusion of the contact lens. The tarsorrhaphy sutures and contact
lens were removed 24 hours later.
Clinical Evaluation
Before death, the infected eyes of all rabbits were photographed.
Three observers, masked to the identity of the individual rabbits,
examined these photographs and scored the severity of the infection as
follows: grade 0, no infection; grade 1, less than 25% of the central
7 mm (débrided section) of the cornea involved; grade 2, 25% and
50% of the central 7 mm of the cornea involved; grade 3, 50% to 75%
of the central cornea involved; grade 4, 75% and 100% of the central
cornea involved; and grade 5, any amount of corneal involvement with a
hypopyon present.
The observers also ranked all twelve rabbits from 1 to 12 based on severity of corneal disease with 1 representing the least infected cornea and 12 the most severely infected cornea. For this ranking, no ties were allowed.
Postmortem Tissue Dissection
Rabbits were killed by rapid intravenous injection of a 26%
pentobarbital sodium euthanasia solution (Sleepaway, Fort Dodge
Laboratories, Fort Dodge, IA). The infected corneas were removed at the
limbus and bisected longitudinally. One half was cut into small pieces
for quantitative isolate recovery, and the other half was fixed in
formalin for histology.
Quantitative Isolate Recovery of Fungi from Infected Corneas
Corneal halves were placed individually into 3-ml sterile normal
saline in a test tube. They were then processed for quantitative
isolate recovery using our standard method of grinding the cornea and
plating 10 or 100 µl samples of serial dilutions, in triplicate, on
Sabourauds agar plates (Becton Dickinson Microbiology
Systems, Cockeysville, MD). The colonies were then counted, and the
number of colony-forming units recovered from each corneal half was
calculated.3
7
Histology
In previous studies with strain VE175 we have demonstrated a very
high correlation between data derived from a single section taken from
the central cornea and a series of sections throughout the
cornea.33
Examination of three sections from each rabbit
in this experiment, spaced at 100 µm from the center of the corneal
lesion, showed no gross differences between rabbits for either strain.
Therefore, in this experiment we studied a single corneal section in
detail for each rabbit. This 5-µm section was cut from the midcornea,
along the edge of the embedded corneal half. The section was stained
with periodic acidSchiffs reagent, and counter-stained with fast
green stain.
Hyphal Penetration
Pictures were taken of one entire corneal section from each
infected cornea with an exposure control unit (model PM-20; Olympus,
Lake Success NY) attached to a microscope (model BX-40; Olympus). The
final magnification of -150 allowed us to determine and mark the
deepest point of corneal penetration of individual hyphae. We then
measured the corneal thickness and hyphal length for each observed
hypha with a ruler and calculated the depth of penetration as a
percentage of corneal thickness. As a measure of the total fungal
invasion for each cornea, the fungal biomass was calculated by summing
the hyphal penetration measurements for each cornea. All observations
were made by masked observers.
Statistics
Standard analysis of variance methods were used to compare the two
groups of rabbits. A log transformation of the isolate recovery rate
was performed. The numbers of hyphae and the percentage invasion of the
cornea by the hyphae were compared directly without transformation.
These parametric data were analyzed using a general linear model
procedure. Because there were only six rabbits in each group, the
parametric data, as well as the nonparametric data derived from the
clinical scores and clinical ranks, were analyzed using the Wilcoxon
two-sample test with continuity correction. Statistical analyses were
performed using commercial software (SAS for Windows; SAS Institute,
Cary, NC).
| Results |
|---|
|
|
|---|
|
|
|
|
The high magnification histology (Figs. 1E 1F) demonstrates that during corneal invasion the morphology of the two strains was different. Strain VE175 invaded the cornea in a pseudohyphal form characterized by the pinching of the cell wall between cells at the septae. Strain SC5314 invaded the cornea in what appeared to be a true hyphal phase.
After detailed analysis was completed for a single section from each cornea, multiple serial sections for all corneas were scanned for evidence of complete corneal penetration. For strain SC5314, the hyphal penetration was 100% with fungi crossing Descemets membrane into the anterior chamber in three of the six corneas (Fig. 1G ). For strain VE175, the fungus invaded farther than half the corneal thickness in only one of the six corneas and did not reach Descemets membrane in any cornea.
| Discussion |
|---|
|
|
|---|
First, the cornea is not traumatized other than by removal of the epithelium so that spores are not inoculated into the stroma. As we observed in these animals, when the organism is inoculated into the cornea, as occurred in several rabbits when the trephine cuts penetrated the stroma, the whole dynamic of the infection is changed. Because of this observation, we no longer use a trephine to mark the epithelium but instead use a filter paper disc moistened with N-heptanol to remove the epithelium from the cornea.
Second, in our previous models of corneal infection, blastoconidia were injected directly into the corneal stroma. Because some but not all of the blastoconidia germinate, viable fungus is present in the stroma in both growing and resting phases. This leads to confusion of the evaluation of disease by both histology and quantitative isolate recovery.
In this model, organisms invading the cornea are only in the growing phase, mimicking human infection. Blastoconidia that do not attach and invade the cornea appear to be swept away from the corneal surface, probably by the mechanical action of the lids.
Third, the host animal is immunocompetent. We have not found it necessary to use local injection of corticosteroid to initiate infection. With this model, we are able to detect differences in virulence between strains that are capable of establishing disease in the rabbit cornea. Of the three measures used, clinical photography, histologic assessment and quantitative mycology, only quantitative isolate recovery failed to demonstrate a difference between these two strains. The histologic parameters chosen (number of hyphae per section, biomass, and depth of invasion into the cornea) clearly demonstrate aspects of virulence that differ between the two strains. These measurements, although time consuming and tedious, are straightforward to obtain because fungal elements are clearly visible under the microscope, with appropriate staining. Alterations in corneal thickness, as a result of the infection or as a fixation artifact, add some uncertainty to the assessment of depth penetration but the difference in invasiveness between the these strains was sufficiently great to render dependence on precise measurements unnecessary. Although we did not study them, other objective histologic measures deserve further investigation, including measurement of horizontal spread of the fungus through the cornea, and the severity of the host response.
Histologic examination of multiple sections from infected corneas is extremely labor intensive. In previous work, we have shown that analyses of one to three sections, separated by 100 µm, from an infected cornea correlate well with data from multiple sections throughout the cornea.33 This greatly reduces the amount of work involved in each experiment.
Clinical evaluation by grading of photographs of infected corneas produced the same result. Although the scoring method was empirical, it correlated with a severity ranking of the disease including all 12 rabbits for the three masked observers. The failure to detect a difference between the two strains by quantitative isolate recovery was not unexpected, given the inability of this method to differentiate between groups of rabbits with clearly different clinical disease that we have observed in other studies.33 The reason for this insensitivity seems to be the varied colony-forming potential of hyphal fragments of unequal length or cell numbers. Quantitative isolate recovery appears to be most suited to assessment of unicellular phase organisms, particularly when subtle differences are present. For these studies, we endeavored to control for experimental factors that might influence virulence. Each strain was maintained in an identical manner, the preparation of the inoculum was identical, and the concentration of the inoculum paste, although not the same, was similar. Only a small percentage of the inoculum attaches to the cornea and is not eliminated when the tarsorrhaphy sutures and contact lens are removed on the day after inoculation. In previous experiments we have shown that the keratitis caused by C. albicans in Dutch belted rabbits is remarkably consistent from animal to animal, and for this reason this animal species was selected for these studies.34 All observations were made in a masked fashion. Thus, although we cannot exclude extraneous factors that could influence the degree of penetration, and the actual measurements have an element of uncertainty due to edema of the cornea that occurs with the infection, all the evidence from these studies points to an intrinsic difference between these strains to account for the difference in virulence.
Various intrinsic differences between the C. albicans strains can be responsible for the observed marked difference in pathogenicity in vivo. First, fungal genes that control morphogenesis may be involved. Recent work in the signal transduction pathways that lead to hyphal or yeast formation in C. albicans demonstrated that genetic mutations in key virulence factors of the fungus can have profound effect on fungal virulence.21 24 For example, Lo et al.24 showed that null mutations in proteins such as CPH and EFG lead to a locked morphologic state of the fungus in the presumed less virulent yeast form. On the contrary, Braun and Johnson21 demonstrated that knockout deletion mutations in transcriptional repressors such as TUP1 could in fact lock the organism in the presumed more invasive hyphal form. It should be cautioned that the relationship between dimorphism and virulence in C. albicans has not been well established, evidenced by our recent study that the TUP1 homozygous deletion mutant is in fact less invasive than the isogenic strain in our model of contact lensfacilitated keratitis in rabbits.35 In the present work, it is interesting to note that the less virulent strain, VE175, was unable to form true hyphae. In appears, therefore, that although the relationship between dimorphism and C. albicans pathogenicity is not well established, C. albicans strains that fail to form true hyphae in vivo are perhaps less penetrative, and thus cause less disease in tissues such as the cornea. In addition to the investigation of genes controlling C. albicans dimorphism, other genes that are involved in fungal adherence and penetration can also be involved in fungal virulence.36 37 38 39
In summary, we believe the present work demonstrates that intrinsic differences between C. albicans wild-type strains can lead to marked difference in predilection toward clinical infection. Further investigation in the molecular mechanisms responsible for this difference may shed new light on the pathogenic mechanisms of this fungus and suggest new therapeutic targets.
| Footnotes |
|---|
Submitted for publication September 27, 1999; accepted December 1, 1999.
Commercial relationships policy: N.
Corresponding author: Denis M. ODay, Department of Ophthalmology and Visual Sciences, Vanderbilt University School of Medicine, Medical Center East, Room 8032, 1215 21st Avenue, South, Nashville, TN 37232-8808. denis.m.o'day{at}mcmail.vanderbilt.edu
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
B. E. Jackson, K. R. Wilhelmus, and B. Hube The Role of Secreted Aspartyl Proteinases in Candida albicans Keratitis Invest. Ophthalmol. Vis. Sci., August 1, 2007; 48(8): 3559 - 3565. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. E. Jackson, B. M. Mitchell, and K. R. Wilhelmus Corneal Virulence of Candida albicans Strains Deficient in Tup1-Regulated Genes Invest. Ophthalmol. Vis. Sci., June 1, 2007; 48(6): 2535 - 2539. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. M. Mitchell, T. G. Wu, B. E. Jackson, and K. R. Wilhelmus Candida albicans Strain-Dependent Virulence and Rim13p-Mediated Filamentation in Experimental Keratomycosis Invest. Ophthalmol. Vis. Sci., February 1, 2007; 48(2): 774 - 780. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Goldblum, B. E. Frueh, G.-M. Sarra, K. Katsoulis, and S. Zimmerli Topical Caspofungin for Treatment of Keratitis Caused by Candida albicans in a Rabbit Model Antimicrob. Agents Chemother., April 1, 2005; 49(4): 1359 - 1363. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |