Ebook Current topics in medical mycology (Vol.4): Part 1

(BQ) Part 1 book “Current topics in medical mycology” has contents: The infection of human skin and nail by scytalidium species, the use of molecular techniques for epidemiologic typing of candida species, skin kinetics ofazole antifungal drugs,… and other contents.

Current Topics 4

in Medical Mycology

LIBERO AJELLO, Ph.D., Emory University Eye Center, Ophthalmic Research, 3rd Floor, 1327 Clifton Road, N.E Atlanta, Georgia 30322, U.S.A

GARRY T COLE, Ph D., Department of Botany, The University of Texas at Austin, Austin, Texas 78712, U.S.A

REBECCA A Cox, Ph D., Research Immunology, San Antonio State Chest Hospital, San Antonio, Texas 78223, U.S.A

DAVID J DRUTZ, M.D., Biological Sciences, Smith Kline and French Laboratories, Swedeland, Pennsylvania 19479, U S A

KAzuo IWATA, M.D., Hattori Seiko C Ltd., 1-10 Kajicho 2-chome, Chiyoda-ku, Tokyo 101, Japan

GEORGE S KOBAYASHI, Ph.D., Division of Dermatology, Washington University School of Medicine, St Louis, Missouri 63110, U.S.A

C.P KURTZMAN, Ph.D., Culture Collection Research, Fermentation Laboratory, USDA-ARS, Northern Regional Research Center,

1815 Northern University Street, Peora, Illinois 61604, U.S.A

THOMAS G MITCHELL, Ph.D., Department of Microbiology and

Immunology, Duke University Medical Center, Durham, North Carolina

JOHN L RICHARD, Ph D., USDA-ARS, Northern Regional Research Center,

1815 Northern University Street, Peora, Illinois 61604, U.S.A

HISASHI TAKAHASHI, M.D., Department of Dermatology, Teikyo University, School of Medicine, 11 Kaga-2, Itabahiku, Tokyo 173, Japan

H UGO VANDEN BOSSCHE, Department of Comparative Biochemistry, Janssen Research Foundation, B-2340 Beerse, Belgium

London Bridge, London SEl 9RT, UK

ISSN 0177-4204

© 1992 by Springer-Verlag New York Inc

Softcover reprint of the hardcover I st edition 1992

All rights reserved This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer-Verlag, 175 Fifth Avenue, New York, NY 10010, USA), except for brief excerpts in con- nection with reviews or scholarly analysis Use in connection with any form of information storage and retrieval, electronic adaptation, computer software,

or by similar or dissimilar methodology now known or hereafter developed is forbidden

The use of general descriptive names, trade names, trademarks, etc in this publication, even if the former are not especially identified, is not to be taken

as a sign that such names, as understood by the Trade Marks and Merchandise Marks Act, may accordingly be used freely by anyone

While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made The publisher makes no warranty, express or implied, with respect to the material contained herein

Typeset by Asco Trade Typesetting Ltd., Quarry Bay, Hong Kong

9 8 7 6 5 4 3 2 1

ISBN-I3: 978-1-4612-7657-9

DOl: 10.1007/978-1-4612-2762-5

e-ISBN-13: 978-1-4612-2762-5

Series Preface

Current Topics in Medical Mycology is intended to

summa-rize current research areas in medical mycology for medical mycologists and other scientists who are working in microbiol-ogy and immunology Topics to be included in each volume will serve as contemporary reviews, summaries of current advancements and future directions, and mechanisms to en-hance the interdisciplinary use of medically important fungi

in understanding pathogenesis, epidemiology, mycotoxins, taxonomy, and other areas where basic, applied, and clinical sciences are used

Marcel Borgers Roderick Hay Michael G Rinaldi

L POLONELLI, G MORACE, S CONTI,

NEIL S RYDER AND HUBERT MIETH

HUGO DEGREEF

vii

8 Molecular Approach to the Toxic Action of

Quinone Mycotoxins-Chemical Structure

KIYOSHI KAWAI, KAZUO HISADA, HIDEKI MORI,

AND YOSHINORI NOZAWA

Microbial Diseases Laboratory, California State Department

of Health Services, Berkeley, CA 94701, USA

M GERLONI, M.D

Istituto di Microbiologia, Universita degli Studi di Parma, Parma, Italy

ix

CARLYN HALDE, Ph.D

Department of Microbiology, University of California, San

Francisco, CA 94143-0414, USA

KAzuo HISADA, Ph.D

Department of Biochemistry, Gifu University School of

Medicine, Gifu, Japan

HIDEOIKEMOTO, M.D

Department ofInternal Medicine, Juntendo University

School of Medicine, Tokyo, Japan

KIYOSHI KAWAI, Ph.D

Department of Food and Nutrition, Chukyo Women's

University, Aichi, Japan

Mycology Department, Institute of Dermatology, United

Medical and Dental Schools of Guy's and St Thomas'

Hospitals (University of London); Mycology Unit,

Department of Microbial Diseases, St Thomas' Hospital,

London, England, United Kingdom

G MORACE, M.D

Istituto di Microbiologia, Universibi Cattolica del Sacro

Cuore, Rome, Italy

HIDEKI MORI, M.D

Department of Pathology, Gifu University School of

Medicine, Gifu, Japan

RICARDO NEGRONI, M D

Faculty of Medicine, Centro de Micologia del Departamento

de Microbiologia, Buenos Aires, Argentina

P NELSON, Ph.D

The Fusarium Research Center, Pennsylvania State

University, University Park, PA, USA

YOSHINORI NozAwA, Ph.D

Department of Biochemistry, Gifu University School of

Medicine, Gifu, Japan

Department of Dermatology, Sandoz Forschungsinstitut, Vienna, Austria

MIRIAMVALESCO, Ph.D

Alameda County Public Health Laboratory, Oakland,

CA 94607, USA

I-The Infection of Human Skin and Nail by Scytalidium Species

MARY K MOORE

Introduction

The Form-genus Scytalidium was established by Pesantel in 1957 to accommodate dematiaceous hyphomycetes producing fission arthroconidia The nomenclatural type for the genus was an isolate from rotting Platanus,

small, hyaline fission arthroconidia together with larger, brown conidia, which in the original description are termed chlamydoconidia During the last 20 years a number of fungi currently classified within this genus have been reported to be human pathogens

The first of these was Hendersonula toruloidea Nattrass, a coelomycete characterized by the production of pycnidial and stromatal conidiomata, but with a well-recognized accompanying Scytalidium synanamorph producing hyaline and dematiaceous arthroconidia In 1970, Gentles and Evans2 isolated this fungus from the toe webs and toenails of eight patients in the United Kingdom The difficulties faced in assessing the significance of the isolation of a fungus not generally recognized as a human pathogen were fully appreciated by these workers The possibility that the isolate repre-sents not a pathogen but a transient or a secondary invader or even a labora-tory contaminant must always be considered in such cases However, Gen-tles and Evans2 felt that in the patients they were describing, sufficient evidence had accumulated to suggest that H toruloidea might be the cause

of the clinical condition observed

Seven of the eight patients described were suffering from lesions tive of infection by a dermatophyte, a closely related group of fungi com-prising members of the form-genera Trichophyton, Microsporum, and

toe webs Microscopical examination of skin and nail samples revealed hyphae indistinguishable from those observed in dermatophytosis, yet cul-tures of the material yielded H toruloidea The fungus was isolated only on

1

media free of cycloheximide, an antibiotic that is often routinely added to culture media when the isolation of a dermatophyte is attempted, specifical-

ly to inhibit the growth of nondermatophyte molds The possibility that H

was investigated in two ways First, clinical material was cultured both on media supplemented with cycloheximide and on media free of this anti-biotic, and it was demonstrated that no dermatophyte could be isolated, even when the growth of H toruloidea was inhibited Second, H toruloidea

was inoculated beside the dermatophytes Epidermophyton floccosum,

cycloheximide, and it was demonstrated that growth of the dermatophytes was unaffected by the contiguous growth of the nondermatophyte Further evidence that H toruloidea might actually be the fungus observed in the tissue samples was the consistent reisolation of this fungus from five of six patients reexamined

As H toruloidea is not endemic in the United Kingdom but has been isolated from plants and soil samples in many parts of the tropics and subtropics,3 it was considered significant that seven of the eight patients were immigrants to the United Kingdom from East Africa, the Indian sub-continent, or Fiji

Overall, the weight of evidence led Gentles and Evans2 to suggest that other cases of H toruloidea infection might have been misdiagnosed, as their first cases initially were, as dermatophyte infections in which the causa-tive organism had failed to grow This suspicion has since been confirmed by reports of H toruloidea infection not only from other parts of Europe but also from Africa, Asia, the Pacific, and the Americas The majority of publica-tions describe single cases or small groups of patients (Table 1-1), but more ambitious surveys of the prevalence and distribution of the infection have also been reported, both from areas where the infection is probably im-ported, such as the United Kingdom24.25 and Canada,26 and from areas where the fungus may be endemic among both human and plant material, including Trinidad and Tobago,27 Thailand,28 Gabon,29 and Nigeria 30

In 1977 a new species of filamentous nondermatophyte, Scytalidium

At that time it was noted that the infections showed many remarkable lels to those caused by H toruloidea The patients were immigrants from the West Indies or West Africa and presented with lesions suggestive of tinea on the hands and feet Hyphae observed on microscopical examination

paral-of skin or nail material resembled those paral-of dermatophytes, although some were of variable diameter The fungus was sensitive to cycloheximide in the culture medium; on media free of cycloheximide an arthroconidial mold was isolated, which differed from H toruloidea, however, by the lack of produc-tion of dark pigment Despite the fact that the genus Scytalidium is placed among the dematiaceous hyphomycetes, the authors considered that the similarity in the mode of conidial production and in the irregular arrange-

TABLE 1-1 Case reports of H Toruloidea infection of thick skin and nails

East Africa, Indian subcontinent, Fiji, United Kingdoma

Cameroons, Ivory Coast

a Four U K patients had traveled to possibly endemic areas

b One patient had no evidence of fungus on microscopy-significant?

ment of the hyphae justified the placing of this hyaline mold within the genus Further cases of S hyalinum infection of the skin and nails have subsequently been described from Europe7•32-34 Africa,3o and America 2o Examination of the various reports of H toruloidea and S hyalinum infec-tion demonstrates that a number of inconsistencies appear with regard to the naming of the pathogens isolated from clinical material This has, in part, arisen from the polymorphism of H toruloidea, possessing both coelomycete and hyphomycete synanamorphs The earlier reports of infection by this fun-gus identify the isolates as H toruloidea; in fact, this is the name which strictly refers to the more complex coelomycete synanamorph The arthro-conidial hyphomycete synanamorph, which is considered less developed, should be referred to as the Scytalidium synanamorph of H toruloidea As it

is this hyphomycete synanamorph which is invariably the primary isolate from clinical material, more recent publications tend to report the fungus under this name

The situation is complicated by the fact that some workers would argue that this Scytalidium synanamorph of H toruloidea should actually be given

a specific epithet, resulting in two binomials to describe the different morphs of a single fungus Thus, there have been several reports of S ligni-

the fungus isolated is the Scytalidium synanamorph of H toruloidea, but the authors believe it to be the same fungus described by Pesante.1 A further complication arises from a report of S lignicola infection in which it is stated that the unpigmented mold S hyalinum is the Scytalidium anamorph of the

dematiaceous H toruloidea.23 Most recently, Sutton and Dyko36 have vised the genus Hendersonula and not only regard the Scytalidium synana-morph of H toruloidea as identical with S lignicola but also suggest a new binomial for this fungus, S dimidiatum, and believe that the coelomycete synanamorph should be placed in a new genus as Nattrassia mangiferae

re-The purpose of this review is to discuss both the clinical and mycological features of infections caused by these fungi For the sake of clarity in the initial clinical discussion, infections will be referred to as H toruloidea when the isolate was of the dematiaceous Scytalidium synanamorph of H toru-

rela-tionship of the Scytalidium synanamorph of H toruloidea to S lignicola and

FIG 1-1 S hyalinum infection of palm

FIG 1-2 H toruloidea

infection of sole

squamous hyperkeratosis of the palms and soles (Figs 1-1 and 1-2), with scaling particularly along the skin creases, is indistinguishable from that seen

vari-able; although in some cases scaling may be severe, resulting in a classical

"moccasin foot" appearance, in others lesions are minimal and the patient is unaware of any abnormality.2,5,15,27 It is noteworthy, however, that spread

to the thin skin of the dorsum of the hand or foot does not occur As in dermatophytosis, unilateral involvement of one palm is not infrequently reported 5,7,19,21,32 Vesicular lesions have been recorded in only one in-stance,8 although an inflammatory response with a clear margin of erythema along the lateral or medial border of the foot has been observed in some cases 32 Toe web infection is characterized by scaling and, to a lesser extent,

by maceration In a survey of 128 cases of H toruloidea or S hyalinum

infection, Hay and Moore32 recorded only 4 instances of toe web maceration;

a recent survey of H toruloidea infection in Thailand,28 however, observed maceration in some 11 % of infected toe webs, and several individual case reports also report toe web maceration 14,20,23 Itching is variable Although

skin lesions may be minimal, infected nails are often severely abnormal In affected toenails, the appearance is most often that of a distal or lateral sub-ungual hyperkeratosis, with onycholysis and thickening and opacification of the nail plate Brown to black discoloration of the nails has frequently been

observed in H toruloidea infection5.6.12.17 -22 and may be particularly tressing in European-type caucasoids Rush-Munr037 noted that the darken-ing of nails seen in patients of European ancestry in the Pacific Islands was especially disfiguring, and Jones et al.lO reported that the dark pigmentation seen in four European caucasoids was particulate and characteristic

dis-It has been suggested that some features of nail invasion by H toruloidea

and S hyalinum, seen particularly in fingernails, are pathognomonic

Camp-bell et al 5 noted that among patients with H toruloidea infection, invasion started at the side of the nail, and Hay and Moore32 also reported that onycholysis of the lateral nail plate was often the predominant form of invasion (Fig 1-3) This led to involvement of the whole nail plate without thickening, followed by transverse fracture and loss of the major portion

FIG 1-3 Characteristic lateral invasion of finger-nail by H torulnidea

TABLE 1-2 H toruloidea and S hyalinum infections: culturally positive sites

% Isolations

Feet

The overall pattern of infection, with invasion of both skin and nail tissues, closely parallels dermatophytosis and is radically different from that seen in skin and nail infections by other molds, where either skin alone is involved,

as in tinea nigra, or nails alone are colonized, as in Scopulariopsis brevicaulis

infection Analysis of the sites infected in the various case reports and larger surveys (Table 1-2) demonstrates that although the skin of the foot and toenails are most commonly reported, involvement of the palms and/or fingernails occurs in a substantial percentage of patients and that examina-tion of these sites in all patients with proven infection of the feet may be worthwhile

Reports of the infection of thin skin sites or hair by H toruloidea are rare One case of groin infection was reported during a large survey carried out in Thailand,28 but unfortunately no details are given The fungus has been iso-lated from scalp hair in two instances Camerlynck et al.16 isolated the fun-gus from an African child temporarily resident in Dijon, France However, repeated microscopical examination of scales and hairs failed to reveal hyphae, and the isolation was regarded as a contaminant More recently, Frankel and Rippon21 reported the isolation of H toruloidea from the scalp

of a 15-month-old girl in Chicago and suggested that this was the first case of "tinea capitis" caused by the fungus As no details of the microscopi-cal examination of scales or hairs are given, the significance of this isolation

is impossible to assess

FIG 1-4 Hyphae of H toruloidea in skin scales; 30% potassium hydroxide mount

Phase contrast; X 460

Results of Microscopical Examination of Skin and Nail Samples

Although H toruloidea is a dematiaceous mold, the original report of tion by this fungus observed that the hyphae seen on microscopical observa-tion of skin and nail samples were hyaline and indistinguishable from those

infec-of a dermatophyte, 12 and hyaline hyphae have been reported in the majority

of later publications Certain characteristic differences between these hyphae and those of dermatophytes have now been recognized, however, including the irregularity of width of the hyphae5 (Fig 1-4), their sinuous nature,25 and a double-contoured appearance caused by the apparent with-drawal of the cytoplasm away from the hyphal wall.25 Together, these dif-ferences are sufficient for the experienced observer to distinguish der-matophyte infections and those caused by H toruloidea by examination of skin and nail scrapings microscopically The arthroconidia, seen particularly

in material from the toe spaces, have also been reported to differ from those

of the dermatophytes due to a thicker wall and long-oblong shape5 or an irregular shape and the occasional presence of septa.25 The hyaline hyphae observed in material from cases of S hyalinum infection have in every in-stance been described as indistinguishable from those seen in H toruloi- dea infections 7,25,31,33

In a relatively small number of cases,4,5, 10, 12, 13,23,25 pigmented, times rough-walled hyphae (Fig 1-5) have been observed in material from

saprophy-tic mode of invasion might not be unexpected, i.e., toe webs and nails In

FIG 1-5 Pigmented hyphae in toe web scales infected by H toruloidea; 30%

potas-sium hydroxide mount Brightfield; x 360

many instances, however, the characteristic distorted hyaline hyphae were also observed

Culture of H toruloidea and S hyalinum

Both of these molds will grow well on the majority of commonly used culture media At St John's Hospital, Sabouraud's dextrose agar is the medium used for routine isolation,25 but the fungi have also been shown to grow well on malt extract agar.5,31 Growth on cornmeal agar and potato dextrose agar pro-duces colonies with less aerial hyphae and surface growth, which are there-fore more difficult to recognize; growth on Czapek-Dox agar may produce very atypical, pale, folded colonies 38 A recent survey in Canada26 recom-mends Littman's oxgall agar as another suitable medium Cycloheximide is not completely inhibitory to all strains of these fungi,25,26,37 but at the con-centrations commonly used (100 to 500 JLglml only about 5% of isolates are capable of primary growth from skin or nail material Suitable incubation temperatures are 26 to 30°C, and plates should be held for a minimum of 3 weeks if direct microscopy is positive

Prevalence of Infection

Studies of the prevalence of H toruloidea and S hyalinum infections fall into two diverse groups: those conducted within areas where the disease is be-

lieved to be imported and occurs almost entirely among immigrants from the tropics and subtropics, and those from presumed endemic areas

The first indication that H toruloidea infection was not a rarity came in

1977, when Clayton39 reported that this fungus accounted for 2.5% of skin and nail infections during 1976 among patients with tinea pedis or manuum diagnosed at St John's Hospital in London A more detailed study,24 carried out in the same department, observed that among 200 immigrant patients with fungal lesions of the hands, feet, or nails, H toruloidea and/or S

not randomly chosen, but selected because they showed features suggestive

of nondermatophyte infection, including the observation of characteristic hyphae in skin and nail samples, a previous history of positive direct micro-scopy followed by negative culture on media supplemented with cyclohex-imide, and a failure to respond to griseofulvin therapy A further, completely random survey was therefore performed 25 In all patients originating from outside Western Europe, material from fungal lesions of the feet, hands, or nails was cultured in duplicate, both on media supplemented with cyclohex-imide and on media free of this antibiotic Over a period of27 months, a total

of 399 patients were included in the survey and an overall prevalence of nondermatophyte infection of 11 % was demonstrated H toruloidea was iso-lated from 30 patients, S hyalinum was isolated from 9 patients, and both fungi were isolated from 2 patients It was noteworthy, however, that certain geographical or racial groups showed a particularly high degree of infection: among 19 patients from Bangladesh, no less than 10 (53%) proved to suffer from H toruloidea infection, and among the 44 culturally positive negroid patients examined, 21 (44%) were infected by one or both of these molds

A recent abstract suggests that a similar prevalence of infection occurs among the immigrant population of France 34 During the period 1981 to

1987, no less than 79 cases of H toruloidea infection and 62 cases of S

The patients were immigrants from the West Indies, Guyana, and larly the former French colonies of Africa

particu-The recent influx of immigrants from the West Indies, West Africa, and southern Asia into Canada may result in a similar picture Summerbell et

al 26 have recently surveyed the nondermatophytes implicated in skin and nail infections in Toronto, Ontario More than 4,000 patients were included, and the overall isolation rate of less than 2% sounds almost insignificant However, this represents 28 cases of H toruloidea infection and 5 isolations

of S hyalinum among the relatively small group of immigrant patients cluded in the sample

in-Overall, these reports from Europe and North America suggest that H

ende-mic areas, and this has been confirmed by several surveys

The first attempt to assess the prevalence of these infections in an endemic

area was carried out on the island of Tobago 27 Forty-five medical inpatients receiving treatment for disorders other than skin disease were included Scrapings from the soles and toe webs were examined microscopically and cultured on cycloheximide-free medium Ten patients yielded a non-dermatophyte-H toruloidea in three instances and S hyalinum in seven instances-and a further nine patients showed hyphae characteristic of these infections on direct examination of skin scales This implies an overall preva-lence of infection in the general population of no less than 45%

A very similar level of infection has been reported from Thailand,28 where the fact that some 20% of cases of tinea pedis proved recalcitrant to treat-ment inspired a survey of possible H toruloidea infection Among 145 sol-diers examined, atypical hyaline hyphae suggestive of non dermatophyte in-fection were observed in 68 instances (47%) and H toruloidea was isolated from 60 subjects (41%) Dermatophytes, in contrast, were isolated from only

12 cases (8%), 4 of these in mixed infections with H toruloidea

Recent surveys carried out in West Africa also record a significant degree

of infection 29.30 Of 250 coal miners examined in Nigeria, 75 (30%) had nicallesions of the feet suggesting mycotic infection and 66 (26.4%) yielded skin and nail samples positive on direct microscopy Hendersonula toru-

micro-scopically positive cases and also present in three mixed infections (4.5%)

from one case of mixed infection (1.5%) Dermatophytes alone were isolated from 35 (53%) of the 66 patients Looking at the whole sample, this leads to

an overall nondermatophyte prevalence of 47% among those with positive direct microscopy

An abstract from the nearby region of Gabon29 reports a survey of 184 patients with foot epidermomycoses and shows a rather lower prevalence of infection than that recorded in the West Indies and Thailand H toruloidea

was isolated from some 21 % of toe webs, 18% of soles, and 17% of toenails

Geographical Distribution of Infection

In the majority of publications, the distribution of human H toruloidea fection is presumed to be closely linked to the distribution of this fungus as a plant pathogen Although of relatively recent interest to medical mycol-ogists, H toruloidea has been investigated by plant pathologists since the fungus was first fully described by Nattrass40 in 1933 Found on apple, plum, and apricot trees in Egypt, this mold· has since been recorded on a wide range of plant hosts, many of them of economic importance The isolation of

gummosis, branch wilt and dieback, and in some instances has been ated with severe disease, yet the fungus is regarded as a weak or secondary

associ-invader, growing on dead plant tissues and diseased or unthrifty hosts 3 ternal factors such as trauma and sunburn are believed to play an important role in the establishment of disease 42-45

tropi-cal, subtropitropi-cal, and some more temperate zones, but in northern Europe is not regarded as indigenous Samples from Portugal demonstrate the pre-sence of the fungus in southern Europe,36 and the mold may be identical with that described in 1887 by Penzig in Italy as Torula dimidiata 46 In 1923, Fawcett47 isolated a fungus from frost-damaged grapefruit and orange trees

in California, which he identified as T dimidiata The original specimens, together with new grapefruit isolates and others from walnut, lemon, and chestnut were later examined by Wilson,41 who described them as a new species, Exosporina fawcetti, believing that although the fungus was similar

information to prove the two fungi synonymous At that time, Wilson41 noted the similarity of the Torula stage of H toruloidea to E fawcetti, but as

no pycnidial or stromatal conidiomata were observed he did not feel justified

in regarding the isolates as strains of H toruloidea In 1949,48 Wilson ported that the production of pycnidia characteristic of H toruloidea by strains of E fawcetti had subsequently occurred, demonstrating that these two fungi were identical A fungus determined to be the Torula stage of H

view that the Scytalidium anamorph of H toruloidea and T dimidiata are identical has been accepted by several workers, including Punithalingam and Waterston,3 Ellis,49 and Sutton and Dyko 36

TABLE 1-3 Geographical areas where plant infection by H toruloidea has been

reported

Representative

In most of the known or presumed areas of endemic human infection by

and 1-3), and the geographical range of the fungus as a plant pathogen will doubtless be extended by future investigations It is interesting, however, that in two areas where the fungus has been extensively investigated as a significant cause of commercial crop losses, not a single case of human infec-tion has been recorded In the Middle East, the relative unimportance of skin and nail diseases might explain the failure to detect H toruloidea infec-tion On the West Coast of the United States, where a much more sophis-ticated medical service is available and extensive damage to crops such

as citrus fruits and walnut have been reported, the failure to diagnose this mycosis is more intriguing

The geographical distribution of S hyalinum infection appears more limited than that of H toruloidea There is no record of the isolation of this fungus from the environment, and as yet endemic human infection appears limited to the Americas, Caribbean, and West Africa

Other Factors Influencing the Distribution of Disease

There is as yet relatively little information available as yet about the sex and age distributions of these infections Examination of the literature dem-onstrates an overall predominance of male patients, but this is undoubted-

ly influenced by the reluctance of many immigrant-and particularly Asian-women to seek medical advice for relatively minor ailments A sur-vey in London25 showed a large majority of male patients infected by the two molds (the ratios of male/female infection were 6.75/1 for H toruloidea and 4.5/1 for S hyalinum), but statistical analysis showed no difference in the actual isolation rates between male and female patients However, as the groups compared were not matched for either geographical origin or racial type, this result must be regarded as very tentative It is unfortunate that the survey/s published from Thailand28 and Nigeria30 were confined to investiga-tions of male workers so that no information on the prevalence of these infec-tions in the two sexes is available from an endemic area

Similarly, it is difficult to assess the age distribution of infection Mycoses caused by H toruloidea and S hyalinum have been shown to be extremely chronic, and reisolation of these fungi over a period of years is not un-common;38 as a result, the age at first diagnosis may bear very little rela-tionship to the actual age at first infection These infections are not particu-larly prevalent in the elderly, however, as so many other mold infections of the nails tend to be, 75, 76 but may occur in patients in their teens and are common in adults between 20 and 50 years 0Id.25,30

Other factors which are believed to predispose to mold infection of the nails, such as trauma,76,77 have only rarely been reported in infection by H

However, three of the four cases of H toruloidea infection diagnosed in India12 were stated to be associated with contact with water; this has not been noted elsewhere

The significance of other concurrent disease or underlying abnormality is hard to assess The majority of cases have been diagnosed in skin clinics, and

it is well recognized that results from such clinics may actually bear very little relationship to the prevalence of disease in the general population 78; certainly, a very high prevalence of nonfungal skin and nail disease has heen noted among patients with H toruloidea or S hyalinum infection Three of the eight patients yielding H toruloidea described by Gentles and Evans2 and six of the ten recorded by Campbell et al 5 presented with skin lesions that were believed to be unrelated to their mycosis, and a similar picture is shown by the original reports of S hyalinum infection 7,31 Among 128 patients with these mycoses diagnosed in London,32 no less than 84 (66%) presented with other skin conditions, and the fungal infection was a secon-dary finding Other diseases included eczema, generalized pruritus, palmo-plantar hyperkeratosis, lichen simplex, urticaria, lichen planus, and leprosy, However, examination oflesions suggestive of these conditions38 showed no evidence of fungal infection except in the small minority of instances in which the soles, palms, toe webs, or nails were involved, and in these cases

it was impossible to determine whether fungal infection preceded or lowed the nonfungal disease Only two patients had serious underlying dis-ease, chronic active hepatitis in one instance and systemic lupus erythemato-sus in the other; in both, the clinical features of infection were unremark-able Diabetes has been diagnosed among only a handful of H toruloidea

fol-patients.5,8,32,38 Mixed infections, in which more than one fungus has been isolated from the same or different sites in a single patient, have frequently been reported in subjects with H toruloidea 2•5,25 or S hyalinum 7 ,25.31 infec-tion Among the 128 patients described by Hay and Moore,32 62 (48%) yielded more than one pathogen on culture, 10 (7.8 %) yielding both non-dermatophytes and 52 (41%) yielding a dermatophyte in addition to one or both nondermatophytes The recent surveys from Thailand28 and Nigeria, 30 however, demonstrated fewer mixed infections, only four patients (2.8%)

in Thailand yielding both H toruloidea and a dermatophyte and only four patients (6%) in Nigeria yielding a dermatophyte in addition to either of the two nondermatophytes

countercur-with only 9% of a control group composed of patients from endemic areas

and laboratory staff exposed to the organisms (P< 0.01) In every case,

cross-reactivity was demonstrated between the positive patients' sera and the heterologous non dermatophyte antigen

More recently, Kotrajaras et al.28 have reported the preparation of a skin test antigen from H toruloldea, which was used in 20 infected volunteers and shown to elicit reactions between 11 and 25 mm in diameter after 24 h

Treatment

resistance to many of the antifungal drugs so far tested (Table 1-4) It has become widely recognized that these infections are extremely recalcitrant to

TABLE 1-4 Reported MIC estimations and clinical response in superficial dium infections

treatment, and at the present time no consistently effective therapy has been reported Both fungi have been shown to be resistant to griseofulvin in vitro

at levels up to 100 mg/liter, and a corresponding lack of in vivo response has been extensively recorded A similar lack of response to ketoconazole has also been described Among the newer oral antifungals available, in vitro resistance and in vivo failure have been recorded for itraconazole and clinical experience with terbinafine is lacking, although in vitro sensitivity has been demonstrated

Topical therapy has also proved disappointing Although both H

polyene and imidazole compounds in vitro, clinical response has been able In many instances some initial clinical improvement has been noted with skin lesions, but mycological investigations have demonstrated the con-tinuing presence of the organisms, and cessation of therapy has been quickly followed by relapse.2•6 ,7,19,21 Less specific therapies such as Whitfield's ointment and iodinated alcohol have also produced little or no sustained effect It may be that the predilection of these molds for very thick keratin-ized tissues leads to a failure of adequate penetration

vari-Ironically, the few records of successful treatment published involve nail infections by H toruloidea Hay et al.85 reported the clinical and mycologi-cal cure of one of three patients included in an open trial of 28% tioconazole solution; the remaining two patients, however, showed little or no response

A good response was also noted in a patient treated with topical zole in Nigeria,17 although other workers have reported failure with the same treatment Rollman and Johansson19 described the clearing of a finger-nail infection following avulsion of the nail and long-term occlusive applica-tion of ciclopiroxolamine This was attributed to the avulsion, which resulted

clotrima-in a relatively thclotrima-in tissue layer that was more easily penetrated by the drug The thicker skin lesions treated at the same time with glutaraldehyde and ciclopiroxolamine failed to respond

Deep Infections

In contrast to the increasing recognition of both H toruloidea and S num as relatively common agents of superficial mycoses, reports of sub-cutaneous or deep infections caused by these molds remain rare

hyali-The first appeared in 1975,86 although the pathogen was not identified as

known defect of cell-mediated immunity At the age of 15 he had presented with a generalized Trichophyton verrucosum infection, involving superficial and deeper tissues Treatment with griseofulvin had improved the condi-tion, but it was not cured and successive relapses occurred The bizarre, lichenified suppurating facial lesions that he presented with 15 years later were clinically very different; lesions of the nails were also present

Histological examinations revealed hyaline fungal cells within an matory dermal granuloma Short, vesicular hyphae and unicellular forms up

was cultured on media free of cycloheximide

Treatment with amphotericin B led to healing of the face; the nails

verrucosum alone, for at the start of therapy both dermatophyte and

nonder-matophyte were isolated from them, but the latter mold may have been acquired from scratching of the facial lesions The nails finally healed upon treatment with griseofulvin, iodinated alcohol, tolnaftate, and avulsion

59-year-old West Indian male The isolate, in contrast to those reported from superficial lesions (Table 1-4), was reported to be sensitive to ketocona-zole in vitro (minimum inhibitory concentration [MIC], 0.36 mg/liter) After

4 months of treatment the lesion was apparently cured, but it relapsed after

12 months

In the same year, a case of subcutaneous phaeohyphomycosis was

had a previous history of myocardial infarction and was taking medication to control diabetes and hypertension Aspirated material from one of several abcesses on the ankle revealed hyphae and "yeast-like" structures in a Gram-stained smear Histological examination of subcutaneous tissues demonstrated thick-walled, occasionally branching, dematiaceous hyphae The mold isolated on potato dextrose agar was reported to be sensitive in vitro to ketoconazole (MIC, 1.6 to 6.4 mg/liter), 5-Huorocytosine (MIC, 0.8 to 6.4 mg/liter), amphotericin B (MIC, 0.1 mg/liter), and miconazole (MIC, 0.4 to 0.8 mg/liter) Two months of treatment with 5-Huorocytosine

at 150 mglkg produced some clinical response and the lesions healed after surgical debridement

a 50-year-old male diabetic suffering from zygomycete infection Initial dies of material following surgical drainage had revealed evidence of hyphae considered suggestive of candidosis Transitory amphotericin B treatment had been discontinued because of increasing renal dysfunction and been followed by 3 months of ketoconazole therapy Clinical improvement was later followed by a severe relapse, with a new lesion of the posterior wall of the sinus, that necessitated rehospitalization Reexamination of the initial material demonstrated the presence of scanty aseptate hyphae characteristic

stu-of zygomycosis A more drastic surgical intervention was made and yielded deeper tissue and bony fragments, which were cultured and examined histo-

in diameter, which presumably were those of the hyphomycete Further treatment with amphotericin B was well tolerated by the patient and re-sulted in drastic improvement

This case is of particular interest as it represents the first published in which a direct link between human and plant infection has been demon-strated The patient had never left France but was a keen carpenter, and

he had used some 5 to 6 months previously 90

Only one case of subcutaneous infection caused by S hyalinum has been published to date Zaatari et al 91 isolated this mold from multiple cysts on the feet of a 54-year-old man with Reiter's syndrome, liver cirrhosis, and coronary artery disease He was receiving azathioprine and prednisone for arthritis Four cysts, filled with necrotic debris, were excised, and hyaline septate hyphae were histologically observed They were irregular, frag-mented, and 4 to 6 JLm in width Occasional large globose cells were also observed The fungus was isolated on cycloheximide-free medium, and in vitro sensitivity tests performed The isolate was sensitive to ketoconazole (MIC, 0.125 mg/liter), 5-fluorocytosine (MIC, 3.0 mg/liter) and amphoteri-cin B (MIC, 1.0 mg/liter) and moderately sensitive to miconazole (MIC, 7.0 mg/liter)

Overall, these cases of deeper infection show several interesting features First, in contrast to those with superficial infection, these patients clearly show a number of factors, such as immunosuppressive therapy, which might predispose them to opportunistic infection Second, the in vitro sensitivity test results, particularly to ketoconazole, show quite marked differences from those reported with superficial infections, although this may, of course, reflect differences in methodology

Mycological Studies

The recent revision of the genus Hendersonula 36 and the impending revision

of the genus Scytalidium 92 will inevitably lead to confusion among clinicians and laboratory technologists Unfortunately this is unavoidable The recog-nition of key differences in conidial ontogeny and morphology between H

and Dyk036 to establish the new genus Nattrassia to accommodate this mold The genus name recognizes the work of Nattrass,40 who, although perhaps not the first to isolate this fungus, did produce the earliest complete description, while the specific epithet follows one of the earlier descriptions

of this mold as Dothiorella mangiferae in 1916.36

The situation with respect to the genus Scytalidium is confusing As pointed out by Sigler and Wang,92 this genus contains forms producing strictly hyaline arthroconidia, forms producing strictly dematiaceous arthro-

conidia, and forms with both arthroconidial and chlamydosporic morphs In this review, only those species currently witbin the genus that have been reported as human pathogens will be considered: S lignicola, the Scytali-

The recent suggestion by Sutton and Dyk036 that the Scytalidium morph of N mangiferae should be given the specific epithet dimidiatum is unnecessary in the present author's view, and their suggestion that this fun-gus is synonymous with S lignicola is considered mistaken

ana-Morphology

Scytalidium lignicola

The information presented here is largely based on the original description

by Pesante,l the study of arthroconidial hyphomycetes by Sigler and Carmichael,93 and the recent description of Scytalidium circinatum by Sig-ler and Wang 92 Only one isolate has been examined by the present author,

a subculture of that deposited at the Centraalbureau voor Schimmelcultures

by Pesante in 1957 (CBS 233.57) It is this isolate which is illustrated, grown

on Sabouraud's dextrose agar (dextrose, 40 g/liter; Oxoid mycological tone, 10 g/liter; agar, 15 g/liter) at 30°C

pep-Colonial Morphology

This is a relatively fast-growing mold; on phytone yeast extract agar, Sigler and Carmichael93 noted a diameter of 70 to 90 mm after 14 days of incuba-tion On this medium, and on Sabouraud's agar (Fig 1-6), the ex-type is white, darkening to gray and black More recent studies by Sigler and Wang,92 however, examining a wider range of cultures on potato dextrose agar and malt agar, have shown that colony color may range from black or dark gray to off-white, light gray, or yellowish gray, depending on the num-ber of dematiaceous conidia produced Tufts of aerial hyphae are produced, but they are generally shorter and scantier than those produced by many isolates of the Scytalidium anamorph of N mangiferae

Microscopical Observations

As illustrated (Fig 1-7), two distinct conidial forms are produced Small, hyaline arthroconidia are 2 X 4.5 to 8 /-Lm on phytone yeast extract agar93; these fragment readily In addition, intercalary and terminal chains of de-matiaceous conidia are formed on lateral branches These are initially hyaline but become brown and sometimes one-septate as they mature Non-septate conidia measure 4 to 7 X 7.5-12 /-Lm; septate conidia are 4 to 7 X 12-

been termed arthroconidia, these conidia do not fragment readily; when

FIG 1-6 S lignicola colony after 10 days of incubation

broken chains of brown conidia are observed, the fracture is seen to have taken place in adjoining small hyaline cells, a feature shown in Fig ge of the

and they still fit in with the current usage of this term Examination of a

arthroconidia are the most distinct and stable morph in this species, the number of chlamydoconidia being formed varying according to strain and medium

The ScytalidiumAnamorph ofN mangiferae

Campbell,94 in a study of 25 isolates from 16 patients infected by this mold, was the first to delineate the two major colonial forms seen in skin and nail cultures Fast-growing isolates completely filled a 90-mm diameter petri dish in a few days and produced a high aerial mycelium; these were desig-nated type A isolates Slow-growing isolates grew at a rate similar to that of

a total of 118 patients, confirmed these observations, although, perhaps takenly and because of a dislike of the word "type" used in this context, she termed her isolates Form 1 and Form 3 rather than type A and type B Both authors also described intermediate forms

FIG 1-7 S [ignico[a Small, regular hyaline arthroconida and larger brown doconidia are present x 170

chlamy-rates and distinctive broad bands of aerial growth Campbel194 had described annular bands oflight and dark growth in two of his type A isolates; however,

not similar to the Form 2 isolates and indeed shows characteristics more

in-termediate forms which resembled the type A isolates in having abundant aerial growth but grew markedly more slowly; unfortunately, these isolates

re-ported several banded isolates from West Africa but gave few details Colonial Morphology

Descriptions are based on Moore's38,95 examination of isolates cultured on Sabouraud's dextrose agar at 30°C Colors are matched with those illustrated

by Rayner 96 The different morphological forms were stable on repeated subculture Among 118 patients, only 4 produced more than one colonial form from the same or from different sites

completely within 3 days of central point inoculation The high, cottony, aerial mycelium is initially white but develops olivaceous gray, mouse gray, and fuscous black tones as the colony matures (Fig 1-8) In a few isolates, slight central folding is observed, and not infrequently drops of clear brown exudate form Intermediate Form 2 isolates (11 patients) grow more slowly

FIG 1-8 Cultures of the Scytalidium anamorph of N mangiferae after 10 days of

incubation Examples of type AlForm 1 (1), Form 2 (2), and type B/Form 3 (3) onies are shown

col-The aerial mycelium in these isolates is produced in broad bands col-The initial white colony develops pale olivaceous buff tones as it matures, and rings of white, smoke gray, or olivaceous gray aerial mycelium gradually darken to pale mouse gray, umber, grayish sepia, or dark mouse gray Some isolates produce rings of green and red tones on the reverse, but this is not a con-

more variable, although they are all characterized by the lack of production

of aerial mycelium, the colonies being of a velvety or, less frequently, wool texture Most isolates (46, or 84%) produce relatively compact, rather domed colonies with occasional circular or radial folds, but a few isolates (9,

wire-or 16%) are mwire-ore spreading The edges of the mwire-ore compact colonies are often lobed and/or fringed by subsurface mycelium Colors range from smoke gray to pale olivaceous gray on the surface and isabelline to olivaceous

on the reverse, but as the colonies age they become mouse gray or dark mouse gray

Microscopical Observations

The fast-growing type AlForm 1 isolates and the Form 2 isolates are almost

Form 3 isolates are easily recognizable, although they share many of the

FIG 1-9 Type AlForm 1 slide culture Hyphae of various widths and pigmentation fragment into branching arthroconidial chains X 170

same features shown by the faster-growing forms to some degree The observations presented are based on slide cultures incubated for 7 to 14 days

In all three colonial forms, variation in hyphal width is present The

1-9) and from 2 to 8 (11) p,m among the Form 2 isolates; the variation is much less marked among the slow-growing isolates, in which the majority of hyphae are 2 to 6 p,m in diameter (Fig 1-10) and a few are up to 8 p,m wide

the faster-growing isolates the broader hyphae are generally the most mented, with dark refractile septa, but all gradations from hyaline to semi-hyaline and pigmented are visible, and some relatively narrow hyphae show pigment, while some broader hyphae lack it This range in hyphal width is

pig-not seen in S lignicola

The conidia produced by the Scytalidium anamorph of N mangiferae are

almost invariably arthroconidia; chlamydoconidia have only occasionally

hyphae, 5 to 9 p,m and above, rarely fragment and remain as a distinctive network All of the narrower hyphae, however, may fragment into holarthric conidia, and enormous numbers of these are produced In the slow-growing

moderate to scanty As the arthroconidia are formed from hyphae that vary

FIG 1-10 Type B/Form 3 slide culture Narrower hyphae produce fewer nidia X 340

arthroco-widely in diameter, it is not surprising that their sizes and shapes vary and that slight differences in the dimensions reported exist 40,93,94

The present author38,95 noted cylindrical conidia measuring 2.5 to 7 X 4 to

16/Lm and globose to subglobose conidia up to 16 /Lm in diameter Initially hyaline, as they mature the arthroconidia become brown and larger, and one-septate and rare two-septate conidia may be formed One-septate arthro-conidia may be versicolored A continuous spectrum from small hyaline to large pigmented conidia is present At the branching points of the arthro-conidial chains, characteristic cuneiform cells40 are seen

Certain other morphological features appear common in this mold hyphalloops (Fig 1-11) and more compact coils (Fig 1-12) have been re-ported in all of the colonial forms.2,3o,94 Moore38,95 suggested that they were relatively scanty in the faster-growing isolates and particularly common in the slow-growing type B/Form 3 isolates, but Gugnani and Oyeka30 have recently reported that in isolates from Nigeria they found no difference in the prevalence of these structures between the fast- and slow-growing iso-lates Similarly, a variety of wall thickenings (Fig 1-13), first described by Campbell,94,97 were reported by Moore38,95 to be most prevalent in the slow-growing isolates, although they were present in small numbers in the other colonial forms Gugnani and Oyeka30 again found these structures to

Loose-be equally prevalent among all of the isolates that they examined The ence of brown slime around the hyphae of this mold30,93,95 may be related

pres-FIG 1-11 Type B/Form 3 slide culture Loose hyphal coil; X 390

to these thickenings; hyphal coils and ropes of fused hyphae are often ered with brown membranous material that may represent solidified slime Distribution of the Different Colonial Forms of the Scytalidium anamorph

cov-of N mangiferae

In his description of the different colonial forms found among human isolates

distribution of the fast- and slow-growing isolates showed no correlation with the site of infection but did appear to be linked to the country of origin of the patient Patients from the West Indies yielded predominantly the type A fast-growing form, and patients from India and Pakistan yielded the slower-

con-firmed these results Among 128 patients, although no statistical difference could be found in the distribution of the three forms in relation to the sex, age, or sites infected, a very significant association was shown with regard to the patients' racial and geographical origins As these two factors are closely linked, it was not possible to determine whether it was actually the racial type or country of origin that was most important Negroid patients from the

Form 1 isolates (Fig 1-14), while Indian-type caucasoids from East Africa

isolates were grown from patients of West Indian or West African origin Since these patients were diagnosed in the United Kingdom, sometimes

FIG 1-12 Type B/Form 3 slide culture Compact hyphal coil; X 370

FIG 1-13 Type B/Form 3 slide culture Hyphal wall thickenings range from small

rounded forms to more membranous structures encompassing the whole width of the hyphae x340

FIG 1-14 Distribution or the three colonial forms of the Scytalidium anamorph of

N mangiferae among patients of different geographical origins Key: SA, South

America; WI, West Indies; WA, West Africa; EA, East Africa; IS, Indian nent; 10, Indian Ocean; HK, Hong Kong; P, Pacific Islands; UK, United Kingdom

subconti-years after their infection began, these data cannot be regarded as sive However, reports from endemic areas are beginning to show the same trends Among the isolates cultured in Thailand,28 the slow-growing form

of use in confirming the identity of stich cultures The production of fertile pycnidial or stromatal conidiomata by isolates of human origin has been re-

group or of Form 2 No pycnidia or stromata appear to have been observed

in the isolates described as slow growing, and identification of these as

production of dematiaceous fission arthroconidia similar to those observed in the Form 1 and Form 2 isolates

A number of media have been used for the production of the coelomycete anamorph, including Sabouraud's agar, malt extract agar, potato dextrose agar, cornmeal agar, Czapek-Dox agar,38,95 soil extract agar,30 and Difco

FIG 1-15 Banana peel, 3 weeks after inoculation Pycnidial conidiomata have discharged their con-idia as white cirrhi

Trichophyton agars 14 Some authors, however, have found that media taining sterilized plant material are particularly useful,16,40,41,48,93 and the present author has found sterilized banana and lemon peels most successful 38,95 The form of con idiom a produced does show some variation according to the choice of medium; on banana peel, the conidiomata were predominantly pycnidial (Fig 1-15 to 1-17); on lemon peel, more complex stromata developed (Fig 1-18 to 1-19) Sutton and Dyk036 have also observed that on monocotyledenous hosts, pycnidial conidiomata are more commonly formed

con-The effect of light on the production of the coelomycete anamorph has been noted repeatedly Peiris et al 7 reported the production of pycnidial conidiomata by a human isolate after 5 weeks of incubation in daylight, and Rush-Munr037 also observed that pycnidia could be obtained by growth -~C> FIG 1-17 Longitudinal section of pycnidial conidioma Walls are made up of pig-mented textura angularis and are 25 to 30 f.Lm thick A few hyaline phialoconidia remain inside the locule x375

FIG 1-16 Pycnidial

con-idioma on banana peel

x 160