Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
Centenary of the genus Cryptosporidium: from morphological
to molecular species identification
Jan Šlapeta
Faculty of Veterinary Science, McMaster Building B14, University of Sydney, New
South Wales, 2006, Australia
Tel: ++61-2-935-12025; Fax: ++61-2-935-17348; e-mail: [email protected]
The biology and species diversity of the genus Cryptosporidium is mystifying for many
protozoologists and even more so for non-specialists. Historically, two morphologically
distinct parasites of the gastrointestinal tract were originally described by Ernest E.
Tyzzer from mice (i.e. Cryptosporidium muris and Cryptosporidium parvum). Later,
these two parasites were thought to parasitize all mammals including cattle. However,
such a scheme became insupportable and human-to-human transmission is now
associated with Cryptosporidium hominis. The causative parasite of the zoonotic animalto-human transmission is Cryptosporidium pestis. Here, all species names applied to the
genus Cryptosporidium are reviewed in the light of historical records and molecular
taxonomy initiatives. A DNA approach to taxonomy stands on the implicit assumption
that the reference databases used for comparison are sufficiently complete and feature
rich with annotated entries. However, the uncertain taxonomic reliability and annotations
in public DNA repositories form a major obstacle to sequence-based species
identification. Finally yet importantly, a huge gap exists between the number of described
names and number of identified genotypes. The closure of this gap represents a prime
challenge for the decades to come.
1
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
Introduction
Species of the genus Cryptosporidium have attracted immense attention over the
past three decades as an object for epidemiological and molecular biological studies.
However, while going through the published records one is quickly confronted with the
masked identity of the organism. There is a dramatic difference in the features of C.
parvum reported in 1980, 2000 and nowadays. The wide number of names and the jargon
used to describe some isolates along with the inability to continuously culture these
parasites in vitro has become a major obstacle for comparing experimental results from
different laboratories over the centenary of the existence of the genus Cryptosporidium
first described by Ernest E. Tyzzer (Figure 1).
Traditionally the taxonomy is based on morphology. However, for microbes like
Cryptosporidium species morphological differences are minute or not existing at all. This
fact has been repeatedly emphasized and additional characteristics including site of
infection and host specificity proved to be useful characteristics for species identification.
While these biological characteristics were useful, they are largely impractical for routine
identification. Thus, an accurate identification using molecular methods has been
particularly welcome. Using molecular techniques once abandoned organisms were
rediscovered and others became topical.
The following is an account of advances in the taxonomy of the genus
Cryptosporidium. The aim is to consider old reports based on morphology in the context
of more recent work on DNA sequencing which are generating barcodes of these life
forms. A consensus is presented within the rules of the International Commission
of Zoological Nomenclature (1999) along with an annotated checklist of each species.
2
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
Additionally, as part of the description available DNA accession numbers of small
subunit rDNA (SSU rDNA), Cryptosporidium oocyst wall protein 1 (COWP1), actin, and
cytosolic 70 kDa heat shock protein (HSP70) gene sequences from reference specimens
are tabulated as molecular identifiers of the respective species.
Mounting complexity: the value of host specificity
Cryptosporidum was an obscure genus and very few reports were published prior to
the 1980s. The genus Cryptosporidium Tyzzer 1907 was established for C. muris
parasitizing mouse gastric glands (Tyzzer, 1907, 1910). A related sporozoan from the
mouse intestine was named C. parvum (Tyzzer, 1912). Tyzzer correctly identified and
experimentally verified the life cycle, and based on his histological findings he correctly
speculated about autoinfection within the host.
It all seemed relatively simple and common wisdom dictated that each host would
have its own species. Providing proof for this concept was very ill-fated. Vetterling et al.
(1971) published a revision of the taxonomy and amended the description of the genus
Cryptosporidium. Ironically, in contrast to Tyzzer’s original precise work, Vetterling et
al. (1971) failed to identify oocysts as a source of infection. At the same time Vetterling
et al. (1971) describes host-specific C. wrairi from the guinea pig. Unfortunately, C.
wrairi has a peculiar nature, since it is a species never recovered from the host in the wild
but only reported on a handful of occasions in guinea-pig laboratory colonies.
Nevetheless, the host specificity within the genus Cryptosporidium is established and as
the solo character for species identification. Follow up papers start to introduce
haphazardly new names solely on different host species.
3
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
The life cycle and transmission were deemed elusive until the end of the 1970s
(Bird and Smith, 1980) despite accumulating information on the significance of the
parasite in immunocompromised hosts. Oocysts, the exogenous stages, as the source of
the infection, were re-demonstrated and the complete life cycle was experimentally
confirmed and validated on C. felis (Iseki, 1979). The genus description was amended to
include the oocyst with 4 sporozoites in accordance with Tyzzer’s (1910) early
description. Using the putative species host specificity, many new species were quickly
described and named (i.e. Barker and Carbonell, 1974). Other evidence however
suggested the existence of only a single species in the genus (Tzipori et al., 1980).
Increasing medical interest due to the recognition of acquired immune deficiency
syndrome (AIDS) caused the accumulation of publications describing numerous aspects
of Cryptosporidium from the early 1980s. Early studies supported the animal-to-human
transmission of Cryptosporidium sp., but some data already suggested a person-to-person
transmission or human infections not acquired directly from infected animals (Casemore
et al., 1985). Host specificity studies and morphological diagnostic techniques definitely
delimitated the life cycle and identified oocysts as a primary cause of infection (Current
and Reese, 1986; Fayer and Ungar, 1986). Experimental infections and ultrastructural
studies provided evidence for the thin-walled, autoinfective oocysts and recycling type I
meronts, thereby explaining overwhelming host infections after inoculums containing
small numbers of oocysts (Current and Reese, 1986).
To simplify the mounting complexity of the taxonomy, Levine (1984) reviewed
existing cross-transmission experiments and hosts of Cryptosporidium spp. and suggested
the existence of only a single species in each of the animal groups: mammals, birds,
4
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
reptiles, and fish. This opinion did not agree with the original description for the
localization of gastric C. muris and intestinal C. parvum in mice (Tyzzer, 1912). A
response to Levine’s was published by Upton and Current (1985) clearly identifying two
morphologically distinct Cryptosporidium spp. in cattle. Both C. parvum and C. muris
were deemed a prototype species for all mammalian species until further evidence
suggested otherwise (Upton and Current, 1985).
Early isoenzyme analysis and later polymerize chain reaction (PCR) based
diagnostic techniques brought new evidence for the existence of human-to-human,
animal-to-human and water-born transmission (Awad-el-Kariem et al., 1995; Morgan et
al., 1995). Introduction of DNA sequencing provided the data to identify genetic variants
affecting different or even same host species. It seemed logical to think that individual
hosts have one unique host specific Cryptosporidium spp., as well as species that are less
host specific affecting a wider spectrum of hosts (Xiao et al., 2004).
Molecular tools provided the techniques for accurate diagnosis and gene sequencing
generated data which allowed the construction of molecular phylogenies mapping the
evolutionary relationships between individual species and isolates. All this effort led to
reopening the question: Is the number of species currently recognized sufficient or do we
need to define more?
Systematics and initial nomenclatural considerations
Systematics is a fundamental discipline in biology consisting of three ideally
inseparable processes. First, the identification of a species with its characters and the
description of a specimen. Second, taxonomy names a species or taxon. And the third, the
5
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
phylogenetics that describes the relationships among species and other taxa. The major
product of systematics is a classification system of species. The system, an
anthropocentric view on the species history, is a flexible structure of arbitrary taxa.
Understandably, there is controversy about the edges of each taxon due to the existence
of several opinions on what constitutes a species, i.e. species concepts. Before one starts
to talk about species in the genus Cryptosporidium, it is important to distinguish two
elementary different subjects:
I. The species; a biological entity that is represented by a distinct population of
individuals. The definition of a species is a controversial issue and many theories
exist, each concept favouring some aspects over another (e.g., biological species
concept, phylogenetic species concept, etc.). Importantly, the species is based on a
convenience of the group of researchers and their own local judgment, experience
and consensus applying the ‘whatever species concept’ in favour. With no strict
rules preset, the biologists in each field tend to be either, those that prefer to group
several possible species into a single species, thus representing the lumpers
category; or those that tend to divide existing species into new species (called
splitters). In case of parasites clinicians often seem to represent the first category
aligning with the conservative less name rich scenario, unlike zoologists that
constantly modify and adjust the systematic scheme.
II. The name; a binominal name of a species consisting of Latin genus and species
name is an arbitrary but very useful identifier of the species. The name must be
unique, to make communication easier between researchers. In contrast to the
6
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
species definition, the rules for naming species are set very strictly in an
international code. In case of the genus Cryptosporidium, as well as for the
majority of parasitic protozoa, it is the International Code of Zoological
Nomenclature (ICZN), specifically the 4th edition (1999) that supersedes all
previous editions from January 1th 2000. It stands on the holotype, the name
bearing specimen of the species the author has used for the original description.
The holotype is deposited in a museum and is the material reference of the
species. However, in the sense of protistan terminology the term holotype (= type)
is substituted with hapantotype, i.e. one or more preparations consisting of
directly related individuals representing distinct stages of the life cycle (Art.
72.5.4, 73.3). Importantly for Cryptosporidium spp., the type (=hapantotype) is
the specimen, specimens illustrated/ photographed or described, but not the
illustration/ photograph or its description (Article 72.5.6). Thus in fact, for the
majority of protozoa there is no original museum material for subsequent
investigators to use for comparative purposes, apart from the original printed
publication. On the other hand, the fact that the specimens illustrated no longer
exist or cannot be traced does not invalidate the type designation (Art. 73.1.4).
The genus Cryptosporidium was traditionally classified within coccidia, where the
oocyst contain in the majority sufficient morphological differences to allow
identification. Such a model worked quite well for the scenario described by Upton and
Current (1985) in respect to defining the stomach versus intestinal species in cattle. The
oocyst morphology detail proved quite practical for species like Eimeria and high quality
7
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
micrographs of sporulated oocysts along with recommended data were considerate to
form a substitute for the type-specimens (Duszynski, 1999). However, taking into
consideration the 4th edition of ICZN (1999), such practice has two issues: (i) illustration
is not a syntype, i.e., name-bearing type (Article 72.5); and (ii) the type must be fixed
(Art. 72.3). Even so, Cryptosporidium oocysts possess very little morphological
characteristics and consequently Cryptosporidium identity cannot be assigned using the
oocyst morphology alone (Fall et al., 2003), thus such illustrations are of limited
diagnostic and descriptive value.
Taking into account all of the above, all existing names in the genus
Cryptospordium were re-evaluated in accordance to ICZN and categorized as:
(i)
The names are reported as nomen dubium (s.) / nomina dubia (pl.) for
those species that are species inquerenda (s.) / species inquerendae (pl.),
that are impossible to recognize and identify again but the name was based
on correct rules of the code. These species remain aside, until a detailed
review is initiated.
(ii)
The names incorrectly defined, without respecting ICZN, are assigned as
nomen nudum (s.) / nomina nuda (pl.). The name is unavailable, but upon
new description can be made available for the same or a different
concept. Species where there is evidence of an original misidentification of
a different organism, the names and species are removed from
Cryptosporidium.
(iii)
Otherwise, the name must be valid.
8
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
So much for the technicalities of taxonomy, let’s apply the above to the
Cryptosporidium species names (Table 1). In total throughout the 100 years of the
existence of the genus Cryptosporidium 37 names were introduced. Out of these, 4 names
are nomina nuda and 5 are removed form the genus as members of different organisms
(e.g. Sarcocystis crotali – syn. Cryptosporidium crotali). The remaining 28 names are
treated here as available names, nevertheless 5 of these currently need redescription (i.e.
C. anserinum, C. agni) or remain aside due to insufficient information (i.e. C. rhesi, C.
garnhami). This leaves us with 23 names to be associated with individual species. Here, I
recognize in total 21 species; thus two names are reported as synonyms (C. tyzzeri, C.
saurophilum). On the one hand the majority of these have a dominant host, but on the
other hand, they are accidentally found in possibly aberrant hosts; as is evidenced by rare
findings of diverse Cryptosporidium isolates in humans and calves faeces. The exception
of this rule is the zoonotic C. pestis that has the potential to affect temporarily the
majority of mammalian hots and C. meleagridis know to be the only species affecting
both mammals as well as birds (Table 1).
9
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
Table 1 - Update on Cryptosporidium spp. names and species status
Available
NN
Name
NAME STATUS/ SPECIES STATUS
Unavailable
Host range
NS
1
C. muris Tyzzer 1907
Available
valid - redescribed by Tyzzer (1910)
1
*
M
2
C. parvum Tyzzer 1912
Available
Valid
2
M
3
C. crotali Triffit 1925
- removed from Cryptosporidium
4
C. vulpis Wetzel 1938
- removed from Cryptosporidium
5
C. baikalika Matschoulsky 1947
- removed from Cryptosporidium
6
C. meleagridis Slavin 1955
Available
Valid
3
MB
7
C. tyzzeri Levine 1961
Available
8
C. lampropeltis Anderson, Duszynski & Marquardt 1968
- removed from Cryptosporidium
4
M
Homo
Yes
Bos
Yes
- subjectively invalid (junior subjective synonym with C. meleagridis)
9
C. ameivae Arcay de Peraza & Bastardo de San Jose 1969
Unavailable
10
C. ctenosauris Duszynski 1969
- removed from Cryptosporidium
- nomen nudum / incomplete description
11
C. wrairi Vetterling, Jervis, Merill & Sprinz 1971
Available
12
C. agni Barker & Carbonell 1974
Available
- subjectively invalid (nomen dubioum/species inqurendae)
13
C. anserinum Proctor & Kemp 1974
Available
- subjectively invalid (nomen dubioum/species inqurendae)
14
C. bovis Barker & Carbonell 1974
Available
valid - redescribed by Fayer, Santin & Xiao (2005)
5
M
15
C. cuniculus Inman & Takeuchi 1979
Available
Valid
6
M
16
C. felis Iseki 1979
Available
Valid
7
M
17
C. rhesi Levine 1980
Available
18
C. serpentis Levine 1980
Available
8
R
19
C. garnhami Bird 1981
Available
20
C. nasoris Hoover, Hoerr, Carlton, Hinsman & Ferguson 1981
Available
9
F
21
C. enteritidis Payne, Lancaster, Heinzman & McCutchan 1983
Unavailable
22
C. baileyi Current, Upton & Haynes 1986
Available
10
B
23
C. curyi Ogassawara, Benassi, Larsson & Hagiwara 1986
Available
- subjectively invalid (nomen dubioum/species inqurendae)
24
C. villithecum Paperna, Landsberg & Ostrovska 1986
Unavailable
- nomen nudum / incomplete description
25
C. varanii Pavlásek, Lávičková, Horák, Král & Král 1995
Available
Valid
11
26
C. cichlidis (Paperna & Vilenkin 1996)
Available
Valid
12
F
27
C. reichenbachklinkei (Paperna & Vilenkin 1996)
Available
Valid
13
F
28
C. saurophilum Koudela & Modrý 1998
Available
Valid
Yes
Yes
Yes
- subjectively invalid (nomen dubioum/species inqurendae)
valid - redescribed by Tilley, Upton & Freed (1990)
- subjectively invalid (nomen dubioum/species inqurendae)
Valid
- nomen nudum / incomplete description
Valid
- subjectively invalid (junior subjective synonym with C. varanii)
10
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
29
C. galli Pavlásek 1999
C. andersoni Lindsay, Upton, Owens, Morgan, Mead &
Blagburn 2000
C. canis Fayer, Trout, Xiao, Morgan, Lal & Dubey 2001
C. blagburni Morgan, Monis, Xiao, Limor, Sulaiman, Raidal,
O'Donoghue, Gasser, Murray, Fayer, Blagburn, Lal &
Thompson 2001
C. hominis Morgan-Ryan, Fall, Ward, Hijjawi, Sulaiman,
Fayer, Thompson, Olson, Lal & Xiao 2002
Available
valid - redescibed by Ryan et al. (2003)
14
B
Available
Valid
15
M
Yes
Available
Valid
16
M
Yes
Available
Valid
17
M
Yes
Yes
34
C. molnari Alvarez-Pellitero and Sitja-Bobadilla 2002
Available
Valid
18
F
35
C. suis Ryan, Monis, Enemark, Sulaiman, Samarasinghe,
Read, Buddle, Robertson, Zhou, Thompson & Xiao 2004
Available
Valid
19
M
Yes
Yes
36
C. scophthalmi Alvarez-Pellitero, Quiroga, Sitjà-Bobadilla,
Redondo, Palenzuela, Padrós, Vázquez & Nieto 2004
Available
Valid
20
F
37
C. pestis Šlapeta 2006
Available
Valid
21
M
Yes
Yes
30
31
32
33
Unavailable
Yes
- nomen nudum / incomplete description
Notes: * - Host range: M, mammal; B, bird; R, reptile; NN - name number; NS - currently recognised species number
11
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
Nucleotide sequences data: reference material - ‘Barcode Approach’
Nucleotide data played a dominant role in unrevealing the epidemiology and
Cryptosporidium population structure. While Cryptosporidium isolates remain refractory
to grow in vitro, the use of PCR and animal models have provided much-needed
characterisation of many former morphologically defined species. Several single locus
analyses provided a congruent foundation for multiple previously uncharacterized
species. The leading role was given to SSU rDNA as the most universal marker, soon
followed by additional conservative markers (actin, HSP70). Supportive evidence has
also come from COWP1. Characterization of the sequence of these four genes from a
diverse selection of isolates has provided the groundwork for follow up studies that
mapped new isolates within the existing diversity of each of these genes. These four
markers are the universal and dominant markers used to characterize individual isolates,
and the availability of existing sequences makes them a favourable entry point for any
surveys. Over the past decade there has been a steady increase in the number of
individual sequences in public databases (Figure 2). The SSU rDNA remains the
preferred marker in recent studies with over 430 available sequences to compare with,
followed by COWP1 with some 130 available sequences. The sequence numbers of actin
and HSP70 are sensibly lower, due to difficulties with amplification of these genes from
some materials.
While the above genes remain extremely useful in primary characterization of the
individual isolates especially from animal hosts, it became apparent that for fine
epidemiological studies these markers are too conservative. A 60-kDa glycoprotein
precursor (GP60) proved to be the ideal marker for the elucidation of the dominating
12
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
human derived isolates. Nowadays, besides SSU rDNA, human epidemiological studies
require characterization of GP60 with a comparison of over 260 available sequences
possible (Figure 2).
These numbers underestimate the real number of genotypes made due to the
absence of published genotyping results in the nucleotide databases. The current practice
of submitting only the unique sequences while referencing the identity to known
sequences should be avoided. Curators of the primary nucleotide databases (NCBI,
EMBL, DDBJ) recommend to submit complete datasets of all sequences regardless of the
identity with already available sequence. Such practice with thorough annotation will
foster a foundation for quantitative meta-analyses with direct reference to features
associated with these sequences, i.e. host species.
The available sequences in public databases are an extremely valuable source of
information, nevertheless they represent second-hand material and thus generally the
quality becomes questionable (Harris, 2003; Morrison, 2006). Besides the quality of the
nucleotides reported, the accuracy and the depth of annotation remains a serious issue
(Morrison, 2006; Nilsson et al., 2006). Firstly, individual submitters provide his/her
preferred species name which is then retained in the database. However, the definition of
a species does differ between researchers and over time. Within the past decade, 11 new
names were introduced (Table 1). For many of these species, DNA has been the primary
source of evidence for their distinct status. First these individuals acquired
type/isolate/genotype status within known species or only as a Cryptosporidium sp. For
example, C. canis sequence accession numbers may be found in 3 distinct organism ranks
despite representing a single valid species, i.e. SSU rDNA is AF112576, but this
13
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
accession number retains its definition as “Cryptosporidium parvum strain CPD1”,
COWP1 is AF266274 defined as “Cryptosporidium sp. 715”, HSP70 AF221529 is
defined as “Cryptosporidium parvum isolate 244” and finally actin AF382340 is correctly
defined as “Cryptosporidium canis isolate 715”. Secondly, the annotation of many
sequences does not provide sufficient information to associate the sequence with its host,
isolate number, country of origin etc. The use of common names for hosts is insufficient,
i.e. monkey or snake. It may be argued that such information appears in the paper,
however in many cases this information is absent there as well.
It is now imperative to take the advantage of the features entry currently available
during the process of sequence submission including: ‘country’, ‘specific_host’,
‘isolate’, ‘dev_stage’. Unfortunately, the popular genotype status is not a
recognized feature but generally, it is acceptable to provide this information in the feature
‘note’.
To reduce ambiguity present in the database derived from a diverse number of
submitters, I provide a table with reference sequence accession numbers (Table 2). This
table aims to list all available full length sequences, complete where possible for each
named species of the genus Cryptosporidium (Table 1). I have taken all precautions to
choose the most representative and complete sequence, if multiple sequences exist in
databases. These sequences are here viewed as barcodes of individual species,
nevertheless as more data becomes available they will be improved and extended.
Preferably, I used only single sequences, to avoid future misinterpretations resulting from
the construction of a consensus sequence. One exception is the SSU rDNA of C. galli,
which is a consensus of two overlapping partial sequences. The four genes SSU rDNA,
14
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
COWP1, aktin and HSP70 are the most commonly used markers used in genotyping
analyses and species delimitation studies. The hypervariability of the GP60 marker
prohibits its use in species identification, moreover it is only available for the study of
dominant populations which infect humans.
15
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
Table 2 - Reference sequences for named Cryptosporidium spp.
Species (alphabetically)
SSU rDNA
COWP1
actin
HSP70
C. andersoni
AF093496
AF266262
AF382352
AF221542
L19068
AF266276
AF382346
AJ310880
C. baileyi
C. bovis
AY741305
n.a.
AY741307
AY741306
C. canis
AF112576
AF266274
AF382340
AF221529
n.a.
n.a.
n.a.
n.a.
AY273775
C. cichlidis
C cf. cuniculus
AY120901
n.a.
AY120924
C. felis
AF108862
AF266263
AF382347
AF221538
C. galli
AF316624+AY168847***
n.a.
AY163901
AY168849
C. hominis
C. hominis draft genome TU502 strain&&
AF108865
chromosome 2: Chro.rrn013 (6112 - 7866)
AF266265
chromosome 6: Chro.60244
AF382337
chromosome 5: Chro.50056
AF221535
chromosome 2: Chro.20010
chromosome 2: Chro.rrn021 (334 - 2082)
chromosome 2: Chro.rrn022 (21763 - 23515)
chromosome 7: Chro.rrn016 (23535 - 25283)
chromosome 8: Chro.rrn005 (599 - 1001)
chromosome 8: Chro.rrn006 (1064 - 1516)
C. meleagridis (syn. C.tyzzeri)
AF112574
AF266266
AF382351
C. cf. molnari
AY524773
n.a.
AY524772
AF221537
n.a.
C. muris
AB089284
AF161579
AF382350
AF221543
C. nasoris
n.a.
n.a.
n.a.
n.a.
C. parvum
AF112571
AF266268
AF382343
AF221530
C. pestis
AF108864
BX538351.1:61269..66140
M86241#
U11761##
C. pestis draft genome IOWA strain&
chromosome 1: cgd1_5 (46-743)
chromosome 6: cgd6_2090
chromosome 5: cgd5_3160
chromosome 2: cgd2_20
chromosome 2*: cgd2_? (200603-202360)
chromosome 7: cgd7_5535 (1278301 - 1278425**)
chromosome 7: cgd7_7 (1582-3330)
chromosome 8: cgd8_5425 (1155551-1156729)
C. reichenbachklinkei
C. scophthalmi
C. serpentis
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
n.a.
AF151376
AF266275
AF382353
AF221541
16
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
C. suis
AF115377
AF266270
AF382344
AF221533
C. varanii (syn. C. saurophilum)
AF112573
AF266277
AF382349
AF221540
C. wrairi
AF115378
U35027+U35028
AF382348
AF221536
Notes: n.a. - sequences are not available; SSU rDNA - small subunit ribosomal DNA, >1500 bp (complete or almost complete) is
desirable for comparative purposes; COWP1 - oocyst wall protein with type I and type II cysteine-rich repeats, oocyst EB module wall
protein (Templeton et al. 2004); HSP70 - 70 kDa heat shock protein, cytosolic form; & - draft genome sequence of C. pestis IOWA
(Abrahamsen et al. 2004, AAEE01000000) has 5 SSU rDNA (some are partial); for the purpose of this table chromosome 7: cgd7_7
sequence serves as reference sequence; Le Blancq et al. (1997) identified the total of 5 copies and two slightly diferent types of rDNA
units per genome of C. pestis KSU-1: refered as Type A (AF015772) and Type B (AF308600); && - draft genome sequence of C.
hominis TU502 (Xu et al. 2004, AAEL01000000) has 6 SSU rDNAs (some partial); for the purpose of this table chromosome 7:
Chro.rrn016 and chromosome 2: Chro.rrn022 as reference sequences; * - on chromosome 2, cgd2_1375 (299599 - 301356) is
currently erroneously annotated as SSU rDNA, a different non-annotated region on the same contig AAEE01000005 is the true SSU
rDNA; ** - cgd8_5425 (1155551 - 1156706); here several additional nucleotides adre added as of the contig; *** - the sequence is
concatenated from two overlapping partial sequences; # - the sequence from chromosome 3 (Kim et al. 1992) but genome projects of
C. pestis Iowa and C. hominis TU502 both identified this gene on chromosome 5 (Abrahamsen et al. 2004, Xu et al. 2004); ## - the
sequence reported by Khramtsov et al. (1995); see also U69698 of Mead et al. (1996) - unpubl. GenBank
17
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
The 21 recognized species are relatively well represented within the four genes
(Table 2). The most problematic species are the fish species (5 species), with only the
assumed C. molnari species represented by SSU rDNA and actin. The COWP1 sequence
is missing for C. bovis, C. cuniculus and C. galli. The list and completeness of the
sequences may seem very strong, but multiple sequence alignments unmask the hidden
incompleteness of many sequences in public databases. The partial sequence
representation is further reflected in Cryptosporidium surveys where only a diagnostic
hypervariable region (400-800bp) is amplified and sequenced, i.e. use of SSU rDNA
primers targeting the hypervariable region by Johnson et al. (1995), CPB-DIAGF and
CPB-DIAGR primers or COWP1 by Spano et al. (1997), Cry-15 and Cry-9 primers
(Figure 3). Many of the sequence entries for HSP70 and COWP1 are short in sequence
length, in contrast to actin where many species are almost fully sequenced. Moreover,
many of the currently recognised genotypes are difficult to further examine due to the
generation of only partial sequences by investigators.
It should be stressed that every effort should be made to generate full-length
sequences for already existing species, genotypes or new genotypes being described.
These sequences represent a permanent record upon which an investigation for biological
characters can be based.
To avoid the confusion that arises from the rapidly expanding complexity of species
names and number of sequences generated in different laboratories over the past decade a
website was developed “Taxonomy of the genus Cryptosporidium http://www.vetsci.usyd.edu.au/staff/JanSlapeta/). The website aims to provide a common
gateway for analysing Cryptosporidium species and to standardise the terminology of
18
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
individual isolates. Alignments of the reference sequences from the named species are
provided with further detailed annotation of individual species. Furthermore currently
under development is a web based analysis (BLAST, multiple sequence alignment, tree
building) as well as an in-house annotation database of sequences to streamline and
standardise the process for identification of Cryptosporidium isolates.
Will the real Cryptosporidium parvum please stand up?
A review of the large amount of information on the human and bovine derived
isolates it became apparent that the mouse species, C. parvum Tyzzer 1912 described and
illustrated by Tyzzer (1912), is different from the commonly encountered species
affecting humans and cattle (Šlapeta, 2006).
Ernest E. Tyzzer described the parasite from the intestine of tame mice in his
Harvard laboratory (Tyzzer, 1912). The species was easily transmissible to mice and was
present in great numbers in the intestine of affected mice.
With the absence of the type material and a lack of data about the host specificity of
the original Tyzzer’s material we need to subjectively align the available data with the
current information. Two studies provided noteworthy details on Cryptosporidium spp. in
mice (Mus musculus). The first study by Klesius et al. (1986) determined up to 30%
incidence of cryptosporidiosis in mice at a calving site and susceptibility of calves to
some of these oocysts from mice faeces. In the second study, Morgan et al. (1999)
genotyped many samples of Cryptosporidium spp. from mice obtained world wide and
genetically characterized a specific “mouse genotype” using several independent nuclear
loci distinct from known genotypes. This “mouse genotype” was genetically different
19
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
from the “bovine” and “human” genotypes previously identified in Australia and
Maryland, USA and lumped under the umbrella name C. parvum (Xiao et al., 2004).
Thus, the genotyping of mice isolates provided additional evidence for the existence of
the “bovine genotype” in wild mice (Morgan et al., 1999). Consequently one can
speculate that it was this “bovine genotype” that was infectious for calves in the study of
Klesius et al. (1986). This “bovine genotype” essentially infects only neonatal mice and
only transiently adult mice (Korich et al., 2000), unlike authentic rodent C. parvum
isolates that produce heavy infections in adult laboratory mice (Bednarska et al., 2003).
No experimental calf infections with the true C. parvum “mouse genotype” were
conducted. Despite this fact, cattle and human isolates of Cryptosporidium spp. have
attracted considerable attention in recent years, but no “mouse genotype” (C. parvum)
was identified (Xiao et al., 2002; Santín et al., 2004). Therefore, C . parvum (C. parvum
“mouse genotype”) appears to be host specific, because there are either no natural calf or
human infections or they may represent an extremely rare case.
It is generally agreed that the species Tyzzer used for his description is, in deed, the
“mouse genotype” (Šlapeta, 2007; Xiao et al., 2007). An urgent step towards the stability
of historical information is to apply the Tyzzer’s name to the species he has used for the
description and use appropriate names for the two dominant human species. The above
data led to the recent taxonomic treatment and support the identity of the “mouse
genotype” with the original description, type host, experimental transmission and
illustrations of C. parvum (Šlapeta, 2006).
Cryptospordium parvum sensu Tyzzer, 1912 is a host-adapted species of Mus
musculus with no documented capacity to infect neither humans nor domestic cattle.
20
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
Consensus, reversibility and universality
Two species which can by supported by morphological characteristics and by
localisation and development in the host are C. muris and C. parvum. The larger C. muris
is found in the stomach whereas the intestine is the location of the smaller C. parvum.
The status quo of only these two species affecting mammals is retained by Upton and
Current (1985), and suggested practical until further evidence shows otherwise.
However, a diverse spectrum of hosts and distinct DNA has provided further
evidence of multiple lineages with different evolutionary history, thus provides only
preliminary evidence for a species. Older names synonymised under the umbrella names
C. parvum and C. muris, were re-erected as full species (i.e. C. felis, C. wrairi). For those
that no name was thought to be available and because its identification was based on
DNA sequence, they acquired a ‘genotype’ status or they were named as new. Mixing
later with the former brought inconsistency to what is C. parvum. In deed, the name C.
parvum has lost its purpose; it is not any more the unique identifier. Genotypes seemed to
be much more useful descriptors, i.e. human genotype, mouse genotype, canine genotype,
snake genotype, bovine genotype etc. Nevertheless, it is all work in progress.
To reinstate some stability to some of the major and clinically important genotypes,
backed up by epidemiological data and experimental information, some were logically
named as new species. Thus, the status quo of Upton and Current (1985) is abandoned
and the following consensus practically applied to cattle species. The abomasum is
parasitised by C. andersoni and at least two named species are recognized to affect the
intestine of cattle, a host specific C. bovis and zoonotic C. pestis (Šlapeta, 2006). This
21
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
approach logically aligns with the recognition of C. hominis, that together with C. pestis
were formerly classified as human and bovine genotypes of C. parvum.
Nevertheless this opinion is reversible to the former conservative opinion, thus
justifying the continuity of the names according to the ICZN, i.e. grouping C. hominis, C.
parvum, C. pestis under the oldest available name which is C. parvum thereby reinstating
the terms “human genotype”, “mouse genotype” and “bovine genotypes”, respectively.
Underling genetic differences have the potential to distinguish among multiple taxa.
While morphologically it is hard to find any differences and host specificity provides
only a partial picture, the use of DNA provides a virtual but very specific and repeatable
label for each individual. If even a minor genetic difference is detected then it may reflect
the niche adaptation, and so should be viewed as a practical tag for identification. Hence,
those populations causing significant disease will stand out from the crowd and
ultimately acquire a unique name. Indeed, C. hominis is now used for the dominant
species transmitted from human-to-human, unequivocally typified by the draft genome of
TU502 (Xu et al., 2004). Similarly, C. pestis is the name for the already know zoonotic
species for which the draft genome of the Iowa strain is available (Abrahamsen et al.,
2004). With the acceptance of both, C. pestis and C. hominis, we are clarifying the
identity of the medically and veterinary important species. The synopsis for the three
species is as follows:
Cryptosporidium parvum Tyzzer, 1912
Syn. Cryptosporidium parvum ‘mouse genotype’ Xiao et al. 2004, 2002
Type host: Laboratory mouse, Mus musculus
22
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
Type locality: Harvard laboratory, Massachusetts, USA
Cryptosporidium hominis Morgan-Ryan, Fall, Ward, Hijjawi, Sulaiman, Fayer,
Thompson, Olson, Lal & Xiao, 2002
Syn. Cryptosporidium parvum ‘human genotype’ Xiao et al. 2002
Type host: Human, Homo sapiens
Type locality: Perth, Western Australia, Australia
Cryptosporidium pestis Šlapeta, 2006
Syn. Cryptosporidium parvum ‘bovine genotype’ Xiao et al. 2004, 2002
Type host: Domestic cattle, Bos taurus
Type locality: Iowa, USA
Future challenges
Historically, two morphologically distinct populations of parasites of the
gastrointestinal tract were originally described by Ernest E. Tyzzer from mice (i.e.
Cryptosporidium muris and C. parvum). Nowadays, DNA sequences are increasingly
seen as primary information sources for species identification in many organism groups
including Cryptosporidium. Such approaches stand on the implicit assumption that the
reference databases used for comparison are sufficiently complete and feature rich with
annotated entries. However, the uncertain taxonomic reliability and lack of annotations in
public DNA repositories form a major obstacle to sequence-based species identification.
The current taxonomic expansion of the genus Cryptosporidium is important relative to
23
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
the dominant host, the pathogenicity and genetic diversity. Unfortunately only few
taxonomically available and named species have so far been molecularly characterized,
which, however, does not invalidate them from scientific usage. On the other hand data
from molecular epidemiological studies constantly reveal new genetic variants (=
genotypes), but the lack of sufficient annotation and additional biological characteristics
prohibit their elevation to species and ultimately taxonomic recognition. Nevertheless
descriptions of new genotypes remain an essential part of our understanding of the host
specificity, the diversity and the epidemiology of the genus Cryptosporidium. Besides
poor annotation of sequences in public DNA repositories, a huge gap exists between the
number of described names and number of identified genotypes. The closure of this gap
represents a prime challenge for the decades to come.
Acknowledgements
I would like to thank the II IGCC organizers for the invitation to present this paper
at the conference. Financial support is in-part provided from the Faculty of Veterinary
Science (University of Sydney) towards the preparation of this article and participating at
the II IGCC is acknowledged. I thank Prof John Ellis (UTS) for editorial suggestions. I
apologize to those authors whose work could not be cited owing to space limitations.
24
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
Legends to figures
Figure 1: Portrait of Ernest E. Tyzzer who has described Cryptosporidium muris from
mouse stomach in 1907. Courtesy of the National Library of Medicine, old negative no.
85-204.
Figure 2: The mounting numbers of individual SSU rDNA, COWP1, actin, HSP70 and
GP60 gene sequences of Cryptosporidium spp. in nucleotide databases over the past 10
years.
Figure 3: Representation SSU rDNA, COWP1, actin and HSP70 sequences available in
primary nucleotide databases (NCBI/EMBL, DDBJ). Individual sequences were
categorized according to their length. The full length of individual genes from C. pestis
and C. hominis draft genomes is given above the graph.
25
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
References
Abrahamsen, M.S., Templeton, T.J., Enomoto, S., Abrahante, J.E., Zhu, G., Lancto, C.A.,
Deng, M., Liu, C., Widmer, G., Tzipori, S., Buck, G.A., Xu, P., Bankier, A.T.,
Dear, P.H., Konfortov, B.A., Spriggs, H.F., Iyer, L., Anantharaman, V., Aravind,
L., Kapur, V. (2004) Complete genome sequence of the apicomplexan,
Cryptosporidium parvum. Science 304, 441-445.
Awad-el-Kariem, F.M., Robinson, H.A., Dyson, D.A., Evans, D., Wright, S., Fox, M.T.,
McDonald, V. (1995) Differentiation between human and animal strains of
Cryptosporidium parvum using isoenzyme typing. Parasitology 110, 129-132.
Barker, I.K., Carbonell, P.L. (1974) Cryptosporidium agni sp.n. from lambs, and
Cryptosporidium bovis sp.n. from a calf, with observations on the oocyst.
Zeitschrift fur Parasitenkunde 44, 289-298.
Bednarska, M., Bajer, A., Kulis, K., Sinski, E. (2003) Biological characterisation of
Cryptosporidium parvum isolates of wildlife rodents in Poland. Annals of
Agricultural and Environmental Medicine 10, 163-169.
Bird, R.G., Smith, M.D. (1980) Cryptosporidiosis in man: parasite life cycle and fine
structural pathology. Journal of Pathology 132, 217-233.
Casemore, D.P., Sands, R.L., Curry, A. (1985) Cryptosporidium species a "new" human
pathogen. Journal of Clinical Pathology 38, 1321-1336.
Current, W.L., Reese, N.C. (1986) A comparison of endogenous development of three
isolates of Cryptosporidium in suckling mice. Journal of Protozoology 33, 98108.
26
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
Duszynski, D.W. (1999) Revisiting the code: clarifying name-bearing types for
photomicrographs of protozoa. Journal of Parasitology 85, 769-770.
Fall, A., Thompson, R.C., Hobbs, R.P., Morgan-Ryan, U. (2003) Morphology is not a
reliable tool for delineating species within Cryptosporidium. Journal of
Parasitology 89, 399-402.
Fayer, R., Ungar, B.L. (1986) Cryptosporidium spp. and cryptosporidiosis.
Microbiological Reviews 50, 458-483.
Harris, D. (2003) Can you bank on GenBank? Trends in Ecology and Evolution 18, 317319.
International Commission on Zoological Nomenclature (1999) International Code of
Zoological Nomenclature. International Trust for Zoological Nomenclature,
London, pp. 306.
Iseki, M. (1979) Cryptosporidium felis sp. n. (Protozoa: Eimeriorina) from the domestic
cat. Japanese Journal of Parasitology 28, 285-307.
Johnson, D.W., Pieniazek, N.J., Griffin, D.W., Misener, L., Rose, J.B. (1995)
Development of a PCR protocol for sensitive detection of Cryptosporidium
oocysts in water samples. Applied and Environtal Microbiology 61, 3849-3855.
Klesius, P.H., Haynes, T.B., Malo, L.K. (1986) Infectivity of Cryptosporidium sp.
isolated from wild mice for calves and mice. Journal of the American Veterinary
Medical Association 189, 192-193.
Korich, D.G., Marshall, M.M., Smith, H.V., O'Grady, J., Bukhari, Z., Fricker, C.R.,
Rosen, J.P., Clancy, J.L. (2000) Inter-laboratory comparison of the CD-1 neonatal
27
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
mouse logistic dose-response model for Cryptosporidium parvum oocysts.
Journal of Eukaryotic Microbiology 47, 294-298.
Levine, N.D. (1984) Taxonomy and review of the coccidian genus Cryptosporidium
(Protozoa, Apicomplexa). Journal of Protozoology 31, 94-98.
Morgan, U.M., Constantine, C.C., O'Donoghue, P., Meloni, B.P., O'Brien, P.A.,
Thompson, R.C. (1995) Molecular characterization of Cryptosporidium isolates
from humans and other animals using random amplified polymorphic DNA
analysis. The American Journal of Tropical Medicine and Hygiene 52, 559-564.
Morgan, U.M., Sturdee, A.P., Singleton, G., Gomez, M.S., Gracenea, M., Torres, J.,
Hamilton, S.G., Woodside, D.P., Thompson, R.C. (1999) The Cryptosporidium
"mouse" genotype is conserved across geographic areas. Journal of Clinical
Microbiology 37, 1302-1305.
Morrison, D.A. (2006) Phylogenetic analyses of parasites in the new millennium.
Advances in Parasitology 63, 1-124.
Nilsson, R., Ryberg, M., Kristiansson, E., Abarenkov, K., Larsson, K., Koljalg, U. (2006)
Taxonomic reliability of DNA sequences in public sequence databases: a fungal
perspective. PLoS ONE 1, e59.
Santín, M., Trout, J.M., Xiao, L., Zhou, L., Greiner, E., Fayer, R. (2004) Prevalence and
age-related variation of Cryptosporidium species and genotypes in dairy calves.
Veterinary Parasitology 122, 103-117.
Šlapeta, J. (2006) Cryptosporidium species found in cattle: a proposal for a new species.
Trends in Parasitology 22, 469-474.
28
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
Šlapeta, J. (2007) Response to Xiao et al.: Further debate on the description of
Cryptosporidium pestis. Trends in Parasitology 23, 42-43.
Spano, F., Putignani, L., McLauchlin, J., Casemore, D.P., Crisanti, A. (1997) PCR-RFLP
analysis of the Cryptosporidium oocyst wall protein (COWP) gene discriminates
between C. wrairi and C. parvum, and between C. parvum isolates of human and
animal origin. FEMS Microbiology Letters 150, 209-217.
Tyzzer, E.E. (1907) A sporozoan found in the peptic glands of the common mouse.
Proceedings of The Society for Experimental Biology and Medicine 5, 12-13.
Tyzzer, E.E. (1910) An extracellular coccidium, Cryptosporidium muris (gen. et sp. nov.)
of the gastric glands of the common mouse. Journal of Medical Research 23, 487509.
Tyzzer, E.E. (1912) Cryptosporidium parvum (sp. nov.), a coccidium found in the small
intestine of the common mouse. Archiv fur Protistenkunde 26, 394-412.
Tzipori, S., Angus, K.W., Campbell, I., Gray, E.W. (1980) Cryptosporidium: evidence
for a single-species genus. Infection and Immunity 30, 884-886.
Upton, S.J., Current, W.L. (1985) The species of Cryptosporidium (Apicomplexa:
Cryptosporidiidae) infecting mammals. Journal of Parasitology 71, 625-629.
Vetterling, J.M., Jervis, H.R., Merrill, T.G., Sprinz, H. (1971) Cryptosporidium wrairi sp.
n. from the guinea pig Cavia porcellus, with an emendation of the genus. Journal
of Protozoology 18, 243-247.
Xiao, L., Sulaiman, I.M., Ryan, U.M., Zhou, L., Atwill, E.R., Tischler, M.L., Zhang, X.,
Fayer, R., Lal, A.A. (2002) Host adaptation and host-parasite co-evolution in
29
Šlapeta J. - Cryptosporidium centenary: species identification (II IGCC)
Cryptosporidium: implications for taxonomy and public health. International
Journal for Parasitology 32, 1773-1785.
Xiao, L., Fayer, R., Ryan, U., Upton, S.J. (2004) Cryptosporidium taxonomy: recent
advances and implications for public health. Clinical Microbiology Reviews 17,
72-97.
Xiao, L., Fayer, R., Ryan, U., Upton, S.J. (2007) Response to the newly proposed species
Cryptosporidium pestis. Trends in Parasitology 23, 41-42.
Xu, P., Widmer, G., Wang, Y., Ozaki, L.S., Alves, J.M., Serrano, M.G., Puiu, D.,
Manque, P., Akiyoshi, D., Mackey, A.J., Pearson, W.R., Dear, P.H., Bankier,
A.T., Peterson, D.L., Abrahamsen, M.S., Kapur, V., Tzipori, S., Buck, G.A.
(2004) The genome of Cryptosporidium hominis. Nature 431, 1107-1112.
30