Investigating Cryptosporidium spp. Using Genomic and Proteomic Techniques: Current Progress and Future Directions

1. Introduction

Cryptosporidium spp. is a protozoan parasite capable of infecting various vertebrates, including humans, causing gastroenteritis with symptoms such as watery diarrhea, abdominal pain, and weakness [1]. Immunocompromised individuals are particularly vulnerable to this parasite, which has a complex life cycle involving asexual and sexual reproduction within a single host and transmission through direct and indirect contact - animal-to-human, water, and food [2]. Of the over 39 species within the genus Cryptosporidium, C. parvum and C. hominis are the most common in humans, responsible for 90% of cases worldwide. The distribution of Cryptosporidium species varies across different regions globally, with C. hominis being the main causative agent in developing and industrialized nations, and C. parvum more frequently detected in humans from Middle Eastern countries. Some species, such as C. parvum and C. meleagridis, can infect both mammals and birds, while others, including C. parvum, C. ubiquitum, and C. muris, can cause zoonotic mixed infections [3].

Cryptosporidium was first identified in mice in 1907, and it emerged as a significant causative agent of severe diarrhea in patients with AIDS in the 1980s and 1990s [4]. Cryptosporidiosis is a rising infectious disease of global public health significance, with prevalence estimates ranging from 2.6 to 21.3% in Africa and from 0.3 to 4.3% in North America [5]. Diagnosis of cryptosporidiosis is based on microscopic identification of parasite oocysts, oocyst antigens, or oocyst DNA in fecal samples. However, due to the lack of specific morphological differences between parasite species, molecular tests using genetic markers such as SSU rRNA, COWP, hsp70, and gp60 are widely accepted as the standard for species or genotype identification. It is important to note that relying solely on a few genetic markers for diagnosis may be insufficient and inaccurate, especially when identifying mixed infections with different species and genotyping. Therefore, newer molecular tools such as whole-genome sequencing are being employed to better understand the genetic structure of Cryptosporidium and develop new diagnostic methods [6].

In addition, identifying proteins involved in the host-pathogen interaction and discovering potential drug targets can be helpful in reducing the parasite’s presence in the environment. Despite extensive efforts to develop experimental models and evaluate nearly 1,000 chemotherapeutic agents, there is still no fully effective anti-Cryptosporidium therapy. [7]. The reason for the lack of drug efficacy is likely multifaceted and may include the parasite’s location within the host cell, distinct structural and biochemical composition of drug targets, or its ability to block the import or rapidly efflux drug molecules. Therefore, understanding the basic mechanisms by which drugs are transported to the parasite and identifying unique targets are critical steps in developing effective therapeutic agents. In this regard, proteomic studies can be beneficial.

This review highlights the significant milestones in Cryptosporidium research that have been made possible through the application of genomics and proteomics techniques over the years.

2. Genomics of Cryptosporidium spp.

2.1. Exploring Cryptosporidium Diversity and Evolutionary History through Whole Genome Sequencing

The study of Cryptosporidium spp. was not a prominent research area until the 1980s, and as a result, genome studies began relatively late. The first attempts at genome analysis were made in 1999 by Liu et al.[8], this is considered to be an initial step but was not entirely successful in providing a comprehensive understanding of this parasite on the genomic level. The genome survey identified over 250 kb of the C. parvum Iowa (isolate originally obtained from C. Sterling, University of Arizona, Tucson) genome by identifying genes independent of their developmental expression using a random sequence analysis. A total of 442 clones were selected and subjected to automated sequencing using one (230 clones) or two primers (212 clones) flanking the cloning site. This method generated 654 genomic survey sequences that were assembled into 408 contigs containing over 250 kb of unique sequence, representing approximately 2.5% of the C. parvum genomic sequence. The comparison of these sequences revealed that 107 contigs (26%) displayed similarity to previously identified proteins, rRNA and tRNA genes, such as putative genes involved in the glycolytic pathway, DNA, RNA, and protein metabolism, and signal transduction pathways. Additionally, diverse microsatellite sequences were identified, constituting less than 1% of the genome and having potential as genetic markers for strain typing. These studies provided first comprehensive molecular insights into the fundamental biology and cellular metabolism of this apicomplexan parasite, which has been traditionally difficult to study experimentally. Unfortunately, the authors’ analysis was constrained by the limited availability of gene sequences in the public databases at the time, which restricted the amount of data they could analyze.

The first cDNA sequence survey of C. parvum sporozoites was published one year later by Strong et al. [9]. To expedite gene discovery and identify potential drug and vaccine targets, the authors constructed cDNA and genomic DNA sequencing libraries from the Iowa isolate of C. parvum and generated approximately 2000 sequence tags by single-pass sequencing of random clones. Analysis of the resulting 567 expressed sequence tags (ESTs) and 1507 genome survey sequences (GSSs) yielded one megabase (1 mb) of unique genomic sequence, representing approximately 10% of the 10.4 mb C. parvum genome. The results of the studies were crucial and provided a basis for further analysis, particularly as they were conducted before the entire genome was known. Despite this, BLAST analyses revealed that 32% of the ESTs exhibited similarity to sequences already present in publicly available databases, indicating a significant proportion of the sequences had a known homology and function. Furthermore, some ESTs encoded proteins with signal peptides or multiple transmembrane spanning segments, indicating their potential of being located in membrane-bound compartments, the cell surface, or extracellular secretion. The authors identified several tags encoding proteins with clear therapeutic potential, including S-adenosylhomocysteine hydrolase, histone deacetylase, polyketide/fatty-acid synthases, various cyclophilins, thrombospondin-related cysteine-rich protein, and ATP-binding-cassette transporters. The identification of these molecules was groundbreaking because many of them e.g., transporters, may be involved in the rapid efflux of drugs as well as nutrient uptake.

To identify new genetic markers for future epidemiological studies and to expand molecular diagnostics of Cryptosporidium spp., whole genome sequencing projects were conducted in the following years. The first two projects involved harvesting oocyst DNA from infected germ-free neonatal calves and utilizing the random shotgun Sanger sequencing approach. Notably, both studies served as the starting point for the genomic era of Cryptosporidium research, paving the way for extensive future analyses.

In 2004, Abrahamsen et al. [10] conducted a comprehensive survey of the Cryptosporidium parvum Iowa type II full genome using a shotgun approach with Sanger sequencing. Notably, the genome of the parasite was fully assembled at the chromosome level by using a HAPPY map to create scaffold and ordering the sequence contigs. The analysis revealed a highly compact genome with a total size estimated at 9.1 Mb, which was comprised of eight chromosomes ranging from ∼0.9 to 1.4 Mb. The genome had a 13-genome coverage with five gaps. Gene prediction and annotation of open reading frames (ORFs) facilitated the identification of 3.807 protein-encoding genes in the C. parvum genome. Interestingly, only about 5% of genes possessed introns, and the parasite had smaller intergenic spaces than related microorganisms. One of the key takeaways from this study was that the parasite does not have extranuclear genomes, which was confirmed for the first time. Furthermore, the sequence analysis strongly suggested that Cryptosporidium lacks an apicoplast, which differs from other apicomplexans. Similarly, a degenerate mitochondrion that had lost its genome was identified. Moreover, the analysis revealed novel classes of cell-surface and secreted proteins with potential roles in pathogenesis, such as var and rifin molecules, SKSR, MEDLE, WYLE, FGLN, and GGC. The studies demonstrated that the parasite have a unique metabolic profile characterized by a limited number of genes associated with anabolic and catabolic processes. These rely solely on anaerobic glycolysis to generate ATP through the degradation of simple sugars and lack both a mitochondrial genome and numerous nuclear genes related to the Krebs cycle or electron transport chain. Additionally, the parasite has largely lost the ability to synthesize amino acids de novo. The authors also identified a set of four paralogous proteins belonging to the sbmA family that are involved in the transport of peptide antibiotics in bacteria and which can be potential novel routes for drug target and drug discovery. Furthermore, C. parvum lacked the group of variant surface antigens used for immune evasion by other Apicomplexa.

In 2004, Xu et al. [11] published a results of a genomic analysis of Cryptosporidium hominis isolate TU502, which revealed striking similarities in gene complement with C. parvum, exhibiting only 3-5% sequence divergence at the level of micro- and minisatellite repeat lengths and differential gene regulation. The C. hominis genome comprises eight chromosomes, with a higher GC content of 31.7% compared with 30.3% for C. parvum. Gene prediction identified 3994 genes, of which 60% exhibit similarity to known nucleotide sequences and 5-20% possess introns. Repeated DNA sequences of 2-50-base-pair (bp) were found towards the chromosome ends. Like C. parvum, C. hominis uses a glycolysis-based metabolism, but it lacks enzymes for the synthesis of key biochemical processes such as simple sugars, amino acids, and nucleotides. However, the parasite can produce more complex compounds such as amylopectin, fatty acids, and starch from precursor molecules, utilizing enzymes that have minimal similarity to known biosynthetic enzymes and are thus potential therapeutic targets. It is worth to noticed that Abrahamsen et al. and Xu et al. studies showed that C. parvum and C. hominis have reduced or absent genes from metabolic pathways, organelles, and genes that are common in eukaryotes generally, or Apicomplexa specifically. This reduction in metabolic capacity forces the parasite to rely on the host for nutrient acquisition. The study conducted by Xu et al. confirmed Abrahamsen et al’s data that both C. parvum and C. hominis have a diminished spliceosomal pathway, characterized by a reduced U6 spliceosomal complex and the absence of identifiable homologues to DICER and argonaute. This suggests that small interfering RNAs may play a distinct role in gene regulation in these microorganisms, as noted by Xu et al.

In 2003, CryptoDB was created as an online database of known Cryptosporidium genomes, following the completion of two ground breaking genomic projects by Abrahamsen et al. and Xu et al. These projects provided the foundation for the development of CryptoDB, which currently includes fifteen genomic sequences from nine different Cryptosporidium genotypes. One of the genomes deposited in CryptoDB is Cryptosporidium muris, whose genome was sequenced using a hybrid sequencing approach. Although the results of the study are unpublished, it is known that the C. muris genome is similar in nucleotide composition and gene content to C. parvum and C. hominis, and encodes around 3900 protein-coding genes. Unlike C. parvum and C. hominis, C. muris relies on anaerobic metabolism for energy production as it lacks a cytochrome-based respiratory chain and a mitochondrial genome [12].

Over time, as techniques for sequencing have become more sophisticated and accessible, new whole-genome sequencing data from additional isolates have been published. Several recent studies have utilized draft genome sequencing to improve our understanding of Cryptosporidium parasites.

For instance, Ifeonu et al. [13] conducted a study on C. hominis isolates TU502 2012 and UKH1, C. meleagridis UKMEL1, and the avian parasite C. baileyi TAMU-09Q1. The authors resequenced the TU502 2012 genome of and improved the genome assembly by reducing the number of contigs by 119, compared with previous results. Although the genome size of 9.1 Mb remained unchanged, gene length increased by 500 bp compared with the 2004 annotation. However, the current gene set in genome TU502 2012 was found to be biased in the opposite direction. Although it contains 3745 predicted proteins, only 63% of them are present in all other annotated Cryptosporidium genomes such as C. parvum IOWA II, C. meleagridis UKMEL1, C. baileyi TAMU-09Q1 and C. muris RN66. The study also revealed that 110 predicted protein-coding genes were present in the three newly sequenced genomes but not observed in the C. parvum predicted proteome. Moreover, the researchers identified a total of 10,526 SNPs in the genome TU502 2012, while 4,394 were identified in C. hominis UKH1, mapped to the reference C. hominis. These findings suggest that there is still a need for further optimization of the assemblies of Cryptosporidium genomes, as well as additional analyses to support genome annotations and to identify and characterize differences among these genomes that can explain the differing phenotypes observed in these parasites, such as host range, infectivity, and pathogenesis.

Single-cell genomics is a novel approach to genome sequencing that has been particularly useful in understanding the cellular composition of complex tissues, discovering new microbial species, and advancing diagnostics. This technique is especially useful for analyzing uncultivable unicellular organisms, such as Cryptosporidium. Troell et al. [14] conducted an important study in which they sequenced 10 single-cell genomes of Cryptosporidium-infected hosts from clinical samples, using a robust procedure for preparing single cells. Fluorescence-activated cell sorting (FACS) was initially used to sort cells by using fluorescent antibodies specific to oocyst surface antigens. The oocysts were then screened using 18SrRNA. The single genomes were sequenced using Illumina MiSeq, with coverage ranging from 67% to 95% of the reference genome (Cryptosporidium parvum Iowa II). The combined data from the 10 genomes resulted in a high coverage of 99.8% at 20x. The Cryptosporidium sequences were distributed across the 10 single-cell genomes, with an average of 1.3 Gbp per sample (range 0.3-1.9 Gbp). Interestingly, the metagenome covered 97.8% of the genome, and comparison between the metagenome and single-amplified genomes revealed a small number of SNPs ranging from 13 to 51. These results suggest that high-quality genome data can be obtained from unicellular parasites with good precision and high coverage, making this approach promising for future studies. Furthermore, the exceptional advantage of a single-cell sequencing approach is the ability to examine mixed infections and test for the presence of multiple species/subtypes.

Another interesting study was conducted in Bangladesh, where unexpected high diversity within C. hominis was confirmed using whole-genome sequencing with long-reads (Pacific Biosciences) on 63 samples from infants over a 2-year period [15]. Of the 63 samples, 32 were derived from C. hominis and had >80% genome coverage at 10x. The authors used an improved WTSI reference genome (Cryptosporidium parvum Tyzzer ATCC PRA-67D) assembly for C. parvum to map the sequences. A total of 36,780 SNPs was found to vary between the Bangladesh C. hominis isolates, but only 1,582 (4.3%) occurred with a frequency >20 in the 32 whole genomes sequenced. Seven regions of particularly high nucleotide diversity were identified, including the gp60 gene, the ortholog of the C. parvum protein COPS, and the sequences encoding the insulinase-like peptidase. The authors also found that recombination in the pathogen was proven by the decay of linkage disequilibrium in the genome over <300 bp. The discovery of highly polymorphic regions in the parasite genome implies that these regions are under strong selective pressure to diversify and potentially underlie host-parasite interactions. The polymorphic regions contain open reading frames for membrane and secreted proteins, suggesting that they play a critical role in the parasite’s adaptation to the host or immune evasion.

In 2018, Nash et al. [16] published an announcement with results from the whole genome sequencing of C. parvum UKP1, which was isolated from a patient suffering from cryptosporidiosis. The C. parvum IIc gp60 UKP1 type IIaA17G1R1 is considered a human-adapted subtype, as it has only been identified in humans and, to a lesser extent, European hedgehogs. The authors aimed to identify markers that could be used to distinguish transmission routes and potential virulence traits. The researchers purified and extracted DNA from fecal samples, and performed whole-genome sequencing using both 454 GS FLX Titanium and Illumina HiSeq 2500 sequencing technologies. The analysis resulted in 1.6 million reads, which were mapped against a reference C. parvum isolate, Iowa II. The eight chromosomes of the parasite were assembled into 14 contigs, with a total genome size of 8,881,956 bp, a G+C content of 30.20%, an N50 value of 1,092,230 bp, and a largest contig length of 1,333,759 bp. To provide a clearer perspective on the sequencing and results generated in Cryptosporidium genomics research, the following table (Table 1) summarizes the key aspects of these studies.

2.2. Unraveling Cryptosporidium’s Secrets through Comparative Genomics

The genomic sequencing of C. parvum conducted by Abrahamsen et al. revealed crucial insights into the intraspecies differences and host specificity of Cryptosporidium. In continuation and increasing this direction, Widmer et al. [17] employed a whole genome sequencing approach that utilized single nucleotide polymorphisms (SNPs) to investigate the anthroponotic C. parvum isolate TU114 (subtype IIcA5G3b) in detailed comparison to the reference C. parvum IOWA (subtype IIaA15G2R1) and C. hominis TU502 (subtype IaA25R3). The analyses identified 12,000 SNPs of C. parvum TU114 which were significantly more similar to C. hominis than to C. parvum IOWA, despite the two C. parvum genome sequences being tenfold more similar to each other than either was to C. hominis overall. This suggests that host preference might be associated with some divergent genetic loci. Highly divergent regions were found near the ends of many chromosomes, containing genes that were almost twice as large as the genome-wide average. These genes were mainly transporters, particularly ABC transporters, and proteins with signal peptides. Furthermore, the identification of three genes (cgd1_650, cgd3_3370, and cgd6_5260) exhibiting significant allelic similarity between TU114 and C. hominis suggests that their evolution may be attributed to the parasite’s adaptation to distinct host species.

Guo et al. [18] conducted comparative studies on Cryptosporidium chipmunk genotype I, a newly emerged zoonotic pathogen in some industrialized nations. The authors utilized WGS to investigate the transmission of the parasite from rodents to humans. Their aim was to develop new tools to compare the genetic similarity among Cryptosporidium chipmunk genotype I isolates from humans, wildlife, and water. To achieve this goal, the authors used the Illumina TruSeq (v3) library protocol to sequence DNA and identified the gp60 gene and nucleotide sequence encoding another mucin protein, the ortholog of cgd1_470 in C. parvum, by mapping to the reference genome of C. parvum. These molecular markers are useful for subtyping within the species due to their high sequence polymorphism. In their study, the authors characterized chipmunk genotype I in 25 human samples received from U.S. states and Sweden, one specimen each from an eastern gray squirrel, a chipmunk, and a deer mouse, and four water samples from New York. Sequencing data revealed the presence of 9,509,783 bp in 853 contigs, resulting in a 98.98% coverage of the genome after mapping. Interestingly, the remaining contigs were mostly small and from the C. hominis IbA10G2 and C. parvum subtype, likely the result of mixed infections. Based on the WGS results, genotyping succeeded in generating 15 subtypes among the 30 isolates analyzed from humans, wildlife, and storm water. At the mucin locus, two subtypes were obtained: MCI and MCII, with differences in the number of a 30-bp minisatellite repeat. The authors’ findings provide valuable information for understanding the transmission dynamics of Cryptosporidium chipmunk genotype I and may aid in the development of effective control strategies.

C. hominis subtypes IbA10G2 and IaA28R4 are known to be highly virulent and have a worldwide distribution. Guo et al. [19] conducted a study in which the genomes of two field specimens of these subtypes were sequenced using both 454 and Illumina technologies and compared. The reference genome used for the analysis was Cryptosporidium parvum IOWA. The results revealed that 8.59-9.05 Mb of Cryptosporidium sequences in 45–767 assembled contigs were generated from four DNA isolates, with a coverage of 94.36-99.47%. The genomes showed almost 97% nucleotide sequence identity with the genome of C. parvum. Major insertions and deletions (InDels) between C. hominis and C. parvum genomes were found in the telomeric regions. Comparative analysis of the four genomes of C. hominis and C. parvum revealed differences in the 5′ and 3′ ends of chromosome 6 and the gp60 region; such differences were largely the result of genetic recombination.

In 2017, Feng et al. [20] expanded our understanding of the genetic factors involved in host adaptation by C. parvum, specifically those related to proteins encoded by certain genes. The genomes of three C. parvum isolates from China and Egypt were sequenced using the Illumina Genome Analyzer IIx. The authors used the C. parvum Iowa II genome as a reference and were able to sequence over 99.3% of the genomes of the C. parvum isolates. Significant genomic differences were found between the sequenced genomes and the reference IOWA genome with the 5,191-5,766 single nucleotide variants and the majority occurring in subtelomeric regions including those encoding SKSR secretory proteins, the MEDLE family of secretory proteins, and insulinase-like proteases of chromosomes 1, 4, and 6. To infer the role of subtelomeric gene duplications in host adaptation by C. parvum subtype families, it is necessary to conduct a direct comparative genomic analysis of specimens from regions with known distinct distributions of these two subtype families among calves and lambs. Mucin proteins and other families of secretory proteins encoding invasion were found to be the most polymorphic between C. parvum isolates. These proteins play a role in sporozoite invasion and are highly immunogenic, making genes encoding them naturally polymorphic.

In a further comparative study of Cryptosporidium genomes in 2020, the authors used a comprehensive phylogenomic approach to analyze C. hominis, C. parvum, and C. meleagridis genomes that occur in humans [21]. They assembled de novo and compared raw whole genome sequencing data from 23 publicly available genomes of C. hominis, C. parvum, and C. meleagridis. The results of the analysis revealed that most genomes have a size of 9.0 Mb. Only C. parvum from humans (UKP14) and two C. hominis isolates (SWEH2 and SWEH5) have a genomic size less than 9.0 Mb. C. parvum genome identity within the species, in comparison to the C. parvum Iowa type II isolate used as a reference genome, was found to be 99.51-99.93%. The genetic similarities between C. hominis genomes were around 96.8%, whereas C. meleagridis had a lower global identity of 91.5%. C. hominis and C. meleagridis had the highest number of single nucleotide variants in coding regions, with more than 150,000 and 400,000, respectively. Single nucleotide variants detection also revealed that C. hominis and C. meleagridis had the highest number of these, with C. meleagridis accounting for 50% of all identified variants. Additionally, the highest number of deletions was found in C. meleagridis, followed by C. hominis. Although some indel genes have undergone partial annotation, the majority of them still encode uncharacterized proteins, posing a challenge for future research.

The three common intestinal Cryptosporidium species in cattle exhibit significant differences in host range, pathogenicity, and public health implications. To gain insight into the genetic determinants of biological differences among C. bovis and C. ryanae, Xu et al. [22] sequenced their genomes and mapped them to reference C. parvum IOWA. Interestingly, the authors proved that C. bovis and C. ryanae have gene organization and metabolic pathways similar to C. parvum. However, in their genomes lacks of invasion-associated mucin glycoproteins, insulinase-like proteases, MEDLE secretory proteins, and other SPDs which suggest the narrower host range of these two species. Furthermore, the loss of some other SPDs such as FLGN, SKSR and NFDQ proteins might contribute to the reduced pathogenicity of C. bovis and C. ryanae which should be further investigated. Comparative genomic analysis of these three intestinal species reveals reductions in secreted pathogenesis determinants in bovine-specific and non-pathogenic Cryptosporidium species.

2.3. Overcoming Challenges in Isolating Cryptosporidium DNA from Clinical Samples

Cryptosporidium is a challenging parasite to study due to the lack of effective in vitro cultivation methods for completing its life cycle. As a result, extracting high purity DNA from oocysts obtained from fecal specimens remains difficult. In this review, we discuss recent advances in methods for isolating and enriching Cryptosporidium genomic DNA from clinical specimens for whole-genome sequencing.

Guo et al. [23] developed a method that combined sucrose and cesium chloride density gradient separation with immunomagnetic separation (IMS) for purifying oocysts from fecal samples of humans and animals with six Cryptosporidium species or genotypes. WGS was used to verify the genome coverage and contamination. The results showed that the generated sequence data covered 94.5% to 99.7% of Cryptosporidium genomes, with minor contamination from bacterial, fungal, and host DNA. Twenty WGA products with low CT values were submitted to whole-genome sequencing, generating sequence data covering 94.5% to 99.7% of Cryptosporidium genomes with mostly minor contamination from bacterial, fungal, and host DNA. These results suggest that this strategy can effectively isolate and enrich Cryptosporidium DNA from fecal specimens for whole-genome sequencing.

Anderson et al. [24] performed a simplified version of the above method using IMS combined with PCR to sequence a positive C. hominis sample (subtype IbA9G3). The results showed a 91.25% target DNA coverage, with an improvement in parameters compared with the previously published C. hominis reference strain TU502.

Similarly, Hadfield et al. [25] developed a preparatory method based on a combination of salt flotation, IMS, and surface sterilization of oocysts prior to DNA isolation directly from human fecal samples. The IMS method was the most efficient and was applied to the sequencing of 17 fecal samples. The results showed eight new whole genome sequences, including two C. hominis and six C. parvum subtypes. The authors concluded that their method could be useful for clinical sample analysis. To provide whole perspective of the genomic studies, the following table (Table 2) summarizes the key aspects of the research during the 20 years of Cryptosporidium spp. genomics.

3. Proteome

The first publications about the Cryptosporidium genome presented the possibility of investigating the proteins expressed by this parasite; its proteome can provide comple-mentary information about the biology of this complex organism. However, despite the development of proteomic techniques, one of the significant challenges in studying the proteome of Cryptosporidium is the difficulty in cultivating the parasite, resulting in inad-equate material for analysis. Accordingly, only a few proteomic analyses have been con-ducted on Cryptosporidium, emphasizing the need for more research in this area.

Magnuson et al. [26] conducted the first such study before the genomic era of Cryptosporidium investigations; in their study they implemented Matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS) for proteomic analysis of the parasite. The important effect of these investigations was that they achieved obtainment of reproducible patterns of spectral markers with increasing sensitivity after lysing the oocysts with a freeze-thaw procedure. However, because of the lack of genomic data, their analysis focused mainly on differentiating between C. parvum and C. muris, based on the spectral peaks patterns specific to the genus. Furthermore, the authors proved that the disinfection of the oocysts resulted in the reduction in and/or elimination of the patterns of spectral markers. Using the same MALDI-TOF MS approach, Glassmeyer et al. [27] conducted a study seven years later and received spectral peaks specific to the oocyst and as well as sporozoites. In this study, the authors focused mainly on sample preparation, including mixing and standing for 45 minutes before spotting. As a result, the mass spectra of intact oocysts showed similar peaks to those of sporozoites, indicating that the laser ablates both the oocyst wall and its internal constituents.

The availability of the Cryptosporidium parvum genome sequence has allowed for large-scale global analyses of its expressed proteome in oocysts and sporozoites. Additionally, the widespread adoption of liquid chromatography coupled with mass spectrometry technology (LC-MS/MS) has facilitated more in-depth protein analysis, opening new avenues for future studies on the biology of Cryptosporidium.

In the first such study, Snelling et al. [28] aimed at the known proteome of C. parvum in the non-excysted (transmissive) and excysted (infective) forms. The authors used LC-MS/MS with a stable isotope N-terminal labeling strategy to identify and quantify proteins in soluble fractions of both non-excysted and excysted sporozoites. The authors confirmed the expression of many previously hypothesized proteins, as well as the presence of several secreted or surface proteins in sporozoites that are unique to either the Apicomplexa or the Cryptosporidium genus. The analysis identified a total of 303 C. parvum proteins, of which 26 proteins were found to be significantly upregulated during excystation. These findings suggest that future vaccine development strategies could potentially focus on selecting antigens that block the recognition and attachment of the parasite to host cells, thereby preventing infection. Of particular interest, analysis revealed the presence of seven proteins that are unique to the genus Cryptosporidium and/or the phylum Apicomplexa. Notably, five of these proteins contain a signal peptide at their N-terminus, indicating that they are secreted proteins. These proteins are likely to play a specialized role in the invasion machinery of the parasite, and further investigation of their functions could provide insights into the pathogenicity of the parasite. In addition, the parasite is likely protected against stress and apoptosis by the abundant presence of heat shock proteins (HSP) and ribosomal proteins, which act as protective mechanisms. Three Apicomplexa-specific proteins and five Cryptosporidium-specific proteins were found to be augmented in excysted invasive sporozoites. The authors also discovered eight promising targets for developing vaccines or chemotherapies that could prevent parasite entry into host cells e.g. glycosylinositol phospholipids and glycosylphosphatidylinositol protein anchors which are abundant in the surface membranes and are increasingly recognized as important modulators of the immune function during infection. Moreover, this study provides further evidence of the expression of antigenic proteins, namely Cpa135 (N64) and GP900 (N88). These proteins not only induce the production of specific antibodies but also elicit a cellular Th1 response. Furthermore, the outcome of the studies was the identification of proteins that represent around 8% of the proteome, predicted based on the whole genome sequencing by Anderson et al. and Xu et al.

Sanderson’s paper [29] presents in-depth analysis of the expressed protein repertoire of Cryptosporidium parvum. The authors used three independent proteome platforms (2-DE LCMS/MS; 1-DE LC-MS/MS; and multi-dimensional protein identification technology (MudPIT analysis)) to identify more than 4800 individual proteins representing 1237 nonredundant proteins. The peptide data were mapped to the respective locations on the C. parvum genome, and a publicly accessible interface was developed for data-mining and visualization in CryptoDB. The data provide a valuable resource for improved annotation of the genome, verification of predicted hypothetical proteins, and identification of proteins not predicted by current gene models. The expressed proteome indicates the expression of proteins important for invasion and the intracellular establishment of the parasite, including surface proteins, constituents of the remnant mitochondrion, and apical organelles. Proteomic analysis of the C. parvum genome revealed that between 25% and 40% of the potential gene products were classified as "hypothetical proteins" due to a lack of sufficient similarity to known proteins. However, the study provides proteomic evidence for 482 of these hypothetical proteins in C. parvum, which represents 39% of the total proteome detected. It is estimated that 69%-74% of the coding region of the C. parvum genome contains genes based on imperfect algorithmic prediction programs. Furthermore, the study suggests that an additional 72 regions of the genome scaffold may contain coding exons that were not previously recognized by current gene prediction models. In this study, the dataset of transporter-like proteins associated with the oocyst/sporozoite life- cycle stages has been expanded to include 64 proteins. The proportion of the total proteome that these proteins represent is estimated to be 9%, which is consistent with the prediction based on genome analysis. It is worth noting that, despite using such advanced techniques, identified proteins corresponded to 30% of the predicted proteome which highlight the current limitations in understanding Cryptosporidium biology.

Another example of a proteomic study aimed at exploring the stage-specific proteome of Cryptosporidium parvum is [30]. The authors paid particular attention to the presence of C. parvum hydrophobic and membrane proteins. Therefore, the study aimed to reduce sample complexity by protein fractionation, enabling the detection of proteins present in lower abundance in a complex protein mixture. In this study, therefore, they used SDS–PAGE to fractionate the sporozoite protein of C. parvum, followed by an LC–MS/MS approach. A total of 135 protein hits were recorded from 20 gel slices, with many hits occurring in more than one band. Excluding non-Cryptosporidium entries and proteins with multiple hits, 33 separate C. parvum entries were identified. These included structural, metabolic and hypothetical proteins covering a wide range of pHs and molecular weights. Several metabolic enzymes, including the glycolytic enzyme glyceraldehyde-3-phosphate dehydrogenase and protein disulphide isomerase, were identified in this study. These findings support the hypothesis that glycolysis is the primary energy source in C. parvum and that the parasite relies mainly on the anaerobic oxidation of glucose for energy production. The presence of these enzymes in the proteome is notable and provides further evidence of the reliance of the parasite on glycolysis.

With the development of mass-spectrometry as a protein discovery tool, studies on the Cryptosporidium proteome have become more specific and focused on characterizing specific fractions of the parasite. The first such study, using mass spectrometry, aimed at identifying the components of the oocyst wall and affinity purification to isolate glycoproteins from the sporozoites, in order to investigate how sporozoites are tethered to the oocyst wall[31]. The investigation revealed that COWPs comprise 75% of oocyst wall proteins, with COWP1 being the most abundant. COWP8 and COWP6 were also present. Affinity-purified glycoproteins included known mucin-like glycoproteins, a GalNAc-binding lectin, and novel glycoproteins, suggesting their involvement in tethering sporozoites to the oocyst wall. These results suggest that mucin-like glycoproteins may contribute to the fibrils and/or globules that tether sporozoites to the inner surface of oocyst walls.

Another such examination focused on organelles called rhoptries and their contents which are crucial in establishing productive infection during invasion [32]. In order to identify rhoptry proteins and explore the parasite’s invasion pathway and pathogenic mechanisms subcellular fractionation and mass spectrometry analysis were used. The authors detected 22 potential novel rhoptry proteins by fraction analysis, using LC–MS/MS and online software. It is worth noting that these novel candidates could serve as targets for future research on the C. parvum invasion pathway and the function of rhoptry proteins.

In the recent years, studies have mainly focused on identifying and characterizing proteins involved in the host–parasite interaction, elucidating the parasite’s virulence mechanisms, and discovering potential drug targets for the development of effective therapeutic agents. Li et al. [33] investigated the excystation process in Cryptosporidium andersoni by collecting and purifying oocysts from naturally infected adult cows. Using both immunological and molecular methods, this study identified several proteins as putative virulence factors. Their study yielded interesting data: they identified 1586 proteins in C. andersoni, and found 17 differentially expressed proteins (DEPs) after excystation, including 10 upregulated and seven downregulated proteins. This is a larger number of proteins than previously reported for C. parvum, suggesting that C. andersoni may have a more complex proteome or different proteins involved in excystation. This study identified several proteins as putative virulence factors using both immunological and molecular methods. For instance, serine protease and aminopeptidase are associated with excystation, while P23 and P30 are involved in adhesion. TRAP and thrombospondin-related anonymous protein are implicated in parasite gliding and cell penetration. Cp2, Cap135, and secreted phospholipase are associated with invasion, while HSP70 and HSP90 are involved in stress protection. These proteins play key roles in various stages of the C. parvum life cycle and are potential targets for future drug and vaccine development. These findings provide new insights into the protein composition of C. andersoni oocysts and potential excystation factors, which could be useful in identifying genes for diagnosis, vaccine development, and immunotherapy for Cryptosporidium.

In the field of Cryptosporidium research, limited peptide and protein data are available for various isolates of C. parvum and other Cryptosporidium species, such as C. hominis, due to the lack of easily accessible animal models. As a result, proteomic studies have made their way into the analysis of clinical samples.

Gathercole et al. [34] developed a refined protocol for purifying oocysts from clinical fecal samples of human patients using salt flotation and potassium bromide density centrifugation. The purified oocysts were then analyzed using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) to identify specific spectral markers unique to Cryptosporidium hominis and C. parvum. This study successfully demonstrated the potential utility of the purification method in clinical samples, providing high-quality reference spectra for further peptide and protein discovery and biomarker identification. The results also showed the applicability of MALDI-TOF MS in analyzing Cryptosporidium spp. from natural infections, broadening the range of microorganisms that can be studied using mass spectrometry techniques.

In 2021, Kacar et al. [35] conducted a study to investigate the effectiveness of proteome method analysis on the sera of Holstein calves naturally infected with Cryptosporidium spp., after being treated with high-quality colostrum and paromomycin. The study analyzed sera samples collected at the 0th and 3rd days post-treatment across three different groups of animals: the first group received only paromomycin (PC); the second group received paromomycin and colostrum (PCOL); and the third group received paromomycin, colostrum, and sodium bicarbonate (PBCOL). The study’s results indicated that colostrum treatment had the most positive effect, as significant changes in several proteins were observed. These proteins played important roles in binding/transporter, catalytic activity, regulation of molecular functions, and regulation of structural–molecular activity. Additionally, the study revealed down-regulation of many inflammatory mechanisms and processes in the colostrum-treated groups, which provides insights into the mechanism of action of colostrum in treating the infection.

Unfortunately, there is currently no fully effective drug or vaccine available to treat cryptosporidiosis, despite the reemergence of Cryptosporidium infection. A large-scale label-free proteomics approach was employed to understand the detailed interaction between the host and Cryptosporidium parvum [36]. Among the 4406 proteins identified, 121 proteins were differentially abundant (>1.5-fold cutoff, P < 0.05) in C. parvum-infected HCT-8 cells compared with uninfected cells. Specifically, 67 proteins were upregulated, and 54 proteins were downregulated 36 hours post-infection. Analysis of the differentially abundant proteins revealed that the host cells mounted an interferon-centered immune response against C. parvum infection and that the parasite extensively inhibited metabolism-related enzymes in the host cells. Several proteins were further verified using quantitative real-time reverse transcription polymerase chain reaction and Western blotting. This systematic analysis of the proteomics of C. parvum-infected HCT-8 cells identified a range of functional proteins that participate in host anti-parasite immunity or act as potential targets during infection, providing new insights into the molecular mechanism of C. parvum infection.

Karpe et al. [37] investigated the biochemical interactions underlying Cryptosporidium parvum infection in C57BL/6J mice. To do so, they collected various specimens from the mice 10 days post-infection, including fecal samples, blood, liver tissues, and luminal contents. They then analyzed the proteomes and metabolomes of these specimens using high-resolution liquid chromatography and low-resolution gas chromatography coupled with mass spectrometry. Their aim was to shed light on the biochemical mechanisms of cryptosporidiosis, which could have implications for clinical diagnosis. Univariate and multivariate statistical analysis revealed altered host and microbial energy pathways during infection, including depleted glycolysis/citrate cycle metabolites, increased short-chain fatty acids and D-amino acids, an abundance of bacteria associated with a stressed gut environment, and upregulated host proteins involved in energy pathways and Lactobacillus glyceraldehyde-3-phosphate dehydrogenase. Liver oxalate also increased during infection. They observed that microbiome-parasite relationships were more influential than the host-parasite association in mediating major biochemical changes in the mouse gut during cryptosporidiosis. Defining this parasite-microbiome interaction is the first step towards building a comprehensive cryptosporidiosis model to discover biomarkers and develop rapid and accurate diagnostics.

The advent of genomic and proteomic techniques has enabled a deeper analysis of the genomes and proteomes of the major Cryptosporidium species, leading to improved insights into their biology and pathogenicity. It is worth noting that these surveys address a wide range of questions, spanning basic biology, drug target identification, taxonomy, and population biology. Table 3 and Table 4 offers a detailed overview of Cryptosporidium proteomics studies, encapsulating critical information about the specimens, applied techniques, search engines utilized, and reference genomes, providing a comprehensive context to the review.However, further work is needed to fully explore the genomes and proteomes of Cryptosporidium, and challenges remain in studying this pathogen due to its intracellular nature, genetic diversity, and difficulties in sample collection and preparation. Nonetheless, ongoing research efforts and advances in technology are likely to lead to new strategies for controlling and preventing Cryptosporidium infections.