Cryptococcal secretomes are controlled by several
Sec expressions, which encode factors that orchestrate the secretory pathways of extracellular proteins, especially the glycosylated proteins.
S. cerevisiae has been well-studied regarding the mechanism of exocytic secretion of proteins. Several haploid double
Δsec mutants of
S. cerevisiae on invertase secretion showed that extracellular secretion by yeast required several forms of
Sec factors, ATP, temperature, and a moderate level of glucose (<1%) [
107]. The endoplasmic reticulum-Golgi Body migration of secretory protein required energy for posttranslational modification. The final vesicular packaging and plasma membrane fusion for budding also required energy, permissive temperature, and several other
Sec factors.
Extracellularly released proteins could be inducible/constitutive depending on the signalling factors. Proteases are released to degrade proteins, lipases degrade triglycerides, phospholipases degrade phospholipids and sphingolipids, urease degrades urea to cytotoxic ammonia and CO2, laccase is involved in pigmentation, and chitin synthase is involved in cell wall chitin formation. Each of these enzymes is examined further as follows:
4.1. Proteases (Proteinases/Peptidases)
Various hydrolytic enzymes, including endoproteinases and exoproteinases, have been implicated in the cryptococcal invasion and colonisation of the CNS (neurotropism). Vu et al. discovered that extracellular secreted metalloproteinase, which is an M36 protease class of fungalysins, is highly required for fungal colonisation of the CNS and brain endothelium but not dissemination and the deletion of the
Mpr1 gene incapacitated the fungal penetration of the BBB endothelial layer to establish infection [
108]. This prolongs the survival of the infected host due to reduced brain pathology.
The nature of the media determines the type of secreted protease. Animal tissue culture induces metalloproteinase and serine endoprotease as major peptidases, but aspartyl endoprotease activity dominates in minimal media. Remarkably, the deletion of
Qsp1 reduced aspartyl endoprotease in the minimal media but elevated the secretion of metalloproteinase and serine endoprotease in animal tissue culture more than the
wt [
109]. Likewise, Brueske discovered that media supplemented with NH
4+ and glucose could inhibit the total secreted proteolytic enzymes of
C. neoformans, while growth in essential salts medium supplemented with bovine serum albumin only enhanced the extracellular protease profile [
110].
The involvement of
C. neoformans extracellular proteinases in the degradation of host immunological proteins has been demonstrated, and this is responsible for fungal invasion and dissemination. Such proteinases could degrade
β-casein, murine IgG
1, bovine IgG, human C
5, haemoglobin, and
γ-globulin in media containing carbon and nitrogen sources [
111,
112]. Regarding the potential to cause infection, a relatively low proteinase profile of
C. neoformans var.
gattii may have reduced tissue invasion, dissemination, and systemic spread of this serotype compared to the
C. neoformans var.
neoformans. However, there is no difference in the total proteolytic activities of the environmental isolates compared to the clinical isolates of
C. neoformans [
113]. Proteases are released at permissive and non-permissive temperatures, irrespective of their sources. A positive correlation exists between the capsule size and the protease activity in the clinical isolates, which was not found in the environmental/bird-dropping isolates [
114]. The majority of these proteinases are serine proteases having optimal activities between pH 7 – 8 at 37
oC [
111]
In another related study, aspartyl endopeptidase (putatively encoded by
May1) and carboxypeptidase D1 (putatively encoded by
Cxd1,
Cxd2 or
Cxd3) have been identified as the major extracellular endoprotease and carboxypeptidase, respectively, in
C. neoformans that help cope with the acidic environment and significantly enhance virulence [
115]. Specifically, the total proteolytic activity in the mutants lacking metalloendopeptidase or metalloproteinase
(encoded by
Mpr1), serine carboxypeptidase
(encoded by
Scx1), carboxypeptidase C
(encoded by
Prc1), serine endopeptidase (encoded by
Prb1), carboxypeptidase D
(encoded by
Cxd2, Cxd3), and aspartyl endopeptidase
(encoded by
Pep4) was not significantly different from the
wt in an enriched medium. However,
Δmay1 and
Δcxd1 mutants had significantly reduced activity of the extracellular proteases because they encoded major secreted protease tagged major pepsin-like aspartyl endopeptidase with optimal expression in an acidic medium.
This aspartyl endopeptidase, by mechanisms, can be inhibited by pepstatin A and antiviral protease inhibitors [
115,
116,
117]. Mutation of these extracellular protease-encoding genes might not affect melanin production. Still, since melanin synthesis is also a cell wall localised process enriched with mixed GPI-anchored proteins, chitin, chitosan, and glycan, the deletion of
Prb1 has shown hypomelanisation in such a mutant [
115].
Targeting the proteolytic activities of pathogens is a significant way of therapeutic approach against infections. A few works have shown that antiretroviral drugs in AIDS patients could antagonise some key extracellularly released hydrolytic enzymes and virulence factors in pathogenic fungi such as
C. albicans and
C. neoformans, and this invariably inhibits fungal growth. Unlike Saquinavir, Nelfinavir, Lopinavir, Darunavir, Tipranavir, Atazanavir, Amprenavir, Brecanavir, Darunavir, Indinavir, Ritonavir, and Saquinavir as well as macrocycles have been shown to inhibit fungal growth, attenuate capsule production, and inhibit urease and proteases but not phospholipase and melanin production in
C. neoformans [
115,
116,
117].
The use of specific and effective antiretroviral drugs in a co-morbidity patient with cryptococcosis may be responsible for the higher clearance of the cryptococcal cells because of the increased susceptibility of the weakened cryptococcal cells to the killing effect of innate effector cells, such as monocytes, neutrophils, and macrophages [
116]. Unlike
C. albicans, which are easily cleared by antiretroviral drugs,
C. neoformans may be relatively difficult.
C. neoformans aspartyl protease is one of the proteases not inhibited by highly active antiretroviral therapy, such as Indinavir, Lopinavir, and Ritonavir. Apparently, the transcriptional level of aspartyl protease is strategically independent of the media used [
118].
4.2. Urease
Urease (encoded by
Ure1) is another hydrolytic metalloenzyme that hydrolyses the urea to ammonia and carbamate. Apart from Ure1, other transcription factors like Ure4
, Ure6
, and Ure7 have been associated with the production of accessory proteins required for the nickel-dependent functional protein complex [
119]. Via the nickel/cobalt transporter (encoded by
Nic1), Ni
2+ is shuttled into the cytoplasm to combine with the apo-urease (encoded by
Ure1) via Ure7p to form an active metallocentric urease. The urease reaction may increase the media pH to create an alkali environment. While this may be deleterious to the infected host cells,
C. neoformans deploy the pH-sensing signalling factors, such as Rim21-Rim8-Rim20-Rim13-complex, to counter the pH surge via the activation of Rim101 signalling pathway [
29].
The
Δure1 mutants displayed comparable phenotypic and virulence features with the
wt; however, evidence existed that deleting this gene may attenuate the progression of infection caused by
C. neoformans. This is corroborated by the outcome of Cox et al. that murine infected with
Δure1 mutant, whether intranasally/intravenously, failed to show pulmonary distress, displayed delayed hydrocephalus, and wasting, and significantly lived longer than the
wt-infected mice [
120]. Hydrocephalus and wasting are delayed manifestations of meningoencephalitis caused by
Δure1 mutants after dissemination into the brain, but the infection caused by urease-positive strains proceeds rapidly in the lungs to result in acute pneumonitis capable of killing the mice before the brain encephalitis sets in [
120]. Notwithstanding, wholistic urease activation is essential for the microcapillary sequestration, dissemination, and brain invasion of
C. neoformans because
Δure1, Δure7, and
Δnic1 mutants, which lacked urease activities, significantly displayed defective growth in the presence of urea with poor blood-to-brain invasion in an animal model [
119,
121]. Though urease seems to be essential to the dissemination, invasion, and survival of
C. neoformans in vivo yet, clinically infectious urease-negative isolates capable of dissemination and invasion have been reported in AIDs patients [
122,
123], which means this enzyme may be playing second fiddle in the pathogenesis of
C. neoformans [
124].
4.3. Laccase
Laccase is a cell-wall-associated pigment-forming enzyme in
C. neoformans. This iron/copper-dependent metalloenzyme is encoded by
Lac1 (major) and
Lac2 (minor) genes [
62,
125]. Four major transcription factors, Bzp4, Usv101, Mbs1, and Hob1, are involved in the induction and regulation of the
Lac1 gene [
126]. The enzyme can act on several neurotransmitters and other amino compounds to synthesise melanin that confers camouflaging [
127,
128], resistance (against fungicides, photolysis, thermolysis, acidolysis, oxidants, denaturants) [
129,
130,
131,
132,
133],[
134,
135], and elongating propagule lifespan [
136] in
C. neoformans. Several mutants of cryptococcal cells displayed impaired melanin formation. Mutating the
Snf7 gene in cryptococcal cells impaired polysaccharide secretion, capsule, and melanin formation, especially at physiological temperature, with a complete loss of virulence in an intranasal model of murine cryptococcosis [
137].
Low glucose level facilitates the nuclear translocation of Bzp4 and Usv101; and, together with the residential nuclear transcription factor ( Mbs1), initiates the transcription of the
Lac1 gene [
126]. Conversely, Hog1 provides a repressive regulation of
Bzp4 and
Lac1 to coordinate melanin production. The
Lac1 expression was particularly shown to be upregulated in
Δssk1, Δhog1, and
Δskn7 mutants [
84]. In the early investigation,
Snf5 and
Mbf1 have been implicated in regulating laccase transcription, poor growth observed in non-glucose media, bilateral sterility, and mating defect in
C. neoformans [
138]. Also, because melanin production is anchored by the cell wall chitin and governed by the cytoplasmic ionic homeostasis, Chs3, Ccc2, Atx1, Cuf1, and Sit1 have been further identified as important transcription factors for melanisation [
138].
Two oppositely regulated signalling pathways govern melanin production – cAMP/Pka and Hog1 [
139,
140] (
Figure 4). Low glucose concentration activates cAMP/Pka-repressing pathways while repressing the Hog1 signalling pathways. This condition induces the
Lac1 gene for melanin production. Hog1 expression could hinder the Pka downstream activity leading to melanin production in serotype A,
Hog1 deletion effectively restored the melanin production in hitherto non-melanized
Δgpa1, Δcac1, or
Δpka1 mutant cultured in 0.1% glucose media (
Figure 4); however,
Hog1 mutation failed to restore melanin production in serotype D
Δpka2 mutant [
140]. From the two-component system of the Hog1 pathway, deletion of
Tco genes did not contribute significantly to melanisation, but
Δtco1 or
Δtco1Δtoc2 mutants showed enhanced melanisation at 37
oC even in up to 2% glucose media, and this melanin content was considerably higher than the
Δhog1 mutant [
141]. This shows that Tco1 kinase is a key repressor of
Lac1 expression, just like Ssk1, Pbs2, Hog1, and Skn7 proteins. However, complementing the
Δtco1 mutant (
Δtco1::Tco1) reduced the melanin to the
wt level.
The Ypd1 is also a component of the Hog1 pathway, which regulates melanin and capsule formation. At 0.1% glucose,
Δypd1Δhog1 mutant produced a highly significant level of melanin just like
Δhog1, Δssk1 and
Δskn7 mutants at 30
oC, and at 37
oC, only the
Δypd1Δhog1 and
Δskn7 could retain the melanin [
142]. Similarly, Bahn et al. reported comparable melanin formation in the
wt,
Δhog1, Δpbs2, Δssk1, and
Δskn7 mutants at 37
oC [
141]. In 1.0% glucose medium,
Δssk1, Δhog1, and
Δskn7 mutants melanised as
wt, but all mutants, including the
wt, failed to retain this melanin at 37
oC. This shows that the efficiency of melanin formation is more negatively affected by increasing glucose levels than by increasing temperature.
Surprisingly, the deletion of
Ypd1 represses melanin formation in the
Δhog1 background mutant irrespective of the temperature [
142] (
Figure 4). Contrary to this,
Δhog1, Δpbs2, Δssk1, and
Δskn7 mutants still retained pigmentation in 1.0% glucose media at 37
oC while melanin formation had already been lost in the
wt and reconstituted
Δssk1. Yet, in 2% glucose media, pigmentations are still found in the same set of mutants with significant melanisation in the
Δskn7 than the rest of the mutants [
141]. Because of the regulatory effect of Hog1p on the expression of
Mbs1 (encoding DNA-binding basic helix-loop-helix protein 1), then Song et al. showed significant melanin and capsule defect in
Δcac1 compared to
Δmbs1 mutants at 37
oC, notwithstanding
Δmbs1 mutant still showed a reduced virulence with low degree tissue fungal burden and titanisation compared to the
wt or complemented
Δmbs1 mutant (
Δmbs1::Mbs1) [
143].
While it is reasonable to summate that increasing cAMP production would elevate melanin synthesis, D'Souza et al. showed that high levels of cAMP led to overexpression of Pka1 and the repression of melanin production as compared to the
wt [
144] (
Figure 4). However, capsule production appeared highly induced [
144]. This means that capsule production may be more critical to virulence than melanin. Though the same upstream regulatory factors may control melanin and capsule productions, different transcription factors are involved in the terminal responses. Evidence exists that melanin is produced in the infected tissue isolates, but not as much is produced
in vitro. Again, the fact that larger capsules have been produced in the infected tissue isolates than culture media means that capsules are more involved in the
C. neoformans in vivo virulence and pathogenesis than melanin. Notwithstanding, an appreciable level of melanin is necessary for effective virulence, tissue invasion, antifungal resistance, and macrophage survival within the infected tissues [
145,
146,
147,
148].
Ironically, the presence of melanin in the infected tissues has been proposed with uncertainty. Liu et al. discovered that laccase-dependent catecholamine oxidative products such as pyrrole-2,3,5-tricarboxylic and pyrrole-2,3-dicarboxylic acids rather than melanin might be produced in the mouse brain during infection to induce oxidative cytotoxic effects within the infected tissue [
149]. The accrued evidence came from the work of Ito et al. that dopamine-
o-quinone is one of the reactive catecholamine oxidative intermediates, which are formed during melanin synthesis, and this intermediate strongly attacks the sulfhydryl groups of protein cysteine [
150]. This can be further oxidised to pyrrole acids in alkaline peroxide [
151,
152]. Therefore, this contrary discovery means that if melanin is actually produced during infection, then it may be stringently controlled for moderate pigmentation against oxidative damage caused by heterologous quinone-like derivatives, which are cytotoxic catecholamine oxidative intermediate products channelled to destroy the host cells.
Mutation of the
Pka1 gene in
C. neoformans usually leads to sterility and avirulent because the strain failed to produce melanin and capsules; however, overexpression of Ste12α in the
Δpka1 mutant restored the mating but was unable to restore the virulence of
Δpka1 mutant [
144]. Contrarily, the mutation of the
Pkr1 gene, encoding the regulatory Pka subunit protein, failed significantly to affect melanin production in serotype A
MATα, unlike the mutation of
Gpa1. Interestingly, it was shown that the
Δpkr1Δgpa1 double mutant produced a comparable level of melanin to the
wt better than the
Δgpa1 mutant [
144]. This indicates that melanin and capsule production is under the multifactorial transcription factors downstream of the Pka protein. Furthermore, mutation of the
Crg1 gene has been shown to enhance melanin production in the
MATα strain, which depends on the
Ste12α expression. This mutation directly increased the virulence in the animal study in a Cpk1-independent manner [
153].
Deleting the
Sit1 gene (encoding siderochrome-iron uptake transporter) in the serotype D background enhances melanisation better than the serotype A
Δsit1 mutant. The melanin production seemed better with increasing DOPA concentration in the medium—however, 1.0% glucose repressed melanin formation in the
Δsit1 mutants of the two serotypes. Additionally, the
Sit1 expression appears to control copper availability to Cu
2+-binding pocket of laccase. So, loss of
Sit1 gene apparently attenuates ionic homeostasis and increases laccase activity more significantly in the mutants than the
wt. This means laccase in the mutants has more access to copper because of the absence of Sit1p, which ordinarily could have distributed the copper among other copper-dependent proteins. In this case, Tangen et al. exogenously supplied copper and discovered an elevated laccase activity in the wt and reconstituted Δsit1 strain, comparable to the
Δsit1 mutant [
125]. Notwithstanding, an exogenous supply of copper or iron in the absence of glucose generally seemed to favour melanisation in serotype D
ΔSit1 mutant.
4.5. Chitin Synthase
Though the presence of chitin and other unique cell wall components predisposes the cryptococcal cells to the host PAMPs-PRRs (pathogen-associated molecular patterns-pattern recognition receptors) recognition to induce anti-cryptococcal defence strategies [
155] but with the capsule formation and vesicular secretion of capsular components,
C. neoformans could shield the chitin and circumvent the host’s PAMPs effect [
28,
156]. Not only this, but
Cryptococcus also has intrinsic chitin deacetylase encoded by seemingly redundant
Cda1, Cda2, and
Cda3 functional genes (the fourth,
Fdp1, remained undetermined), which can convert cell wall chitin produced by chitin synthase (encoded by
Chs3), in the presence of its regulator (encoded by
Csr2), to chitosan (a non-rigid soluble polymer) [
157,
158]. Interestingly,
C. neoformans prefers the Cda1 gene, which is highly upregulated in the recovery strain from mammalian lungs and primarily selected for fungal proliferation [
159].
Eight genes encode chitin synthase,
viz Chs1 to
Chs8, but only three chitin synthase regulators,
Csr1 – Csr3, are functionally active. Unexplainably, as many of these encoding genes are in cryptococcal cells, a compensatory expression for each mutant rarely occurs except for
Δchs3, where a slight increase in
Chs5 and
Chs7 was detected [
157]. Furthermore, the Chs3-Csr2-Cda1 complex is essential for chitosan production for the cryptococcal cells to grow, disseminate, and invade the host at physiological conditions [
160,
161]. Under vegetative growth,
Chs7 expression is entirely repressed, while
Chs6 is remarkably and minimally detected in contrast to
Chs1 and
Chs4 expressions. However,
Chs2, Chs3, Chs5, and
Chs8 are highly induced during vegetative growth [
157]. Surprisingly, none of these chitin-associated genes or regulators is essential for cryptococcal viability at 30
oC, but
Δchs3 and
Δcsr2 mutants are particularly hypersensitive to non-permissive temperatures and cell wall stressors such as Calcofluor White (CFW), Congo red, SDS, and caffeine.
4.9. Extracellular Vesicles
Extracellular vesicles (EVs) described as “virulence bags” are exosomal secretory vesicles containing capsular components (GXM, GalXM, and MP88) and enzymes like laccase
(Lac1), urease, chitin deacetylase
(Cda1, Cda2, and
Cda3), glyoxal oxidase
(Gox1, Gox2, and
Gox3), multicopper oxidase
(Cfo1 and
Cfo2), ricin-type lectin-domain containing proteins
(Ril1, Ril2, and
Ril3), Sul7/PalI family motif protein (
Tsh1, Tsh2, and
Tsh3), Pr4/Barwin domain protein family
(Blp2 and
Blp4), and phosphatase [
28,
173,
174,
175,
176]. Studies showed that the key components of EVs are virulent and antigenic, capable of activating macrophages to produce pro-inflammatory cytokines and inducing protection in animal studies [
174,
177,
178,
179]. As a standalone component, it could enhance cryptococcal cell infectivity and BBB penetration [
180]. The production, packaging, and secretion of these vesicles are controlled by
Sec genes, such as
Sec6 [
181].
The proteomic analysis of the
C. neoformans secretory vesicles showed diverse functional proteins and enzymes classified as antioxidative and heat-shock proteins (Hsp70 and Hsp90), signal transduction and nucleotide salvage enzymes, ribosomal proteins, metabolic enzymes (for sugar, lipid, protein, and amino acid), nucleoproteins, membrane proteins (transporters, carriers, pumps, channels, adhesins, inositol-3-phosphate synthase, Rab proteins), cytoskeleton proteins (actin and actin-binding proteins, tubulin, and annexin), and a few mitochondria membrane proteins [
28]. In addition, the presence of UDP-glucose dehydrogenase (encoded by
Ugd1) and UDP-glucuronate decarboxylase (encoded by
Uxs1) in the extracellular vesicles of
C. neoformans showed the presence of functional enzymes that not only promote capsule formation but also enhance cell wall integrity and thermotolerance [
182]. Cryptococcal EVs witnessed a considerably wide range of studies recently and have opened a new way of presenting the component of these vesicles as potential vaccine candidates against cryptococcal infections [
183]. Many more studies will be needed to characterise the strain specificity in the antigenic properties of these vesicles.