Human Pathogen Associated Molecular Pattern Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Chitin is an abundant polysaccharide in nature found as a component of fungal cell walls and the exoskeletons of arthropods. Chitin may possibly act as a pathogen-associated molecular pattern (PAMP), enabling pathogen recognition by the innate immune system (similarly to LPS recognition). However it is totally insoluble in aqueous media and detection by other organisms has been shown to be a complex process, involving the release of exogenous chitinases followed by release of soluble chitin fragments and subsequent receptor-mediated recognition. The hypothesis underlying this project is that a similar process occurs in humans - the constitutive manufacture of chitinases allows the breakdown of chitin to form various chitooligosaccharides. These may subsequently be recognised by the innate immune system, and induce receptor-mediated cell signalling and an immune response. Recent studies have found that certain sized fragments could indeed act as a PAMP. However, these fragments are still insoluble. Chitotriosidase (chit1), the chitinase expressed in human peripheral blood macrophages, has been shown in previous studies to prevent fungal growth, a role analogous to the action of chitinases within the plant kingdom. We have therefore asked whether whole chitin or the soluble fragment, chitohexaose, up-regulates chitotriosidase along with the putative peptidoglycan-binding domains that may function as part of the receptor responsible for chitin recognition. We have also analysed whether there is an induction of cytokine production as a function of the immune response. Human peripheral blood monocytes were isolated, cultivated in vitro into monocyte-derived macrophages (MDMs), and stimulated with chitohexaose fragments or whole chitin (α or β). A reverse transcription-polymerase chain reaction (RT-PCR) method was used to assay gene expression and analysis was carried out using gel electrophoresis. The levels of mRNA were measured quantitatively by real time-PCR (RT-qPCR) using the SYBR-Green I fluorescence method. Neither whole chitin nor chitohexaose were shown to up-regulate Chit1 or any tested cytokines, suggesting that this size of fragment does not act as PAMP within humans. The lack of cytokine induction by the fragment is inconsistent with current views on chitin-induced macrophage activation (however these other studies have not stated which oligosaccharide was used in experimentation so it is difficult to make comparisons).


1.1 Chitin and Chitinases

Chitin is a polysaccharide of N-acetylglucosamine that is found as a major structural component in fungal cell walls, and in the exoskeleton of arthropods such as insects and crustaceans. It is an insoluble and chemically stable polymer with a resilient structure that confers rigidity to the organism in which it is found. It is resistant to chemical and enzymatic degradation due to its crystalline structure. There are three variants in the microcrystalline structure of chitin: alpha, where the sheets are anti-parallel, beta where they are parallel, and gamma in which there is a mixture of both. The alpha form is found as part of calyces within molluscs, plankton and also as part of arthropods. The beta form, the more degradable and less stable form, is found in squid pen, cuttlefish bone and fungal cell walls.

The chitinases within chitin-containing organisms are believed to have a role within growth and development.1 Although humans do not have the capability of synthesising chitin, two chitinase analogues are produced that contain DXXDXDXE as a conserved sequence, and have the last glutamate and aspartate as the catalytic residues (ref). They have been suggested to degrade pathogens that contain chitin as a defence against invasion; this is due to homologous chitinases in plant, fungi and eukaryotes also fulfilling this role.2 This may be deemed to be an anti-fungal role.

The chitinases are organized into two families named 18 and 19, determined on the sequencing of the amino acids in the catalytic domain.3 The two families each have a different tertiary structure and mechanism of catalytic activity.4,5 Family 19 chitinases are those found within plants and Streptomyces griseus.6 Those from family 18 generally consist of a catalytic domain, and linker region rich in serine and threonine, and a chitin-binding domain.7 The first mammalian chitinase to be found, chitotriosidase, was discovered in patients diagnosed with Gaucher disease, where expression in the plasma is substantially higher than in healthy individuals (1000-fold). Its name derives from the ability of the enzyme to hydrolyse the synthetic substrate 4-methylumbelliferyl-β-chitotrioside.8

Chitotriosidase belongs to the family 18 of glycosylhydrolases, and the gene (chit1) is located on chromosome 1 in the q32 region. The second chitinolytic enzyme, acidic mammalian chitinase (AMCase) belongs to the same family, and is encoded by the gene Chia, also located on chromosome 1, though in the p13.1-21.3 region. Other proteins in this family exist in man that are structurally homologous to these chitinases - the hydrophobic binding cleft is preserved they but lack chitinolytic activity, due to the mutation of one or both active site residues.9 These are known as the chitinase-like lectins, and include chitinase-3-like-1, a protein with unknown function but which has been associated with inflammatory disorders such as asthma.10

Chitotriosidase expression can be increased in human macrophages (which constitutively secrete it and store it within lysosomes) by granulocyte macrophage colony-stimulating factor (GM-CSF). This cytokine stimulates the production of further enzyme, a finding consistent with an analysis of gene expression which showed that GM-CSF matured macrophages had a greater expression of the chit1 gene compared to those matured with macrophage colony stimulating factor (M-CSF).11 Because of this, it has been reported that the administration of GM-CSF has favourable effects in patients with fungal conditions.12, 13

Chitotriosidase is found as a 50kDa form and a 39kDa form within the human body. The larger form contains a binding domain, a hinge region and the 39kDa region in which the chitinolytic activity resides. This is the form that is seen within the circulation - it is within the lysosomes of macrophages that the proteolytic processing occurs yielding the smaller form of the enzyme.14 Certain ethnic groups have a mutation of the chitotriosidase gene causing a 24-base pair duplication within exon 10 which prevents the proteolytic conversion. This leads to a deficiency in the [functional] enzyme in homozygous individuals. In the majority of ethnicities, around 35% of the people are carriers, and 5% have the homozygous phenotype.15 However, some populations, for example the sub-Saharan Africans, do not have any individuals with the mutation, or any carriers. This may possibly suggest the importance of chitotriosidase in areas that have high incidences of parasitic infections.16 This is further supported by the discovery that infections by Wuchereria bancrofti, a parasitic nematode, tend to occur in those with chitotriosidase deficiency.17 However, despite the high incidence of deficiency, there are very little consequences in terms of ability to acquire disease. This may suggest that this chitinase has a redundant role in humans, but the function of homologous enzymes within plants as an immunological defence against fungal pathogens has been repeatedly confirmed.

It is possible that the second human chitinolytic enzyme, AMCase, (or other related enzymes with similar substrate specificities, such as lysozyme) may compensate in chitotriosidase defiency.18

1.2 Chitin sensing and degradation in the marine environment

Chitin is a major nutrient source of nitrogen, carbon and energy, and some bacteria have evolved to be able to utilize this, particularly in the marine environment. Approximately 1011 tons are produced annually in marine waters, mainly by planktonic crustaceans, and its consumption by various organisms is essential for the maintenance of the carbon and nitrogen cycles. Early studies have shown that very little chitin is present within the ocean floor despite the continuous settling of chitin-containing 'marine snow' derived from old exoskeletons and cuticles of chitinous organisms. 20 This is due to the presence of marine bacteria that rely on chitin as their sole source of nutrients which degrade the snow before it is able to settle to the bottom.21

The way in which specialised chitinolytic bacteria recognise and degrade chitin is complex. As chitin is such a large, crystalline molecule, it is not able to generate a concentration gradient that can be sensed by bacteria for chemotaxis. Instead, it has been shown by Li and Roseman (2004) that starved Vibrio furnissii release large quantities of chitinases that interact with sources of chitin to produce a polysaccharide gradient [most commonly the disaccharide of N-acetylglucosamine (G1cNAc)2].22 This allows chemotaxis towards the source, and further induces the expression of the chitinase gene along with other genes associated with the catabolic degradation of chitin.

Following attachment via a chitin-specific lectin, the chitin is degraded. The bacteria release large amounts of extracellular chitinases which cleave the chitin to produce various oligosaccharides. The endochitinases then breakdown the fragments further, ultimately producing the disaccharide, chitobiose, for establishment of the chemical gradient. One particular species, Serratia marcescens, exports a protein that is capable of increasing the rate of chitin degradation. CBP21 acts to allow the chitinases to have a greater ease of access to the polymer by invading crystal structure and removing some of the monomers.23 The final degradation products of chitin are fructose-6-P, acetate, and ammonia.

Chitin in pathogens and recognition by the innate immune system

All multi-cellular organisms require the ability to detect the invasion of pathogens. This is achieved by the presence of pathogen-associated molecular patterns (PAMPs) on the pathogen. These are small, diverse molecular motifs that alert host cells to the presence of intruders that are generally recognised by toll-like receptors (TLRs) or other pattern-recognition receptors (PRRs) on immune cells within plant and animals.26

As chitin is a widespread component of fungal cell walls, it could play a role in the recognition of pathogens (similarly to the well established role of LPS recognition in bacterial recognition). Its presence within macrophages and neutrophils, and also its secretion, suggests it has a role in innate immunity. This is supported by inherent chitinase activity in guinea pig blood which increases upon systemic infection with Aspergillus fumigatus .24 A recombinant form of human chitotriosidase has been shown in in vitro studies to prevent fungal growth and hydrolyze chitin analogues, and prolonged survival when administered to mice with a systemic candidiasis or aspergillosis infection.18, 25

Various chitinolytic enzymes or chitin-binding proteins are expressed constitutively in macrophages or the epithelia of the lungs and digestive tract, where they are believed to have a role in the first-line of defence against chitinous organisms that enter the body.27, 28 Additionally, many studies have found that some are induced as part of an inflammatory or allergic response. The chitin-binding protein human cartilage glycoprotein-39 (HC-gp39) is induced significantly in a range of inflammatory disorders including osteoarthritis and rheumatoid arthritis, and levels are seen to decrease upon treatment.29 Patients with Gaucher disease, whereby an inherited deficiency of a lysosomal hydrolase produces lipid-laden macrophages, are shown to have significant chitotriosidase induction, as are patients with atherosclerosis.30, 31

A number of studies have been carried out that detail the effects that chitin has on cytokine production by macrophages as part of the innate immune response. Shibata et al (1997) conducted a study whereby chitin was fractionated into phagocytosable particles (1-10µm) and administered to the lung. This was shown to activate alveolar macrophages to produce various type 1 pro-inflammatory cytokines, namely IL-12, IL-18 and TNFα, leading to the subsequent production of IFNγ by natural killer cells.32 IFNγ is a cytokine known to classically stimulate macrophages. This process was later shown to occur by mannose-receptor mediated phagocytosis - mannose receptors located on the plasma membrane cause the internalization of the particles which then are degraded by lysozyme and N-acetyl-β-glucosaminidase within the cell.33

A study conducted by Reese et al (2007) found that chitin, when administered to the lungs of mice, is able to induce accumulation of immune cells that express IL-4, such as eosinophils and basophils. It is not mentioned how the insoluble chitin was recognised by the immune system. These studies were conducted in toll-like receptor 4 deficient cells and so were unresponsive to LPS if it was present. Chitin was also shown to produce alternatively activated macrophages and subsequent leukotriene-B4 production. This macrophage response was essential for accumulation of eosinophils, as depletion of macrophages using clondronate liposomes prevented this occurrence. The allergic-type inflammation that resulted from chitin administration was abolished upon the treatment with the IL-4- and IL-13-inducible mammalian chitinase, AMCase, or when given to over-expressing transgenic mice.34 This would suggest that the chitinases have a protective function against chitin-containing pathogens. However, other studies that have been conducted have found a significant induction of chitinases and binding proteins in other models of an allergic response caused by non-chitinous allergens such as ovalbumin.19,35 Induction is also seen within the lungs of IL-13 over-expressing mice, and chitinase inhibition using allosamidine or specific antibodies considerably reduces ovalbumin or IL-13-induced inflammatory responses.19 This shows that the chitinases or binding proteins have a possible role in immune response, regardless of the presence or absence of chitin, and so does not prove that chitin is capable of generating an immune response itself.

The studies produced by Shibata et al and Reese et al are not consistent in terms of macrophage activation - one details the classical activation of macrophages, the other, alternative activation. A further study by Da Silva et al (2008) has shown an additional adaptive effector pathway involving IL-17A. It suggests that chitin and chitin fragments (40-70µm) act as PAMPs, and have a direct action on macrophages to cause the expression of IL-17, IL-23 and TNFα, mediated by TLR-2- and MyD88-dependent mechanisms.36

A second study by Da Silva et al (2009) looked further into the importance of the size of the chitin fragments by characterizing their ability to stimulate IL-10 and TNFα production in the murine lung. It was found that 'intermediate chitin' (IC) defined as being in the range of 40-70µm and 'small chitin' (SC) defined as <40µm (mostly 2-10µm) were both potent stimulators of TNF production in lung macrophages. IL-10 was induced strongly by SC only. The size of the fragment was also shown to determine what mechanism is used during chitin stimulation. The study demonstrated that IC caused TNF production via a TLR-2, dectin-1 (a C-type lectin) and nuclear factor-κβ pathway that is independent of phagocytosis. SC was shown to stimulate TNF production via TLR-2 dependent and independent pathways which may involve the mannose receptor and phagocytosis. IL-10 production upon SC stimulation was mediated via TLR-2 and dectin-1 pathways, and is prevented by phagocytosis and the mannose receptor. 'Big chitin', that is fragments in the range of 70-100µm, and 'super small chitin' <2µm in size did not have an effect on macrophage production of IL-10 or TNF. The conclusions that the authors derive from this study are firstly, that chitin acts a PAMP to activate TLRs on immune cells, and secondly, that this interaction is size dependent whereby the size of the fragment determines not only if it acts as a PAMP, but also what mechanism is used to elicit a response.37

Is there a receptor for chitin?

Little is known about chitin signalling, especially within humans. Studies conducted in other organisms have identified conserved sequences that have a critical role in chitin recognition. The LysM motifs were originally identified as a structural feature of enzymes involved in prokaryotic cell wall degradation, and also as part of the chitinolytic enzymes within the yeast Kluyveromyces lactis.38 The chitinases within certain eukaryotes, for example the nematode Caenorhabditis elegans, contain LysM domains suggesting that they, like bacteria, bind to chitin or other G1cNAc oligosaccharides.39

These domains contain a series of four conserved cysteine residues which form disulphide bridges with one another, a sequence not found in bacterial analogues and so may account for binding specificity.

Chitin-like Nod factors released by the bacterial species Rhizobium bind to the LysM domains within leguminous plant receptors allowing uptake for the symbiotic relationship whereby the bacteria fix nitrogen as a nutrient source for the plant and in return obtain proteins and carbohydrates for its own sustenance.40 The Nod factor receptor proteins, nuclear factors 1 and 5, within the plant are LysM domain-containing receptor-like kinases and are seen to be the most likely candidate for chitin signalling and fungal pathogen recognition.40

Figure 1. LysM domain from E. coli lytic murein transglycosylase D (MltD). The ribbon cartoon depicts the characteristic ¢¡¡¢ secondary structure of this domain. The two helices are stacked onto one side of a plate formed by the two-stranded anti-parallel beta-sheet (ref)

As well as having involvement in the symbiotic relationships between leguminous plants and rhizobial bacteria, chitin has the ability to elicit defence responses in a variety of plants. Chitin Elicitor Binding Protein (CEBiP) is a 75kDa LysM-domain containing plasma-membrane protein with an extracellular domain found in rice plants (Oryza sativa). It lacks a cytosolic domain so is considered to be part of a greater chitin receptor complex. The interaction of CEBiP with chitooctaose (G1cNAc8) causes the up-regulation of proteins involved in immune responses, most notably the chitinases.41

Chitin binding proteins structurally similar to CEBiP in rice are found in other plants as well, indicating that this chitin-recognition system of defence is conserved among many plant species. As CEBiP has no intracellular domain, it requires a second protein to produce a cellular signal upon interaction with chitin. This protein is most likely to be a kinase. It has been found within Arabidopsis species that the presence of a LysM domain-containing receptor-like kinase 1(LysM RLK1)/

chitin-elicitor receptor kinase 1 (CERK1) is vital for this signal transduction.42, 43 It was observed that contact with chitooctaose affects the expression of around 890 'chitin-responsive' genes - insertional mutations within the CERK gene prevents these changes, and shown to increase the plant's susceptibility to fungal pathogens. Arabidopsis has five LysM RLK-encoding genes (RLK 1-5) within its genome, similar to the NFR1 and 5 Nod receptor proteins found in legumes. Mutations in all but LysM RLK 1 have been shown to have no effect on gene induction, which implies a specific role of this particular kinase in chitin recognition.43

Within humans, a number of genes that encode LysM domains have been identified in the human genome.44 A search of PfAM for proteins containing this motif reveals 16 genes in Homo sapiens (see Annex 1). Among these, LysM and putative peptidoglycan-binding domain-containing proteins 1-4 (LysMD1-4) are four genes that may have a role in pathogen recognition. Inputting of their amino acid sequences into a predictor of membrane spanning regions (TMHMM 2.0) reveals that LysMD3 and D4 contain a transmembrane domain, which suggests that upon interaction with a ligand, these LysM motifs could initiate intracellular signalling.45

In accordance with this suggested role, The UniProt Knowledge Base suggests that LysMD3 is a single pass transmembrane protein with the LysM domain facing the extracellular side and a shorter cytoplasmic tail (Fig. 2)

Fig. 2 Prediction of topology for LysMD3 from the UniProtKB interface (ref)

Chitin, because of its crystalline structure, is totally insoluble in aqueous media, and so this begs the question as to how chitin is recognised by the immune system. Many previous studies of chitinase activity have not used crystalline chitin, and instead have used, for example, chitosans. These are a family of chitin derivatives with a much greater degree of water solubility, achieved by removing parts of the N-acetyl groups leading to a reduction in size of the amino groups within the polymer.46

Although in some contexts, such as the quantitative evaluation of chitinase activity, the use of chitin derivatives are appropriate, their use is not a true reflection of chitin-chitinase binding as it would occur in nature.47 Modifications that occur in nature can also reduce the ability for chitinases to interact with chitin. After production in various organisms, chitin enters the extracellular space where it can associate with components of cell walls or the exoskeleton. Covalent interactions occur between proteins and chitin - these proteins then sclerotize, that is they harden by the formation of adducts with the products from oxidation of diphenolic substrates.48 This could limit access to chitinases and may interfere with the ability of the immune system to recognise chitin. Additionally, chitin is also known to associate with β-glucans and galactomannans within fungi.48

Our hypothesis is that constitutive production of an exogenous chitinase by macrophages could lead to the release of chitin breakdown products which would be soluble and could be subsequently sensed by the immune cells. This could then induce receptor-mediated cell signalling to initiate the processes involved in an immune response.

2.5 Aims and Objectives

The Aim of this project is to determine whether crystalline chitin or the soluble oligomer chitohexaose are able to activate human macrophages. This will be addressed by assessing whether:

chitin or chitin fragments induce further chitotriosidase expression,

cytokine production is initiated upon macrophage stimulation.

chitin can modulate expression of LysM domains within macrophages

2 Materials and Methods

2.1 Isolation of human peripheral blood monocytes and cultivation into MDMs

Human peripheral heparinised blood was diluted using DPBS (BioWhittaker) (with Ca2+ and Mg2+) to a 1:1 ratio, and layered over 15 ml of Ficoll (Sigma-Aldrich) in a 50 ml sterile tube. Erythrocytes were separated by centrifugation at 600 - g for 20 minutes. Leukocytes, along with platelets, were subsequently removed from the separated blood suspension and resuspended with DPBS without Ca2+ and Mg2+ (Sigma) (). The platelets were removed by centrifugation at 120 - g for 10 minutes, and discarding of the resulting supernatant. The leukocytes were washed in DPBS (Sigma) (without Ca2+ and Mg2+) and centrifuged for 6 minutes at 450 - g. The supernatant was discarded, and the cell then resuspended in 0.5ml RoboSep buffer (StemCell Technologies, Inc) - phosphate buffered saline (PBS) with 2% fetal bovine serum (FBS) and 1mM EDTA (without Ca2+ and Mg2+). EasySep Human Monocyte Enrichment Cocktail without CD16 Depletion was added at a concentration of 50 µL/mL of cells and allowed to incubate on ice for 10 minutes. EasySep D Magnetic Particles were vortexed for 30 seconds to dislodge any aggregates, and added at a concentration of 50 µL/mL of cells and allowed to incubate on ice for 5 minutes. The cell suspension was made up to a volume of 2.5mL using RoboSep buffer, and inserted into an EasySep magnet which was set aside for 2.5 minutes for separation. The monocytes were then removed by pouring off the desired fraction leaving the magnetically labelled unwanted cells bound to the tube (negative selection).

Collected monocytes were resuspended in RPMI 1640 medium (Lonza) supplemented with 5% autologous serum, 1% L-glutamine (Sigma-Aldrich), 1% penicillin-streptomycin (Gibco) and 5 ng/mL filter-sterilised recombinant human GM-CSF (Peprotech). Autologous serum was derived from a further blood sample which was centrifuged at 1000 - g for 10 minutes. Cells were added to a tissue culture plate and incubated at 37°C with 5% CO2 for one week. The medium was replenished after 3-4 days.

Stimulation of MDMs with chitin

Following this, MDMs were treated with1 mg/ml α-chitin (Sigma), 1 mg/ml β-chitin (purified from squid pen and kindly donated by Prof. George Roberts, Nottingham) ), or a serial dilution of chitohexaose ( ). A 50ng/ml LPS 055:B5 (Sigma-Aldrich) treatment was also produced as a positive control. A single batch of cells was left untreated for use as a control. Both the treated and untreated cells were incubated for 2 hours at 37°C with 5% CO2. Stimulations with chitohexaose were at the following concentrations: 500µg/ml, 50µg/ml, 5µg/ml, 0.5µg/ml, 0.05µg/ml and 0.005µg/ml with the exception of MDMs from the second donor where the final dilution was not tested.

mRNA extraction and cDNA synthesis

Following incubation, mRNA was extracted from the macrophages using the µMACS mRNA isolation kit (Miltenyi Biotech) following the manufacturer's instructions. From this, cDNA was yielded using the µMACS One-step cDNA kit (Miltenyi Biotech) following the manufacturer's instructions.

Primer design for expression of LysM domains, chitinases and lysozyme (with APRT as a reference gene) and RT-PCR

Primers for LysMD1-LysMD4 were designed and tested by Nana Serwaa Ampem Amankwah (previous year's MPharmIV student). A reverse transcription-polymerase chain reaction (RT-PCR) method was used to assay gene expression. Each RT-PCR reaction contained the following: 10µL ReddyMix (Thermo Scientific), 50nM forward and 50nM reverse primers (Sigma), 0.5µL of cDNA and DEPC-treated water to a final volume of 20­l. A PTC-200 Peltier Thermal Cycler (MJ Research) was used with the following conditions:

(PAR58-35) program. The products of amplification were analysed by gel electrophoresis on 1% agarose (Invitrogen) stained with ethidium bromide .

Primer design for expression of cytokines (with GAPDH as a positive control) and RT-PCR

PCR primers were designed using Primer3.49 To prevent the amplification of genomic DNA, primers were designed to span an exon-exon junction. They were subsequently submitted to NCBI Primer-BLAST to analyse regions of similarity and so prevent incorrect hybridisation to unwanted regions on the cDNA and primer-dimer formation. Suitable primer pairs were selected based on similarity of melting temperatures to allow efficient annealing - primers with similar guanine-cytosine (GC) content were chosen as this is a prediction of melting temperature. Repetitive mononucleotide sequences were avoided to prevent internal loop and hairpin formation.

[Table 1.]

RT-PCR was again used to assay the expression of various interleukins and interferon-γ. Each RT-PCR reaction contained the following: 7.5µL ReddyMix (Thermo Scientific), 37.5nM forward and 37.5nM reverse primers (Sigma), 0.53µLof extracted cDNA and DEPC-treated water to a final volume of 15­l.

[Figure 1.]

The optimal annealing temperature was determined from a temperature gradient PCR assay. Each gradient PCR reaction contained the following: 10µL ReddyMix (Thermo Scientific), 2µL of a combined solution of 5µM forward and 5µM reverse primers (Sigma), 1µLof extracted cDNA from a combined mixture of 5µL cDNA from each condition of each donor, and DEPC-treated water to a final volume of 20­l. The gradient was run using the GRA48-66 program, and applied the temperatures as follows: 48.0°C, 48.5 °C, 49.6 °C, 50.9 °C, 53.0 °C, 55.6 °C, 58.6 °C, 61.2 °C, 63.2 °C, 64.6 °C, 65.6 °C and 66.0 °C.

The RT-PCR was run using the PAR63-35 program.

[Figure 2.]

A separate gradient to determine the annealing temperature for interleukin-6 was carried out using 7.5µL ReddyMix (Thermo Scientific), 1.5µL of a combined solution of 5µM forward and 5µM reverse SYBRgreen primers (Sigma), 0.5µLof extracted cDNA from a combined mixture of 5µL cDNA from the LPS treated MDMs of donors 1 and 3, and DEPC-treated water to a final volume of 15­l. The gradient was ran using the GRA48-66 program

The RT-PCR was repeated for each condition of each donor with IL-6 using the PAR58-35 program.

Stimulation of monocytes

Monocytes from a fourth donor were isolated using previous methods with the exception that no GM-CSF was added to the medium. No 7-day incubation was undertaken. Cells were stimulated with either 1 mg/ml α-chitin or 10µg/ml chitohexaose and incubated for a maximum of 4 hours at 37°C with 5% CO2. At time periods of 0.5, 2 and 4 hours, treated cells from both conditions were removed and analysed using the above procedures used for MDM analysis.

Quantitative cDNA analysis using RT-qPCR

RT-qPCR based on SYBR-Green I fluorescence50 was carried out on the cDNA extracted from the MDMs of donor 3. All conditions were tested with exception of the 1 mg/ml β-chitin and 0.005µg/ml chitohexaose stimulated cells. Each qPCR reaction contained the following: 10µL SYBRgreen Master Mix (KAPA), 80nM forward and 80nM reverse primers (Sigma), 1µLof extracted cDNA, and DEPC-treated water to a final volume of 20­l. GAPDH and APRT were used as reference genes to normalise gene expression. The change of target gene expression (chit1 or TNF α) was normalised to the reference genes and was relative to expression in the untreated control cells. Calculation of the fold changes were achieved using the Pfaffl mathematical model.51

The PCR protocol was as followed:

Cycle 1: (1X)

Step 1: 95.0°C for 15:00

Cycle 2: (45X)

Step 1: 95.0°C for 00:15

Step 2: 60.0°C for 00:20

Step 3: 72.0°C for 00:20

Cycle 3: (1X)

Step 1: 95.0°C for 01:00

Cycle 4: (1X)

Step 1: 55.0°C for 01:00

Cycle 5: (70X)

Step 1: 58.0°C for 00:10

3 Results

3.1 mRNA expression in monocyte-derived macrophages

[Fig. 3-5]

A reverse transcription-polymerase chain reaction (RT-PCR) method was used to assay gene expression and analysis was carried out using gel electrophoresis. MDMs stimulated with both chitohexaose and whole chitin (α and β) produced bands on the gel that correspond with expected sizes for LysMD2 and 3, Chit1, lysozyme and Chi-3-L1 expression. LysMD4 expression was also seen, but the band produced was significantly weaker. The bands generated from analysis of the untreated MDMs were comparable to those obtained in the stimulated cell cultures. This is consistent with the knowledge that chitotriosidase is expressed constitutively in macrophages.26 LysMD1 was not shown to be expressed constitutively, nor induced upon stimulation. Acidic chitinase (Chia) was also not expressed, and this is consistent with studies that show it is abundant within the gastrointestinal tract and lung epithelia, but not within macrophages.17 Stimulation with LPS produced comparable results to the chitin stimulated cultures.

3.2 mRNA expression in monocytes

[Fig. 6 and 11]

For an additional study, gene expression in monocytes was also considered over an incubation period of four hours. Whole (α) chitin and chitohexaose-stimulated monocytes produced the same results on analysis. An expression of the LysMD2 and Chi-3-L1 gene was observed, though relatively weaker to that seen in the MDMs. Lysozyme was expressed to the same extent. All other genes, including Chit1, were not expressed. This again is consistent with previous studies that have found that chitotriosidase is absent in human monocytes.53 These results are analogous to those obtained with unstimulated monocytes, suggesting that expression of lysozyme, LysMD2 and Chi-3-L1 is constitutive.

3.3 Cytokine expression in MDMs

[Fig. 7-9 and 11]

To determine whether chitin acts as a PAMP in humans, the expression of macrophage derived inflammatory cytokines was analysed. As before, treatment with either whole chitin or chitin fragments produced similar results in MDMs. Strong expression of IL-10 and IL-23A and weak expression of IL-6 and IL-17A was observed in both treated and untreated cells - a quantitative comparison could not be made based on results from gel electrophoresis alone. IL-12A and interferon-γ were able to produce a very weak band in the untreated culture; however these bands were lost upon chitin stimulation. This suggests that within MDMs, there are constitutive levels of interleukins 6, 10, 12A, 17A, 23A and interferon-γ. MDMs incubated with LPS were shown to have a relatively stronger expression of IL-12A and interferon-γ; bands produced by other cytokines were analogous to those seen upon the addition of chitin, and again because of the technique employed, it cannot be said as to whether there are quantifiable differences in the levels of expression.

3.4 Cytokine expression in monocytes

[Fig. 10-11]

The studies carried out on monocytes found that there is very weak expression of IL-6, IL-10, IL-23A, and also of interferon-γ. This was observed in both the treated and untreated cultures.

Quantitative analysis of Chit1 and TNFα expression

[Fig. 12 and 13]

The effect of chitohexaose, chitin α and LPS on chitotriosidase expression in MDMs was measured quantitatively. Levels were measured using RT-qPCR and normalised to two reference genes, GAP-DH and APRT. Figure 12 shows the level of fluorescence (generated by the binding of SYBR green to the DNA) against the number of cycles. The threshold at which amplification becomes exponential is set at 466.9. Results are given as a fold change in expression relative to an untreated control (see Materials and Methods). TNFα levels were also measured as a positive control, and also as an indicator of chitin recognition leading to the generation of an immune response. The cDNA from one donor was analysed and all conditions were tested with exception of the β-chitin and 0.005µg/ml chitohexaose (final dilution) stimulated cells. All experiments were carried out in triplicates with the exception of LPS relative to GAP-DH and LPS relative to APRT due to a lack of cDNA. Results within the triplicates were comparable. To validate the primers, a melt curve was produced. Incorrect hybridisation, the presence of contaminants, and primer-dimer formation can cause unwanted fragments to be amplified. The PCR products for a pair of primers should have similar melting temperature, and this is dependent on the length of the sequence and the base composition. As the dye cannot distinguish between the correct and incorrect sequences of DNA, the melt curve acts as a method of quality control to identify additional unwanted amplification products. Figure 13 shows distinct peaks, indicative of the formation of the correct PCR product only. Each peak corresponds to one specific primer pair.

[Fig. 14]

Figure 13 shows the expression levels of both Chit1 and TNFα mRNA normalized to GAP-DH. Both chitin α and chitohexaose stimulated MDMs did not show a significant up-regulation of either the chitotriosidase or TNFα genes. As expected, LPS showed a strong induction of TNFα with an approximately 154-fold increase in expression. TNFα levels were also shown to increase by approximately 4-fold in MDMs stimulated with chitin α, but this is thought to be biologically irrelevant (see Discussion).

[Fig. 15]

Figure 14 shows the expression levels normalized to APRT. These results are consistent to the findings observed with GAP-DH -no up-regulation of either gene occurs upon stimulation with chitohexaose, and similarly, a small induction of TNFα is seen in the chitin α-stimulated cells which possibly due the presence of contaminants rather than the chitin itself.

The stimulation with 50ng/ml LPS was included in the design of the experiment as another positive control, due to previous studies demonstrating Chit1 induction upon endotoxin treatment.54 The data obtained shows that LPS is able to significantly up-regulate TNFα expression in MDMs, but does not induce chitotriosidase expression (Fig. 14 and 15).

4 Discussion

To determine whether chitin or chitooligosaccharides have the ability to act as pathogen-associated molecular patterns in humans, we evaluated the effects of both whole chitin and the fragment chitohexaose on macrophage expression of particular genes in vitro. The studies demonstrate that both forms of the polymer do not initiate an immune response nor do they up-regulate the genes associated with its catabolic degradation i.e. Chit1. They also show that there are constitutive levels of chitotriosidase, the binding lectin Chi-3-L1, and the putative peptidoglycan-binding LysM domains 2-4. This suggests that while whole chitin or chitohexaose does not modulate macrophage function, a different chitooligosaccharide (formed by the action of constitutive chitinases on chitin) may.

Recently published data show an increased murine macrophage expression of IL-10, 12A, 17A and 23 when stimulated with certain sized fragments.36, 37 Within our study, these cytokines were found to be expressed in macrophages both stimulated and unstimulated, however there appears to be no apparent difference in the intensity of the bands produced from gel electrophoresis analyses suggesting that these cytokine levels are constitutive, and not increased upon treatment with chitin. Without further experimentation (i.e. RT-qPCR carried out on other cytokine gene expression additionally to TNF α), it cannot be said as to whether there are quantifiable differences in the levels of expression between the two conditions.

TNFα is a pro-inflammatory cytokine that is normally produced by macrophages upon recognition of a PAMP. Quantitative results show a marked increase in levels of TNFα expression in the cells stimulated with LPS, a well established PAMP, but not with chitohexaose. This, to some extent, agrees with the results obtained by Da Silva et al (2009) who found that certain sizes of chitin fragments are able to stimulate TNFα production within murine macrophages (see Introduction).

However, it is difficult to make comparisons between the studies. A clear conclusion derived by Da Silva is that immune cell recognition is size-dependant, and so in order to make an informative comparison, it would be appropriate to define the degree of polymerization that induced the inflammatory response. The chitin preparations used for the earlier studies are defined purely on particle diameter and are in the micrometre size range - there is no indication on the size of polysaccharide in terms of monomer length. Using the approximate size of a molecule of glucose as a reference (0.84nm), we can estimate the size of a molecule of chitohexaose to be six-times that amount, approximately 5nm.52 Thus it can be said that there is approximately a 1000-fold difference between the chitohexaose used in our experimentation, and the smallest chitin fragment that elicited a response in the Da Silva paper. These fragment sizes are comparable to or much larger than the average size of a macrophage (approximately 10-15 µm). It was previously discovered that fragments less than 2µm (SSC) fail to induce any macrophage activation, however it is known that much smaller chitooligosaccharides are able to initiate gene induction and defence responses in various other organisms [refs]. It is possible that the chitin-induced TNFα production occurred by the recognition of much smaller fragments formed by the release and action of constitutive exochitinases which we have confirmed exist in macrophages. However, this would not explain why chitin fragments larger than 70µm and less than 2µm remain inert.

Secondly, the acquisition of macrophages differs - the Da Silva studies obtained their macrophages either by in vivo recovery from bronchoalveolar lavage fluid or thioglycollate-elicited peritoneal macrophage wash-out, whereas our experimentation utilized human peripheral blood MDMs. It has been demonstrated that macrophages from different tissues possess a distinct phenotype with regard to their role within the immune system, morphology and cell-surface protein expression (other ref, or ref 55?). Additionally, the choice of growth factor used to stimulate monocyte differentiation in macrophages also controls macrophage heterogeneity.55 A study conducted by Akagawa (2002) has identified the generation of two phenotypically-distinct macrophage cultures derived from human monocytes when stimulated with either macrophage colony-stimulating factor (M-CSF) or granulocyte-macrophage colony-stimulating factor (GM-CSF). The two cultures were shown to differ in function, morphology and cell-surface markers.56 The study on chitin-induced TLR-2 and IL-17A activation produced by Da Silva used in vitro prepared cultures of bone-marrow derived macrophages grown in M-CSF, which may account for the increased expression of IL-17 and IL-23 upon stimulation with chitin. The MDMs used in our study were cultivated in GM-CSF, and it is unknown as to whether this would account for differences in our results - additional studies will be required using various macrophage cultures to establish if gene expression and response to chitin is affected.

Our data shows that TNFα expression seems to be up-regulated in MDMs stimulated with chitin α, however we believe this to be untrue based on the insolubility of the chitin particles that would make sensing by the macrophages extremely difficult (fig. 16) The increase in expression is possibly due to the presence of endotoxin within the chitin α preparation - endotoxin levels should be below the limits of detection in a Limulus amebocyte lysate (LAL) assay, however may not be below the limits for macrophage recognition. The LAL assay (Sigma-Aldrich) is used for the quantitative measurement of endotoxin in a solution, and has a minimum detection limit of 0.05EU (endotoxin units)/ml, equivalent to 5pg/ml. While to the best of our knowledge there is no data detailing the minimum detection limit of LPS in human macrophages, it cannot be said that undetectable levels of LPS being the cause of the up-regulation is anything more than a mere possibility. We have shown that TNFα levels increase upon MDM stimulation with LPS in a separate PCR experiment, and so this favours the validity of this theory. It should also be noted that chitin in its purest form is white. The preparation used to produce the chitin α suspension however was yellow-brown, suggesting the presence of impurities which may have initiated a macrophage response.

We have shown that chitotriosidase is not up-regulated upon MDM stimulation with chitin α or chitohexaose. It was found to be expressed constitutively, a result consistent with earlier studies on expression.26 Current data suggests a role for chitotriosidase within innate immunity and in the defence of fungal pathogens, for example, activity is shown to increase in guinea pigs with systemic infection of Aspergillus fumigatus .24 It is possible that a larger chitooligosaccharide is required to produce this increase in enzymatic activity, and additional studies with such polymers could help determine this. The cells were stimulated with the chitin for 2 hours - it is possible that a longer incubation was needed to allow the catalytic degradation of the polymer into a smaller fragment that could be sensed. It can be seen in figures 16 and 17 the size differences between the chitin crystals and the cells themselves. A similar study conducted over a time course would be needed to identify whether this would generate a macrophage response.

The stimulation with 50ng/ml LPS was included as a positive control as it t has previously been demonstrated that chitotriosidase expression can be induced in MDMs by LPS. 54 As previously discussed, LPS was shown to significantly up-regulate TNFα expression, however no such effect is seen on Chit1 induction. WHY?

Due to time restraints and budget restrictions, quantitative experimentation was only able to be carried out on the cDNA obtained from one donor. For reassurance on the validity of the results, repeat RT-qPCR experimentation on the cDNA from further donors would be required to see whether the data collected in this study is indicative of a 'normal' macrophage response to chitin.

With the discovery that the human genome encodes various proteins containing LysM domains, we tested for their presence within MDMs. The data shows that LysMD2-4 are all expressed constitutively in MDMs, however LysMD1 is not. In plants and prokaryotes, the LysM motif is found is various parts of the cell such as the cell wall and membrane, and binds the N-acetylglucosamine moiety of peptidoglycan and chitin as part of pathogen detection. The presence of LysM domains in the chitinases of certain eukaryotes has suggested that it is this particular motif that is responsible for chitin recognition in the eukaryotic kingdom.23 Our discovery of the existence of such motifs in human macrophages favours this hypothesis, however further experimentation is needed to elucidate their precise role in man and whether these are indeed involved in chitin detection and subsequent cell signalling.


Chitin and its smaller fragments have been demonstrated in many studies to have a diverse range of biological activities. The discovery of chitinolytic enzymes within man that share a significant sequences similar to that found within other organisms has prompted consideration of chitin to act as a modulator of the innate immune system. Recent studies have shown that chitin contains PAMPs that allow the interaction and activation of macrophages in a size-dependent fashion. Due to its total insolubility in aqueous media, and the size of the chitin particles in these studies being comparable (if not larger) than the average size of macrophages, we hypothesized that recognition by immune cells requires the prior release of chitinases which would interact with the chitin to form various chitooligosaccharides for immune detection. In summary, our studies demonstrate that neither crystalline chitin nor chitohexaose cause an immune activation or enhance the expression of genes encoding any members of the chitinase protein family. Further studies that investigate other sizes of fragments over a time course are warranted to confirm whether chitin is indeed a human PAMP. These studies also show the existence of constitutive LysM domains within human MDMs which may possibly be part of a putative receptor that allows chitin recognition, cell signalling and generation of an innate immune response, in analogy to what has been described in the plant kingdom. Additional investigations are required to establish whether chitin does indeed cause macrophage activation via these motifs.