Isolation and characterization of chitin-degrading micro-organisms from the faeces of Goeldi's monkey, Callimico goeldii


  • C. Macdonald,

    1. School of Life, Sport & Social Science, Edinburgh Napier University, Edinburgh, UK
    2. Animal Department, Edinburgh Zoo, Edinburgh, UK
    Current affiliation:
    1. Life Sciences Department, Twycross Zoo, Warwickshire, UK
    Search for more papers by this author
  • S. Barden,

    1. School of Life, Sport & Social Science, Edinburgh Napier University, Edinburgh, UK
    Search for more papers by this author
  • S. Foley

    Corresponding author
    1. School of Life, Sport & Social Science, Edinburgh Napier University, Edinburgh, UK
    • Correspondence

      Sophie Foley, School of Life, Sport & Social Science, Edinburgh Napier University, EH11 4BN, Edinburgh, UK. E-mail:

    Search for more papers by this author



The objective of this study was the isolation and characterization of chitin-degrading micro-organisms from the faeces of the insectivorous Goeldi's monkey, Callimico goeldii.

Methods and Results

Faeces samples were screened for chitin-degrading bacteria using basal medium in which chitin was included as the carbon and energy source. Of fifteen bacterial isolates with chitin-degrading activity, fourteen were also capable of degrading cellulose. All isolates were either aerobes or facultative anaerobes.


Phylogenetic analyses of those isolates exhibiting strongest activity, as determined by the most distinctive zones of clearing in chitin-supplemented medium, were identified as Cellulosimicrobium spp., Arthrobacter spp., Staphylococcus spp. and Enterobacteriaceae.

Significance and Impact of the Study

This study reports on the isolation of chitin-degrading microflora from nonhuman primates. Considering that chitin and cellulose are the most abundant naturally occurring polymers, it is of interest to note that the majority of isolates are capable of digesting both substrates. This may be of significance given that omnivorous primates live in seasonal environments, where the availability of food items varies with the seasons. Furthermore, given the presence of a chitin-degrading microflora, this may have implications, in terms of the inclusion of fungi and/or insects in the diets of these animals in captivity, whether as part of medical research or conservation programmes.


In the last decade, there has been a rapid increase in the applications for chitin, chitosan and chitin-degrading enzymes in agriculture (plant protection), animal husbandry (nutrition) and human health care (Hamid et al. 2013). Chitin-degrading enzymes have been isolated from animals and plants, as well as from a diversity of bacteria and fungi originating from soil and plant material, aquatic environments and the gastrointestinal tract of a range of animal species including invertebrates, fish, reptiles, birds and mammals (Jackson et al. 1992; Marsh et al. 2001; Souza-Neto et al. 2003; Molinari et al. 2007; Paoletti et al. 2007). Chitin-degrading enzymes in animals are therefore either endogenous to the animal itself, originate from ingested food or are associated with the commensal microflora. The confirmed presence of chitin-degrading micro-organisms in the digestive tracts of animal species such as cetaceans (Olsen et al. 2000) and pinnipeds (Sugita et al. 1996), which consume chitinous prey, is suggestive of a link between the chitin content of the diet and the presence of chitin-degrading microflora within the digestive tract. While a key role of chitin-degrading enzymes may be in the digestion of chitin to provide nutrients and energy, it is hypothesized that chitinases in the mammalian digestive tract may contribute to a defence mechanism against fungal pathogens (Overdijk et al. 1999).

Callitrichids, which includes marmosets and tamarins, are omnivores, consuming a range of dietary components including invertebrates, fruits, fungi, plant exudates and vertebrate prey. The precise composition of the diet varies according to species and seasonal availability of the components (Wolda 1978). A major component of insects is their chitinous exoskeleton, and as invertebrates can make up a large component of the diet of some callitrichid species, such as buffy tufted-eared marmosets, Callithrix aurita, consuming animal prey for 38·5% of feeding observations (Martins and Setz 2000) and the Panamanian tamarin, Saguinus oedipus geoffroyi, which makes up 40% of their diet with invertebrates (Garber 1980), it can be expected that callitrichids may harbour chitin-degrading micro-organisms within their digestive tract.

While insects form a significant component of the diet of wild callitrichids (Garber 1980; Martins and Setz 2000; Smith 2000), a limited number and range of invertebrates may be presented in the diets of callitrichids in captivity, for example in zoos and medical research facilities. Given that captive callitrichids historically have exhibited poor nutritional and digestive health (Clapp and Tardif 1985; Lewis et al. 1987), the ability of callitrichids to digest chitin, found within their invertebrate prey, warrants further investigation. Furthermore, the diet of C. goeldii in the wild has been shown to include fungi, which contains chitin in their cell walls, making up 29% of the overall diet for one study group (Porter 2001). With the inclusion of both fungi and invertebrates in their diet, C. goeldii is a strong candidate for the investigation of chitin-degrading micro-organisms.

The objective of this study was the isolation and characterization of chitin-degrading micro-organisms from the faeces of the insectivorous Goeldi's monkey, Callimico goeldii.

Materials and methods

Preparation of chitin for incorporation into agar media

1·5 g of crabshell chitin (practical grade, Sigma-Aldrich, Dorset, UK) was placed in the grinding bowl of a planetary micro mill (Pulverisette 7, Fritsch, Idar-Oberstein, Germany) along with 15 ml dH2O. This mixture was milled using five ceramic balls, as described in the manufacturer's manual, for 2 h at a rotational speed of 4·5, as marked on the mill. An additional 8·5 ml of dH2O was added, and the mixture was milled for a further 3 h resulting in a finely ground chitin paste with a chitin concentration of 6% w/v.

Preparation of cellulose for incorporation into agar media

For incorporation into agar medium, a cellulose solution was prepared as described by Egal et al. (2007). 5 g of Sigmacell® cellulose was dissolved in 100 ml ice-cold 10% NaOH by stirring for 1 h. After overnight storage at −20°C, the solution was thawed and added to 100 ml dH2O and then syringe-filtered to eliminate insoluble residues. The pH was neutralized with 50% hydrochloric acid and made up to a final volume of 1 litre with dH2O. The cellulose solution was then centrifuged for 2 min at 1250 g to eliminate NaCl. This was repeated three times, discarding the supernatant and replacing with water.

Faecal sample collection

Faecal samples from a group of Goeldi's monkeys, residing in a naturalistic enclosure at Edinburgh Zoo (UK), were used for the analysis of the faecal microflora as this species is known to consume insects in the wild, and they are fed with insects at Edinburgh Zoo as part of their routine diet. This group was composed of five animals at different life stages: adults, juveniles and infants. These animals had not been on any antibiotic or anthelmintic treatment for at least 4 weeks prior to faecal collection. Furthermore, they readily consumed the invertebrate portion of their diet. The daily diet per individual monkey included: 5 g concentrated pellet ‘Marex’ (Special Diet Services, UK); 145 g fruit and vegetables (Paterson Bros, Total Produce, and Fyffes Multifresh, UK); ad lib gum Arabic (J.Flach and Son, UK); 5 live insects (Eurorep and Livefood Direct, UK); and vitamin supplements (Dunlops Veterinary Suppliers and Woodstocks Nutritional Supplements, UK). The insects were themselves fed on bran, cuttlefish bone and vegetables.

Fresh faecal boluses, collected within 4 h of defecation, were placed in a resealable bag and the air expelled. Samples were transported to the laboratory and further processed within 2 h of collection in the anaerobic chamber (MACS MG500 + MG Airlock, Don Whitley Scientific; 10% CO2, 10% H2 and 80% N2). As numerous individual boluses were collected at each sampling session, an amount was taken from each bolus and pooled. The sample was carefully removed from the soft interior of the bolus using a sterile loop. The sample was homogenized in a hand-held sterile homogeniser (Safe-Seal homogeniser, Jencons Scientific, Bedfordshire, UK), and when homogenized to a paste, diluent (per litre 5 g peptone, 2·5 g sodium chloride, 0·5 g l-cysteine hydrochloride, adjusted to pH 7 with 1 mol l−1 NaOH) was added. The sample was carefully mixed, and serial dilutions were carried out before plating on the appropriate medium.

Screening for chitin- and cellulose-degrading bacteria

Screening of faeces samples for the presence of chitin- and cellulose-degrading bacteria was undertaken using a basal medium containing per litre, 75 ml mineral solution I, 75 ml mineral solution II, 10 ml trace element solution, 10 ml vitamin solution, 0·5 g yeast extract, 1·0 g tryptone, 1 ml resazurin (0·1%), 1 ml hemin (0·1%), 20 ml cysteine sulfide reducing agent (containing 1·25% w/v each cysteine-HCl·H2O and Na2S·9H2O) and 2·5 g NaHCO3 (Robert and Bernalier-Donadille 2003) supplemented with a carbon/energy source – glucose (5 g l−1), chitin (5 g l−1) or precipitated cellulose (133 ml l−1) prior to autoclaving. The chitin and cellulose preparations are described above. Faecal sample dilutions were plated in the anaerobic chamber, and colonies were counted after 3 days of incubation at 37°C under aerobic conditions or 33°C in the anaerobic chamber.

Genomic DNA isolation

Genomic DNA was isolated from overnight cultures using the Wizard® Genomic DNA Purification Kit (Promega, Madison, WI, USA). The presence of DNA was checked by agarose gel electrophoresis using molecular-grade agarose (0·8% w/v, Fluka, Dorset, UK) in Tris-acetate buffer (TAE; 40 mmol l−1 Tris-base, 20 mmol l−1 acetate and 2 mmol l−1 EDTA, pH 8·0) and ethidium bromide at a final concentration of 0·3 μg ml−1. DNA samples were mixed with loading buffer (0·25% bromophenol blue, 30% glycerol in water). The samples were electrophoresed at 100 V, and gels visualized using the Molecular Imager FX Pro and Quantity One software (Bio-Rad, Hemel Hempstead, UK).

PCR amplification and cloning of 16s rRNA gene

PCR amplification of the 16S rRNA gene was carried out with the specific forward primer S-*-Eub-0339-a-A-20 (5′-CTC CTA CGG GAG GCA GCA GT-3′) and reverse primer S-*-Univ-1385-b-A-18 (5′-GCG GTG TGT ACA AGR CCC-3′) (Mangin et al. 2004), 1 ng of template DNA, 1 U Phusion® Hot Start DNA polymerase (Finnzymes, Vantaa, Finland). The thermocycling conditions used were as follows: a denaturing period of 10 min at 95°C, followed by 25 cycles of 1 min denaturation at 97°C, 1 min annealing at 59°C and 1 min elongation at 72°C and then a final extension step at 72°C for 10 min, followed by cooling down to 4°C.

PCR-generated partial 16s rRNA genes were cloned using the pCR 2·1-TOPO vector and the TA cloning system (Invitrogen, Paisley, UK) as per the manufacturer's instructions. Due to the proof-reading activity of the Phusion® Hot Start DNA polymerase, PCR products were incubated in the presence of 0·25 μl Taq polymerase (Promega), 0·2 mmol l−1 dATP at 70°C for 5 min prior to ligation.

Phylogenetic analysis on the basis of 16s rRNA gene sequence

Plasmid DNA was purified from recombinant colonies using the QIAprep Spin Miniprep kit (Qiagen, Manchester, UK) as per manufacturer's instructions, and the cloned 16S rRNA gene sequence was determined. Sequences were compiled for each clone using the CAP3 sequence assembly program ( (Huang and Madan 1999). To ascertain the closest relatives of the examined strains, analysis of the 16S rRNA gene sequences was undertaken using BLASTN (Zhang et al. 2000). Using the MEGA 5.05 software (Tamura et al. 2011), phylogenetic trees were constructed using the UPGMA (unweighted pair group method with arithmetic mean), neighbour-joining, maximum parsimony and maximum-likelihood methods.


A culture-based approach was taken to screen faeces samples from Goeldi's monkey for the presence of chitin-degrading bacteria. Since the recovery of bacteria from faeces was comparable on basal medium containing glucose (BM–glucose) to that on the complex medium brain heart infusion agar (BHI) (data not shown), the basal medium supplemented with chitin was used in the recovery of chitin-degrading microflora from the faeces samples. The ability of the basal medium to support bacterial growth in the absence of supplementation with a carbon/energy source (glucose, chitin or cellulose) was also tested. Escherichia coli was inoculated on to basal medium without glucose, chitin or cellulose. The E. coli did not grow on the medium, indicating that no carbon sources were available. Furthermore, E. coli only grew on the glucose-supplemented medium. This confirmed the value of using the BM-chitin medium in screening faeces for chitin-degrading micro-organisms as it can be assumed that any microbial growth observed is capable of actually utilizing the chitin provided.

Faecal samples were plated in duplicate on BM-chitin and incubated under anaerobic and aerobic conditions. Following the appearance of colonies, plates were stored at room temperature, and within 2 weeks, chitin-degrading activity became visible as clearings around colonies on plates stored aerobically. Clearing was not visible on plates stored anaerobically. From one faeces sample, 15 colonies exhibiting distinctive zones of clearing in the chitin-supplemented medium were isolated and purified. Due to the known ability of some micro-organisms to degrade both chitin and cellulose (Reguera and Leschine 2001), all isolates were then cross-screened on BM-cellulose. This revealed that 14 of the 15 isolates were capable of both chitin- and cellulose-degrading activity. Nine isolates exhibiting greatest chitin-degrading activity were selected for preliminary identification.

Phylogenetic analysis on the basis of 16s rRNA gene sequence

To proceed with the identification of the chitin-degrading isolates from Goeldi's monkey faeces, a 1000-bp region of the 16S rRNA gene was PCR-amplified, cloned and sequenced. The data were subjected to BLASTN analysis, the results of which are presented in Table 1.

Table 1. BLASTN analysis of the 16S rRNA gene sequences of 9 isolates from Goeldi's monkey faeces
IsolateClosest matched identityGenBank accession no.% Nucleotide matchOther significant matches



Cellulosimicrobium funkei strain W6123 AY523787.1 100

Oerskovia xanthineolytica

Promicromonospora sp.

Cellulomonas sp.



Cellulosimicrobium cellulans strain NFR1 GQ496666.1 100

Actinobacterium sp.

Cellulomonas cellulans

X10.1.1Raoultella sp. clone HSL78C HM461206.1 99 
Enterobacter aerogenes NC4211 AB244472.1 99 
Kluyvera cryocrescens HQ540321.1 99 
X20.1.2Arthrobacter sp. TS22 EU073088.1 99 
Arthrobacter nicotianae GJA819 HM209738.1 99 
Arthrobacter arilaitensis strain DS37 EU834260.1 99 
Brevibacterium liquefaciens AJ251417.1 99 
Arthrobacter bergeri isolate Re127 NR025612.1 99 
X5.1.1 Enterobacter aerogenes AB244302.1 99

Raoultella sp.; Pantoea sp.

Kluyvera cryocrescens

X10.1.2 Uncultured bacterium AM697374.1 100 

Staphylococcus hominis


HE578786.1 99 

Isolates X2.1, X5.2.1, X7.1.2, X9.2.1 and X5.1.2 were all identical – Gram-positive, nonmotile rods, and 100% sequence identity (based on a partial 16s rRNA gene sequence of 1047 nucleotides) was obtained with Cellulosimicrobium funkei strain W6123. The phylogenetic trees generated using the UPGMA, neighbour-joining, maximum parsimony and maximum-likelihood methods were nearly identical with the same clusters being recovered in each. As can be seen from Fig. 1a, in the phylogenetic tree based on the UPGMA algorithm, bootstrapped with 1000 replications, isolate X2.1 joins the cluster comprising C. funkei.

Figure 1.

Unweighted pair group method with arithmetic mean (UPGMA) phylogenetic trees based on 16S rRNA gene sequences showing the position of isolates characterized in this study and related taxa (largely representing type strains of related genera and species as indicated by T). Bootstrap percentages (based on 1000 replications) are shown at the branch points. Bar, number of substitutions per nucleotide position. (a) Isolate X2.1. Tree is rooted using Jonesia denitrificans DSM 20603T as the outgroup. (b) Isolate X20.1.2. Tree is rooted using Mycobacterium smegmatis ATCC 19420T as the outgroup.

Isolate X20.1.2 produced round smooth colonies that were lobate and yellow-white on glucose-containing medium, while convex and yellow on medium containing chitin or cellulose. Based on a partial gene sequence of 1047 nucleotides, the 16s rRNA sequence obtained had a 99% match with Arthrobacter species. Phylogenetic trees were constructed as described above using all recognized species of the genus Arthrobacter. In each case, isolate X 20.1.2 was consistently associated with the Arthrobacter mysorens and Arthrobacter nicotianae species (Fig. 1b).

Isolate X5.1.1 produced white colonies on glucose and yellow colonies when grown on chitin or cellulose medium. The BLASTN search revealed a 99% match with Enterobacter aerogenes. When grown on medium containing glucose, isolate X10.1.1 produced round, white colonies with an entire margin, which was flat with a smooth surface. On medium containing chitin or cellulose, the colonies were also round and smooth with an entire edge, but were also umbonate and yellow. The 16S rRNA gene sequence showed a 99% match with the partial 16S rRNA sequences of Raoultella species, Ent. aerogenes and Kluyvera cryocrescens. Isolate X10.1.2 is Gram-positive motile coccus producing round white colonies with an entire edge and a flat, smooth surface on all media tested. The 16S rRNA gene sequence showed a 99% match with that of Staphylococcus species. It should be noted that the 16S rRNA sequence is considered to be of limited value for the phylogenetic analysis of the genus Staphylococcus and of the Enterobacteriaceae, of which the genera Enterobacter, Raoultella and Kluyvera are members (Mollet et al. 1997).


This investigation has successfully isolated chitin- and cellulose-degrading micro-organisms from the faecal samples of Goeldi's monkeys. The screening technique of using chitin-based medium required initial investment into the optimization of the incorporation of chitin into agar medium. Various methods were attempted, and while acidic degradation (Hsu and Lockwood 1975) and the production of colloidal chitin (Chernin et al. 1995) have produced suitable media for other studies reported in the literature, the opacity of the agar plates produced was inconsistent, and in many cases, the large particles of chitin were clearly visible. This led to concerns that the large chitin particles may not be biologically available to chitin-degrading micro-organisms that may be present in the faecal samples. Therefore, the method chosen for screening in this study was that of mechanically pulverizing the chitin in a ball mill (Barrett-Bee and Hamilton 1984). The degree of pulverization was judged by assessing the appearance and texture of the resulting chitin powder and the opacity of the plates when it was incorporated into agar. This resulted in plates that contained chitin-based agar, which was dense and sufficiently opaque to show clear halos, if any of the isolates had chitin-degrading potential, without having a particulate appearance.

The Cellulosimicrobium funkei type strain was isolated from human blood, but the organism has since been recovered from duck faeces (Murphy et al. 2005) and in airborne samples from duck housing (Martin et al. 2010). Other members of the genus Cellulosimicrobium have been isolated from a range of sources including Cellulosimicrobium terreum isolated from soil (Yoon et al. 2007) and Cellulosimicrobium cellulans that was isolated from Antarctic snow (Antony et al. 2009). While Schumann et al. (2001) state that most members of the genus Cellulomonas, reclassified as Cellulosimicrobium, show cellulolytic activity, there has been little research conducted with C. funkei to assess chitin-degrading potential. A closely related species, Cellulosimicrobium variabile, was isolated from the hindgut of termites where it is likely to play a role in cellulose digestion (Bakalidou et al. 2002). Oh et al. (2008) described a strain, Cellulosimicrobium sp. HY-12, isolated from the digestive tract of the mole cricket, Gryllotalpa orientalis, and Kim et al. (2009) described another isolate, Cellulosimicrobium HY-13, isolated from the gut of the earthworm Eiseniafetida, both of which lack cellulolytic potential, but can digest xylan. These studies therefore demonstrate that members of the genus Cellulosimicrobium are involved in the digestion of complex polysaccharides.

The 16S rRNA partial gene sequence obtained for isolate X20.1.2 showed a 99% match to that of members of the genus Arthrobacter. Arthrobacter species have been isolated from a number of environmental and animal-based samples including penguin guano (Wang et al. 2009), Antarctic marine sediment (Pindi et al. 2010), soil (Zhang et al. 2010), sewage (Kim et al. 2008), sea water (Chen et al. 2009) and the nasal passages of the common seal (Collins et al. 2002). Of particular interest is the organism Arthrobacter psychrochitiniphilus, which originates from penguin guano (Wang et al. 2009). This organism was found to be closely related to isolate X20.1.2 using the BLASTN search, and it is also able to degrade chitin.

The remaining isolates present an interesting case. Chitin-degrading activity has previously been identified in Enterobacter, including Ent. aerogenes (Brzezinska and Donderski 2006), Enterobacter agglomerans (Chernin et al. 1995) and Enterobacter spp. strain NRG4 (Dahiya et al. 2005), while cellulolytic activity has been reported for numerous Klebsiella species including Klebsiella pneumoniae (Anand et al. 2010), Klebsiella trevisanii (Okeke and Lu 2011) and Klebsiella oxytoca (Anand and Sripathi 2004), and some other members of the Enterobacteriaceae family including Proteus vulgaris, Proteus mirabilis, Citrobacter freundii and Serratia liquefaciens (Anand and Sripathi 2004). Some Staphylococcus species have the ability to degrade colloidal chitin (Wadstrom 1971), although there is limited evidence available in the literature.

It should be noted that all of the isolates described above are either aerobes or facultative anaerobes and were all recovered on agar plates incubated in an aerobic atmosphere. While growth was observed on chitin-supplemented medium under anaerobic conditions, colony size remained small (compared with growth on a more complex medium) and zones of clearing in the medium did not develop on storage for a 1–2 week period. It may therefore be necessary in future studies to optimize the medium used for recovery of chitin-degrading strict anaerobes.

Since this study led to the isolation of a diversity of chitin-degrading bacteria from the faeces of the Goeldi's monkey, it will be of interest in future studies to determine the exact nature of the enzymatic activity undertaken by each of these organisms, recognizing that the processes involved in chitin degradation include chitinases/N-acetylglucosaminidases, deacetylases, deaminases and chitosanases/cellulases (Beier and Bertlisson 2013). It will also be of interest to determine the extent to which they contribute to the microflora of Goeldi's monkey and indeed that of other insectivorous and mycophagous callitrichids. The approach taken in this study cannot account for the interactive nature of a full microbial community, the composition of which can fluctuate and change. Therefore, a longitudinal study recording, and where possible, controlling dietary and environmental changes, would allow comparisons between subjects on diets with and without chitin. This may have implications for the welfare of these animals and the development of clearer husbandry guidelines for these animals in captivity, whether that is for the purpose of medical research or in conservation programmes.

It is noted that the organisms isolated in this study, for the most part, are capable of digesting both chitin and cellulose. Cellulose and chitin are the most abundant naturally occurring polymers on earth and are structurally very similar, with the acetyl amine group of chitin being replaced by a hydroxyl group in cellulose. Reguera and Leschine (2001) showed that cellulolytic micro-organisms from soil also had the ability to degrade chitin and speculated that these organisms might have a selective advantage in an environment where cellulose may not always be available in abundance. Thus, omnivorous primates, consuming both animal and plant materials, may harbour micro-organisms in their digestive tract, which have the potential to degrade both chitin and cellulose. This could present them with a selective advantage when living in an environment where the availability of food items varies with the seasons.


We would like to acknowledge the support of the Royal Zoological Society of Scotland and the provision of a studentship by Edinburgh Napier University.

Conflict of Interest

No conflict of interest declared.