The Tannase Gene: Metaphylogenomics, Global Distribution and Presence in the Midgut Flora of the Forest Tent Caterpillar Malacosoma disstria Hübner Michael Gasse A Thesis In The Department of Biology Presented in Partial Fulfillment of the Requirements for the Degree of Master of Science (Biology) at Concordia University Montreal, Quebec, Canada January 2014 © Michael Gasse, 2014 CONCORDIA UNIVERSITY School of Graduate Studies This is to certify that the thesis prepared By: Michael Gasse Entitled: The Tannase Gene: Metaphylogenomics, Global Distribution and Presence in the Midgut Flora of the Forest Tent Caterpillar Malacosoma disstria Hübner. and submitted in partial fulfillment of the requirements for the degree of Master of Science (Biology) complies with the regulations of the University and meets the accepted standards with respect to originality and quality. Signed by the final Examining Committee: ___Selvadurai Dayanandan__________Chair ___Ian Ferguson____________________Examiner ___David Walsh____________________Examiner ___Dan McLaughlin_________________External Examiner ___Emma Despland_________________Supervisor Approved by ___Patrick Gulick______________________________ Chair of Department or Graduate Program Director ___January 8th ___2014 ___Joanne Locke_____________________________ Dean of Faculty ABSTRACT The Tannase Gene: Metaphylogenomics, Global Distribution and Presence in the Midgut Flora of the Forest Tent Caterpillar Malacosoma disstria Hübner Michael Gasse Tannase enzymes hydrolyze tannins, a class of plant polyphenolics that defend against herbivory. Prior to the 1980’s, most studies were focused on the tannase gene in fungi owing to the interest surrounding their industrial value. Since then a large number of bacterial tannase genes have also been discovered. I performed phylogenetic analysis on 110 fungal and bacterial tannase reference sequences in an effort to observe the relationships between fungal and bacterial tannase. The generated maximum likelihood tree shows eight strongly supported tannase clades, with a rift among fungal tannases, which either align with proteobacterial tannase or actinobacterial tannase. Metagenomes were used to assess the biogeographical distribution of the tannase clades, revealing that they may have environmental specificity. An unpublished observation made by the Despland lab of Concordia University suggested that forest tent caterpillars (Malacosoma disstria) from western Canada are unable to survive on tannin-rich sugar maple foliage (Acer saccharum), whereas populations in eastern Canada develop on sugar maple. In Chapter 2 I examine this observation, and explore the possibility of tannase-expressing secondary symbionts in eastern M. disstria. For the first time, the microbiome of M. disstria has been described using 16S rRNA gene sequencing and is shown to contain several genera known for expressing the tannase gene. iii Acknowledgements Firstly, I would like to thank my family, lab mates and supervisor Dr. Emma Despland for their support and guidance. Special thanks to my colleague Dr. Jessica Ethier and co- supervisor Dr. David Walsh who both acted in a supervisory capacity during Dr. Emma Despland’s maternity leave. Secondly, I would like to thank Dr. Ian Ferguson for heroically stepping in as a committee member at the absolute last minute. Lastly I would like to thank the Lagavulin distillery. iv Table of Contents Chapter 1: Metaphylogenomics and Global Distribution of the Tannase Gene 1.1 Introduction 1 1.2 Methods 6 1.2.1 Reference Phylogeny 6 1.2.2 Metagenomics 8 1.3 Results 9 1.3.1 The T/FE Reference Phylogeny 9 1.3.2 Functionally Characterized Tannase 13 1.3.3 Udatha et al. (2011) Feruloyl Esterase Superfamilies 13 1.3.4 Xenologs 17 1.3.5 Global Biogeography of the T/FE Superfamily 17 1.3.6 Phlorotannins 23 1.3.7 Conserved Domains 23 1.3.8 Conserved Cysteines in Tannase 26 1.4 Discussion 28 Chapter 2: Tannase Producing Symbionts in M. disstria 2.1 Introduction 39 2.2 Methods 49 2.2.1 Insect Rearing 2011 49 2.2.2 Insect Rearing 2012 50 2.2.3 Caterpillar Dissection 51 2.2.4 Culturing of Midgut Bacteria 51 2.2.5 DNA Extraction and Bacterial 16S rRNA Amplification 52 2.2.6 Sequencing 53 2.3 Results 54 2.3.1 Insect Rearing 2011 54 2.3.2 Abundance and Taxonomy of Culturable FTC Gut Bacteria 56 2.3.3 16S rRNA Sequencing of Midgut Communities 57 2.4 Discussion 59 References 69 v Chapter 1 - Metaphylogenomics and Global Distribution of the Tannase Gene 1.1 Introduction Tannins are important plant polyphenols and are the fourth most plentiful biochemical component of plants after cellulose, hemicellulose and lignin (Swain, 1965). As a polyphenol, tannins consist of multiple phenol structural units. They are located in leaves, bark and wood and defend against herbivores as well as microbial and fungal pathogens (Lekha and Lonsane, 1997; Aguilar and Gutierrez-Sanchez, 2001). Tannins produce a bitter taste when ingested and in many cases have toxic effects, however the exact mechanism for toxicity depends on the herbivore. In vertebrate herbivores, the toxicity is based on the ability of tannins to bind to useful macromolecules like proteins, rendering them indigestible (Aguilar et al., 2007). For Lepidopteran herbivores, toxicity may instead be caused by the alkaline pH of the gut oxidizing tannins to form semiquinone radicals, quinines and other reactive oxygen species (Barbehenn and Constabel, 2011). In addition to tannins, another plant phenolic, ferulic acid, is also thought to be used in antiherbivore defense. It is found in cell walls and binds to useful polysaccharides rendering them unusable to herbivores (Bunzel et al., 2005). Tannins are divided into four major groups; gallotannins, ellagitannins, complex tannins and condensed tannins. Gallotannins are typically gallic, digallic and chebulic acids bound to glucose through ester bonds, ellagitannins are made of ellagic acid bound to glucosides and complex tannins are ellagitannins bound to a flavonoid. Condensed 1 tannins are constructed from solely flavonoids and are not easily hydrolysable (Ramirez- Coronel et al., 2004). Aguilar et al. (2007) described the basic structure of the four tannin groups (see Figure 1). The type of tannin used commercially is tannic acid. Tannic acid has as many as five digallates bound to glucose. It is used frequently as clarifying agents in alcoholic drinks and fruit juices (Lekha and Lonsane 1994; Belmares et al., 2004). Gallic acid, one of the breakdown products of tannase action, is a major compound in the synthesis of trimethroprim, an antibacterial drug (Sittig, 1988). Figure 1 - The four tannin groups, as illustrated by Aguilar et al. (2007). All but condensed tannins are hydrolysable by tannase, an enzyme belonging to the tannase/feruloyl esterase family (Aguilar, 2007). Tannase (E.C. 3.1.1.20) and feruloyl esterase (E.C. 3.1.1.73) belong to the same protein family (IPR011118 and Pfam PF07519), but have different functions. This enzyme family is expressed by certain bacteria and 2 fungi, many of which are plant pathogens. The enzymes hydrolyze the ester bonds of hydrolysable tannins and feruloyl-polysaccharides, releasing the bound macromolecules that were previously indigestible. Tannase cleaves ester bonds between gallate and macromolecules (like carbohydrates) or complexes with multiple gallates (digallate). Ferulic acid, a hydroxycinnamic acid, is produced by plants and binds polysaccharides, creating the complex feruloyl polysaccharide. These complexes bind to lignin monomers that are active in cell wall development and help plants resist attack by microorganisms (Ishii, 1997). Once feruloyl esterases cleave the ester bond allowing the release of these polysaccharides, leftover ferulic acid may act as an antioxidant by reacting with free radicals like reactive oxygen species (Kampa et al., 2004). Tannase/feruloyl esterase enzymes are expressed by a variety of aquatic as well as terrestrial microbes. Tannins inhibit microbial growth in aquatic environments through metal ion chelation (Hagerman, 2002) in addition to substrate deprivation. The tannase enzyme allows aquatic microorganisms to breakdown and absorb nutrients from dissolved organic matter more efficiently (Hagerman, 2002). At the sequence level, tannase and feruloyl esterase are currently indistinguishable, causing confusion when annotating the genes without functional characterization. The majority of tannase sequences on record are often given both names but while both have similar function, they act on different plant polyphenols. 3 Traditionally, tannase and other esterases that act on similar substrates have been grouped under feruloyl esterases. Crepin et al. (2004) proposed 4 sub-classes for feruloyl esterases, types A through D, based on sequence data and the various substrates the enzymes use. Of these, tannase was designated as feruloyl esterase sub-class C for sharing sequence similarity with Aspergillus niger and Talaromyces stipitatus esterases that use methyl esters MFA (methyl ferulate), MSA (methyl sinapinate), MpCA (Methyl p- coumarate) and MCA (methyl caffeate) as substrates. Unfortunately, class C is characterized by only six fungal sequences, the majority of which are from Aspergillus species. As well, tannase is lumped into class C even though type C tannases were not tested on methyl gallate, digallate or glucogallin, three signature substrates of tannase activity. Without bacterial representatives and without functional characterization using gallate esters, it is unlikely that class C enzymes are truly representative of tannases. This idea is supported by the fact that the fungal and bacterial tannase sequences made available since 2004 do not fit neatly into the class structure defined by Crepin et al. (2004). In our reference phylogeny, the Crepin et al. (2004) class structure breaks apart, with members of each class ending up in very different tannase clades. Following on the work by Crepin et al. (2004), Benoit et al. (2008) proposed 7 subfamilies rather than 4 sub-classes, but their analysis only included fungal sequences, creating a fungal bias for their particular classification method. Furthermore, their analysis only restructured the ABCD categories from Crepin at al. (2004) and did not account for new 4 sequences. The system proposed by Crepin et al. (2004) was replaced recently by the phylogenetic analysis performed by Udatha et al. (2011). Udatha et al. (2011) proposed 12 feruloyl esterase families including selected sequences described by Crepin et al. (2004) as feruloyl esterase types A through D. What they observed was a scattering of feruloyl esterases A through D across their 12 families. The naming system proposed by Crepin et al. (2004) had fallen apart with the use of new sequences calling for a new classification system. Udatha used machine learning of physicochemical properties and sequence derived descriptors. The 12 families proposed by Udatha et al. (2011) were based on the molecular weight of the protein, the number of amino acids in the sequence and the percentage of certain types of residues in the sequence (charged, aliphatic, aromatic, polar, non-polar). The downside is that these descriptors provide little information about the function of the protein or its evolutionary history. Proteins can share similar descriptors yet still be completely different, as the order of amino acids is very important for folding and function. As an example, it is not enough to know how many polar residues are in a sequence, what matters is precisely where those polar residues are. Creation of a family structure for a protein based purely on the length, weight and percent composition of certain amino acids is not sufficient to develop families that may share meaningful evolutionary information or function. 5