**1. Introduction**

Keratin is a fibrous and recalcitrant protein belonging to a large family of structural proteins that constitute hair, wool, feathers, nails, bristles, and horns of several animals [1]. This protein is densely packed in α-helix or β-sheet structures (α/β-keratin, respectively) and their high degree of cross-linkages by disulfide and hydrogen bonds confers high mechanical stability and resistance to common proteolytic enzymes such as pepsin, trypsin, and papain [2]. Its recalcitrant structure is a significant challenge for degradation, a process that naturally can take as long as many years [3]. Indeed, there are many known examples of preserved hair, skin, and feathers on archeological materials [4–6]. Its accumulation can be a major problem, especially for poultry processing farms, where feather keratin is a by-product of industrial activities. Therefore, finding microorganisms that present the ability to degrade this protein will become an eco-friendly alternative to improve the management of their waste resources. So far, several microorganisms have been reported

**Citation:** Valencia, R.; González, V.; Undabarrena, A.; Zamora-Leiva, L.; Ugalde, J.A.; Cámara, B. An Integrative Bioinformatic Analysis for Keratinase Detection in Marine-Derived *Streptomyces*. *Mar. Drugs* **2021**, *19*, 286. https://doi.org/ 10.3390/md19060286

Academic Editor: Max Crüsemann

Received: 14 April 2021 Accepted: 18 May 2021 Published: 21 May 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

to produce keratinases, including Fungi [7,8] and Bacteria, such as *Bacillus* [9–11] and *Streptomyces* [12–16].

Keratinases (EC 3.4.21/24/99.11) are a particular class of proteases that possess keratinolytic activity, being able to degrade several insoluble keratin substrates [3]. They are primarily extracellular; however, cell-bound [17] and intracellular [18] enzymes have also been reported. Most of the keratinases known to date are serine proteases [14,16,19], while a few have been classified as metalloproteases [20]. Nevertheless, it remains unclear which features are required for keratin degradation [21]. Most studies are focused on the purification and characterization of a unique keratinase or protease with keratinolytic activity [16,22–26]. However, purified keratinases known to date cannot completely solubilize native keratin, which confirms that one enzyme alone cannot fully decompose its recalcitrant structure [3,27]. For this reason, it has been hypothesized that a pool of proteases is needed to penetrate and break down keratin structure, instead of a single enzyme. Indeed, in natural environments, this degradation could result from the combined action of enzymes from various organisms, such as Fungi and Bacteria [3]. Moreover, due to the high cysteine content in keratin, and therefore, their respective disulfide bonds, it is believed that other enzymes, such as disulfide reductases, are likely to be involved in this process, acting in addition to keratinolytic proteases [28]. Although there are many publications describing keratinase characterization, the molecular mechanism of microbial keratinolysis is not yet completely elucidated.

Genomic analysis of keratinolytic bacteria is seldom reported [29–34]. *Streptomyces* species are one of the main producers of keratinases [35], but so far, there is only one recently reported genome of keratinolytic-degrading *Streptomyces* [36]. Moreover, according to our knowledge, there are no comparative studies between keratinolytic and non-keratinolytic bacteria, which may unveil the underlying genetic factors contributing to its degradative capacity. Recently, our Chilean marine actinobacterial culture collection [37–39] was screened for extracellular enzyme activities analyzing 75 strains [40], providing evidence of promising keratinolytic activity. To see whether we could bioinformatically predict the genes encoding putative keratinases, in this study, we sequenced the genome of three of these streptomycetes, selected for showing varying levels of keratin degradation: Strains G11C, CHD11, and Vc74B-19, with high, low, and no keratinolytic activity, respectively. We present an integrative comparative analysis between the universe of putative proteases belonging to these strains, complementing information from orthogroups of proteases, peptidase families, cell location prediction, and phylogeny, to finally deliver a bioinformatic prediction consisting of a set of genes considered to encode potential keratinases. Efforts addressing a thorough *in silico* investigation of putative proteases and peptidases, to obtain those likely to be involved in feather degradation, have been accomplished for the first time. Interestingly, 18 of 24 proteases predicted by this pipeline (6 of seven potential keratinases (predicted by phylogenetic analysis), and 12 of 17 unique genes of the G11C genome ("unassigned p-orthogroup" peptidases)) and thought to be involved in keratinolytic activity were validated in our previous study by being present in the secretome of strain G11C [40]. This study provides a not so far described bioinformatic multi-step pipeline, that helps decipher potential genetic factors that enable some strains to have higher efficiency in keratin degradation.
