Appl Microbiol Biotechnol (2005) 69: 404–410 DOI 10.1007/s00253-005-1993-3 BIOTECHNOLOGICAL PRODUCTS AND PROCESS ENGINEERING Balázs Bálint . Zoltán Bagi . András Tóth . Gábor Rákhely . Katalin Perei . Kornél L. Kovács Utilization of keratin-containing biowaste to produce biohydrogen Received: 23 December 2004 / Revised: 1 April 2005 / Accepted: 4 April 2005 / Published online: 22 April 2005 # Springer-Verlag 2005 Abstract A two-stage fermentation system was constructed to test and demonstrate the feasibility of biohydrogen gen- eration from keratin-rich biowaste. We isolated a novel aerobic Bacillus strain (Bacillus licheniformis KK1) that displays outstanding keratinolytic activity. The isolated strain was employed to convert keratin-containing biowaste into a fermentation product that is rich in amino acids and peptides. The process was optimized for the second fer- mentation step, in which the product of keratin fermentation —supplemented with essential minerals—was metabolized by Thermococcus litoralis, an anaerobic hyperthermophil- ic archaeon. T. litoralis grew on the keratin hydrolysate and produced hydrogen gas as a physiological fermentation byproduct. Hyperthermophilic cells utilized the keratin hy- drolysate in a similar way as their standard nutrient, i.e., bacto-peptone. The generalization of the findings to pro- tein-rich waste treatment and production of biohydrogen is discussed and possible means of further improvements are listed. Introduction Keratin is the insoluble structural protein of feather, wool, and animal hair. It resists biodegradation due to a large number of intramolecular disulfide bonds and a unique three-dimensional structure (Lynch et al. 1986). Keratin does not accumulate in nature as several bacteria (Lucas et al. 2003), e.g., Streptomyces pactum (Bockle et al. 1995), Bacillus licheniformis PWD-1 (Lin et al. 1992; Williams et al. 1990), Chryseobacterium sp. kr6 (Riffel et al. 2003), Streptomyces fradiae (Yu et al. 1969), Bacillus halodurans (Takami et al. 1992, 1999), the hyperthermophilic Fervi- dobacterium species (Friedrich and Antranikian 1996; Nam et al. 2002), and several dermatophyta fungi (Kunert 1973; Yu et al. 1972) produce keratinolytic proteases, which ap- pear ubiquitous, albeit slowly acting enzymes. However, in the poultry, wool, fish, and meat industry millions of tons of keratin waste is produced annually (Williams et al. 1991). Chemical treatment of the concentrated waste by reducing agents (Onifade et al. 1998) and various com- binations of physical and chemical degradation by heat and NaOH are commonly employed (Williams et al. 1990), although these techniques are hardly environmentally acceptable. Utilization of proteinaceous biowaste as animal feedstuff is not feasible when one takes food safety consid- erations into account. In principle, this material might serve as a valuable feed source for microbes, producing useful bioproducts such as biohydrogen. This possibility has not been explored yet. There is an increasing interest in the utilization of renew- able sources to satisfy the exponentially growing energy needs of mankind. Research on biological hydrogen pro- duction is propelled by the possible use of biohydrogen as the cleanest energy carrier and raw material (Benemann 1996). Biohydrogen is part of a broader concept of devel- oping zero-emission technologies employing production of H 2 from biomass in photobiological or heterotrophic fer- mentation routes (Cammack et al. 2001). Both processes depend on the supply of organic substrates and could be therefore ideally suited for coupling energy production with treatment of organic waste. In the present study, we report on a two-step dark fer- mentation process joining together the keratin degradation ability of a Bacillus strain isolated in our laboratory with the hydrogen-producing capabilities of an anaerobic ar- chaeon from the Thermococcales order. The Bacillus isolate has a remarkably sound keratinolytic activity, whereas the The principle, the strain and the results presented in the article together with the potential applications have been submitted for patenting to the Hungarian Patent Office (Ref. No. P0203998) B. Bálint . Z. Bagi . G. Rákhely . K. Perei . K. L. Kovács (*) Department of Biotechnology, University of Szeged, Temesvári krt. 62, 6726 Szeged, Hungary e-mail: kornel@brc.hu Tel.: +36-62-544351 Fax: +36-62-544352 B. Bálint . A. Tóth . G. Rákhely . K. L. Kovács Institute of Biophysics, Biological Research Center, Hungarian Academy of Sciences, Szeged, Hungary