In developing countries, poultry industry plays a vital role to meet the protein requirements for human nutrition. For this reason, the production efficiency of poultry to convert feed into meat is essential [1,2]. The problem is that quality poultry feed with reasonable prices are lacking and low production is the result in developing countries [3]. Maize and soybean meal are the major ingredients as energy and protein sources, respectively. But also human beings in developing countries consume different grains and soybean to cover their daily nutritional requirements [2]. As a consequence, to reduce the feed costs and to overcome the unavailability of quality poultry feed, some low quality agricultural by-products are used along with main ingredients [3]. Among the available sources, wheat bran (WB) and rice bran (RB) are the abundant and cheap agricultural by-products in rice and wheat producing countries [4] that are used as components in poultry feed [3]. Although these are inexpensive, high fiber content limits their use in poultry feed as the high fiber content reduces the digestibility of nutrients [5]. In addition, brans also contain some anti-nutritional substances e.g. β-glucans [6]. In RB and WB, the neutral detergent fiber (NDF) content varies from 33% to 40% and 48% to 51%, respectively [7]. β-Glucans adversely affects the utilization of other nutrients, especially protein and starch utilization and produce highly viscous conditions in the small intestine of the chicken [8] that contribute to delayed gastrointestinal absorption. On the other hand, the digestive system of poultry is not capable to produce enzymes to hydrolyze non-starch polysaccharides [5] present in brans. Therefore, the availability of nutrients is reduced if poultry consumes these fibrous feed and that depresses the production performance [4]. Different exogenous feed enzymes have been used in poultry diet to increase the digestibility of nutrients [9]. However, the use of specific enzyme is costly in one hand and on the other hand, the effectiveness of these enzymes depends on many factors. One of the most important factor is pH and the catalytic activity of different enzymes depends on the pH level. A varied range of pH in the digestive tract of poultry becomes the first physiological limitation for the activity and stability of exogenous enzymes [10]. Additionally, the feed spends relatively short time in the digestive tract of poultry which is another limiting factor for ineffectiveness of exogenous feed enzymes [10]. For these reasons, it is very important to reduce the fiber content of the brans by any processing method. In this regard, fermentation with rumen liquor (containing microbes) could be an effective method to overcome these problems. Rumen microbes can synthesize β-glucanases, cellulases, and hemicellulases the enzymes required for the breakdown of cellulose, hemicelluloses and phenolic polymers [11,12]. Fiber is degraded by a combination of ruminal bacteria, fungi and protozoa [11], where approximately 80% of this degradation performed by bacteria and fungi and 20% by protozoa [13]. To the author’s best knowledge, the use of specific fibrolytic enzymes or microbes are common methods of fermentation but limited data are available on fermentation with rumen liquor by maintaining appropriate environment for microbes to decrease the fiber content to produce high quality feed components for poultry.

First some experiments (single-step fermentation) were performed to find optimum fermentation conditions. Fermentation of WB was carried out in different composed mixtures under different environmental conditions. Fermentation was started with 48 h then reduced gradually (48, 36, 24, 12, 6, and 3 h) to 3 h due to the instability of the pH for fiber degrading bacteria. It was found that, optimum pH (6.5) was maintained up to 3 h then decreased to 5.1, 4.7, 4.2, 4.0, 3.9, and 3.8 after 6, 8, 12, 24, 36, and 48 hours, respectively. As optimum pH could not be maintained, further nutrient analyses were not performed. But few samples (from 3 h fermentation) when analyzed it was observed that, after 3 h the fiber content did not decrease satisfactorily then a method with two-step fermentation was performed.

It would be feasible to implement the developed method at the industrial scale production. The new fermentation technology will be of great benefit and would probably be provided to the rice/wheat miller for the production of fermented brans in a large scale to introduce a quality product to the feed market for economic poultry ration formulation. It might also be possible to use this technology to ferment other kinds of fibrous feed for feeding other non-ruminants. A designed fermenter would be useful for industrial scale production and a fermenter might be designed to facilitate all the conditions suitable for proper growth of rumen microorganisms. In developing countries, it is promising to collect rumen liquor as well as rumen content from different slaughtered cows and goats that would be a good sources of inoculum. It might be useful to reuse the effluent after fermentation as a inoculum source for large scale production. It might also be feasible to grow rumen bacteria in a laboratory that might be useable for large scale fermented feed production.

Considering all these factors, the present study (two-step fermentation) was undertaken to systematically evaluate and compare the effect of fermentation on nutritional improvement of WB and RB after fermentation with rumen liquor. The hypothesis is that, fermentation of different kinds of brans with rumen liquor will improve their nutritive value by decreasing the fiber content. Thus the aim of this present study was to develop an easy and cheap method for optimum fermentation of high fiber feedstuffs to improve their nutritive value due to lower fiber content.

In the present investigation, a two-step fermentation was performed after examining the results of some initial experiments (single-step fermentation) of this study that were conducted based on some in-vitro studies [22,23] such as rumen simulation technique (RUSITEC). However, RUSITEC systems are not useful for commercial feeding purpose or industrial use. A major problem to continue the single-step (initial experiments) fermentation for a longer period of time is the decrease of pH of the fermented mixture with increasing the time of fermentation. Due to these facts, the single-step fermentation was not continued. However, it seemed to be necessary to increase the fermentation time to increase the chances of microbe growth and with this the production of fiber degrading enzymes [11]. To overcome these problems, a two-step fermentation method was developed. It was assumed that most of the components that require less time to degrade, would be fermented during the 1st step of fermentation and that the amount of fiber would be reduced in the 2nd fermentation step. The 1st fermentation step was stopped when the pH decreased to a level of <6 that is known to be incompatible with the growth of fiber degrading bacteria.

The pH is one of the most important factors for ruminal microorganisms for the fermentation of fibrous components [24,25]. The appropriate range of rumen pH should be between 6 and 7 if cow maintain with proper ratio of grass and concentrate with good physiological conditions and a pH lower than that decreases the extent of fiber degradation process [26,27]. In the present study, the pH was reduced significantly after each step of fermentation and continued to decrease with increasing fermentation time. Also rumen pH (in live animal) changes constantly and depends on several factors e.g. the production of saliva, the generation and absorption of volatile fatty acids (VFA), the type and level of feed intake, and the exchange of buffer through the ruminal epithelium [28]. In the used in-vitro system most likely the pH was reduced due to the formation of VFA [27–29] and lactic acids [27,28] during fermentation and accumulation of these acids [30]. As a weak acid, these organic acids rapidly dissociate and release a proton and thereby decrease pH under most circumstances. As our fermentation system was completely closed, absorption or removal of VFA or other acidic components were not possible. For this reason, additional buffer substances NaHCO₃ were added to increase the pH and with this the fermentation capability of the system. As an appropriate pH (6 to 6.5) could maintained only during the 3 h fermentation, the additional buffer substance was added to control pH during the 6 h fermentation. NaHCO₃ is a potent buffer to prevent acidic conditions in the rumen or used in in-vitro system [31,24] and with this produce appropriate fermentation conditions. Although the procedure used for fermentation was the same in RB as in WB, different pH values from WB to RB were measured during the whole fermentation process. This might be due to the compositional variation of WB and RB. At the end of the 2nd step of 6 h fermentation, pH was not as much reduced as after the 1st step and remained near to 7. This may indicate that most of the components of brans that degraded quickly were fermented during the 1st fermentation step. In future, the question should be answered if a longer 2nd fermentation would be useful to improve the nutritional value of the brans.

In the described fermentation system, fiber was degraded more during 6 h compared to 3 h fermentation and the fiber was decreased more in the 2nd fermentation step. This could be expected as fermentation time is a very important factor for fiber degradation and Wizna et al [32] reported that the fermentation time had a greater influence than the inoculum dose. Longer fermentation time at ideal temperature and pH condition means a higher amount of microbes within the fermentation process. A too short fermentation time might be the reason for the slow degradation of fiber during the 1st step of fermentation and also in the 3 h fermentation compared to 2nd step and 6 h fermentation. Jazi et al [33] reported that fermentation of cotton seed meal with Bacillus subtilis, Aspergillus niger and A. oryzae for 7 days significantly (p<0.05) reduced CF (34.73%, from 12.58% to 8.21%). Also during fermentation of RB with Bacillus amyloliquefaciens and humic substances for 3 and 5 days, CF was reduced from 10.62% to 8.37% and 8.36%, respectively [4]. Conversely, Aanuoluwapo and Gbenga [34] found that CF increased from 6.25% to 6.89% when WB was fermented with rumen liquor where the mixture of WB and rumen liquor were put in a polythene bags and kept under the soil for 3 days. They did not mention anything about the pH and temperature of the fermented mixture that are very important determinants for cellulose degrading bacteria. In the present investigation, the appropriate environments (pH, temperature, anaerobic condition) were maintained during the whole fermentation period. In this study, NDF content was clearly reduced in the 2nd fermented WB after 6 h compared to fresh bran. The fiber reduction was higher in WB compared to RB. Probably, the reasons for these differences are the different compositions of WB and RB. RB contains less crude protein (CP) than WB. Previous studies reported that cellulolytic bacteria increased significantly (p<0.05) with increasing level of CP content in the diet whereby increased fiber degradability [35,36]. Also in beef cattle and dairy cattle in other studies [37,38], the NDF degradability tended to increase with increasing dietary CP levels. Another reason could be the higher crude fat content in RB compared to WB which ranges from 4.07% to 19.31% depending on the sources of collection [6]. This might be another reason for slower degradation of fiber in RB compared to WB. A high fat content coats the dietary fiber and therefore interferes with the fiber degradation [39,40]. Additionally, RB has more lignification than WB. Due to the presence of strong covalent bonds between lignin and the cell wall polysaccharides, it reduces the accessibility for the enzymatic hydrolysis secreted by rumen microbes [41] and thereby reduces fiber degradation. In the present study, the lignin content was not changed significantly by fermentation because of rumen microbes can not synthesize lignin degrading enzymes. This is in agreement with van Soest [41]. However, it was observed that the ADL content of fermented bran was higher compared to fresh bran after both 1st and 2nd fermentation step. The reason for this could be a relative increase of lignin as other fiber components were degraded during fermentation and with this the ADL content that not degraded by fermentation, increased relatively [41].

The results obtained in this study suggest that fermentation of WB and RB with rumen liquor improved their nutritional value by decreasing the fiber content. Fiber degradation was strongly influenced by the time of fermentation and a 2nd fermentation step had a significant effect on fiber reduction compared to the 1st fermentation step. Further the procedure of fermentation should be adjusted according to the fermented feed-stuffs. In all in-vitro fermentation of feed stuffs to improve their nutritive value seems to be an easy but effective method. However, in future, the method should be optimized by a longer fermentation time and the improvement of nutritive value, especially the changes of amino acid (AA) profile should be investigated. Additionally, the method should be adapted for an industrial use for the improvement of the nutritional quality of different kinds of fibrous feed to be able to produce meat at affordable price. This would help to provide humans in the third world with AAs in a sufficient amount.