The addition of yeasts to the diet of ruminant animals exerts favorable effects on the microbial population and on the fermentative indicators of the rumen and as consequence; it improves health and animals productivity (Bruno et al. 2009 and Al Ibrahim et al. 2012).

The Animal Science Institute (ICA) from the Republic of Cuba has a collection of yeast isolated from the ruminal ecosystem, which do not belong to Saccharomyces genus, showed potential as ruminal additive (Marrero et al. 2013). It was demonstrated that Levica 27 from the above collection and belong to Pichia guillermondii specie stimulated , principally, gas production in vitro from fibrous substrates compared with the rest of the strains of the collection and with S.cerevisiae (Marrero et al. 2014).

These results justify the search of an appropriate culture medium, economically feasible, for the obtaining of an additive product to pilot scale, that allow to show the activator effect of the strain on animals in production.

It is known that the design of the culture medium is one of the most important tasks inside the biological technology. According to Winkler (1998), in the total cost of the biotechnological products, raw matters can represent between 30 and 80 %.The composition of the culture medium it also has to satisfy all the nutritional requirements from the microorganism. In this sense, yeasts are able to survive under adverse conditions, behaving as facultative anaerobes. They are also easy to cultivate whether in the laboratory as to industrial scale, using a culture medium that includes sugars, minerals salts, and a small quantity of nitrogen source (Alvarez 1995).

The objective of this study was to evaluate the growth of Pichia guilliermondii strain Levica 27 in different energy sources and nitrogen, for a later design of a culture medium.

Biological material and culture media. The strain Levica 27, belonging to Pichia guilliermondii specie, located in Gen Bank, with number JF894143 was used. This strain belongs to yeasts collection from ICA (RYCASI), with registration number 980 in the World Data Center for Microorganisms (WDCM).

The NRF medium is used for ruminal yeasts culture therefore was eliminated from its composition the cellobiose, soluble starch, sodium lactate, hemin, volatile fatty acids, HCL-cysteine, Na₂S.9H₂O and Na₂ HCO₃.The composition of the medium used is showed in table 1.

The mineral solution A contains, per 1,000 mL, K2HPO₄, 3.0 g. The mineral solution B contains, per 1,000 mL: KH₂PO₄, 3.0 g; SO₄ (NH4)₂, 6.0 g; NaCl, 6.0 g; MgSO₄.7H₂O, 0.6 g and CaCl2, 0.6 g.

Preparation of yeast inoculum. A yeast culture obtained at 24h was used, from which a lop was taken, and was diluted in 10mL of malt solution extractmedium . It was incubated at 30 ºC during 24 h for their reactivation.Then, from this culture, 0.5mL was inoculated in an erlenmeyer of 100mL with 50 mL of the same medium and was incubated under the same conditions. This constituted the inoculums for the studies and shows a concentration of 57.1 x 10^-7 cel.mL^-1 of live cells.

Experimental procedure. The growth of the yeast strain in the extract malt medium and the modified NRF solution was first compared. The first mean was used, specifically, for fungi and yeasts culture, but it not have in its formulation the minerals solutions of the second one. For this, 0.1mL of the inoculum above cited in test tubes that contained 10mL of both media was added. Four tubes per treatment for every study hour were used. Tubes constituted the experimental unit. At 0, 4, 8,12,24,36 and 48 h the pH and optical density (O.D) were determined. Cell counts were also made in Neubauer camera. For this, yeasts were tinted according to Painting and Kirsop (1990) method.

Later on the strain growth was studied in NRF medium when changing sugars and nitrogen sources, according to the study. In the first case, glucose was substituted by sucrose and lactose and for nitrogen sources, the ammonium sulfate found in the B mineral solution was removing to included tryptone (144 mg), casein (144 mg) and urea (24.4 mg).For this, same as in the curves study, 0.1mL of the inoculums was added in test tubes that contained 10 mL of both media. Four tubes per treatment for every study hour were used. Tubes constituted the experimental unit. At 0, 48 and 72 h the pH and O.D were determined.

The pH was measured with a portable potentiometer HANNATM., model HI 9017, ± 0.1 precise units. The O.D was measured by colorimetry, with an espectrophonometer HACH, model DR 5000, at 530 nm.

Statistical analysis. A completely randomized design was used, with four repetitions per treatment in every study hour. The experimental results were analyzed with the InfoStat statistical package, version 1.0 (Balzarini et al. 2001).In the microorganism counts, data were transformed according to LN, to assure variance normality. The Duncan (1955) test was applied, in necessary cases to distinguish differences between means.

When analyzing the curves that create the cell count of Levica 27 in malt extract and NRF media (figure 1) was proven the similarity between them. The statistical analyses showed that there was not interaction between treatments and the study time. There were not differences in the number of cell that grew in both media (table 2), showing that the NRF medium is appropriated for the culture strain in study and the evaluation of its growth before different carbohydrates and nitrogen sources. However, in table 3 are show the significance differences (P Â 0.001) that were found in the time, logically by the increase of cell numbers that shows the microorganisms, due to its reproduction, when they are in a culture medium with the necessary nutrients for this process.

The previous results, in both culture media, verified the existence of two from the four phases described by Mateos (2005) for the microorganism growths in general: lag phase and exponential phase or logarithmic.

The existence of the stationary and decline or death phases was not verified, since there was increase of cell counts up to the 48h. Apparently, the nutrients quantity in the medium was enough to achieve a yeast growth up to this hour.

The growth Lag phase of Levica 27 was extended up to the 8h. In the adaptation and recovery periods in which this phase is divided, first the cells are adapted to the new medium conditions, and could happen decrease in viable cells percentage. Later on begins to increase the cellular division speed.

The exponential growth phase is show from 8h up to the 48 h. To this regard, Mateos(2005) show that, in this period, nutrients are used quickly and the number of cells is increased. As consequence, final metabolites are accumulated and can bring changes in the pH. This statement is corroborated in table 4, which shows the pH and OD of Levica 27 culture in the study means. In both indicators, there were interaction between time and treatments. Decrease of the pH was observed, while OD increases.

It should highlight that yeast grew at high pH in the NRF medium (from 6.0 to 4.41) regarding to malt medium (from 4.22 to 4.54).To this regard, Carrillo (2003) explained that yeasts accept a pH range between 3 and 10, but prefers a lightly acid habitat, with pH from 4.5 to 6.5, corresponding with the obtained values in both cultures.

The microorganism growth can be verified by direct cell count method, and by indirect methods, as the turbidity measurement through O.D (Pandey et al. 2001).For this reason, studies of O.D lineal regression adjusted to cell count for media in study were carried out and the corresponding equations were stated:

Malt extract smedium

Cell count =3.21 x 10⁶ + 67.32 x 10⁶ (O.D)

Cell count= 2.84 x 10⁶ + 110.07 x 10⁶ (O.D)

The results showed high regression coefficients of R²= 0.90 in the malt extract medium and R²= 0.60 in NRF solution, with P Â 0.001in both cases.

This showed the possibility to use the O.D as indicator of Levica 27 growth in these culture media.

Microorganism growth dynamics depends on multiple factors. Besides the genus, specie and, specifically, the strain, variability is observed when different conditions, culture media and inoculum doses are used. This statement is confirmed by Tai Shin et al. (2002), Castillo (2009) and Angulo (2012), who showed heterogeneous results in the growth dynamics of Candida norvegensis, due to the use of culture medium and different inoculums doses. Everything seems to indicate that doses of inoculum inclusion were decisive in the differences among the three studies. The first two studies included 20 and 5% respectively, and the third one, 0.1%.

Although this study was carried out with a strain of the P. guilliermondii specie, the inoculum dosis was the same used by Angulo (2012). Probably the amount of inoculum was very low, which caused that Lag growth phase were extended up to the 8 h.

Results showed the possibility to study the growth of Pichia guilliermondii strain Levica 27, in NRF medium with different energy sources and nitrogen. Figure 2 shows the OD values of the strain culture in NRF medium for the three carbohydrates used as energy sources. All sugars were used by the yeast, but the best growths were reached up to the 48 h, when glucose and sucrose were used, although with sucrose at 48h the same O.D values were reached as with glucose at 72h.This should be taken into account due to the lower time used by the strain for growing with this carbohydrate.

Similar studies, in those the NRF medium was used to determine the strain requirements from the same genus, showed that this one mostly used glucose as energy source, when was compared with sucrose and lactose (Angulo 2012).This shows the specificity that exits in yeast species, as for the use of carbohydrates sources.

The previous has been confirmed by Kurtzman and Fell (1998), whose describe the yeast capacity to use a wide range of compounds, among them carbohydrates that lend them a significant value to be use in the industry and agriculture.

In the media that contained glucose and sucrose had pH decrease (table 5) at 48h, which extended with sucrose medium at 72h.However, this one stayed in the optimal range described for yeast growth(Carrillo, 2003) and the described result is confirmed in table 4.

Contrary to previous researches, with yeasts from the same genus, but from the I. orientalis specie, 92% of survival to PH 3 was maintained (Tai Shin et al. 2002).From there, the importance of the specific studies in each strain, when is intended to use yeasts in ruminant animals feeding. The hardies to low pH could be more effective in diets with concentrates, in those the presence of carbohydrates of easy fermentation decrease the rumen pH.

Calderón et al. (2005) suggested that yeasts can get nitrogen to synthesize unicellular protein and the common sources are ammonium, urea and amino acids.This was evidenced in this study, with Levica 27 growing in different nitrogen sources (figure 3).It can verify that O.D values, when using casein, urea and triptane, they duplicate those that are obtained in ammonium sulfate presence. Regarding, Villamil and Zapata (1999) mention that all yeasts can use nitrogen in the weight ammonium ion, either in chlorides, nitrates, sulphates, ammonium phosphates, and can use the sulfur from the ammonium sulphate for the synthesis of some amino acids(Sarmiento and Herrera 2003).

If the results of figure 2 and 3 are analyzed it could recommend different combinations of carbohydrates and nitrogen sources taking into account their market prices.

The presence of all nitrogen sources caused gradual decrease of the pH, at 48 and 72 h (table 6), and the urea maintained values over the optimal for yeasts. Although the growth was not affected, this aspect should be kept in mind if it chosen the urea as source in culture medium to the studied strain. Nevertheless, Sánchez et al. (2007) designed a mean of simple culture for P. guilliermondii, without yeast extract, that contained molasses as carbon source and urea as nitrogen source. These authors obtained high biomass levels in submerged cultures as in solid state fermentation.

It is concluded that the growth of Pichia guilliermondii strain Levica 27 was higher in presence of glucose or sucrose and casein or urea, when was compared with other energy and nitrogen sources, respectively. It is recommended to carry out studies to design an economical culture medium with energy and nitrogen national sources that allow the obtaining of a preparation with Levica 27 to make in vivo studies and check the additive strain effect.

Al Ibrahim R. M., Gath V. P., Campion D. P., McCarney C., Duffy P. & Mulligan F. J. 2012. ``The effect of abrupt or gradual introduction to pasture after calving and supplementation with Saccharomyces cerevisiae (Strain 1026) on ruminal pH and fermentation in early lactation dairy cows''. Animal feed science and technology, 178 (1), pp. 40-47.

Angulo C. 2012. Crecimiento y requerimientos nutricionales de la levadura candida norvegensis como activador ruminal. PhD Thesis, Facultad de Zootecnia y Ecología, Universidad Autónoma de Chihuahua, Chihuahua, México.

Angulo C., Ruiz-Barrera O., Castillo Y., Marrero Y., Elías A., Arzola C., Mancillas P. C. & Díaz D. 2013. ``Growth dynamics and metabolites produced in the fermentation of Candida norvegensis yeast''. Cuban Journal of Agricultural Science, 47 (2), Available: <http://www.lebas.com.mx/files/7.pdf>, [Accessed: November 7, 2014].

Balzarini M. G., Casanoves F., Di Rienzo J. A., González L. A. & Robledo C. W. 2001. InfoStat. version 2001, Córdoba, Argentina: Universidad de Córdoba.

Bruno R. G. S., Rutigliano H. M., Cerri R. L., Robinson P. H. & Santos J. E. P. 2009. ``Effect of feeding Saccharomyces cerevisiae on performance of dairy cows during summer heat stress''. Animal Feed Science and Technology, 150 (3), pp. 175-186.

Calderón J. O., Elías A. & Valdivié M. 2005. ``Dinámica de la fermentación en estado sólido de la camas de cascarilla de café en inicio de ponedoras inoculadas con vitafert''. REDVET. Revista Electrónica de Veterinaria, 6 (5), Available: <http://www.veterinaria.org/revistas/redvet/n050505/050521.pdf?>, [Accessed: November 7, 2014].

Caldwell D. R. & Bryant M. P. 1966. ``Medium without rumen fluid for nonselective enumeration and isolation of rumen bacteria''. Applied microbiology, 14 (5), pp. 794-801.

Carrillo L. 2003. ``Levaduras. Cultivo, identificación, ambiente y sinonimia''. In: Los hongos de los alimentos y forrajes, Universidad Nacional de Salta, ISBN: 987-9381-19-X, Available: <http://www.unsa.edu.ar/matbib/ hongos/09htextolevaduras.pdf>, [Accessed: July 12, 2010].

Casadesus L., Rojas T., Brizuela A. L. & Sánchez A. Y. 1985. Micología. La Habana, Cuba: Universidad de la Habana, Facultad de Biología.

Castillo Y. 2009. Fermentación in vitro para obtener la levadura Candida norvegensis en mezclas de alfalfa y paja de avena con bagazo de manzana fermentado y sus efectos sobre la actividad microbiana ruminal. PhD.Thesis, Facultad de Zootecnia y Ecología, Universidad Autónoma de Chihuahua, Chihuahua, México.

Duncan D. B. 1955. ``Multiple range and multiple F tests''. Biometrics, 11 (1), pp. 142.

Elias A. 1971. The rumen bacteria of animals fed on a high molasses-urea diet. PhD. Thesis, Aberdeen.

Kurtzman C. P. & Fell J. W. 1998. The yeast, a taxonomic study. 4th ed., ELSEVIER, 13-19 p.

Marrero Y., Burrola-Barraza M. E., Castillo Y., Basso L. C., Rosa C. A., Ruiz O., González-Rodríguez E. & others. 2012. ``Identification of Levica yeasts as a potential ruminal microbial additive''. Czech Journal of Animal Science, 58 (10), pp. 460-469.

Mateos P. F. 2005. Crecimiento microbiano. Departamento de Microbiología y genética, Universidad de Salamanca, Available: <http:/nostoc.usual.es>, [Accessed: April 1, 2005].

Nogal P. J. Á. 1995. ``Los probióticos como complemento alimenticio''. Mundo ganadero, (11), pp. 38-46.

Painting K. & Kirsop B. 1990. ``A quick method for estimating the percentage of viable cells in a yeast population, using methylene blue staining''. World Journal of Microbiology and Biotechnology, 6 (3), pp. 346-347.

Pandey A., Soccol C. R., Rodriguez-Leon J. A. & Nigam P. 2001. ``Factors that influence solid state fermentation''. In: Solid State Fermentation in Biotechnology: Fundamentals and Applications, Asiatech.

Sánchez M. I., Santos A., Dustet J. C., Guerra G., León T., Argüelles J., Ramos-Leal M., Manzano A. M., Casado G. & Gómez B. 2007. ``Estudio fisiológico de una cepa de levadura con potencialidades para el enriquecimiento proteico del bagazo de caña de azúcar''. Revista CENIC Ciencias Biológicas, 38 (1), Available: <http://revista.cnic.edu.cu/revistaCB/sites/default/files/articulos /CB-2007-1-039-043.pdf>, [Accessed: November 7, 2014].

Sarmiento A. & Herrera J. 2003. Obtención y caracterización de un banco de levaduras con potencial aplicación probiótica. Degree Thesis, Pontificia Universidad Javeriana, Bogotá, Colombia.

Shin H. T., Lim Y. B., Koh J. H., Kim J. Y., Baig S. Y. & Lee J. H. 2002. ``Growth of Issatchenkia orientalis in aerobic batch and fed-batch cultures''. Journal of Microbiology, 40 (1), pp. 82-85.

Villamil Y. & Zapata Y. 1999. Caracterización de levaduras fermentadoras aisladas de frutas en descomposición con potencial aplicación productora de etanol. Degree Thesis, Pontificia Universidad Javeriana, Bogotá .

Winkler M. A. 1988. ``Optimisation and time-profiling in fermentation processes''. Progress in industrial microbiology, 25, pp. 91-150.