Journal of IndustrialMicrobiology, 9 (1992) 69-72 Elsevier SIM00379 Short communication 69 Some factors influencing the proportion of periplasmic hepatitis B virus pre-S2 antigen in the recombinant yeast Hansenula polymorpha Marie-Ren6e de Roubin 1, Michael D. Cailas 2, Shi-Hsiang Shen I and Denis Groleau ~ National Research Council of Canada, Bioteehnology Research Institute, Montreal, Quebec, Canada and z Environmental and Occupational Health Sciences, University of Illinois, School of Public Health, Chicago, Illinois, U.S.A. (Received 18 April 1991; revision received 7 september 1991; accepted 9 September 1991) Key words: Secretion; Periplasmic pre-S2 antigen; Recombinant protein; Experimental design; Methylotrophic yeast SUMMARY A central composite design(CCD) was used to evaluate,for the purpose of future process optimization, the influence ofpH, yeast extract and ammonium chlorideconcentrations on the proportion ofperiplasmic hepatitis B pre-S2 antigen in the recombinantyeast Hansenulapolymorpha. Each factor was tested at five levels, and a second order polynomial model for the proportion of periplasmic antigen was fitted to the treatment combinations,pH showed the greatest effect: the proportion of periplasmic antigenwas greatlyincreased at the higherpH levels. At the higherpH levelsused, the proportion ofperiplasmic antigen was enhanced by a high concentration of ammonium chloride. Additional experimentshave confirmedboth the validity of the selected model and the optimal conditionsfound. A significant correlation was found betweenthe proportion ofperiplasmic antigen and the total yieldof antigen. These results indicated that it should be possible to modulate the distribution of the pre-S2 antigen between the periplasm and the cytoplasm of the yeast. INTRODUCTION The hepatitis B virus pre-S2 antigen (middle surface antigen), described by Blum et al. [1], has been recently produced in significant amounts by a recombinant strain of the methylotrophic yeast Hansenula polymorpha under control of the methanol oxidase promoter [5,7]. The antigen was shown to be present as 22-nm particles identical to those found in the blood of patients infected with hepatitis B virus [5,7]. The pre-S2 antigen contains internal signal sequences able to direct the antigen across the cytoplasmic membrane; however, due to their large size, the 22-nm particles remain trapped in the peri- plasmic space. The periplasmic protein, nevertheless, can be easily released by treating the cells with a mixture of lyric enzymes, which simplifies the recovery and the puri- fication of the antigen. Only a fraction of the total antigen produced is found in the periplasm [7]; the rest remains inside the cells. Correspondence: M.-R. de Roubin, National Research Council of Canada, Biotechnology Research Institute, 6100 Royalmount Avenue, Montreal, Quebec, Canada H4P 2R2. It has been observed that the proportion of peri- plasmic pre-S2 antigen produced by H. polymorpha varied to a certain extent with the physico-chemical conditions used for cultivation and expression. Many factors may be expected to influence the accumulation of a periplasmic protein, and these factors may also interact. A second order response surface technique was employed to effi- ciently determine the relationship between the proportion of periplasmic pre-S2 antigen (response variable) and three pre-selected experimental factors presumed to affect the response. MATERIALS AND METHODS Strain, media and growth conditions The recombinant Hansenula polymorpha strain, the medium and culture conditions used were the same as described previously [5], except for the following modifi- cations: no phosphoric acid was included; pH was adjust- ed to 4 with KOH and then to the desired value with NaOH; the levels of the three factors chosen for optimi- zation are given in Table 1. 0169-4146/92/$03.50 9 1992 Society for Industrial Microbiology Downloaded from https://academic.oup.com/jimb/article/9/1/69/5987632 by guest on 22 December 2023