FIELD OF INVENTION
The present invention relates to a fermentation process using a methylotrophic yeast. More specifically, the invention relates to a fermentation process of enhanced expression level due to exogenous addition of some nitrogenous nutrients and change of process parameters like pH during induction. The present invention also relates to a process for the production of proinsulin without use of pure oxygen As a result, it is a highly economic process. Also a method of removing colored pigment produced by Pichia pastoris extracellularly into medixim, which has impact in subsequent purification steps, has been described.
Several publications are referenced in this application. Full citation of these publications is found at the end of the specification. These publications relate to the state-of-the-art to which the invention pertains. However, there is no admission that any of these publications is indeed prior art.
BACKGROUND OF THE INVENTION & PRIOR ART
Recombinant DNA technology has enabled the expression of foreign (heterologous) protein in microbial and other host cells. A vector containing genetic material directing the host cell to produce a protein encoded by a portion of the heterologous DNA sequence is introduced into the host, and the transformant host cells can be fermented and subjected to conditions which facilitate the expression of the heterologus DNA, leading to the formation of large quantities of the desired protein.
The advantages in using a recombinantly produced proteins in lieu of isolation from a natural sources include the ready availability of raw materials, high expression levels, which is especially useful for proteins of low natural abundance and the ease with which a normally intracellular protein can be secreted into the fermentation medium facilitating the purification process.
However, the aforementioned benefits of r-DNA technology are also accompanied by several disadvantages, when E. coli is used as the host system. Posttranslational modification (i.e., glycosylation) may not be carried out in the expression system; proteolytic degradation of newly formed protein may result upon expression in host

cells; the formation of high molecular aggregates often referred to as " inclusion bodies" or "refractile bodies" which result from the inability of the expressed proteins to fold correctly in an unnatural cellular environment. The recombinant protein cannot be secreted into the culture media upon formation of inclusion bodies. These inclusion bodies contain protein in a stable non-native conformation and difficult to process to its native form and as a result, the yield is less.
There are other associated problems related to E. coli when it is used as expression system like endotoxin presence, plasmid instability, difficulty in reaching very high cell density, deleterious effect of acetate accumulation (greater than 5 g/L) on cell growth and recombinant protein expression.
Yeast is well suited to the expression of heterologous protein of pharmaceutical importance. An example of the importance of yeast in biotechnology is the production of insulin for the treatment of diabetes mellitus; approx. half of the world's need for insulin is met by production process that employ S. cerevisiae. An alternative yeast species has become important expression system for the production of recombinant heterologous proteins (Thomas Kj eldsen et al., 1999). Among these are methylotrophic yeast species like Pichia pastoris, Hansenulla polymorpha, P. methanolica and the budding yeast Kluyveromyces lactis (1-6). In particular, the yeast Pichia pastoris has become successful in the production of high level of a broad range of heterologous proteins (2, 3, 7, 8).
The advantages oi Pichia pastoris as expression system are: clone stability (the gene is integrated into host's genome), very high cell density, easy scale up, strong and tightly regulated promoter, limited or absent hyperglycosylation, absence of endotoxin and choice of intracellular or extracellular expression.
The peptide hormone insulin is essential for glucose homeostasis, and recombinant insulin is important in the treatment of diabetes mellitus. Insulin is normally synthesized as prepro-insulin in the pancreatic p-cells and is processed to the two chain 51 amino acids globular peptides hormone in the secretory pathway.
The insulin precursor (proinsulin) is expressed in Pichia pastoris in fermentation media containing different ingredients (synthetic) using glycerol as sole carbon

source. A number of different processes including recombinant as well as non-recombinant methods have been developed for the production of insulin (9)
Human insulin can also be made with recombinant microorganisms that produce intact proinsulin (insulin precursor) instead of the old method wherein A and B chains are produced separately. The proinsulin route is the current method of choice for insulin production (Kroeff et al., 1989, ref 10) and entails one sequence of fermentation rather than two sets of sequences (i.e. one for the A chain and one for the B-chain).
Watson et al. in 1983 (11) suggested that the ideal situation for recombinant protein would be to have large amounts of the foreign protein efficiently secreted into the medium by the bacteria. S. cerevisiae was used to secrete insulin as a single chain insulin precursor (Diers et al. 1991, ref 12). Another process for producing human proinsulin intracellularly in the yeast S, cerevisiae has been described (Tottrup and Carlsen, 1990, ref. 13). Using the yeast system in an optimized fed-batch fermentation, yield of the fusion protein of superoxide dismutase-human proinsulin (SOD-PI) were reported to be 1500 mg/L. In this fermentation, yields of the final product, human insulin have not been reported. US patent 4745047 (Beckage and Ingolia, 1988, ref 14) describes a process of aerobic culturing of yeast, followed by anaerobic and then back to aerobic conditions. This method is reported to result in a high expression of product, beta-galactosidase and proinsulin produced intracellularly in S. cerevisiae, but the yield was not reported.
Majority of references on expression of insulin precursor by fermentation are reported using E, coli as expression system. Darrin J. Cowley et al. (15) also reported the expression of proinsulin in bacteria. In this fermentation of E. coli culture using LB media, the proinsulin was expressed by inducing the culture with IPTG (Darrin J. Cowley et. al, 1997). The final yield of properly folded proinsulin was 1-2 mg/L of culture only. There are also many reports on S. cerevisiae as expression system but this expression system utilizes a constitutive promoter. The S. cerevisiae expression system use episomal vectors. The fermentation yield of secreted heterologous protein expressed in yeast is a result of not only gene transcription and mRNA level, but also the over all capacity of the secretory pathway (Thomas Kjeldsen et al., 1999, ref 16).

Another advantage oi Pichia pastoris seems to be a Kex2 endoprotease analogue that is significantly more efficient that the S. cerevisiae Kex2 endoprotease in processing the pro-leader/insulin precursor fusion protein (Thomas Kjeldsen et al., 1999). An important aspect of any cellular production system is its easy and efficient adaptation to large-scale fermentation conditions used for the production of recombinant proteins. Clearly, the potential of Pichia pastoris for easy up-scaling and high cell density is an attractive feature in the production of recombinant proteins.
T. Kjeldsen (2000, ref 17) uses yeast secretory system for extra cellular expression of insulin precursor. In this art, development of the insulin secretory expression system in S, cerevisiae and its subsequent optimization is described. Expression of proinsulin in S. cerevisiae does not result in efficient secretion of proinsulin. However, expression of a cDNA encoding a proinsulin like molecules with deletion of threonine B as a fusion protein with the S. cerevisiae alpha-factor prepo-peptide (leader), followed by replacement of himian proinsulin C-peptide with a small c-peptide (e.g., AKK) results in the efficient secretion of folded single chain proinsuln-like molecules to the culture supernatant. Fermentation of insulin precursor expression in S. cerevisiae by different leaders (secretory pathways) varied from 16.5 to 79.2 mg/L determined by RP-HPLC of the culture supernatant.
Y. Wang et al. (2001, ref 18) used Pichia pastoris for over-expression of human insulin precursor. In this high cell density fermentation using a simple culture medium composed mainly of salts and methanol, the expression level reached was 1.5 g/L.
Patent number US 5,395,922 and WO0149870 (issued to Novo Nordisk AS), both of which are herein specifically incorporated by reference, describe a novel insulin precursors and its analogues comprising of a connecting peptide (mini C-peptide) because of which improvement of fermentation yield in insulin precursor molecules and its analogues when expressed in a transformed microorganisms, in particular in yeast, Saccharomyces cerevisiae has been reported.
Patents CA2392844 (issued to Novo Nordisk AS, 2001), US2001041787 (issued to Svend, 2001) and US 2002137144 (issued to Svend, 2002) describe the expression of proinsulin in yeasts.

Patent No. US 5962267 describes a process for preparing human insulin, which comprises the inserting the DNA into a vector to construct an expression vector culturing the resulting transformant under a condition that allows the expression of the human proinsulin derivatives thereof
Patent CN 1099416 describes the invention, which adopts high-density fermentation of II328-GLP-I (7-36) gene engineering strain. By using this technology, the secretion of insulin from animal pancreas can be significantly increased.
There are three major methods used for the production of human insulin in microorganisms. Two involve E. coli with either the expression of a large fusion protein in the cytoplasm [Frank, et al. (1981) in peptides: proceedings of the 7* American peptide chemistry symposium, Rich & Gross, eds.]. Pierce chemical co., Rockford, IL pp 729-739) or use a signal peptide to enable secretion into the periplasmic space [Chan et.al. (1981) PNAS 78; 5401-5404]. A third method utilizes S. cerevisiae to secrete an insulin precursor into the medium [Thim et al (1986) PNAS 83: 6766-6770]. The prior art discloses a limited number of insulin precursors, which are expressed in either E. coli or S. cerevisiae^ vide US 5,962,267, WO 95/16708, EP 0055945, EP0163529, EP0347845 and EP 0741188.
Patent EP1211314 and CA 2357072 by Beta Lab SA (AR) describes the expression of human insulin precursor in Pichia pastoris extracellularly. These patents describe about the use of methylotrophic recombinant yeast for producing human insulin precursor, the strain comprising in its genome, a copy of a DNA construction and a second DNA construction, wherein the constructions are capably directing the expression of secretion of human insulin precursor of the formula B (1-30) - Yi - Y2 -A (1-21).
Patent US 5612198 describes the expression of insulin-like growth factor-1 in methylotrophic yeast, Pichia pastoris as secretory product wherein the protein is expressed at a lower pH between 2 and 5 during induction phase, but yield of fermentation is not reported.

REFERENCES
1. Weydemann, U., Keup, P., Piontek, M., Strasser, A. W., Schweden, J., Gellissen, G and Janowicz, Z. A (1995) Appl. Microbiol. Biotecnol. 44, 377-385
2. Romanos, M. (1995) Curr. Opin. Biotechnol. 6, 527-533
3. Faber, K. N., Harder, W., Ab, G. and Veenhuis, M. (1995) Yeast 11, 1331-1344
4. Hollenberg, C. P. and Gellissen, G. (1997) Curr. Opin. Biotechnol. 8, 554-560
5. Raymond, C. K., Bukowski, T. Holderman, S. D., Ching, A. F. T., Vanja, E. and Stamm, M. R. (1998) Yeast 14,11-23
6. Hollenberg,C.P. and Gellissen, G. (1997) Gene 190, 87-97
7. Cregg, J. M., Vedvick, T. S. and Raschke, W. C. (1993) Bio/technology 11, 905-910
8. Scorer, C. A., Clase, J. J., McCombie, W. R., Romanos, M. A. and Sreekrishna, K. (1994) Bio/Technology 12,181-184
9. Michael R. Ladisch and Karen L. Kohlmarm (1992) Biotecnol. Prog. 8, 469-478.
10. Kroeff, E. P., Owens, R. A., Campbell, E. L., Johnson, R.D. and Marks, H. I. (1989) J. Chromatgr., 461,45-61
11. Watson, J. D., Tooze, J., Kurtz, D. T. Scientific American Books, W. H. Freeman Co. NY, 1983; pp 231-235
12. Diers, I. V., Rasmussen, E., Larsen, P. H., Kjaersig, I. L. In Drug Biotechnology Regulations (Scientific Basis and Practices); Chiu, Y. H., Gueriguian, D. L., Eds., Marcel Dekker, Inc. New York, 1991, pp 167-177
13. Tottrup, H. v., Carlsen, S. A. (1990) Biotechnol. Bioeng., 35, 339-348
14. Beckage, C. A. Ingiloia, T. D. US Patent 4,745,057, May 17,1988
15. Darrin J. Cowley and Robert B. Mackin, (1997) FEBS Letters, 402, 124-130
16. Thomas Kjeldsen, Annette Frost Pettersson and Marten Hach (1999) Biotechnol. Appl. Biochem. 29, 79-89
17. T. Kleldsen, Appl. Microbiol. Biotechnol. (2000) 54, 277-286
18. Wang,Y., Liang Z-H, Zhang Y-S, Yao S-Y, Xu Y-G, Tang Y-H, Zhu S-Q, Cui D-F, Feng Y-M. (2001) Biotechnol. Bioeng. 73, 74-79
SUMMARY OF THE INVENTION
The present invention describes a novel fermentation process, which gives an increased production yield of insulin precursor, and a downstream process including removal of colored pigment from fermentation culture broth when insulin precursor is expressed in a transformed microorganism, in particular methylotrophic yeast. Such insulin precursor of least colour or colourless fermentation supernatant can be converted into insulin by suitable, well-known conversion steps.

The present invention describes the growth and induction of insulin producing recombinant yeast in three phases comprising of batch and fed-batch growth phases and methanol induction phase of fermentation process in a bioreactor (Fermenter) with all kinds of process controller / instrumentation like automatic pH, temperature, dissolved oxygen control, cascade control of dissolved oxygen involving air, oxygen and rpm of agitator for process monitoring. In addition to this, automatic control of feed to the fermenter can be used for different feeding strategies of carbon source as well as supplements like yeast extract, peptone, casamino acids and skimmed milk powder during induction phase for improving expression level of insulin precursor, up to approximately 15-30% as compared to control (without supplement, pH at 5.0 and sparging of pure oxygen during entire fermentation period). Prior to induction by methanol, pH is to be increased stepwise up to 7.0, which increases another approximately 15% of insulin precursor. The entire fermentation process can be carried out without any pure oxygen usage during growth as well as induction phase. The overall improvement in proinsulin yield is approx. 30-45% as compared to control process.
Yet another aspect of the present invention relates to usage of peptone and yeast extract in different strengths to enhance the expression level over 15-30% as compared to control process without peptone and yeast extract. In addition to this, casamino acids can also be used but nitrogenous sources comprising peptone and yeast extract has better improvement in the yield of insulin precursor.
A further aspect of the present invention relates to the removal of color produced during methanol induction phase of Pichia fermentation by using active charcoal. The intensity of pigment is so high that it binds to capturing matrix. As a result, capacity of the matrix for the insulin precursor is drastically reduced which increases the cost of downstream processing and may even affect the quality of finished insulin. The novelty of this invention is that active charcoal is added to the fermentation culture broth directly instead of adding the same to supernatant after centrifugation. The clarified supernatant is almost colorless which enhances the loading capacity of capturing matrix over 60% as compared to supematant without charcoal pretreatment.

OBJECTIVE OF INVENTION
1. The first objective of the present invention is to develop a cost-effective fermentation process with improved expression level of proinsulin using methylotrophic yeast, Pichia pastoris extracellularly by optimizing parameters such as pH, etc. during growth and induction.
2. The second objective is to improve the expression of the heterologous protein in particular human insulin, by a methylotrophic yeast such as Pichia pastoris using nitrogenous supplements during induction.
3. The third objective is to develop a highly efficient fermentation process without oxygen use for the production of heterologous protein, human proinsulin in Pichia pastoris.
4. The fourth objective is to remove coloured pigment produced by Pichia pastoris from direct fermentation culture broth using a cheap source of activated charcoal.
DESCRIPTION OF THE INVENTION
In the present invention, a methylotrophic recombinant yeast Pichia pastoris Mut+ (methanol utilizing fast) clone was used for the production of proinsulin into the fermentation media. Human proinsulin gene used for expression was constructed synthetically by annealing and ligation of oligonucleotides. The proinsulin gene was designed such that the protein coded by it could have an amino acid sequence identical to human insulin B chain (amino acid 1-29); a spacer with sequence Ala, Ala, Lys followed by human insulin A chain (amino acids 1-21). The synthetic proinsulin gene obtained as described above was prepared for expression in Pichia pastoris. For this, S. cerevisiae alpha-mating secretory signal coding sequence was amplified by PCR and ligated in front of human proinsulin gene. A sequence coding for a spacer peptide (EEAEAEAKR ) for efficient processing of alpha-mating secretory signal peptide was introduced between the signal sequence and the proinsulin gene.

The present invention uses the above-mentioned Pichia pastoris as expression system for production of proinsulin by submerged fermentation. The fermentation process describes the seed development, biomass generation (high cell density culture) using completely synthetic media where glycerol is used as the sole carbon source by batch and fed-batch strategy and thereafter induction of high cell density P. pastoris culture with methanol as the sole carbon source. The fermentation process was conducted in an in-situ sterilizable fermenter with all kinds of process controller/instrumentation to control pH, temperature, dissolved oxygen, agitation and methanol feed. The high cell density was achieved using fed-batch strategy with precise profiling of feed so that the entire growth phase and induction phase (expression of human proinsulin) doesn't need any supply of pure oxygen even at a very high cell density, 400 g/L wet weight, 112 g/L dry weight or 500 OD600 (Fig. 1 and Fig. 2). The enhancement of protein expression was achieved by using some nitrogenous supplements like yeast extract and peptone in different combinations and strengths during induction phase. In addition to these, the P. pastoris was grovm to high cell density at pH 5.0 which thereafter is increased and is maintained at pH 5.8 - 6.5 prior to and during induction of methanol responsive promoter (expression of proinsulin).
In the present invention, a change in pH boosts the proinsulin expression level and also there is an improvement in expression of proinsulin when nitrogenous ingredients such as yeast extract and peptone in combinations of specific strengths used during induction phase The mechanism that may contribute to the much higher accumulation of product in process is reduced proteolysis effect due to extracellular protease to the medium resulting in increased synthesis rate. The overall improvement of expression level of insulin precursor during induction phase of fermentation is approx. 45% as compared to a control fermentation process wherein induction was carried out without nitrogenous supplement and below pH 5.8 using pure oxygen along with air. The novelty of the process is such that the entire fermentation process is carried out without the use of pure oxygen during growth phase (batch and fed-batch) to high cell density and induction/production phase. The process is manipulated with respect to feeding strategy of glycerol during grov^h phase and methanol during induction phase in combination v^th back pressure in order to completely eliminate oxygen use.

Consequently, most established fermentation protocol/processes for the expression of recombinant proteins from either bacteria or yeast is the use of pure oxygen along with air to maintain the growth and expression of high cell density culture. Additionally, a result of the expression of recombinant proinsulin is the accumulation of high concentration of proteolytic enzymes in the media during the induction phase. The deleterious effect of proteolytic enzymes on degradation of expressed proteins into media has been well documented in the literature.
In the present invention, a novel fermentation process has been developed wherein combination of pH, use of peptone, yeast extract and casamino acid have been made to enhance the proinsulin expression without pure oxygen. The expression of proinsulin into the media is achieved with a level of 3.5% of the total dry cellular protein when 'no oxygen' is used, but the level of 3,42% of the total dry cellular protein is achieved when pure oxygen along with air is used during growth as well as induction phase. The specific product yields are 20 mg and 19.7 mg of proinsulin per gram of dry cell mass respectively in case of without oxygen and with oxygen fermentation processes (Fig.3). The volumetric productivities are 3.5 g and 3.8 g per liter of culture supernatant in the above two cases respectively (Fig. 4). In all cases, induction with methanol is carried out with nitrogenous supplements and at pH 6.3. On the contrary, the improved process and control process produce 3.5g and 2.4-2.5g per liter of culture supernatant with an overall improvement of 40 to 45% expression level. The specific productivities are 20 and 14.5 mg of proinsulin per gm of dry cell weight and 3.5% and 2.52% of the total dry cellular protein respectively in case of improved and control processes.
In another aspect, the present invention also relates to the removal of the colour produced in the Pichia fermentation medium during induction phase with methanol. The Pichia pastoris when it is grown in a suitable media produces lot of brown/brownish yellow/greenish yellow/green/bluish green pigment during growth phase as well as induction phase whether it is methanol utilization fast (Mut^ , either AOXl or AOXl and A0X2 are functional) or methanol utilization slow (Mut^ when A0X2 is only functional). Majority of the pigment/color is normally produced during the induction phase with methanol of Pichia fermentation whether the recombinant protein is expressed intra/extracellularly. In the present fermentation process when

Pichia pastoris expresses the proinsulin during induction with methanol into media, the colour of culture supernatant after removal of cell debris is intense. It also interferes in protein estimation by lowry's method. It binds to capturing matrix along with proinsulin. The intensity of pigment is so high that the pigment component in the fermentation supernatant reduces the loading capacity of capturing matrix towards proinsulin, which is supposed to be captured on the matrix. As a result, the economy of the process is not favorable and it increases the cost towards further downstream processing and quality of the finished insulin may not be satisfactory.
The novelty of this invention is the usage of active charcoal for colour removal from the fermentation culture broth directly. Normally, the charcoal is for colour and smell removal, but its use for removal of pigment produced by Pichia fermentation is not available in the literature. In the present invention, the novelty is that active charcoal is added to the fermentation culture broth directly instead of adding to the supematant after concentration. The clarified supematant is almost colorless which enhances the loading capacity of capturing matrix, over 50% as compared to supematant without charcoal treatment. The specific activity of proinsulin of supematant before and after active charcoal treatment is 0.219 gram and 0.358 gram per gram of Lowry protein respectively. The Lowry protein of treated supematant is reduced by 43% over untreated fermentation supematant whereas quantity of proinsulin is reduced by 8% (measured by RP-HPLC) after activated charcoal treatment. In order to remove the coloured pigment from the centrifuge supematant different depth filters have been tried but a large amount of target protein, proinsulin, was adsorbed on to the filter in the order of 15-45% protein loss (estimated by RP-HPLC). Subsequently, there is a blocking of depth filter at a little bit higher pressure as well as flow rate. With regard to cost involved in this unit operation, treatment using depth filter is much costlier than the charcoal treatment process and not at all economic as compared to price of insulin product. Scale-up with depth filter for removal of colour pigment and cell debris is also problematic as compared to charcoal treatment, where charcoal along with cell debris can be separated out using simple batch / continuous centrifugation step.

BRIEF DESCRIPTION OF THE FIGURES
Figures 1 shows the comparison of dry cell weight of different batches during induction phases (with oxygen, vdthout oxygen and control). The batch with oxygen is defined as the batch where batch is carried out at pH 6.3, sparging of oxygen and supplement of nitrogenous ingredients under back pressure. The batch without oxygen (improved fermentation process) is defined as the batch where batch is carried out at pH 6.3, without sparging of oxygen and supplement of nitrogenous ingredients under back pressure. This graph shows that there is no appreciable change in dry cell weight during induction phases which indicate higher specific productivity during induction phase (figure 3).
Figure 2 shows the fermentation time vs. dry cell wt., pH, glycerol and methanol feed rate, cumulative feed of glycerol and methanol, air saturation (%), back pressure and agitation speed of improved process (YEP addition, pH 6.3 during induction phase and without use of pure oxygen under back pressure).
Figure 3 shows the comparison of specific productivities of processes with pure oxygen, without pure oxygen and control. This graph shows that specific productivities in all cases increase with induction time and thereafter is reaches to plateau at 108-120 h of MIP. From this graph, it can also be concluded that improved process does not differ from other two processes.
Figure 4 shows the comparison of the volumetric productivities without (improved) and with pure oxygen [YEP addition, pH 6.3 during induction]. This graph depicts that there is not difference in volumetric productivities of proinsulin expression with and without oxygen when fermentation process is with YEP, pH 6.3 under back pressure, but there is an improvement of approximately 45% expression level over control batch.
Figure 5 depicts the estimation of pro-insulin by HPLC (BP method) and SDS-PAGE. In this figure, one standard HPLC (using ActrapidO and only culture supernatant (120 h post induction) is shown and SDS-PAGE of different samples at different time intervals of induction is also shown.

Figure 6 shows the removal of coloured pigment using active charcoal as adsorbant and their treatment conditions. 'A' depicts the treatment of culture supernatant (after cell separation by centrifiigation) with charcoal and approximately 50% colour is removed. 'B' indicates the colour removal of HP2MG elute (capturing matrix) and almost 80% colour is removed. 'C shows the same treatment of 'B' but duration of treatment is 2 hours and there is not much change of colour removal as compared to 'B'. 'D' shows the treatment of direct culture broth with different amounts of active charcoal (w/v) for 1 hour at room temperature. There is not much colour change between 3% and 9% (w/v). Later we find that 1% charcoal is optimum to remove almost entire coloured pigment (data not shown) and approximately 60-66% reduction in Lowry's protein with 5-10% loss in proinsulin.
EXAMPLES
A better imderstanding of the present invention and the enhancement of expression level of proinsulin by submerged fermentation will be had from the following examples, given by way of illustrations. In addition to these examples, examples will be given for the removal of colour pigments from fermentation culture broth directly.
Example 1
The inventive process for the fermentative production of heterologous protein, proinsulin in methylotrophic yeast, Pichia pastoris and an inventive process for improving the expression level by fermentation comprises the following stages:
L Seed development: Grown on MGY media up to 10-12 ODeoo at 30°C
temperature, 250 rpm for 20-22 hours. Seed was in mid-logarithmic stage.
IL Transfer of seed to the Fermentor aseptically at the rate of 5% (v/v)
in. Fermentation process consists of
a. Biomass generation
i. Batch glycerol growth phase (GB) ii. Glycerol fed-batch growth phase (GFB)
b. Starvation phase
c. Methanol induction phase (MIP)

IV. Harvesting of fermentation culture to a separate mixing vessel
V. Treatment of culture broth with activated charcoal
VI. Clarification of charcoal treated fermentation culture broth by batch or continuous centrifugation
VIL Collection of clarified colourless supernatant
EXAMPLE 2
BIOMASS GENERATION
Experimental results showed that high biomass generation could be accomplished by two growth phases comprising of glycerol batches (GB) and Glycerol fed-batch (GFB). In glycerol batch phase, fermentation media consisting of complete synthetic media along with trace elements and D-biotin was inoculated with seed developed in minimal glycerol buffered media (MGY) at a concentration of 5% (v/v). When glycerol is completely exhausted (22-24 h) from the medium, there is an indication of sudden increase in the DO spiking. The time needed to consume all the glycerol will vary with the amount of initial seed and its ODeoo- The biomass yield of 33-38 g dry cell weight/L and cellular yield coefficient is of 0.65 to 0.75 gram of dry cell weight per gram of glycerol are expected in this phase.
Once all the glycerol is consumed during the GB phase evidenced by sudden DO spiking, a programmed glycerol feed is initiated to increase the required biomass under carbon limiting condition. The programmed feeding strategy was conducted through a computer-controlled Fermentor in auto/profile mode. In order to keep the DO level near the critical dissolved oxygen concentration of Pichia pastoris, feeding profile of glycerol is adjusted accordingly in combination with back pressure, rpm and aeration rate. The profile of feed was set at a rate from 13.8 to 24 ml/h/L of initial fermentor batch volume. The feeding was carried out under 'auto' mode where glycerol feed rate is to be increased stepwise manually depending upon DO spiking within one minute or "profile" mode where the feed rate is linearly increased automatically depending upon the programme. The pH of the fermentation growth phase was controlled at 5.0 automatically via a pH control loop using 30% ammonia. The residual glycerol level in the fermentation broth could be maintained at very low

to zero level (carbon limiting condition). Sampling is performed in growth phase at 6 h intervals and analyzed for cell growth, OD600, NTU (Nephleometric Unit), wet and dry cell weight, pH and contamination.
Under these conditions, dry cell weight of up to 112 to 116 g/L was obtained and cellular yield coefficient was up to 0.41 gram of dry cell per gram of glycerol was achieved.
EXAMPLE 3
Starvation and methanol induction phase (expression ofproinsulin)
The level of expressed protein, proinsulin depends on the cell mass generated during the GFB phase. The length of the phase (example 2) can be varied to optimize protein yield. The specific productivity is to be calculated instead of volumetric productivity.
Once the required biomass is achieved in GFB (example 2), methanol induction phase (MIP) commences after glycerol is exhausted (starvation). During this phase, on gradual depletion of carbon source, DO shoots up to a range of 70-80% of air saturation indicating the exhaustion of residual carbon source. After this stage of fermentation the actively grown cells are subjected to methanol induction thereby diverting the process towards the expression of proinsulin by ftilly inducing the AOXl promoter on methanol Approximately 1 hour before the biomass reaches the required level in GFB, pH is increased in auto mode to 6.3 using ammonia.
It is very important to introduce methanol slowly to adapt the culture to grow on methanol. This addition of methanol can be done in auto mode (stepwise manually) or by on-line methanol sensor (feedback control, make: Raven Biotech Inc., Canada) based on a particular set point value (mV) equivalent to a particular residual methanol concentration (off-line GC). It is important to use DO spike to analyze the state of culture and toxic time point over the course of methanol induction to optimize protein expression. The following stages are to be followed.

1. Terminate glycerol feed, allow to exhaust residual glycerol (DO spiking) and initiate induction by starting a 100% methanol feed containing 12 ml each of PTMi (Pichia trace elements and D-biotin per liter of methanol. Set the feed rate to approx. 2.4 ml/h/L of initial GB medium in auto mode (incremented manually) or use on-line methanol sensor (based on set point milli volt).
2. The methanol feed rate profile is from 2.4 ml/h/L to 13 ml/h/L of GB medium based on residual methanol, determined by GC (Off-line). The residual methanol was maintained at 0.05 to 0.2% (v/v). Instead of manual operation, it is better to maintain the residual methanol by using on-line methanol sensor and set point is incremented manually based on earlier fermentation batch.
3. If the DO reduces below a particular level stop the methanol feed, wait for the DO to spike and continue with the methanol feed rate (when operated manually). Increase the back pressure to a maximum of 2 bar to maintain the required DO. The novelty of the process is that a highly aerobic organism like Pichia pastoris was grown and induced without pure oxygen.
4. When the culture is fully adapted to methanol utilization (5-6 hours) and is limited on methanol, it will have a steady reading and a fast DO spike time (generally under 1 minute). Maintain lower methanol feed rate under limited conditions for at least another 2-4 hours after adaptation before increasing the feed. The feed rate is then increased depending on the residual methanol concentration to be maintained (< 0.2% v/v/) as measured by GC or by methanol sensor.
5. The entire methanol fed-batch phase lasts approx. 96-108 hours with a total of approx. 1.5 L methanol feed per litre of initial GB volume. Sampling is performed in this stage at every 6 h intervals and analyzed for cell growth (OD600, NTU, wet and dry cell weight) and expression level of proinsulin by HPLC and SDS-PAGE (Fig. 5).

EXAMPLE 4
Effect of pH and supplement in Example 3 to enhance protein expression.
The productivity of the fermentation could be increased substantially (approx. 45% over control) by addition of nitrogenous supplements like yeast extract and peptone continuously during induction phase. For further enhancement of expression level of proinsulin observed when the induction was carried out at 5.8 to 6.5 pH, preferably at pH 6.3 by 15-20% over the induction conducted at pH 5.0 and below. Overall improvement in expression level was approx. 52% as compared to controlled batch when induction was at pH 5.0 without yeast extract and peptone supplement. Casamino acid has also the ability to improve the expression level but combination of yeast extract and peptone has higher ability to enhance the expression level.
EXAMPLE 5
Color removal by charcoal from direct fermentation culture broth
Once the fermentation (induction) is over (plateau of expression level), the broth is taken in a vessel having a stirrer. The culture broth was treated with activated charcoal at different temperatures using different amount of charcoal for different time periods of incubation. The best result was observed when 3% (w/v) charcoal was treated at room temperature for 1 hour only. Before and after pretreatment of fermentation culture broth, the proinsulin in fermentation supernatant was measured by RP-HPLC and it was found to be 8% loss in proinsulin and 66% loss in total Lowry protein value which gives overall rise in specific activity of 54% over untreated fermentation supernatant. The centrifliged supernatant after charcoal treatment was loaded after adjusting the conductivity into column packed with capturing matrix.
Having thus described in detail preferred embodiments of the present invention, it is to be understood that the invention defined by the listed claims is not to be limited to particular details set forth in the above descriptions as many apparent variations thereof are possible without departing form the spirit or scope of the present invention.

CLAIMS
I / we claim:
1. A fermentation process of improved yield (expression level) of proinsulin (insulin precursor) from genetically engineered Pichia pastoris due to addition of nitrogenous sources in methanol induction phase.
2. The improved fermentation process of claim 1, wherein the nitrogenous source is yeast extract
3. The improved fermentation process of claim 1, wherein the nitrogenous source is peptone.
4. The improved fermentation process of claim 1, wherein a combination of both yeast extract and peptone is used.
5. The improved fermentation process of claim 4, wherein the strength of yeast extract and peptone (YEP) is varied from IX to 7X, where IX of yeast extract and peptone are 1 and 2% (w/v) respectively.
6. The process of claim 5, wherein the optimum strength of YEP is 5X
7. The process of claims 1 to 6, wherein yeast extract, peptone or a combination of both yeast extract and peptone is added continuously till the end of induction phase of fermentation.
8. The process of claim 7, wherein the mode of continuous addition of YEP is either maintained constant, or increased stepwise or exponentially.
9. The process of claim 6, wherein the volume of YEP ranges from 0.5 to 7.5% (v/v) of induction culture volume.
10. The process of claim 9, wherein optimum volume of YEP added to the fermentation is 3.3% (v/v) of induction culture volume.
11. The improved process of claim 1, wherein the nitrogenous source is casamino acids.
12. The process of claim 1, wherein the concentration of casamino acids ranges from 0.01% to 0.1% (w/v) of methanol used.
13. The process of claim 12, wherein the volume of casamino acid is from 0.5 to 7.5% (v/v) of induction culture broth.

14. A fermentation process for producing insulin precursor from genetically engineered Pichia pastoris wherein the pH during induction phase ranges from 5.8 to 7.0.
15. The process of claims 1-13, wherein the pH during methanol induction phase ranges from 5.8 to 7.0
16. The process of claims 14 and 15, wherein the pH is increased stepwise or by auto mode from 5.0 during growth phase of glycerol fed-batch to the desired pH of induction phase ranging from 5.8 to 7.0 prior to induction and maintained the same during the entire induction phase.
17. The process of claims 14 and 15, wherein the optimum pH during induction is 6.3
18. A fermentation process for the production of insulin precursor using a methylotrophic yeast comprising of growth phase (glycerol batch and glycerol fed-batch) and production phase (methanol induction phase), wherein the required dissolved oxygen for high cell density culture can be maintained without using pure oxygen by suitably adjusting the flow rate of carbon source during growth and induction phase and the pressure in the fermentor.
19. The fermentation process of claims 1-18 wherein the required dissolved oxygen for high cell density culture can be maintained without using pure oxygen (a) by adjusting the rate of glycerol feeding and the back pressure during growth phase, (b) by adjusting the rate of methanol feeding and the fermentor pressure during induction phase
20. The process of claims 18 and 19, wherein the rate of feeding of 50% glycerol in fed-batch mode is varied from 5.0-35.0 ml/h/L of initial batch volume, more preferably 10.0 to 30 ml/h/L of initial batch volume and most preferably 15.0 to 27.0 ml/h/L of initial batch volume.
21. The process of claims 18 and 19, wherein the rate of feeding of 100% methanol is varied from 1.1 to 20 ml/h/L, preferably 3.0 to 15.0 ml/L/h of initial batch volume during methanol induction phase.

22. The process of claim 21, wherein the methanol concentration in the culture
broth is maintained at <0.2% (v/v) as estimated by GC and On-line methanol
sensor (Raven Biotech, Inc., Canada).
23. The process of claims 18 and 19, wherein the back pressure is varied from 0-
2.0 kg/cm depending upon dissolved oxygen requirement of the process.
24. A process for the removal of a pigment from the culture broth wherein
fermentation culture broth resulted from the genetically engineered Pichia
pastoris containing proinsulin by treating with active charcoal
25. The process of claim 24 wherein the percentage of active charcoal ranges from 0.5 to 5.0% (w/v).
26. The process of claim 25, wherein the optimum charcoal percentage is 1% (w/v).
27. The process of claim 24, wherein the incubation temperature ranges is from 10
to 32° C.
28. The process of claims 24 to 27, wherein the duration of treatment is from 0.5 h
to 8h.