Note: Descriptions are shown in the official language in which they were submitted.
CA 022~1014 1998-10-06
WO 97138122 PCT/US97/05799
TITLE OF THE INVENTION
METHOD FOR IMPROVING CULTURE MEDIUM FOR
RECOMB~NANT YEASTS
5 CROSS-REFERENCE TO RELATED APPLICATIONS
This is related to Merck Case 19033, U.S.S.N. 086,216,
filed July 1, 1993, now published as WO 95/01422.
STATEMENT REGARDING FEDERALLY-SPONSORED R&D
Not applicable.
REFERENCE TO MICROFICHE APPENDIX
Not applicable.
15 FIELD OF THE INVENTION
Not applicable.
BACKGROUND OF THE INVENTION
Production of compounds of pharmaceutical significance by
20 cultivation of recombinant yeasts is an expanding field of science and
commerce. Purified recombinant hepatitis B surface antigen (HBsAg) is
used as a vaccine for hepatitis B viral disease and i.s a well-known
example of a pharmaceutically-significant recombinant protein.
Recombinant HBsAg is produced by cultivation of yeast
25 cells in complex or chemically-defined (synthetic) culture media.
Generally, complex media contain crude sources of nitrogen such as
yeast extract and peptones. Although high yields of cells and crude
HBsAg are achieved in these complex culture media, overall
perforrnance is frequently variable, and sometimes unacceptably
30 inconsistent. Inconsistencies in fermentation performance adversely
affect downstream purification steps and may also increase costs for the
purified product.
CA 022~1014 1998-10-06
WO 97/38122 PCT/US97/05799
- 2 -
Regulated expression systems are commonly used for the
production of recombinant proteins. One type of regulated ,system
provides tight nutritional control of the production of heterologous
protein. This type of system maximizes biomass.production and
5 product stability while minimizing the adverse effects of heterologous
protein expression on the host cell, e.g., Zabriskie et al., Enzyme
Microhial Tec~hnol. ~:706-717 (1986).
For convenience, applicants employ a recombinant S.
c~erevisiae ~strain for the production of Recombivax HB(~) (a trademark
10 of Merck & Co. Inc.), which strain harbors a plasmid composed of the
coding se4uence for HBsAg linked to the glyceraldehyde-3-phosphate
dehydrogenase (GAP) promoter, as well as an origin of replication
from the yeast 2~ plasmid, and the LEU2 gene for selection in yeast
cells. The strain is an adenine auxotroph, i.e., re4uires adenine for
15 growth. Other adenine auxotrophs of yeast are typically used as
recombinant host~ for heterologous protein expression, for example
~strains bearing mutations at the ADE 1 or ADE 2 loci. See, e.g.,
Kniskern, P. et al. in Expression Systems for Pro~ esses for
RecomhinaMt DNA Product~ (Hatch et al., eds.) ACS Symposium
20 Series No.447 (ch.6) pp.65-75 (1991), and Schultz, L. et al. Gene 61,
123 (19~i7).
It would be desirable to identify the component(s) of
complex media that affect fermentation performance, especially yields.
Advantages of such di.scoveries would include a more reproducible
25 fermentation process and a more predictable purification process.
Yeast extracts are commonly used in the media for yeast
fermentations as the source for vitamins, trace elements and nitrogen
nutrient. In many fermentation processes the nutrient which becomes
limiting during the course of fermentation is the carbon source. The lot-
30 to-lot variation of yeast extract due to variations in vendor's
manufacturing processes dramatically affect recombinant yeast
fermentation productivity and consistency, e.g. Recombivax HB(E) (a
. . .
CA 022~1014 1998-10-06
WO 97138122 PCT/US97/05799
trademark of Merck & Co., Inc.) fermentation. The problem was
partially solved in the past by the "brute-force" fermentation screening
("use-te~st") of new yeast extract lots. As a re.sult, additional manpower
and facilities had to be tied up, and sometimes "good" lots could not be
5 secured due to delay in decision while other times "poor" lots were
purchased and had to be thrown away. This disadvantage can be
overcome by first identifying the critical and varying components in
yeast extract that affect Recombivax HB(~ fermentation, and establishing
rapid assay methods for these component,s. After a ,sufficiently
10 representative database is built, the analytical results can be used to
evaluate whether a particular yeast extract lot is desirable for
Recombivax HB(~) fermentation.
The invention relates to a method to rapidly determine
whether a yeast extract lot will be "good" for recombinant yeast
15 fermentations, including that which produces HBsAg (Recombivax
HB(~)), by measuring the contents of critical varying components such as
adenine, trehalose and lactic acid. This simple and rapid screening
procedure elimin~e,s lots with sub-optimal levels of these components
and allows in mo,st ca,ses (about gO% of lots) superior and consistent
20 fermentation productivity. The method also enable,s the improvement
of fermentation yield by rational supplementation of those components
to "poor" yeast extract lot,s.
Applicants have identified adenine and two metabolizable
carbon sources (trehalose and lactate) as critical components in yeast
25 extract causing fermentation inconsistency. Adenine is required for
growth while the slowly metabolized trehalose supplies energy after
growth phase for recombinant gene expre,ssion in the synthesis of
expression product. The rapidly utilized lactate exerts a positive effect
indirectly by sparing more ethanol as the carbon source for product
30 synthesis. These effect,s on growth and production are mutually-
dependent. A relatively high level of carbon sources (trehalose plus
lactate, > 4 g/42 g) and a mid level of adenine (0.06 ~ 0.1 g/42 g) are
CA 022~1014 1998-10-06
WO 97/38122 PCT/US97/05799
necessary characteristics of a good yeast extract lot for yeast cultivation
and crude HB,sAg production.
SUMMARY OF THE INVENTION
A method for improving the culture medium useful for the
cultivation of recombinant yeasts and the production of recombinant
proteins is provided. The medium is particularly useful for the
cultivation of recombinant strains of Saccharomyces cerevi~iae which
produce HBsAg.
BREF DESCRIPTION OF THE DRAWINGS
Not applicable.
DETAILED DESCRIPTION OF THE INVENTION
The present invention is related to a general ferrnentation
process for the production of recombinant proteins by yeast cells. The
process of the present invention is demonstrated with the production of
HBsAg by batch fermentation of strains of Saccharomyces c~erevisiae
transformed with a plasmid comprising the gene for HBsAg. As will be
appreciated by one of ordinary skill in the art, the process of the present
invention has a more general application to cultivation of other strains
of S. cerevisiae and the production of other recombinant products and is
not limited to HBsAg.
In general, yeast batch fermentation in complex medium is
either a growth-limited proce.ss or a carbon source-limited process,
depending on the adenine and trehalose/lactate contents of the YE (yeast
extract) lot used. The concentration of these critical components in YE
can vary dramatically due to variations in vendors' manufacturing
processes. These inconsistencies contribute to fluctuations in
ferrnentation performance, e.g., the amount of HBsAg produced. The
analytical tools for adenine, trehalose and lactate in YE have been
developed. Adenine content deterrnines biomass production while
CA 022~1014 1998-10-06
WO 97/38122 PCT/US97/05799
carbon source (trehalose plus lactate) content affects antigen (HBsAg)
product synthesis, and these two effects are related to each other. A
mid-level adenine (0.06 ~ 0.1 g/42 g YE) and a high level trehalose plu.s
lactate (> 4 g/42 g YE) are the necessary requirements for a good lot,
5 provided that the concentration of lactate does not exceed about 4.0 g/42
g YE. Concentration.s of lactate exceeding about 4.0 g/42 g YE will
cause significant change in fermentation pH profile. Many poor lot.s are
improved by rational supplementation of adenine or trehalose or lactate
or their combination.
ln this invention, there i.s provided a method for improving
culture medium with limiting carbon source for a recombinant yeast
prototroph, comprising the steps of:
a) providing a quantity of a given lot of yeast
extract to be tested;
b) measuring the concentrations of trehalose and
lactate;
c) adjusting the concentration of trehalose plus
lactate to more than or equal to about 4.0
g/42g of yeast extract, provided that the
concentration of lactate is les.s than or equal to
about 4.0 g/42 g yeast extract.
In one embodiment of this invention, there is provided a
method for improving culture medium with limiting carbon source for
a recombinant yeast prototroph, comprising the steps of:
a) providing a quantity of a given lot of yeast
extract to be tested;
b) measuring the concentrations of trehalo.se and
lactate;
c) adjusting the concentration of trehalo.se plus
lactate to between about 5.0 g/42 g of yeast
extract and about 8.0 g/42 g of yeast extract,
provided that the concentration of lactate i~s
CA 022~1014 1998-10-06
WO 97/38122 PCT/US97/05799
less than or equal to about 4.0 g/42 g yeast
extract.
This invention also provides a method of identifying bad
lots of yeast extract for fermentation with limiting carbon source for a
S recombinant yeast prototroph, compri.sing the steps of
a) providing a quantity of a given lot of yeast
extract to be tested;
b) measuring the concentrations of trehalose and lactate; and
c) identifying bad lots as those lots with sub-
optimal concentrations of trehalose or lactate.
In another embodiment of this invention, there is provided
a method for improving culture medium with limiting carbon source
for recombinant yeast adenine auxotrophs, comprising the steps of:
a) providing a quantity of a given lot of yeast
extract to be tested;
b) measuring the concentration of one or more ofadenine, trehalose and lactate;
c) adjusting the concentrations of adenine to
between about 0.06 to about 0.10 g/42g of
yeast extract, and of trehalose plus lactate to
more than or equal to about 4.0 g/42g of yeast
extract, provided that the concentration of
lactate is less than or equal to about 4.0 g/42 g
yeast extract.
In another embodiment of this invention, there is provided
a method for improving culture medium with limiting carbon source
for recombinant yeast adenine auxotrophs, comprising the steps of:
a) providing a quantity of a given lot of yeast
extract to be tested;
b) measuring the concentration of one or more of
adenine, trehalose and lactate;
CA 022~1014 1998-10-06
WO 97/38122 PCT/US97/05799
c) adju~sting the concentration.s of adenine to
between about 0.06 to about 0.10 g/42g of
yeast extract, and of trehalose plus lactate to
between about 5.0 g/42 g of yeast extract and
about P~.0 g/42 g of yea.st extract, provided that
the concentration of lactate is less than or equal
to about 4.0 g/42 g yea.st extract.
Another embodiment of this invention provides a method
of identifying bad lots of yeast extract for recombinant yeast adenine
10 auxotroph fermentation with limiting carbon source, compri.sing the
steps of
a) providing a quantity of a given lot of yeast
extract to be tested;
b) mea.suring the concentration of one or more of
lS adenine, trehalose and lactate; and
c) identifying bad lots as those lots with sub-
optimal concentrations of adenine, trehalose or
lactate, or combination thereof.
Another embodiment of this invention is a method for
20 improving culture medium with limiting carbon source for recombinant
yeast adenine auxotrophs for the synthe.sis of recombinant Hepatitis B
surface antigen, comprising the steps of:
a) providing a quantity of a given lot of yeast
extract to be tested;
b) measuring the concentration of one or more of
adenine, trehalose and lactate;
c) adjusting the concentrations of adenine to
between about 0.06 to about 0.1 g/42g of yeast
extract, and of trehalose plus lactate to more
than or equal to about 4.0 g/42g of yeast
extract, provided that the concentration of
CA 022~1014 1998-10-06
WO 97/38~22 PCTIUS97/OS799
lactate is less than or equal to about 4.0 g/42 g
yeast extract.
Another embodiment of this invention is a method for
improving culture medium with limiting carbon source for recombinant
S yeast adenine auxotrophs for the synthesis of recombinant Hepatitis B
surface antigen, comprising the steps of:
a) providing a 4uantity of a given lot of yeast
extract to be tested;
b) measuring the concentration of one or more of
adenine, trehalose and lactate;
c) adjusting the concentrations of adenine to
between about 0.06 to about 0.1 g/42g of yeast
extract, and of trehalose plus lactate to between
about 5.0 g/42 g of yeast extract and about ~S.0
g/42 g of yeast extract, provided that the
concentration of lactate is less than or equal to
about 4.0 g/42 g yeast extract.
Another embodiment of this invention is a method of
identifying bad lots of yeast extract for recombinant yeast adenine
auxotroph fermentation with limiting carbon source in the synthesis of
recombinant Hepatitis B surface antigen, comprising the steps of
a) providing a quantity of a given lot of yeast
extract to be tested;
b) measuring the concentration of adenine,
trehalose and lactate; and
c) identifying bad lots as those lots with
suboptimal concentrations of adenine, trehalose
or lactate, or combination thereof.
It is understood that the yeast adenine auxotrophs are
30 provided as illustrations of the techniques of identifying bad lots and
rational supplementation of yeast extracts. Other yeast auxotrophs~ as
well as yeast prototrophs provide suitable sources for yeast extract
CA 022~1014 1998-10-06
WO 97/38122 PCT/US97/05799
analytical screening and supplementation for the purpose of synthesizing
recombinant proteins.
In thi~s invention, one preferred sum of the trehalose plus
lactate content is more than or equal to about 4.0 g/42 g YE, provided
5 that the concentration of lactate does not exceed about 4.0 g/42 g YE.
This upper limit in lactate concentration avoids suboptimal yields from
high fermentation pH. The concentration of trehalo~se is in principle
unlimited, but at levels above about ~S.0 g trehalose/ 42 g YE, it is
typically not metabolized. At higher concentrations, no toxicity effect
10 of trehalose has been ob,served. It is preferable to have at least both
trehalose and lactate in the medium since they are providing an
additional carbon source at different stage,s of fermentation. There are
42g yeast extract (YE) per liter of the medium.
1~ Improvement of fermentation performance of "poor-~rowth" lots
It was observed that, in general, poor growth led to poor
volumetric HBsAg (i.e. antigen) yield; yet abundant growth frequently
also did not support good antigen production. Because the addition of
>0.2 g/L adenine boosted growth to the range of that obtained with a
20 "super-growth, poor-yield" lot, it was possible that the ample biomass
production might have depleted other nutrients/factors related to and
neces,sary for antigen synthesis. Therefore an adenine titration study
was carried out using a "poor-growth" lot, which supported low antigen
titer as expected. The results showed that while the growth increased
25 progressively as the adenine concentration increased (up to 0.2 g/L),
there was apparently an optimal level of adenine for antigen yield. In
this case, adding 0.1 g/L led to a 60% increase in titer. The on-line
respiration profiles of the cultures growing in another yeast extract lot
clearly demonstrated that the original medium was limited in adenine
30 and the addition of 0.04 g/L of adenine boosted growth dramatically. A
40% increase in biomass and 20% increase in antigen titer were
achieved compared to the control batch. The sharp drop of OUR
CA 022~1014 1998-10-06
WO 97/38122 PCT/US97/OS799
- 10 -
(Oxygen Uptake Rate) at ~32 hrs suggests that the higher growth
supported by the higher adenine concentration quickly depleted ethanol
(accumulated from glucose fermentation by the culture), a known
provider of energy source for antigen synthesi.s, resulting in a smaller
5 increase in antigen titer than biomass.
Enzymatic assay for adenine in YE
A method based on Naher (Methods of Enzymatic Analysis
4, 1909 (1974)) was developed, in which adenine is deaminated by
10 nitrous acid to hypoxanthine, and oxidized by xanthine oxidase rapidly
and qunatitatively to xanthine and further to uric acid measurable at 293
nm (see Examples). The conversion of adenine to uric acid during the
assay was complete and quantitative. Finally, no formation of uric acid
was observed when xanthine oxidase was omitted.
The adenine content measured for a YE lot was found to be
insensitive to heat-sterilization conditions, indicating that
adenine/growth relationship established at the 2-L shake-flask scale is
applicable to large scale.
20 Relationship between YE adenine content and fermentation performance
The biomass and antigen production of lots at the 2-L scale
was measured as they relate to adenine content. Good correlation wa.s
obtained between growth and adenine in that biomass incre~sed with
adenine until the measured content reached about 0.12 g/42 g yeast
25 extract(YE): after that the adenine level was no longer the limiting
factor for growth. But no direct relationship between adenine and
antigen yield existed except that most "good-yield" lots (> 3X mg HBsAg
/L) possessed a mid-level of adenine (0.06 ~ 0.10 g /42 g YE), although
some "poor-yield" lots were also found in this range. Thus, a mid-
30 range adenine content is a desirable but not sufficient condition foroptimal antigen (HBsAg) production.
CA 022~l0l4 l998-l0-06
WO 97/38122 PCT/US97/OS799
Identification of trehalose and lactate as metabolizable carbon .sources
in YE
Supplementation of adenine to some YE lots boosted
growth but decreased HBsAg specific production, and some "super-
growth" lots due to high adenine contents supported very poor antigen
yields. The likely explanation was that the abundant growth depleted
the energy source such as ethanol required for antigen synthesis. On the
other hand, a similar amount of ethanol should be produced from
glucose (which is constant in every fermentation), yet in many cases
growth or YE adenine content alone could not predict antigen yield of
the fermentation, and dra,stically different yields were obtained for lots
with very similar adenine or biomass level. A carbohydrate HPLC
analysis was employed to examine YE components in conjunction with
fermentation kinetic analysis. It was discovered that there were two
metabolizable components in essentially every YE lot, a major
disaccharide peak and a smaller "lactate" peak, and their levels varied
lot-to-lot. Since these two peaks decreased or disappeared after
fermentation, the corresponding compounds must have contributed to
the fermentation by serving as carbon/energy sources.
The disaccharide peak was assigned as trehalose, an isomer
of maltose, because treating the YE sample with a specific trehalase
resulted in reduction of this peak and the formation of a glucose peak.
As for the "lactate" component, incubation of the YE sample with L-
lactate 2-monooxygenase led to a decrease in the peak size and the
formation of an acetate peak. In order to confirm the structures of
these two components, their purification from YE was carried out by
hot ethanol extraction followed by preparative HPLC on an analytical
column. The purified compounds were identified as trehalose and
lactate by NMR studies.
Trehalo~se (oc-D-glucopyranosyl (x-D-glucopyanoside) is a
storage material synthesized by baker's yeast in response to
environmental stress. Trehalose content amounts up to 20% on dry cell
CA 022~1014 1998-10-06
WO 97/38122 PCTIUS97/05799
weight basis. Since vendors' cultivation and downstream processes
could not be ab~solutely consistent, trehalose content in various YE lots
was found by HPLC to range widely from < 1 to > 7 g/42 g YE~. As for
the lactate component, since baker's yeast does not accumulate this
~S metabolite, the minor amount detected (mostly < 3 g/42 g YE) is often
present due to lactobacillus contamination during the vendors'
manufacturing processes, a common phenomenon in the baker's yeast
industry.
The utilization of trehalose and lactate from YE during a
yeast fermentation at the 23-L scale was monitored. It was found that
lactate wa.s rapidly metabolized as carbon source for growth after
glucose utilization, which delayed the depletion of the accumulated
ethanol, a known energy source for antigen production. Broth pH
increased during lactate utilization and dropped back down thereafter.
Glycerol was accumulated but not re-utilized due to membrane
impermeability. Trehalose was catabolized slowly during and after the
oxidation of the accumulated ethanol, thus serving as carbon/energy
source for the later phase of the fermentation during which recombinant
product antigen (HBsAg) was being synthesized. Besides being an
energy source, another plausible function of trehalose is the stabilization
of cell membrane structure again~st environmental stress.
Relationship between the level of trehalose plus lactate and fermentation
performance
Various lots which had been evaluated in 2-L yeast
fermentations were analyzed for their trehalose and lactate contents.
The relation.ship between carbon source (trehalose plus lactate) contents
and bioma.ss gave no apparent correlation to relate growth and YE
carbon source content, as most fermentations were limited by adenine.
But there is a readily apparent trend that up to 6 g/42 g YE higher
carbon source content supported higher antigen titers. The majority of
the "good" lots (yielding > 3~ mg HBsAg/L) had ~ 4 g/42 g YE in
CA 022~1014 1998-10-06
WO 97/38122 PCT/US97/05799
- 13 -
carbon source and lots with less than this level were essentially all
"poor". However, not all the lots with respectable carbon source
contents were "good". About 80% of the "good" lots possess mid-level
adenine (0.06 ~ 0.1 g/42 g YE).
s
Effect of lactate supplementation on fermentation performance
There was a positive effect of lactate supplementation at
23-L scale to a YE lot containing high adenine (0.13 g/L) and low
carbon sources (2.7 g tre, 0.6 g lact/L). Since lactate metabolism was
10 found to increase pH, the pH was manually controlled to match the
control. Clearly, the presence of 4.5 g/L more lactate provided carbon
source for growth, thus sparing the ethanol. The resulting delay of
ethanol depletion (as reflected by CO2 Evolution Rate or CER) made
more energy source available for antigen synthesis and hence led to
15 higher HBsAg titer.
New mechani~sm of trehalose effect and improvement of poor lots by
rational ,supplementation
One known function of trehalose is the protection of
20 microbial membrane integrity against environmental stresses because of
its uni4ue characteristics in forming bonds with phosphodiester linkages
in phospholipids. In the yeast fermentation, however, the positive effect
of trehalose was often observed when trehalose was not intact, i.e.,
when it was split into glucose and catabolized. It appeared that trehalose
25 affected yeast ferrnentation through slowly supplying glucose for
growth and product synthesis. The later effect was major in that after
ethanol depletion at 24~36 hrs (depending on the lot) which led to the
cessation of exponential growth, trehalose became the sole
carbon/energy .source available for antigen synthesis, as the glycerol
30 produced from glucose could not be re-utilized, and the lactate brought
in by YE and Hy-soy had been depleted in earlier phase.
CA 022~1014 1998-10-06
WO 97/38122 PCT/USg7/05799
- 14 -
Based on such a new mechanism, a poor YE lot (high
adenine, and low trehalose plu.s lactate content) is improved by
providing additional trehalose. In one example, it was seen from the
control that without additional trehalose, antigen synthesis essentially
S stopped when ethanol had depleted (judged by OUR) and most of the
original trehalo.se was consumed at ~30 hrs. Addition of more glucose
at 0 hr resulted in accumulation of more ethanol (and more non-u.sable
glycerol) for growth, which .slightly delayed the depletion of ethanol,
and thus could only ,slightly increa.se antigen titer. When trehalose was
10 supplemented to the level of about ~ g/42 g YE, similar catabolic
profiles were observed, and trehalo,se utilization provided
carbon/energy during synthesis phase which led to more active cells (as
reflected by OUR profiles) and significantly higher antigen yield. It is
noteworthy that more trehalose did not delay ethanol depletion as seen
15 with more glucose, indicating different mechanisms and the importance
of the slowly-released carbon/energy source which ensured the
availability of energy for antigen synthe.sis.
The effect of trehalose supplementation to various low- to
mid-trehalose lots at 23-L fermentor scale indicated that most of them
20 were improved mainly through the increase in specific production,
while the biomass was increased only slightly compared to antigen titer
ln most cases the on-line OUR profiles showed the distinctive higher
respiratory activities at the synthe~sis phase compared to the respective
controls.
EXAMPLE 1
Culture Inoculum Development and Production Fermentation
The culture source for all the experiment~s was frozen seed
stocks, generated from frozen vials of Saecharamyces c~erevisiae 2150-
30 2-3 (pHBS56-GAP347/33).
The medium for all seed stages was 5x Leu~ containing 90
g/L dextrose. The production fermentation medium was Enhanced
CA 022~1014 1998-10-06
WO 97/38122 PCT/US97/05799
YEHD, comprised of 42 g/L yeast extract (YE), 35 g/L Hy-Soy peptone
and 17 g/L dextrose (sterilized separately), with the presterilization pH
adju~sted to 5Ø Polyalkylene glycol was added as antifoam at 0.5 ml/L
for shake-flask fermentation and 1 ml/L for ,stirred-tank fermentation.
5 Adenine, lactate or trehalose was added prior to sterilization, at the
.
concentratlons speclfied.
A frozen cell suspension (1.5 ml) was thawed at room
temperature and inoculated to a 250-mL Erlenmeyer flask containing 50
ml of medium. After 24-h incubation on a rotary shaker (220 rpm.
10 2~~C), twenty ml of the culture were transferred to a 2-L Erlenmeyer
flask containing 500 ml of medium, and cultivated for 24 h on a rotary
shaker at 1 ~0 rpm and 2~~C. The culture was used as the inoculum for
ferrnentation ~studies in the 2-L shake-flasks and in some 23-L tanks.
For other 23-L scale fermentations, a third seed stage was included
15 which was developed for 24 h in a 23-L tank cont~ining 15 liters of
medium, at 2~~C with an agitation of 600 rpm and aeration of 6 L/min.
For fermentation studie,s carried out at shake-flask scale,
the 2-L baffled flask containing 200 ml of Enhanced YEHD medium
was used. The flasks were inoculated with 4% (v/v) seed culture and
20 incubated at 2~$~C and l~S0 rpm on a rotary shaker for two days. For
23-L stirred-tank fermentations, an inoculum of 5% from the shake-
flask seed or P~% from the third stage seed was used. The tanks were
operated at 28~C with an agitation of 600 rpm, an aeration of 12 L/min,
and a back pressure of 0.6 bar. Respiratory activities ( Oxygen Uptake
25 Rate or OUR, and CO2 Evolution Rate or CER), dissolved oxygen and
pH were monitored on-line, while carbohydrates were monitored off-
line by HPLC.
EXAMPLE 2
30 Analysis
Growth was measured by optical density (OD) at 660 nm
on a spectrophotometer, or by dry cell weight (DCW). These two
CA 022~1014 1998-10-06
WO 97/38122 PCT/US97/05799
- 16 -
method.s gave essentially the same conclusions. Carbon source
compounds .such as glucose, trehalose~ lactate and ethanol were analyzed
by HPLC system. To profile antigen production, cell pellet.s of 50 OD
units were prepared from fermentation broth .samples taken at various
5 time points, washed once with PBS buffer and stored at -70~C till
breakage. The Iysate was prepared by vortexing the cells with glass
beads. The protein content in cell lysate.s wa.s analyzed by the
bicinchoninic acid method, and the HBsAg concentration was
deteImined by enzyme immunoa.ssay (EIA) using the commercially
10 available assay kit. All result,s were back-calculated and expressed a.
fermentation titers (mg/L).
The data was based on the assays carried out at the same
time and under the same conditions for the experimentals and the
respective controls to minimi7e variations from assay kits, standards,
15 and as~say conditions. Similarly, all the comparisons were based on the
same experiment to eliminate differences due to culture conditions.
When two or more measurements were carried out, average results
were used.
EXAMPLE 3
Measurement Of Adenine Content In Yeast Extracts
Adenine content in variou.s YE lots was determined by an
enzymatic assay developed based on Naher (Methods of Enzymatic
Analysis 4, 1909 (1974)) which involves adenine deamination by nitrous
acid and oxidation by xanthine oxidase to give uric acid measurable at
293 nm. The procedure is as follows:
1. Prepare 42 g/L YE sample by adding 24.5 ml of water and 0.2 ml of
2 N HCI to 1.05 g YE powder and mixing throughly to get clear
,solution (the lot giving turbid solution is not desirable). Al.so prepare
adenine standard solutions (0, 0.025, 0.05, 0.10, 0.20, 0.40 g/L) by
CA 022~1014 1998-10-06
WO 97/38122 PCT/US97/05799
diluting with water a 1.0 g/L, pH 2 ,stock solution (stable at 4~C for
month.s).
2. Mix throughly by vortexing 2.0 ml of the YE .sample or the adenine
5 standard with 0.9 ml of 20% (w/v) sodium nitrate and 0.1 ml of
undiluted .sulfuric acid in a 50-mL uncapped tube. lrnmediately put the
mixture into a 37~C water bath to incubate for 60 min with paper towel
covering the uncapped tube.
10 3. After taking out the tube add 1.0 ml of 20% (w/v) sodium hydroxide
solution and mix well to stop reaction. This mixture serves as the assay
solution in the following steps and is found stable at 4~C for at least a
month.
15 4. Saturate Tris buffer (0.1 M, pH ~S.0) with oxygen by sparging air to
the buffer. Add 3.0 ml of this buffer and 30 ,ul of the assay solution to a
~S-mL cuvette. Seal the cuvette with parafilm and invert to mix the
content, and immediately read the extinction (El) at 293 nm on a
spectrophotometer blanked with the standard containing 0 g/L adenine.
20 Two readings should be made for each measurement and the values
should not differ more than 0.002.
5. Add 10 ,ul of 1 :10-diluted xanthine oxidase suspension (15.61 U/ml,
diluted with 3.2 M ammonium sulfate) to the cuvette and seal the cuvette
with parafilm. Invert to mix the content, and read the extinction at 292
25 nm the same way as above on the same spectrophotometer immediately
and then every 5 min until a constant/maximal value (E2) is reached
(generally in less than 30 min).
6. Adenine concentration in a YE lot (g/42 g YE) is estimated from its
30 E value based on a standard curve generated from the authentic
adenine samples (0~ 0.025, 0.05, 0.10, 0.20, 0.40 g/L, treated the same
CA 022~1014 1998-10-06
WO 97/38122 PCTIUS97/05799
- lX -
way and at the same time as the YE samples). E is calculated according
to the following equation ("blank" has 0 g/L of adenine):
E = (E2 - E 1 ).sample - (E2 - E 1 )blank
s
EXAMPLE 4
Measurement Of Trehalose And Lactate
Trehalose and lactate contents in various yeast extract (YE) lots
were determined by HPLC method ulsing an ion-exchange column. The
10 procedure is as follow.s:
1. Prepare 42 g/L YE sample the same way as that for adenine analysis.
Dilute the sample (1:5) with 0.005 M sulfuric acid (mobile phase)
before filtering through a 0.45 11 membrane. Also prepare trehalose (as
15 dihydrate) standard solutions (0 ~ 2.0 g/L) and Na-lactate standard
solutions (0 ~ 1.0 g/L) with the mobile phase.
2. Generate the standard curve,s for trehalose and lactate on an HPLC
,system, and then analyze the YE sample. The equipment includes a
20 solvent delivery pump, an automatic sampler injector and a detector. A
20-~1 sample is injected into column containing a poly~styrene
divinylbenzene cation exchange resin (for organic acids and alcohols)
maintained at 60~C.The sample is eluted isocratically with 0.005 M
sulfuric acid at 0.7 ml/min, and monitored for refractive index (RI)
25 change. Sample peaks are identified and quantified by comparing with
tho,se of authentic compounds. Under these conditions, trehalose eluted
at ~7.3 min and lactate at ~12.6 min.
EXAMPLE ;~
30 Purification Of Trehalose and Lactate
Purification of trehalose and lactate from YE in order to
confirm the structures by NMR wa~s achieved through hot ethanol
CA 022~1014 1998-10-06
WO 97/38122 PCT/US97/05799
- 19 -
extraction followed by preparative HPLC on an analytical column. To
50 g of YE was added 200 ml of ethanol and the mixture was stirred for
30 min in an ~5~90~C water bath. The filtrate was allowed to cool at
room temperature and the resulted precipitate was collected. After
S washing with cold ethanol and dried with air, the precipitate was
dissolved in 2 ml of water. The preparation, estimated to be > 30% in
weight purity in terms of trehalose, was injected and eluted repeatedly
on the above analytical HPLC system for further purification (no prep
column was available). The pooled trehalose and lactate fractions were
10 dried by Iyophilization before NMR structure determination.
While the foregoing specification teaches the principles of
the present invention, with examples provided for the purpose of
illustration, it will be understood that the practice of the invention
15 emcompasses all of the usual variations, adaptations, or modifications,
as come within the scope of the following claims and its equivalents.
