"METHOD OF PRODUCING PNEUMOCOCCAL CAPSULAR POLYSACCHARIDE AND USE THEREOF"

Anae
(Bec
empl bacte cond meni CA nasopharyngeal swab sample was collected from patients in whom a presumptive clinical diagnosis of pneumococcal meningitis had been made. This sample was plated immediately onto both 5% sheep's blood agar and tryptic soy agar containing 5000 units of catalase, and the plates were incubated at 37°C in an air atmosphere containing 5% CO2. Cerebrospinal fluid (CSF) samples were also plated on both blood agar and transparent plates, as were cultured patient blood samples in which pneumococci were observed following standard aerobic incubation for up to 48 hours. Initially, isolates which exhibited typical morphology on blood agar were confirmed as being pneumococci by Gram stain and Quellung reaction. All isolates were stored at-80°C in bacterial preservers, and colony morphology was determined as described herein.
Statistical differences between rates of T and O phenotypes in nasopharyngeal and invasive site cultures were examined using Fisher's Exact Test.
Genetic Transformation Encapsulated and non-encapsulated pneumococci were made competent for natural transformation as described (Weiser et al., 1999). Mutations in two- component signal transduction systems (TCSTSs) were transformed into strain P303 by selection of colonies from medium comprising erythromycin (1 microgram per milliliter). Encapsulated transformants were screened for altered colony morphology, and expression of CPS confirmed by the Quellung reaction using type-specific antisera (obtained from Statens Seruminstitut, Copenhagen, Denmark).
Quellung Reaction Pneumococcal capsules were visualized in bacteria grown to mid-log phase in semi-synthetic medium under various conditions. For the Quellung reaction, an equal volume of culture and a composition comprising type 6 anti-serum and 1 % (w/v) methylene blue obtained from the Statens Seruminstitut (Copenhagen, Denmark) were mixed on a glass slide for one minute, as described (Neufeld, 1902).
Ouantitation of CPS O-and T-form pneumococcal variants were grown in the semi-synthetic medium to mid-log phase (Ab = 0. 3) or for 16 hours (at which time the cells were in the stationary phase). Cells were harvested by centrifugation at 2000 x g, washed in phosphate-buffered saline (PBS), sonicated for three 10-second intervals at 0°C, and stored at-20°C. A capture ELISA technique was used to determine quantities of CPS present in variants grown under selected conditions, as described (Kim et al., 1998).
Type-specific rabbit anti-serum (obtained from Statens Seruminstitut, Copenhagen, Denmark) was used at a dilution of 1 : 5, 000 in 0. 05 molar Na2CO3 (pH 9. 6), and was fixed overnight at room temperature to the walls of wells in microtiter plates. Between each incubation step, the plate was washed five times with Tris buffer (comprising 10 millimolar Tris, 150 millimolar NaCI, 0. 05% Brij 35, and 0. 02% (w/v) sodium azide). Brij 35 (polyoxyethylene (23) lauryl ether C, 2H25 (OCH2CH2) 23OH) is a detergent. Purified CPS obtained from type 6B, 6A, 18C or 9V pneumococci was used at a known concentration, and was purchased from the American Type Tissue Collection (Rockville, MD) for use as a standard. CPS in sonicated cell samples was detected using monoclonal antibodies designated HASP 4 (which binds specifically with type 6A and 6B CPSs), HASP 22 (which binds specifically with type 18C CPS), and HASP 33 (which binds specifically with type 9V CPS), and were used at a concentration determined in pilot experiments. Binding of monoclonal antibodies and CPS was detected using antisera which reacts specifically with mouse IgM and which is conjugated with alkaline phosphatase. This reagent was used as described (Kim and Weiser, 1998). Total cellular protein determination was carried out with sonicated cell extracts using the micro-bicinchoninic acid kit according to the manufacturer's directions (Pierce Chemical Co., Rockford, IL). The amount of CPS in the supernatant fraction was based on protein concentration in the corresponding cell sonicate fraction (i. e. in order to normalize CPS content on a per-unit-protein basis).
The capture ELISA method used to compare amounts of total teichoic acid in cell sonicates has been described (Kim et al., 1998). All experiments were performed in duplicate, and repeated at least three times. Results are expressed as mean values per total cellular protein concentration.
Western Analysis Bacteria were grown on tryptic soy agar supplemented with catalase as described above. After growth under selected conditions forl6 hours at 37°C, cells were removed from the surface using a sterile Dacron swab, re-suspended in PBS, and washed in PBS in order to adjust the cell density such that A620 = 0. 5. After centrifuging, pelleted cells were re-suspended in Laemmli gel loading buffer and heated to 100°C for 5 minutes. The samples were then separated by SDS-PAGE on 12. 5% (w/v) polyacrylamide gels, and then transferred to Immobilon-P membranes as described (Wani et al., 1996).
Immunoblotting was performed using a monoclonal antibody designated 4G10 (Upstate Biotechnology Inc., Waltham, MA) at a dilution of 1 : 2000 in TSBP buffer (Wani et al., 1996), and reactivity of blotted samples with this antibody was detected using mouse IgG anti-sera conjugated with alkaline phosphatase as described (Kim et al., 1999). Loading of approximately equal numbers of bacteria was confirmed by immunoblotting of duplicate membranes using previously described antisera raised against whole pneumococci (Wani et al., 1996).
Northern Analysis Total RNA was isolated from O and T variants of strain P303 cells grown to mid-log phase in semi-synthetic medium under atmospheric and anaerobic conditions and purified as described (Weyand et al., 2000). Following separation of RNA on a formamide gel, ethidium bromide stained RNA was photographed and the image was digitized for comparison with RNA size markers. The RNA was transferred to a membrane using the VacuGene XL System according to the manufacturer's instructions (Pharmacia LKB Biotechnology, Uppsala, Sweden). The membrane was pre-hybridized in a solution comprising 0. 5 molar NaP04 at pH 7. 2, 1. 5 millimolar EDTA, and 7% (w/v) SDS for 15 minutes at 65°C. The probe was prepared from an internal fragment of the cps4D gene, and was amplified by PCR using primers having the following sequences : 5'-CCGGAATTCG TACAAATATA CAGTTGAGCG GAGATAAAC-3' (SEQ ID NO : 1) and 5'-CGCGGATCCT GTTGCTGTTA CCAAGATGGA CG-3' (SEQ ID NO : 2) and genomic DNA obtained from a type 4 strain. The strain is the same strain used in whole genome sequencing by the Institute for Genomic Research. The PCR product was digested using restriction endonucleases BamHI and EcoRI, and then cloned into a vector having transcriptional terminators flanking the polylinker to allow cloning of streptococcal sequences. This vector was constructed from pJDC9 by insertion of the blunt ended kanamycin resistance cassette of pUK4K (Pharmacia, Upsalla, Sweden) into a blunt ended ClaI site to allow selection of this marker in E. coli (Chen and Morrison, 1988). The cps4D fragment was inserted into BamHI and EcoRI digested vector and transformed into E. coli strain DH5a. Total RNA was extracted from pneumococcal O and T variants of strain P303 grown to mid-log phase in tryptic soy medium under atmospheric and anaerobic conditions.
The results of the experiments presented in this Example are now described.
Selection for Phenotypic Variants in Humans In order to test whether selection occurs for different pneumococcal phenotypes in carriage and bacteremia in humans, paired isolates of the same serotype were obtained from the nasopharynx and blood of patients who exhibited signs of sepsis. It was necessary to obtain paired isolates, owing to strain-to-strain heterogeneity in colony morphology unrelated to opacity. Pneumococcus could not be isolated from the nasopharynx of bacteremic patients once antibiotic treatment had been initiated. Cultures of the nasopharynx and blood were therefore obtained substantially simultaneously. This limitation required that the study be carried out in a geographic location having a high incidence of pneumococcal bacteremia, in order that a sufficient numbers of paired isolates could be obtained.
Blood and nasopharynx isolates were obtained from 19 adult patients in Blantyre, Malawi, and were only considered as paired isolates if the isolates were subsequently shown to be of the same type by the Quellung reaction using type specific antisera. The majority of the paired isolates were type 1, the most common type in this location, as indicated in Table II. The majority of patients also tested positive for infection with the human immunodeficiency virus, which is believed to account for the high incidence of invasive pneumococcal disease among these patients. Of 19 paired isolates, 17 (89%) were of the T phenotype in the nasopharynx, whereas 12 (63%) were of the O phenotype in the blood. Of the ten paired isolates in which the phenotype of pneumococci obtained from the two sites were discordant, all exhibited a more T-like phenotype in the nasopharynx (P Table II Opacity Phenotype of Isolates Obtained from Human Carriage and Invasive Infections Colony Phenotype HIV Status No. of Patients Types (No.) | Nasopharynx Blood/CSF1 + - N/A2 T T 7 1 (4), 4, 18, 19 5 1 1 1 O O 2 1 (2) 1 0 1 0 T 0 T 0 10 1 (4), 4 (2), 6, 12, 15, 22 5 1 4 Notes :'4/19 invasive isolates were obtained from cerebrospinal fluid rather than blood.
2 N/A means clinical status not available.
Effect of Oxygen on Amounts of CPS Host and bacterial factors that could promote selection of O variants during infection and T variants during carriage were investigated. It was hypothesized that lower oxygen tension of the sort that would be encountered in consolidated lung would promote a transition to from the T to the O phenotype. Growth of a type 6A clinical isolate, P303, in reduced concentrations of oxygen had no effect on the rate of spontaneous switching from the T-to the O-form. Growth under anaerobic conditions (in the presence of an elevated concentration of carbon dioxide), however, was associated with a shift to a larger and more mucoid colony morphology that was particularly marked for the O variant, as shown in Figure 1A.
Because pneumococci exhibiting the O phenotype express greater amounts of CPS than those exhibiting the T phenotype, it was considered possible that the greater colony size associated with the O form was a result of increased production of this material (Kim et al., 1998). In initial studies, the refractile zone of CPS surrounding bacterial colonies was visualized using the Quellung reaction and type 6 antisera, as shown in Figure 1B. This zone was larger for the O variant grown under anaerobic conditions in the presence of 10% carbon dioxide, compared to either growth of the same phenotype in the presence of oxygen or growth of the T variant under any of the conditions tested including the absence of oxygen. In contrast, when the bacteria were cultured under atmospheric conditions, a zone of capsular material could not be detected around colonies of T-form pneumococci. This result confirmed the ability of the O form to synthesize increased quantities of CPS, and indicated that reduced oxygen tension, higher carbon dioxide tension, or both promotes increased CPS production.
The effect of environmental oxygen and carbon dioxide concentration was assessed quantitatively using an assay developed for measuring amounts of CPS, as indicated in Figure 2 (Kim et al., 1998). The amounts of CPS per milligram of total cellular protein, as assessed using the capture ELISA described herein, correlated with colony surface area and with capsular volume values calculated based on visualization of zones using the Quellung reaction and with colony surface area, as indicated in Table III. 0-form bacteria at either mid-log phase or stationary phase growth exhibited increased amounts of cell-associated CPS compared to the T-form organisms. Growth of the O type 6A variant under anaerobic conditions in the presence of 10% Cor cause a 7. 9-fold rise in the quantity of CPS produced, compared to CPS production in the same cells cultured under atmospheric conditions. Compared to T variants grown under the anaerobic conditions in the presence of 10% CO2, the O variant made 29 times as much CPS. Similar results were obtained using O and T isolates obtained from two unrelated clinical isolates (i. e. using types 6B and 18C).
Table III Characteristics of Pneumococci Grown Under Various Conditions Colony Morphology3 0-form T-form °2 (%) 21 16 4These values are based on colonies observed after 16 hours growth. Values represent the average of determinations made using three single well isolated colonies.
5These values are based on capsular material revealed using the Quellung reaction and the formula for the volume of a spherical ellipsoid, V= (/6) LW2. Values represent the average of determinations for three diplococcal forms divided by two to correspond to a single cell."UD"means that a zone of capsular material was not detectable.
6These values are based on capture ELISA assessment of organisms grown in liquid culture at mid-log phase. Values are the mean of two separate determinations.
Production of CPS in culture supernatant was assessed for type 18C pneumococci after growth to stationary phase. The observation that a greater amount of polysaccharide was produced in the culture supernatant of the 0-form variant as the concentration of oxygen was decreased and as the concentration of carbon dioxide was increased indicates that increased CPS production was attributable to increased synthesis of CPS by cells under decreased oxygen/increased carbon dioxide conditions, rather than being attributable to decreased release of CPS from cells under increased oxygen/decreased carbon dioxide conditions. A type 9V isolate was used to determine whether increased synthesis of CPS was attributable by decreased oxygen or increased carbon dioxide. Type 9V bacteria grown to stationary phase in the presence of 10% carbon dioxide exhibited 12-fold greater CPS production in the absence of oxygen than the same cells grown in the presence of 10% carbon dioxide and in the presence of oxygen. This confirmed that the oxygen tension, rather than carbon dioxide tension, is the predominant environmental factor affecting the degree of CPS production.
Changes in CPS Production Correlate with Changes in cpsD/CpsD The next phase of the study addressed the mechanism by which the effect of oxygen on CPS production is mediated. Involvement of a two-component signal transduction system for sensing and transducing oxygen tension/presence was examined. Previously described marked mutations in 11 of the 12 of two-component signal transduction systems found in the genome of a type 3 isolate of were transformed into strain P303, as described (Throup et al., 2000). 10 constructs harboring mutations in response regulator homologues were made, as were 7 constructs harboring mutations in histidine kinase homologues. For each of the 17 mutant homologues tested in P303, there was no effect on increased colony size or zone of capsule formation, as assessed using the Quellung reaction, in experiments in which the cells were grown under anaerobic conditions. This result indicated that it is unlikely that the effect of oxygen is mediated through a TCSTS, although one of these systems in the pneumococcus could not be examined because the genes appear to be essential.
The genes in a locus known to be required for expression of CPS were examined next. Since the effect of oxygen on amounts of CPS was observed in strains of multiple types, genes conserved among different types were considered the most likely candidates. Among the genes in the capsule loci, only the first four, cpsA-D, are common to pneumococci of different types (Iannelli et al., 1999). Type 3 pneumococci are known to form mucoid colonies and to have large capsules, as assessed by the Quellung reaction, when grown under atmospheric conditions, as indicated in Table I (Knecht et al., 1970). Unlike all other pneumococcal types tested, type 3 strains failed to show an increase in the sizes of colonies or capsules when cultured under anaerobic conditions, relative to the sizes when cultured under atmospheric conditions. Based on this observation, differences in cpsA-D between the type 3 and loci of other types was determined.
Homologs of cpsA, encoding a putative transcriptional attenuator, and cpsB, encoding a protein of unknown function, are present in the type 3 capsulation locus (Genbank accession no. Z47210 ; Arrecubieta, 1995). The gene encoding CpsC in serotype 2 pneumococci is highly conserved in the type 3 capsulation locus, but the predicted CpsD in serotype 3 pneumococci is truncated following amino acid 69 (of 226). CpsC and CpsD are required for expression of the pneumococcal capsule and together are homologous to a single protein (Wzc) expressed in Escherichia coli that is involved in chain length regulation and export of the extracellular polysaccharide designated colanic acid (Morona et al., 1999 ; Stevenson et al., 1996).
In E. coli, Wzc undergoes reversible autophosphorylation at a tyrosine residue (Vincent et al., 1999). A monoclonal antibody which is designated 4G10 which binds specifically with phosphorylated tyrosine residues was used to determine by Western analysis that phosphorylated tyrosine is present in the pneumococcus. A single band of approximately 26 kD, the predicted size of CpsD, was present in whole cell lysates of the type 6A isolate. A single band of the same size was also seen in whole cell lysates of the types 6B, 18C, and 9V strains used in this study, but was absent from the type 3 pneumococci (which have a truncated CpsD). These observations indicated that CpsD is phosphorylatable at a tyrosine residue. Further evidence that 4G10 was recognizing CpsD having a phosphorylated tyrosine residues was obtained by comparing strains D39 (an non-encapsulated mutant) and R6x (a strain having a 7504 base pair deletion spanning cps2A-H, of which only A-D are common to other types ; Iannelli et al., 1999). The mutation in R6x-type pneumococci was corrected in some experiments by transforming the bacteria using DNA obtained from either the (type 2) D39 parent strain or a type 3 strain to yield an encapsulated strain with either type 3 or 2 capsule in the D39 type 2 genetic background.
Failure of 4G10 to react with R6x and the type 3 R6x transformant and reaction of the antibody with the type 2 R6x transformant confirmed that expression of the protein corresponding to this band requires one of the first four common genes in the capsulation locus and an intact copy of cpsD. It was determined that the 4G10 reactive protein is not likely to be CpsA or B based on size. There are only 15 conservative amino acid changes between type 2 and 3 CpsC, of which none involve tyrosine. The lack of the 4G10 reactive band in the type 3 strain indicated that the 4G10-reactive protein is not CpsC. It was therefore concluded that CpsD protein comprises a phosphorylatable tyrosine residue in the pneumococcus.
The effect of growth under various concentrations of oxygen and carbon dioxide on tyrosine phosphorylation of CpsD was examined. Lysates of whole cells of the O and T forms of the type 6A pneumococcal isolate designated P303 which had been grown under various concentrations of oxygen and carbon dioxide were adjusted to an equivalent density and examined by Western analysis using the phosphotyrosine- specific monoclonal antibody 4G10. The results of these experiments are shown in Figure 3. A single band of the expected size for CpsD was detected in the O variant grown in the presence of oxygen. There was no reactivity with the extract obtained from an approximately equal number of cells grown anaerobically, as indicated in lane 1 of Figure 3. There was no apparent effect of differences in carbon dioxide tension from Difference in tyrosine phosphorylation can affect the activity of CpsD and account for decreased expression of CPS in O pneumococci in the presence of oxygen. Neither O nor T pneumococci grown anaerobically express phosphorylated tyrosine ; however the two phenotypic forms exhibit large differences in the degree of CPS production.
Northern analysis was used to determine if differences between O and T variants were due to differences in the level of transcription of cpsD. The results of these experiments are shown in Figure 4. Total cellular RNA was obtained from O and T forms of pneumococcal strain P303 grown either under atmospheric conditions or anaerobically in the presence of 10% carbon dioxide. The RNA was probed using a cloned fragment of cpsD. Quantitative differences in transcriptional activity were determined by assessing hybridization of radio-labeled probe with cpsD in mRNA and adjusting these values for the total amount of RNA present in the hybridization assay mixture. These experiments demonstrated that oxygen concentration had no substantial effect on the level of cpsD transcription. However, cpsD was transcribed at a 3. 4-fold greater level in T-form cells, relative to O-form cells. It was therefore concluded that transcriptional regulation of cpsD determines the opacity phenotype of pneumococci.
Without wishing to be bound by any particular theory of operation, it is believed that CpsD regulates CPS production. Correlation between differences observed in the degree of tyrosine phosphorylation of CpsD and differences in CPS production support this theory. It is believed that CpsD functions as a tyrosine kinase, based in part on its amino acid sequence similarity to a tyrosine autophosphorylating enzyme in E. coli which is involved in chain length regulation and export of the extracellular polysaccharide colanic acid. Cps23FD has 30% identity and 49% similarity over 188 of 229 amino acids to Wzc. CpsD of serotype 23f pneumococcus contains an unusual carboxyl-terminal amino acid sequence (YGSYGNYGNY GKNKK ; SEQ ID NO : 3) which contains a repetitive feature (YGX) 4 including tyrosine residues. This carboxyl-terminal region exhibits considerably greater amino acid sequence heterogeneity among pneumococcal variants than the remainder of CpsD. It is believed that the variability in CPS production among pneumococcal strains is attributable to the variability of this carboxyl-terminal portion among the strains, the sequence variation affecting the phosphorylat-ability of the protein. A number of other bacterial protein kinases have been identified and these have in common a C-terminus with a tyrosine-rich repetitive feature (Ilan et al., 1999).
Inability to generate a pneumococcal strain harboring a mutation in cpsD while maintaining the ability of the strain to produce CPS is further evidence of the importance of this gene in CPS and capsule production.
The following model of a mechanism for regulating CPS production in response to colony opacity phenotype and oxygen tension is proposed. The model involves transcriptional regulation of cpsD and post-translational modification of CpsD.
CpsD is a negative regulatory factor which act on genes involved in biosynthesis of pneumococcal type-specific CPS. Relatively low cpsD transcription accounts for the relatively high amounts of CPS expressed in O-type pneumococci under anaerobic conditions. Apart from transcriptional regulation of cpsD, which accounts for a significant proportion of CPS expression regulation, a second level of regulation is involved, namely phosphorylation or autophosphorylation of tyrosine residue (s) on CpsD. Phosphorylation of CpsD affects only the O variant and requires growth under aerobic conditions. Detectability of CpsD which is phosphorylated at one or more tyrosine residues only in the 0 variant may be due to increased phosphatase activity in T pneumococci. Tyrosine phosphorylation increases the negative regulatory activity of CpsD, such that phosphorylated CpsD decreases synthesis of CPS under aerobic conditions. As a result, less CPS is made by O variants under aerobic conditions than under anaerobic growth conditions. Relatively low CPS production in T-form pneumococci, even in the absence of appreciable tyrosine phosphorylation of CpsD, is attributable to increased expression of cpsD in T-form pneumococci.
What had previously been described as improved growth under anaerobic condition, may in fact be an effect of increased CPS synthesis resulting in larger colony size (Modde, 1978).
Homologues of CpsD with even greater similarity to Wzc are present in many other species of bacteria that express surface polysaccharides. Examples of such species include Streptococcus agalactiae, Staphylococcus aureus, Klebsiella pneumoniae, and Acinetobacterjohnsonii. Since capsules and their size are critical determinants of virulence, the ability to regulate amounts of CPS based on transcriptional and post-translation modification of this class of gene/protein may be important in facilitating the ability of encapsulated bacteria to adapt to the differing requirements of colonization and infection.
The disclosure of every patent, patent application, and publication cited herein is hereby incorporated herein by reference in its entirety.
While this invention has been disclosed with reference to specific embodiments, it is apparent that other embodiments and variations of this invention can be devised by others skilled in the art without departing from the true spirit and scope of the invention. The appended claims include all such embodiments and equivalent variations.
RSS




We Claim
1. A method of producing capsular pneumococcal polysaccharide comprising the steps
of:
a) obtaining an opaque variant of the pneumococcus;
b) maintaining said opaque variant of the pneumococcus in a growth medium in
contact with a gas having either an oxygen concentration less than about 16% or a
carbon dioxide concentration greater than about 5%; and
c)optionally isolating the produced capsular polysaccharide.
2. The method as claimed in claim 1, wherein said gas has an oxygen concentration less
than about 16%.
3. The method as claimed in claim 1, wherein said gas has a carbon dioxide concentration
greater than about 5%.
4. The method as claimed in claim 2, wherein said gas has an oxygen concentration not
greater than about 0.1%.
5. The method as claimed in claim 1, wherein said pneumococcus is an organism of the
genus Streptococcus.
6. The method as claimed in claim 5, wherein said pneumococcus is an organism of the
species Streptococcus pneumoniae.
7. The method as claimed in claim 6, wherein said pneumococcus is a variant of the
species Streptococcus pneumoniae selected from the group consisting of variants 6A, 6B,
18C, and 9V.
8. The method as claimed in claim 2, wherein said gas has a super atmospheric
concentration of carbon dioxide.
9. The method as claimed in claim 3, wherein said gas has a carbon dioxide concentration
of at least about 10%.
10. The method as claimed in claim 3, wherein said gas has a sub-atmospheric
concentration of oxygen.
11. The method as claimed in claim 10, wherein said gas has an oxygen concentration not
greater than about 0.1%.
12. The method as claimed in claim 1, wherein the carbon dioxide concentration of said
growth medium is greater than the concentration of carbon dioxide in the same growth
medium equilibrated at the same temperature with a gas comprising 5% carbon dioxide.
13. The method as claimed in claim 1, wherein the oxygen concentration of said growth
medium is lower than the same medium equilibrated at the same temperature with a gas
which has an oxygen content of 16%.
14. The method as claimed in claim 13, wherein said gas has an oxygen concentration not
greater than about 0.1%.
15. The method as claimed in claim 14, wherein said growth medium is substantially
devoid of oxygen.
16. The method as claimed in claim 13, wherein said growth medium has a carbon
dioxide content which is greater than the same medium equilibrated at the same
temperature with normal air.
17. The method as claimed in claim 16, wherein said growth medium is saturated with
carbon dioxide.
18. The method as claimed in claim 17, wherein said growth medium comprises a
carbonate or bicarbonate salt.
19. A method of assessing whether a test compound is useful for alleviating a
pneumococcal infection in an animal, the method comprising comparing the degree of
phosphorylation of CpsD in pneumococcal cells maintained in the presence of the test
compound and the degree of phosphorylation of CpsD in the same type of cells
maintained in the absence of the test compound, wherein if the degree of phosphorylation
of CpsD in the cells maintained in the presence of the test compound is less than the
degree of phosphorylation of CpsD in the cells maintained in the absence of the test
compound, then the test compound is useful for alleviating the infection.

20. The method as claimed in claim 19, wherein the degree of phosphorylation of CpsD
is assessed by assessing the number of phosphorylated tyrosine residues present in CpsD.
21. The method as claimed in claim 19, wherein the degree of phosphorylation of CpsD
is assessed by assessing the fraction of CpsD having at least one phosphorylated tyrosine
residue.
22. A method of producing a capsular polysaccharide from a pneumococcus substantially
as herein described with reference to the accompanying drawings and examples.
23. A method of producing an immunogenic preparation for administration to an animal
at risk for developing a pneumococcal infection substantially as herein described with
reference to the accompanying drawings and examples.