FIELD OF INVENTION

The invention relates to a method for assessing the Fc- function of immunoglobulin containing preparations. In particular, the invention relates to a method for assessing the Fc- function of intravenous immunoglobulin (IVIG) products.

BACKGROUND ART

Intravenous immunoglobulin (IVIG) products are indicated as the treatment of choice in a number of immune-mediated disorders, including inflammatory diseases and autoimmune diseases (Riepert et al (2004) ) . Interactions with Fc-receptors have been described as an important mechanism of action of IVIG products. The Fc region of IgG mediates biological functions such as binding to Fc- receptors of macrophages for the promotion of phagocytosis, activation of the complement system and antibody-dependent cellular cytotoxicity (Anon (1997) , Schroeder et al (2008) ,

Virella (2005)). Evaluation of Fc-function is therefore, critical for assessing the biological activity of IVIG products and manufacturers are required to demonstrate that Fc-mediated functions are maintained in their IVIG products.

The test for the Fc-function of IVIG products (European

Pharmacopoeia, Method 2.7.9) measures complement-mediated

haemolysis of group 0 red blood cells (RBCs) . The RBCs are tanninized and coated with Rubella antigen prior to contacting with samples of the IVIG product. Complement-mediated haemolysis is initiated by adding complement to the sample and haemolysis measured over time by monitoring absorbance (541 nm) . The index of Fc-function (J _Fc ) is derived from the time dependent absorbance curve. The test method {European Pharmacopoeia Method 2.7.9) employs a large number of preparative steps , a number of which may only be performed shortly before or at the time the test method is performed. As a consequence the test method is cumbersome and lengthy. The convenience and efficiency of the test method has been improved by using frozen RBCs and a microtitre plate format (Georgakopoulos et al (2009)). However, the test method remains problematic, particularly in respect of the preparation of the RBCs coated with Rubella antigen.

It is an object of this invention to provide an improved test method for the Fc- function of immunoglobulin containing

preparations. It is an object of this invention to provide reduced intra- and inter-laboratory variation in assaying the Fc- function of immunoglobulin containing preparations. These objects are to be read disjunctively with the object to at least provide a useful choice.

STATEMENT OF INVENTION

In a first aspect the invention provides a method of assaying Fc- function of a sample of an immunoglobulin containing preparation comprising the steps of: · contacting the sample, in the presence of complement, with red blood cells (RBCs) modified to incorporate a construct of the structure F-S-L; and then

• monitoring the rate of haemolysis, where F is a target antigen for a complement activating

immunoglobulin, S is a spacer covalently linking F to L and selected to provide a construct (F-S-L) dispersible in water, and L is a lipid.

Preferably, where the complement is added to the sample following the step of contacting the sample with the RBCs the concentration of complement is in the range 18 to 28 CH ₅₀ /ml. More preferably, where the complement is added to the sample following the step of contacting the sample with the RBCs the concentration of

complement is 23 (±5%) CH ₅₀ /ml.

Preferably, the concentration of immunoglobulins in the sample is in the range 40 to 60 mg/ml . More preferably, the concentration of immunoglobulins in the sample is 50 (±5%) mg/ml. Preferably, the immunoglobulin containing preparation is serum. More preferably, the immunoglobulin containing preparation is an intravenous immunoglobulin (I IG) product.

Preferably, the target antigen is an antigen of cytomegalovirus (CMV) . More preferably, the target antigen is:

ThrProThrProValAsnProSerThrAlaProAlaProAlaProThrProThrPhe Ala of the biosequence designated SEQ ID NO:l.

Preferably, the construct is of the structure:

where g is the integer 1, 2 or 3 , h is the integer 1, 2, 3 or 4 , is the integer 3, 4 or 5, and M' is a monovalent cation or substituent. An example of a monovalent cation is H ⁺ . An exampl of a monovalent substituent is methyl (CH ₃ ) . More preferably, the construct is of the structure:

Preferably, the spacer (S) is covalently linked to the target antigen (F) via a cysteine residue at the carboxy (C) terminus of the target antigen.

Preferably, L is a diacyl- or dialkyl- glycerophospholipid. More preferably, L is a diacylglycerophospholipid. Preferably, the spacer (S) is covalently linked to the lipid (L) via an amide bond formed between the terminal carboxyl group and a primary amine . Preferably, L is a phosphatidylethanolamine . More preferably, L is a phosphatidylethanolamine of the structure: where M is a monovalent cation and R ^' and R ^" are independently selected from the group consisting of: oleoyl and stearyl .

Typically, M is H. Most preferably, L is dioleoyl- phosphatidylethanolamine (DOPE) .

In an embodiment of the first aspect of the invention the construct (F-S-L) is of the structure designated DOPE-Ad-CMG (2) - hCMV2

In a second aspect the invention provides constructs of the structure :

for use in a method of assaying the Fc- function of a sample of an immunoglobulin containing preparation where F is a peptide of the biosequence designated SEQ ID N0:1, L is phosphatidylethanolamine, g is the integer 1, 2 or 3, h is the integer 1, 2, 3 or 4 , i is the integer 3, 4 or 5, and M' is a monovalent cation or

substituent.

Preferably, g is 2, h is 2, i is 4, and M' is H. In an embodiment of the second aspect of the invention the construct is of the structure:

where R' and R' ' are both oleoyl, i.e. L is DOPE, and designated D0PE-Ad-CMG(2) -hCMV2.

In the description and claims of this specification the following acronyms, terms and phrases have the meaning provided:

"Dispersible in water" means a stable, single phase system is formed in water at a concentration of at least 10 mg/ml in the absence of detergents or organic solvents.

"Complement activating immunoglobulin" means an immunoglobulin capable of activating the complement system.

"In the presence of complement" means the complement is either present in the sample prior to the step of contacting the sample with the RBCs, or added to the sample following the step of contacting the sample with the RBCs. "Kodecytes" means red blood cells (RBCs) modified to incorporate one or more constructs of the structure F-S-L where F is a functional moiety, S is a spacer and L is a lipid.

"Modified to incorporate" means modified by incorporation of an exogenously preprared, preformed construct.

"Serum" means the amber-coloured, protein-rich liquid which separates out when blood coagulates.

" ( ) ^x " , " ( ) _x " , " [ ] χ" and "[ ] ^x " mean the group contained in the parentheses is repeated a number (x) of times . By way of illustration:

means the methylene group (-CH ₂ -) is repeated 4 times and the structure represented is equivalent to:

It will be recognised that the phosphate moiety of a phospholipid is often protonated, but the proton (H ⁺ ) may be replaced by another monovalent cation such as Na ⁺ , K ⁺ , NH ₄ ⁺ or triethylamine ( [NH (CH ₂ CH ₃ ) ₃ ] ⁺ ) . Similarly, the primary and secondary amino functions of a construct may also be protonated. The constructs described in this specification may therefore be prepared as a number of salts, including pharmaceutically acceptable salts.

It should be noted that an asterix (*) is used to represent a point of covalent attachment to another moiety and does not indicate an atom. In the description and claims of this specification the amino acids of peptides are identified in accordance with Annex C, Appendix 2 of the PCT Administrative Instructions (as in force from January 1, 2010) and in accordance with the convention:

["amino (N) terminus"] XaaXaaXaa XaaXaaXaa ["carboxy (C) terminus"]

Where concentrations or ratios of reagents are specified the concentration or ratio specified is the initial concentration or ratio of the reagents. Where values are expressed to one or more decimal places standard rounding applies. For example, 1.7 encompasses the range 1.650 recurring to 7.499 recurring. The terms "first", "second", "third", etc. used with reference to elements, features or integers of the subject matter defined in the Statement of Invention and Claims, or when used with reference to alternative embodiments of the invention are not intended to imply an order of preference. The invention will now be described with reference to embodiments or examples and the figures of the accompanying drawings pages.

BRIEF DESCRIPTION OP DRAWINGS

Figure 1. Structure of the construct designated DOPE-Ad-CMG (2) -hCMV2 where M' and M' ' are both H.

Figure 2. Haemolysis curves of kodecytes (modified at a concentration of 0.8 mg/ml of the construct designated DOPE-Ad-CMG (2) -hCMV2) following incubation with buffer (solid squares) , IVIG (open triangles) or EPBRP standard (shaded circles) and the addition of 153 CH ₅₀ /ml of GPC. The samples were analysed in duplicate and the mean haemolysis curve for each sample is shown.

Figure 3. The effect of different concentrations of complement on the rate of haemolysis of kodecytes (modified at a concentration of 0.8 mg/ml of the construct designated DOPE-Ad-CMG (2) -hCMV2) following incubation with EPBRP standard. Complement concentrations of 46 (solid squares) , 23 (open triangles) and 11.5 (shaded circles) CH ₅₀ /ml were added to the samples in duplicate. The mean haemolysis curve for each concentration of complement is shown.

Figure 4. A comparison between the Fc indices (I _fc ) of various samples of immunoglobulin containing preparations determined using kodecytes

(solid squares) or the modified EP method according to Georgakopoulos et al (2009) (shaded squares) .

Figure 5. Inter-batch variation of kodecytes in the method for assaying Fc-function. Samples of immunoglobulin containing preparations were incubated with batch #1 (solid squares) and batch #2 (shaded squares) . Figure 6. Stability of kodecytes determined by comparing the Fc indices (I _f c) obtained for samples of immunoglobulin containing preparations using kodecytes stored for 1 (dark shaded squares) , 4 (diagonal hash squares) , 6 (horizontal hash squares) , 8 (light shaded squares) and 10 (cross hash squares) weeks at 2 to 8 °C.

DETAILED DESCRIPTION

The invention resides in part in the selection of a target antigen for a complement activating immunoglobulin and its incorporation into a water dispersible construct that will spontaneously incorporate into the membranes of RBCs . Human cytomegalovirus

(hCMV, human herpes virus 5) , the beta herpes virus, is prevalent in human populations. Detection of the virus is part of the standard screening for blood donation. hCMV negative blood is particularly important for infants, organ transplantations and immunocompromised patients (Tutschka (1988) ) . As discussed in the specification accompanying international application no.

PCT/NZ2008/000239 (publication no. WO 2009/035347) the use of kodecytes creates an opportunity to employ established blood typing platforms to detect a range of peptide antigen-antibody interactions with the capital costs associated with establishing a new diagnostic assay being avoided. The use of constructs (F-S-L) comprising the spacer (S) described in the specification

accompanying international application no. PCT/NZ2008/000266 (publ . no WO 2009/048343) are advantageously used in the method of the present invention due to the absence (or at least negligible incidence) in the population of individuals expressing antibodies to the spacer (S) . By contrast individuals expressing antibodies to polyethyleneglycol (PEG) , for example, are known in the population. In the context of assaying the Fc-funtion of immunoglobulin containing preparations, use of the method of the invention has greatly simplified and reduced the time of the assay (cf. European Pharmacopoeia, Method 2.7.9) by eliminating the need to freeze and thaw RBCs, treat with tannic acid and incubate with Rubella antigen. The insertion of the constructs (F-S-L) into the membrane to provide kodecytes is a simple incubation procedure requiring only two hours to perform. Furthermore, the kodecytes can be stored at 2 to 8°C for up to 10 weeks without loss of function, thus providing a readily available source for assaying Fc- function in samples of immunoglobulin containing preparations. In contrast, thawed RBCs, treated with tannic acid and derivatised with Rubella antigen are stable only for two days at 2 to 8°C. A further advantage of the invention is the ability to control the amount of antigen inserted into a cell membrane providing for improved batch-to-batch and laboratory- to- laboratory reproducibility in the preparation of kodecytes for use in the assay.

EXAMPLES

Materials

The construct designated DOPE-Ad-C G (2) -hC V2 was prepared according to the method described in the specification

accompanying international application no. PCT/NZ2008/000266

(publ . no WO 2009/048343) employing the Cys terminated peptide ThrProThrProValAsnProSerThrAlaProAlaProAlaProThrProThrPheAla Cys . For the preparation of RBCs modified to incorporate the construct designated DOPE-Ad-CMG (2) -hCMV2 (kodecytes) the construct was dispersed in a cell preservative solution (CELPRESOL™, CSL

Limited, Melbourne, Australia) . RBCs of blood group O were obtained from the Australian Red Cross Blood Service (ARCBS) . A European Pharmacopoeia Biological Reference Preparation (EPBRP) of human immunoglobulin (batch no. 3) was obtained from Council of Europe (Strasbourg, France) . Guinea pig complement (GPC) was supplied by Harlan (Bicester, England) . Purified and concentrated Rubella virus antigen of no less than 256 haemagglutination units was obtained from Aalto Bio Reagents (Dublin, Ireland) . All samples of immunoglobulin containing preparations tested for Fc- function were obtained from CSL Biotherapies (Broadmeadows ,

Australia) . The samples of immunoglobulin containing preparations tested included hyperimmune tetanus formulated at 6% (w/v) protein, pH 4.25 (IVIG) and hyperimmune hepatitis B and tetanus, RhD for intramuscular administration (IMIG) , and normal

immunoglobulin (NIG) formulated at 16% (w/v) protein, pH 6.8.

Bovine serum albumin (BSA) fraction V was obtained from Calbiochem (La Jolla, CA, USA) , tannic acid was obtained from Sigma (St Louis, MO, USA) , pepsin was obtained from Roche Applied Science

(NSW, Australia) , phosphate buffered saline (PBS) was obtained from Pierce (Rockford, IL, USA) , barbitone sodium was obtained from Merck (Darmstadt, Germany) and CELPRESOL™ was obtained from CSL Biotherapies (Parkville, Australia) . Tris-glycine 4-20% polyacrylamide gradient gel, tris-glycine running buffer, tris- glycine non-reducing sample buffer and SeeBlue standard were obtained from Invitrogen (Mulgrave, Australia) . Coomassie blue R- 250 was obtained from Merck (Darmstadt, Germany) , methanol (AR) and acetic acid (glacial) was obtained from APS Chemicals

(Cheltenham, Australia) . All solutions were prepared with pyrogen free water (PFW) .

Preparation of kodecytes

Packed 0 group RBCs were washed in saline then CELPRESOL™ and resuspended in CELPRESOL™. The Packed Cell Volume (PCV) of the suspension was adjusted to 50% ± 2.5% v/v with the addition of CELPRESOL™. A quantity of stock solution of the construct was added to the suspension and incubated at 37 °C for 2 hours with constant gentle agitation. Following incubation the kodecytes were washed with CELPRESOL™ and the PCV of suspension determined. The PCV of the suspension was adjusted 80% ± 2.5% v/v with the addition of CELPRESOL™ and the kodecytes stored at 2 to 8 °C.

Fc-function assay using kodecytes

The kodecytes were spun at 3400 x g for 5 min at 4 °C. The

CELPRESOL™ was removed and the kodecytes resuspended in albumin barbitone buffer (142 mM sodium chloride, 5 mM barbitone sodium, 0.15 mM calcium chloride, 0.5 mM magnesium chloride, 0.6 mM sodium azide and 0.15% w/v bovine serum albumin in PFW, pH 7.3) such that the absorbance of the suspension when diluted 1:10 was 1.0 ± 0.1 at 541 nm. The human immunoglobulin EPBRP standard and

immunoglobulin containing preparations were diluted to 40, 50 or 60 mg/ml in albumin barbitone buffer solution and if necessary adjusted to pH 7 by addition of 0.5 M sodium hydroxide. Diluted samples (900 μΐ) were transferred to Eppendorf tubes. A negative control was also prepared by adding 900 μΐ of albumin barbitone buffer to an Eppendorf tube. Kodecytes (100 μΐ) were added to each Eppendorf tube and mixed. Samples were incubated for 1 hour before centrifugation at 2600 g for 1 minute. Supernatants were discarded and the kodecytes washed three times with 1 ml of albumin barbitone buffer. The kodecytes were resuspended in a final volume of 700 μΐ albumin barbitone buffer. Haemolysis was monitored in a microtitre plate format. A 240 μΐ aliquot of each sample of kodecytes was transferred in duplicate to wells of a microtitre plate and pre-warmed in a plate reader (SpectaMax 190, Molecular Devices, CA, USA) at 37 °C. A 60 μΐ aliquot of GPC diluted in albumin-barbitone buffer was pre-warmed for 30 seconds at 37 °C and added immediately to each sample well. The optimal

concentration of GPC required for generating haemolyses curves that progressed at an appropriate rate was determined. The plate was returned to the plate reader and absorbance was monitored at 541 nm every 4 seconds for 12 minutes. The observed haemolysis curves were examined by electronic derivative analysis (Sigma Plot, Version 10.0, Systat Software, 2006, Chicago, IL) involving determination of the maximal rate of change of absorbance achieved at the inflection point (Georgakopoulos et al (2009)) . The index of Fc- function was expressed as a percentage of the reference standard .

Fc-function assay using modified European Pharmacopoeia method Determination of Fc- function of samples of immunoglobulin

containing preparations by a modified EP method was performed as previously described (Georgakopoulos et al (2009)) .

Pepsin digestion and analysis of an immunoglobulin containing preparation

Pepsin was added at 1 . 8 mg/ml to a sample of an IVIG product which was then incubated at 37 °C for 19 hours. The sample was adjusted to pH 7.4 with 0.5 M sodium hydroxide to inactivate the enzyme. Any precipitate was removed by centrifugation at 1000 g for 15 minutes. The protein concentration of untreated and pepsin- treated sample was determined by use of a biuret reagent. Untreated and pepsin-treated samples were loaded at 1 mg/ml on a 4 to 20% tris- glycine gradient gel under non-reducing conditions. When electrophoresis was complete the gel was stained with Coomassie blue for 30 minutes then destained and dried.

Statistical analysis

Statistical analysis was performed using the Student's paired two tail t-test. A P value less than 0.05 was considered statistically- significant .

Results

The twenty amino acid peptide (SEQ ID NO: 1) located between amino acids 595 to 614 on the ppl50 structural protein of CMV has been reported to be immunogenic (Novak et al (1991) , Ripalti et al

(1994)) . Figure 2 illustrates the haemolysis curves generated when the EPBRP standard and an intravenous immunoglobulin

preparation (positive control) at 40 mg/ml were incubated for 1 hour with kodecytes prepared with a solution of the construct designated DOPE-Ad-CMG (2) -hCMV2 at a concentration of 0.8 mg/ml. Although, the reaction rate following the addition of 153 CH ₅₀ /ml complement was too rapid the results appeared promising as the haemolysis curves of IVIG and the EPBRP standard were comparable. At a complement concentration of 23 CH ₅₀ /ml, haemolysis was complete within 3 min. Immunoglobulin was used at a concentration of 50 mg/ml in the assay to ensure highest possible availability of anti-CMV for binding to the kodecytes. With appropriate Fc- function assay parameters defined for using kodecytes, studies were undertaken to validate the assay by determining the inter- and intra-assay variation. For validation of a bioanalytical method it is recommended that the precision around the mean value should not exceed 15% of the coefficient of variation (CV) (Anon (not known) ) . The inter-assay precision was determined by testing 16 different immunoglobulin samples in duplicate in six separate assays over a period of two weeks. The Fc indices for all products tested were consistent with the expected values and the CV was between 5 and 12% which satisfied the predefined acceptance criteria of less than 15%. The intra-assay precision was

determined by testing one IVIG sample eight times within the same assay. The Fc index of the IVIG sample reflected the expected value and the CV was found to be 6% which satisfied the acceptance criteria. A comparison of Fc- function determined by the EP method and using kodecytes was performed. The Fc- function of various immunoglobulin products was determined using the two methods. No significant difference was observed in the results by the two methods. Taken together, these results display that the use of kodecytes is a valid approach for determining Fc- function. To establish the inter-batch variation of kodecytes, the Fc- function of various immunoglobulin products was determined using two different batches of kodecytes. The CV for the products tested using different batches of kodecytes was between 1 and 7% which satisfied the acceptance criteria and demonstrated that different batches will generate similar results. The robustness of the kodecytes was established by comparing the Fc indices of various immunoglobulin preparations generated using kodecytes stored for 1, 4, 6, 8 and 10 weeks at 2 to 8 °C. The results revealed comparable Fc indices between each storage time and the CV ranged from 4 to 8% for each product tested. CMV kodecytes can therefore be stored up to 10 weeks at 2 to 8 °C without loss of function. To verify that the Fc- function assay with the kodecytes was specific for intact IgG, an IVIG sample was treated with pepsin overnight to generate F(ab) ₂ and Fc fragments. To confirm fragmentation of IgG, the untreated and pepsin- treated IVIG samples were run on an SDS-PAGE gel (data not shown) . The Fc function of undigested and pepsin- treated IVIG was evaluated using CMV kodecytes. The untreated IVIG sample revealed an Fc index of 99% whereas the pepsin-treated sample showed a significantly lower Fc index of 28%. The low level of complement -induced haemolysis observed with the pepsin- treated sample was most likely due to the presence of a low level of uncleaved IgG (as seen by SDS-PAGE electrophoresis) . These results verified that the Fc- function assay is specific for intact IgG molecules.

Although the invention has been described with reference to embodiments or examples, it should be appreciated that variations and modifications may be made to these embodiments or examples without departing from the scope of the invention. The advantages discussed in the description may be provided in the alternative or in combination in these different embodiments of the invention. Where known equivalents exist to specific features, such

equivalents are incorporated as if specifically referred to in this specification.

REFERENCES

Anon (1997) B cell receptors, immunoglobulin and Fc receptors Klein J, Horejsi V (eds) : Immunology. London, Black ell Science, 204-270.

Anon (2007) Test for Fc function of immunoglobulin European Pharmacopoeia (Monograph 2.7.9), Edition 6, Volume 1. Strasbourg, European Directorate for the Quality of Medicines & Healthcare.

Anon (not known) Guidance for Industry - bioanalytical method validation; http : //www . fda . gov/downloads/Drugs/GuidanceComplianceRegulatorylnforma io n/Guidances/ucm070107.pdf

Beers, M.H. (ed) (2006) Merck Manual of Diagnosis and Therapy 18th Edition, Merck Research Laboratories.

Frame et al (2007) Synthetic glycolipid modification of red blood cell membranes Transfusion Vol 47, 876-882.

Galili et al (1984) A unique natural human IgG antibody with anti-ot- Galactosyl specificity J. Exp. Med Vol 160, 1519-1531.

Georgakopoulos et al (2009a) An improved Fc function assay utilizing CMV antigen-coated red blood cells generated with synthetic function-spacer- lipid constructs Vox Sang Vol 96, 133-137.

Georgakopoulos et al (2009b) An improved method for the determination of Fc function of immunoglobulins Vox Sang 96, 133-137.

Georgakopoulos et al (2011) An improved method for the determination of Fc function of immunoglobulins Vox Sang, 1-7.

Gronski et al (1991) Measurement of Fc function on human immunoglobulin preparations Pharmeuropa 3, 287-292.

Gurevich et al (2006) Determination of Fc function with frozen red blood cells Biologicals 34, 221-222.

Harrison et al (2010) A synthetic globotriaosylceramide analogue inhibits HIV-1 infection in vitro by two mechanisms Glycoconj J Vol 27, 515-24.

Heathcote et al (2010) Novel antibody screening cells, MUT+Mur kodecytes, created by attaching peptides onto red blood cells Transfusion Vol 50, 635-641.

Henry (2009) Modification of red blood cells for laboratory quality control use Curr Opin Hematol Vol 16, 467-72.

Hooper, JA (1997) : Production and properties of intravenous immunoglobulins Lee ML, Strand V (eds) : Intravenous Immunoglobulins in Clinical Practice. New York, Marcel Dekker, 37-55.

Jorguera, JI (2009) Flebogamma 5% DIF development : rationale for a new option in intravenous immunoglobulin therapy Clin Exp Immunol 157, 17 -21. Macher and Galili (2008) The Galal , 3Galbl , 4GlcNAc-R (a-Gal) epitope: A carbohydrate of unique evolution and clinical relevance Biochim. Biophys . Acta Vol 1780, 75-88.

Novak et al (1991) Mapping of serologically relevant regions of human cytomegalovirus phosphoprotein ppl50 using synthetic peptides J Gen Virol Vol 72, 1409-13.

Riepert et al (2004) Evaluating the Fc-Function of Intravenous

Immunoglobulin Products by Flow Cytometry J Allergy Clin Immunol S214, 751

Ripalti et al (1994) Construction of polyepitope fusion antigens of human cytomegalovirus ppUL32: reactivity with human antibodies J Clin Microbiol Vol 32, 358-63.

Schroeder et al (2008) Immunoglobulins : structure and function Paul WE (ed.): Fundamental Immunology. Philadelphia, Lippincott Williams & Wilkins, 125-150.

Virella, G (2005) Biosynthesis, metabolism and biological properties of immunoglobulins Virella G (ed.): Medical Immunology New York, Marcel Dekker, 109-121.

Vlug et al (1994) Nephelometric measurements of human IgG subclasses and their reference ranges Ann Biol Clin 52, 561-567.

von Gunten et al (2009) Intravenous immunoglobulin contains a broad repertoire of anticarbohydrate antibodies that is not restricted to the IgG2 subclass J Allergy Clin Immunol Vol 123, 1268-76.

European Pharmacopoeia (Monograph 2.7.9) Test for Fc function of immunoglobulin in, Edition 6, Volume 1. Strasbourg, European Directorate for the Quality of Medicines & Healthcare, 2007