Neutralizing high affinity human monoclonal antibodies specific to rsv f-protein and methods for their manufacture and theraputic use thereof

Abstract

A highly efficient method for generating human antibodies in particular which are specific to be RSV fusion protein which combines in vitro primary of human spleen cells and antigen boosting in SCID mice is taught. This method provides for very high human antibody titers which are predominantly of the IgG isotype which contain antibodies of high specificity and affinity to desired antigens. This method is well suited for generating human monoclonal antibodies for therapeutic and diagnostic applications as well as for rescue of human cells for generation of combinational human antibody gene libraries. Two human monoclonal antibodies, RF-1 and RF-2 which each possess an affinity for RSV F-protein <=2x10-9 Molar are taught as well as their corresponding amino acid and DNA sequences. These antibodies are to be used therapeutically and prophylactically for treating or preventing RSV infection, as well as for diagnosis of RSV in analytes.

Description

[0001] Respiratory syncytial virus (RSV) is a Parmixovirus of the Pneumovirus genus which commonly infects the upper and lower respiratory tract. It is so contagious that by age two, a large percentage of children have been infected by it. Moreover, by age four, virtually all humans have an immunity to RSV.[0002] Typically, RSV infections are mild, remaining localized in the upper respiratory tract and causing symptoms similar to a common cold which require no extensive treatment. However, in some subjects, e.g., immunosuppressed individuals such as infants, elderly persons or patients with underlying cardiopulmonary diseases, the virus may penetrate to the lower respiratory tract requiring hospitalization and breathing support. In some of these cases, RSV infection may cause permanent lung damage or even be life threatening. In the United States alone, RSV results in about 90,000 hospitalizations each year, and results in about 4500 deaths.[0003] RSV appears in two major strain sub...

AI Technical Summary

Benefits of technology

[0038] It is another object of the invention to provide an improved method for producing EBV immortalized B-cells which favors the formation of EBV immortalized B-cells which predominantly secrete IgG.

[0069] More specifically, the present invention provides two particular human monoclonal antibodies to the RSV F-protein, i.e., RF-1 and RF-2, as well as recombinant human antibodies derived therefrom, which are preferably produced in CHO cells, which cells have been transfected with DNA sequences encoding the variable heavy and light domains of RF-1 or RF-2. These antibodies are particularly useful as prophylactic and / or therapeutic agents for treatment or prevention of RSV infection. Moreover, these antibodies are useful as diagnostic agents because they bind the RSV F-protein with high affinity, i.e., each possess affinity for the RSV F-protein of .ltoreq.2.times.10.sup.-9. They are especially useful as therapeutic agents because of their high affinity and specificity for the RSV F-protein, and their ability to effectively neutralize RSV infection in vitro.

[0090] The subject method for producing human monoclonal antibodies essentially involves the combination of in vitro priming of naive human spleen cells, transferral of these spleen cells to immunocompromised donors, i.e., SCID mice, followed by antigen boosting of SCID mice which have been administered said spleen cells. It has been surprisingly discovered that the combination of these two known methods for producing human antibodies results in synergistic results. Specifically, it results in very enhanced antigen specific responses to the immunizing antigen as well as very high titers of human monoclonal antibodies of the IgG isotype. More specifically, it has been found that this combination results in unprecedented high secondary responses: the human IgG responses in the hu-SPL-SCID serum were 10-fold higher than those resulting from transfer of naive cells in SCID and specific antibody responses were 1000-fold increased. Also, the resulting antibodies are found to be of high affinity and specificity comparable to antibodies produced in experimentally hyperactive immune animals. It has also been found that when using naive spleen cells, to obtain such unexpected results it is necessary to challenge with antigen both in vitro and after introduction into the resultant hu-SPL-SCID mouse. Also, it is preferable but not essential to introduce additional fresh non-primed spleen cells to the hu-SPL-SCID donor just prior to antigen boosting. This has been found to result in still further enhancement of the antibody response.

[0094] Therefore, having established an optimal in vitro primary and boosting protocol for generation of secondary responses from naive human spleen cells; it was conceived to test this protocol in combination with previous in vivo methods for producing human monoclonal antibodies, i.e., the SCID mouse. It was unknown prior to testing what effect, if any, administration of antigen primed spleen cells would have on the resultant production of human monoclonal antibodies to a given antigen by the SCID mouse or the ability of human lymphocytes to be maintained therein. However, it was hoped that this would provide for enhanced antigen boost and enhanced expression of the in vitro antigen primed naive spleen cells.

[0117] These results are highly significant and indicate that it should generally be possible to rescue human cells from the hu-SPL-SCID and use same for generating combinatorial human antibody gene libraries thereby resulting in human monoclonal antibodies of high affinity and specificity that may be used clinically and / or diagnostically.

[0120] As discussed, it has been discovered that the combination of in vitro immunization, in particular of human splenocytes, i.e., which have or have not been previously exposed to the RSV F-antigen and transferred to an immunocompromised animal donor, i.e., SCID mouse which is then boosted with RSV F-protein antigen affords significant advantages relative to conventional methods for making human antibodies in SCID mice. Namely, it provides for very high antibody titers, i.e., the highest anti-F protein titers being about 10.sup.-7, high IgG concentrations, i.e., about 3 mg / ml for the highest responders. Moreover, this method allows for the production of human antibodies having highly advantageous combinations of properties, i.e., which exhibit both high affinity to the RSV F-protein and which moreover display substantial in vitro neutralizing activity.

Problems solved by technology

In some of these cases, RSV infection may cause permanent lung damage or even be life threatening.

In the United States alone, RSV results in about 90,000 hospitalizations each year, and results in about 4500 deaths.

As noted, while RSV infections are usually mild, in some individuals RSV infections may be life threatening.

However, while Ribavarin exhibits some efficacy in controlling RSV infection, its use is disfavored for several reasons.

For example, it is highly expensive and may be administered only in hospitals.

To date, RSV vaccines intended to boost antiviral protective antibodies have been largely unsuccessful.

Therefore, the results reported with the humanized and human anti-RSV antibodies are not directly comparable.

However, while rodent monoclonal antibodies possess therapeutic efficacy, they can present restrictions and disadvantages relative to human antibodies.

This may result in shortened antibody half-life Dillman et al.

(1986), 5, 73-84, Miller et al, Blood, (1983), 62:988-995 and in some instances may cause toxic side effects such as serum sickness and anaphylaxis.

By contrast, in patients with normal or hyperactive immune systems, murine antibodies, at least for some disease conditions may exhibit limited efficacy.

However, although such antibodies have been used successfully clinically Gillis et al.

This is because the understanding of the requirements for optimal antigen recognition and affinity is not yet fully understood.

While human antibodies are highly desirable, their production is complicated by various factors including ethical considerations, and the fact that conventional methods for producing human antibodies are often inefficient.

For example, human subjects cannot generally be adequately immunized with most antigens because of ethical and safety considerations.

Also, isolation of anti-viral human monoclonal antibodies from donor primed cells has proved to be unwieldy.

However, such antibodies often react with intracellular, and thus therapeutically useless antigens Ho et al, In Hybridoma Technology, Amsterdam (1988), 37-57 or are of the IgM class McCabe et al, Cancer Research (1988), 48, 4348-4353, a class of antibodies with lesser ability to penetrate solid tumors than IgGs.

Moreover, since these approaches exploit the testing donor primed B cells, it is clear that these cells are not an optimal source for rescue of useful monoclonal antibodies.

However, unfortunately, the resulting antibodies were typically of the IgM and not the IgG subclass McCabe et al, Cancer Research (1988), 48, 4348-4353, Koda et al.

Therefore, it would appear that these protocols only succeed in inducing a primary immune response but require donor immunized cells for generation of recall responses.

However, despite the great potential advantages of IVI, the efficiency of such methods are severely restricted because of the fact that immune cells grow in monolayers in culture vessels.

However, this protocol required donor primed cells and the yield was very low, only 2 clones were obtained from a library of 370,000 clones.

However, use of both of these protocols are substantially restricted.

However, this method is severely restricted by the limited availability of fetal tissue, as well as the complicated surgical methodology of the protocol McCune et al. L. Science (1988), 241, 1632-1639.

However, this method is disadvantageous because it requires a four month incubation period.

Moreover, both protocols result in very low antibody titers, i.e., below 10.sup.4.

However, EBV transformed cells are typically unstable, produce low amounts of mainly IgM antibody, clone poorly and cease making antibody after several months of culturing.

However, a disadvantage is the poor yield, i.e., .ltoreq.1 hybridomas per 20,000 lymphocytes Boerner. et al.

However, the yield is typically extremely low, on the order of 1 per 370,000 clones Duchosal, et al, Nature (1992), 355:258-262.

Images