Preparation and Use of Hyperimmune Serum for Prophylaxis and Therapy of Ebola Virus Infections
- Natalya M. Kudoyarova-Zubavichene,
- Nikolai N. Sergeyev,
- Alexander A. Chepurnov and
- Sergey V. Netesov
- Reprints or correspondence: Dr. Sergey V. Netesov, State Research Center of Virology and Biotechnology “Vector,” 633159 Koltsovo, Novosibirsk Region, Russia (netesov{at}vector.nsk.su).
Abstract
To obtain hyperimmune serum appropriate for the treatment of filovirus infection, methods were developed to immunize nonsusceptible animals with live Ebola (EBO) virus preparations. Immune plasma with high ELISA and neutralization-specific antibody titers was obtained by multiple immunization of sheep and goats with preparations of live EBO virus. Goat immunoglobulin was prepared by Cohn's method and tested on guinea pigs, using an EBO virus strain that is highly pathogenic for guinea pigs. Prophylaxis with these immunoglobulins within 48 h after infection was effective in challenge experiments, with a log10 prophylaxis index as high as 1.92 ± 0.52. Other studies have shown that equine anti-EBO virus immunoglobulins worked well in baboons. The goat immunoglobulins were also tested in preclinical trials on laboratory animals; after being positively evaluated, they were administered to volunteers in clinical trials for biologic safety and reactivity, and they were administered to researchers suspected of becoming infected with EBO during their experimental work. These immunoglobulins may be useful for the emergency treatment of persons accidentally infected with EBO.
After it was realized in 1976 that Ebola (EBO) virus poses a threat to public health, the virus remained a great challenge to virologists. Although research has been conducted for 30 years, there are still no proven methods of prophylaxis and therapy for persons infected with filovirus infections. Regretfully, all attempts to develop effective vaccines against EBO infection have been unsuccessful [1–5]. Thus, there is a need for the development of other antiviral preparations, such as specific antiviral immunoglobulins, which are conventionally used in emergency cases (e.g., to treat accidental infections in laboratory workers).
Herein, we present the results of protective efficacy studies of anti-EBO virus immunoglobulins that were used to treat investigators who were suspected of becoming infected with EBO virus during their experimental work. We also present data from other Russian researchers on the use of equine anti-EBO virus immunoglobulin in baboons. The data concern the development of antifilovirus immunoglobulin preparations and are based on the authors' investigations and other investigations reported in the Russian literature.
Materials and Methods
For the studies, we used the Zaire subtype of EBO virus (EBOZ). It was obtained from the National Collection of Hemorrhagic Fever Viruses of the Virologic Center of the Institute of Microbiology of the Russian Ministry of Defense (Sergiev Possad, Moscow Region). We also used 8ms, a newly selected strain that is pathogenic for guinea pigs (from the virus collection at Vector; described in [5]).
Virus-specific antibodies and antibodies against liver antigen were determined by ELISA as previously described [6, 7]. As antigen, we used purified and concentrated (by ultracentrifugation in a sucrose gradient) filtrate of 1% homogenate of liver from noninfected guinea pigs and SDS-lysate of EBO virus grown on Vero cell culture.
To measure the neutralization titer, we used newborn outbred mice intracerebrally infected with EBO (in vivo test) or infected Vero cell monolayers under agar cover (in vitro test) as described in [8]. Before the cells and animals were infected, the neutralization reaction was done by incubating 10-fold dilutions of EBO virus with undiluted serum or gamma globulin preparations for 18 h at 4–6°C. The virus—nonspecific serum mixture served as a control. The neutralization index was determined as the difference between a log10 virus titer of the experimental and the control data.
A 10% liver homogenate from infected guinea pigs was used to immunize sheep and goats. Animals were usually immunized two times, with the second immunization done 1–3 months after the first. Occasionally, three immunizations were given. Immune blood (500–900 mL) samples were obtained 15–20 days after the last immunization and every 10–15 days thereafter. Plasma was obtained after the CaCl2 precipitation step. EBO immunoglobulin preparations were prepared from the hyperimmune sera by alcohol fractionation according to Cohn's method [9].
Guinea pigs (weight, 200–250 g), baboons (Papio hamadryas; weight, 3–5 kg), and 1- to 2-day-old suckling outbred mice were used. Titers of EBO-Z in suckling mice and of adapted EBO virus (strain 8ms) in guinea pigs were determined by titration [10]. These 2 strains were also titrated by a plaque assay as previously described [8].
The protection index of the immunoglobulin preparation was determined by infecting guinea pigs intraperitoneally with 1 mL of a series of six 10-fold dilutions (100−2-100−7) of virus-containing suspension (4 guinea pigs/dilution). Immunoglobulin or serum preparations were injected intramuscularly (0.1 mL of 12% immunoglobulin solution/guinea pig). The prophylaxis index was determined as the difference between a log10 LD50 for the experimental and the control data.
Hemosorption was performed, using a hemosorbent (FAS; Akrikhin, Moscow), by the vein-venous route through peripheral shoulder veins. A hemosorption rate of 70–80 mL/min was provided by a peristaltic pump (Unirol, Kirishy, Russia); 1–2 vol of circulating blood was passed through the sorbent during each 1- to 2-h procedure.
Equine immunoglobulin was obtained by multiple immunization of horses with a virus-containing cell-culture fluid and 10% homogenate of EBO-infected monkey liver with subsequent purification by alcohol fractionation according to Cohn's method [11]. The final immunoglobulin preparation was a colorless, virtually transparent liquid, pH 7.0–7.5, with 9%–11% total protein content, and it was apyrogenic, nontoxic, and sterile. The protein consisted of 90%–98% γ-globulin and 2%–10% albumin and αand β-globulins. The liquid also contained 3.1% remaining ethanol. The specific activity (the titer of virus-neutralizing antibodies [VNA]) was 1:4096–1:65536, and 6 mL was administered to baboons.
To determine the potential of the immunoglobulins to cause an anaphylactic reaction, we subcutaneously inoculated guinea pigs with 0.1 mg of sheep, goat, or equine immunoglobulin. After 21 days of sensibilization, 1, 2.5, 5, or 10 mg of the corresponding immunoglobulin was intracardially injected into the guinea pigs (10 animals/dose). Anaphylactic reactions were rated as follows: −, no reaction; +, slight reaction (minor scratching); ++, moderate reaction (frequent scratching and sneezing); +++, severe reaction (spasmodic coughing, falling on one side, urinating, defecating); ++++, fatal reaction (convulsions, acute respiratory distress, death on day 5–7).
Ten guinea pigs were used as negative controls. They were sensibilized with 0.9% NaCl solution, and after 21 days, each guinea pig was intracardially inoculated with 100 mg of immunoglobulin. The anaphylaxis index was calculated according to a modification of Kerber's method [12].
Immunoglobulin was administered to human volunteers according to Bezredko's scheme [13]. The first inoculation, a 0.1-mL preparation dissolved in 0.9% NaCl solution (1:100), was injected intracutaneously into an internal site of the shoulder. The second inoculation, a 0.1-mL undiluted preparation, was injected subcutaneously into the external site of the shoulder 20 min later if there were no local or general reactions to the first inoculation. The third inoculation, a 3-mL preparation, was inoculated intramuscularly into the upper half of the buttock 1 h after the second inoculation if there were no local or general reactions. Control volunteers (untreated group) were inoculated with 0.9% NaCl solution as described above.
Results
The treatment of EBO hemorrhagic fever with immune human convalescent plasma was first reported by Emond et al. [14] in 1977. However, sources of human immune plasma are scarce; therefore, antiviral preparations using animals as a source of immune plasma had to be developed. By the end of the 1980s, we and other Russian scientists began to develop the needed preparations using goats and sheep. First, we tested the possibilities of obtaining animal immune plasma by immunization with inactivated virus preparations. However, the ELISA and neutralization—specific EBO antibody titers we measured were unacceptably low, a finding that was in agreement with previously published data [2, 4–5]. Subsequently, we found that sheep and goats were completely insensitive to EBO virus, and therefore, we could use large quantities of live virus for immunization. We continued using live virus from infected guinea pigs as 10% liver homogenate. Virus-neutralizing activity was tested in virusneutralizing experiments on guinea pigs; a specially selected EBO virus strain that is pathogenic for guinea pigs (strain 8ms) was used for this purpose. The strain was first selected and characterized by M. Smolina at the Vector center [5]. Only batches containing VNA with titers >2.5 log10 were used in gamma globulin preparations. As a result of fractionation, the concentrations of VNA were >3 log10 in the final immunoglobulin preparations, as described in [7, 15].
The anaphylactic activity of these preparations at various doses in guinea pigs was as follows: A 0.5-mg dose of equine immunoglobulin caused severe anaphylaxis in 12% of cases, while a 1-mg dose (i.e., 2-fold higher amounts than equine immunoglobulin) of sheep and goat immunoglobulins produced the same reaction; a 2.5-mg dose of equine immunoglobulin caused severe reactions in 50% of the cases, while sheep and caprine immunoglobulins produced a similar pattern at 5 mg/dose; a 10-mg dose of equine immunoglobulin was 100% fatal, while mortality rates were 70% and 50% when sheep and caprine immunoglobulins, respectively, were used. Thus, the equine immunoglobulin preparation was more anaphylactogenic than sheep and caprine immunoglobulins.
Tests of the storage properties of the final immunoglobulin preparations at 4°C showed that the goat preparations can be stored for >5 years without a considerable decrease in their neutralizing activity (all the other parameters of these preparations remained unaltered).
We also analyzed the neutralizing activities of different fractions of goat immune serum (described in [7, 15]). The fractions were obtained by ion-exchange chromatography, and the data are shown in table 1.. As shown, the IgG2 fraction contains the bulk of antiviral antibodies. The IgG1A and IgG1B fractions mainly contain antibodies against liver antigens of guinea pig (see details in [7]). A similar analysis of the final immunoglobulin preparations obtained using Cohn's method showed that they contain mainly the IgG2 antibodies. Furthermore, IgG2 fraction was less reactive than the other two fractions in anaphylaxis tests.
Table 1
Ebola virus neutralizing activity of fractions of goat immune serum.
Goat immunoglobulin preparation was tested on guinea pigs for its ability to protect against EBO virus infection. Immunoglobulin was injected 24, 48, and 72 h before challenge by intraperitoneal inoculation with EBO strain 8mc, which is pathogenic for guinea pigs. From the results, it is clear that the prophylactic index was maximal when the interval between immunoglobulin injection and virus inoculation was ⩽24 h (table 2.). Table 3 presents the results of similar experiments in which immunoglobulin was injected after inoculation with virus. In these cases, the results were positive up to 72 h after virus inoculation.
Table 2
Prophylactic effect in guinea pigs of goat anti-Ebola immunoglobulin inoculated at various times before animals were challenged with Ebola virus.
Table 3
Prophylactic effect in guinea pigs of anti-Ebola immunoglobulin inoculated at various times after animals were challenged with Ebola virus.
Goat preparations were tested in conformity with national regulations for preclinical trials (anaphylactogenicity, apyrogenicity, acute and chronic toxicity, tumorigenicity, effects on hematologic parameters, liver function). Clinical trials on volunteers were started after formal governmental permission was given. Goat immunoglobulin was injected into 7 volunteers (treated group), and placebo was given to 3 volunteers (untreated control group).
There were no local or general reactions at the time of the injection or later. Recorded body temperature, blood pressure, and urine parameters were normal. The time course of changes in the number of lymphocytes and neutrophils (segmental and rod) was stable, while the number of monocytes in the treated group rose to the upper limit of the normal value on day 1 after immunoglobulin inoculation and remained unaltered during the observation period. The number of eosinophils in the treated group was above normal on days 14 and 28. The increase in the number of monocytes and eosinophils was probably due to sensibilization in response to administered foreign protein. The number of T lymphocytes (percent and absolute value) remained unaltered in both groups of volunteers.
The number of erythrocytes and thrombocytes, the hemoglobin concentration, and the activity of alanine aminotransferase, aspartate aminotransferase, and lactate dehydrogenase did not change considerably in the 2 groups.
On day 7, the serum IgG level rose to 7.8 U in the treated group and remained at this level until day 28. Thereafter, it slowly declined to the initial mean level. The serum IgM level also rose on day 7 and then fell during days 14–21. There were no such variations in serum IgG and IgM in the control group. Table 4 presents the data on gamma globulin circulation. Immunoglobulin circulated in the blood of most volunteers up to 14 days after injection.
Table 4
Reverse ELISA titers of goat anti-Ebola immunoglobulins in sera of volunteers at various days after injection.
On the basis of the results, governmental permission was obtained to use this goat immunoglobulin for emergency prophylaxis against EBO infection in humans. The preparation was used subsequently in 4 persons suspected of being infected with EBO. In 2 of the cases, infection with EBO virus was slightly suspected: The persons' hands were injured while washing (using a bleach solution) cages in which EBO-infected animals had died. In a third case, the persons's left fist was injured by a syringe used to obtain a blood sample from a monkey that had been infected with EBO-Z 4 days earlier. The syringe was treated with bleach for 1–2 min before the injury. In these 3 cases, the immunoglobulin administration was combined with injections of recombinant α-2-interferon (two times daily for 14 days). The only symptom of illness was a rise in temperature to 37.3–37.5°C, probably, because of the injected interferon.
The fourth person with a suspected case of accidental EBO infection was an investigator who pricked his left palm with a thick needle during a plasmapheresis procedure on an ill monkey that was infected with EBO-Z. Analysis of blood obtained from this monkey on the day of the accident demonstrated the presence of live virus at 7.8 log10 LD50 for guinea pigs. According to our calculations, the injured person received a virus dose of at least 102-103 baboon LD50. This fourth case is noteworthy because there were other symptoms related to possible EBO infection.
Just after the injury, the investigator pressed blood from his wound and treated the injury with bleach. He was then referred to a hospital, where he received a 6-mL injection of anti-EBO goat immunoglobulin intramuscularly. Recombinant human α-2-interferon was administered two times daily (3 × 106 IU portions) for 12 days. Laboratory studies of 20 biochemical blood parameters were done daily. Four days after the injury, the serum α-amylase level increased 2-fold. On day 5, the size of the inguinal lymph nodes and the liver increased by 2 cm. A rise in body temperature (to 37.0–37.3°C) was recorded from the first day of injury and was presumably due to the interferon treatment. On day 6, the patient's fibrinogen level started to rise, reaching 5.5 g/L (normal, ⩽4); the same was observed for the prothrombin index, which rose to 112% (normal, 100%). The activities of alanine aminotransferase and aspartate aminotransferase also increased. There was erythema at the site of the immunoglobulin injection. Symptoms of illness included myalgia, headache, arthralgia, and malaise.
On days 8 and 10, a hemosorption procedure was done. All the parameters normalized on day 12 after the injury. The patient's general condition improved, and the symptoms disappeared. The patient recovered, and the virus could not be detected and antibodies were not found in his blood.
A retrospective analysis of his case report revealed a confounding pattern. The course of changes in biochemical parameters showed that a cytolytic process of unknown etiology was developing. Clearly, all the symptoms could not be attributed to EBO infection or to the gamma globulin injection. However, the volunteers showed no similar reactions in trials of the preparation.
Similar immunoglobulin technology was developed and protective efficacy studies were done at the Virologic Center of the Institute of Microbiology of the Russian Ministry of Defense. The work was in parallel and partly independent of our studies. An immunoglobulin preparation from immunized horse serum was successfully developed at the Center [3, 16, 17]. The Moscow virologists used a live virus preparation that was similar to ours to immunize horses (described in [11]), and they used a similar immunization schedule (described in [3, 11, 16]). As aresult, they developed immunoglobulin preparations with VNA titers of 1:8000–1:64000 (see [16]). Table 5 presents the first data obtained from tests of equine gamma globulin preparations in P. hamadryas [3]. The preparations protected baboons against low doses of EBO virus (10–30 LD50 by intramuscular injection). The only fatally infected baboon died after a prolonged incubation period on day 18 after infection, which is quite an unusual course for EBO virus.
Table 5
Results of equine immunoglobulin given as prophylaxis against Ebola virus infection in baboons (Papio hamadryas).
In their second and third papers [16, 17], the Moscow investigators gave a more detailed evaluation of their protective efficacy studies of equine gamma globulin in EBO-infected baboons. Pertinent information is given in table 6 (see [16] for details). The preparation with the highest VNA titers provided >80% protection against an EBO virus injection of 10–30 LD50, and the baboons that were injected with immunoglobulin during the first hour after infection did not develop blood and liver viremia. When the interval between virus injection and immunoglobulin inoculation was >1.0–1.5 h, the percentage of baboons that survived was considerably smaller. It should be noted that in this particular case, the incubation period was up to 26 days, which is much longer than the usual 4–7 days.
Table 6
Evaluation of the effectiveness of anti-Ebola virus gamma globulin in baboons (Papio hamadryas) infected by intramuscular injection of Ebola virus (as presented in [17]).
The immunoglobulin preparation met all national standard requirements, and it was evaluated by the State L. A. Tarassevich Institute of Standardization and Control of Medical Immunobiologic Preparations of the Russian Federation (Moscow). Since 1995, there has been governmental permission to use the product in Russia for the emergency treatment of EBO fever.
Discussion
The United States Army Research Institute of Infectious Diseases (Fort Detrick, Frederick, MD) received doses of the equine anti-EBO immunoglobulin and independently conducted protective efficacy studies (described in [18]). The results obtained in cynomolgus monkeys treated with the immunoglobulin and simultaneous inoculations of EBO virus were thought-provoking: The beneficial effects of IgG treatment were limited to a delay in onset of viremia and clinical signs, compared with the timing of symptoms in untreated controls. The discrepancy between these results and those reported by the Moscow investigators in [16, 17] might be explained by the use of other monkey species and large (up to 1000 LD50) infectious doses [18].
Specific gamma globulin administered early after infection. of guinea pigs and hamadryas baboons is effective presumably because small quantities of EBO virus are completely neutralized by antibodies in vivo before the virus starts to interact with cell receptors. It is well known that human hyperimmune anti-tickborne encephalitis (TBE) virus immunoglobulin is similarly effective against low doses of TBE virus and that this preparation has been successfully used for many years in Austria and Russia for urgent prophylaxis of disease after tick bites. As for EBO virus, its presence was first shown in monocytes and macrophages of infected monkeys 48 h after virus inoculation [19]. Specific antibodies against EBO virus after immunoglobulin inoculation of baboons appear in the bloodstream as early as 30 min after infection, reach maximum concentration after 1 h, and continue to circulate for up to 10 days, decreasing in concentration with time. Neutralization of EBO virus by immunoglobulin takes place within 1 h in vitro. Taken together, these facts may explain why the equine anti-EBO immunoglobulin preparation is effective when administered shortly (no later than 1–2 h) after EBO virus infection.
Conclusion
As detailed here, immunoglobulin preparations have been developed that protect guinea pigs and baboons from infection with fatal doses of EBO virus. Occasionally, medical research investigators, medical personnel, and animal care workers working with EBO virus in the hospital, laboratory, or jungle are accidentally infected with EBO virus. To our knowledge, these immunoglobulin preparations are the first that can be used for the prompt treatment (within 72 h of exposure) of such exposures to EBO virus. The results reported here are milestones in the research leading to the prevention of EBO infection. At the least, these results are applicable to accidental cases of EBO infection occurring in a laboratory or hospital setting.
Acknowledgments
We appreciate the efforts of our colleagues: Larissa M. Dedkova, who participated in the fractionation and analysis of the sera and immunoglobulin preparations; Sergey V. Luchko, who performed most of the neutralization experiments; and Margarita P. Smolina, whose significant contribution to the experimental selection of Ebola virus strains pathogenic for guinea pigs facilitated our work. We also thank anonymous reviewers for helpful comments.
Footnotes
-
All animal experiments were done according to the national regulations of the Russian Federation, and all animals received anesthesia during painful procedures.
-
Financial support: Ministry of Science and Technology of the Russian Federation.
- © 1999 by the Infectious Diseases Society of America
References
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 6.288
5-Yr impact factor: 6.103
For The Media
For Authors
- Services for authors
- Self-archiving policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
