Research ArticleVaccines

DNA vaccine–derived human IgG produced in transchromosomal bovines protect in lethal models of hantavirus pulmonary syndrome

See allHide authors and affiliations

Science Translational Medicine  26 Nov 2014:
Vol. 6, Issue 264, pp. 264ra162
DOI: 10.1126/scitranslmed.3010082


Polyclonal immunoglobulin-based medical products have been used successfully to treat diseases caused by viruses for more than a century. We demonstrate the use of DNA vaccine technology and transchromosomal bovines (TcBs) to produce fully human polyclonal immunoglobulins (IgG) with potent antiviral neutralizing activity. Specifically, two hantavirus DNA vaccines [Andes virus (ANDV) DNA vaccine and Sin Nombre virus (SNV) DNA vaccine] were used to produce a candidate immunoglobulin product for the prevention and treatment of hantavirus pulmonary syndrome (HPS). A needle-free jet injection device was used to vaccinate TcB, and high-titer neutralizing antibodies (titers >1000) against both viruses were produced within 1 month. Plasma collected at day 10 after the fourth vaccination was used to produce purified α-HPS TcB human IgG. Treatment with 20,000 neutralizing antibody units (NAU)/kg starting 5 days after challenge with ANDV protected seven of eight animals, whereas zero of eight animals treated with the same dose of normal TcB human IgG survived. Likewise, treatment with 20,000 NAU/kg starting 5 days after challenge with SNV protected immunocompromised hamsters from lethal HPS, protecting five of eight animals. Our findings that the α-HPS TcB human IgG is capable of protecting in animal models of lethal HPS when administered after exposure provides proof of concept that this approach can be used to develop candidate next-generation polyclonal immunoglobulin-based medical products without the need for human donors, despeciation protocols, or inactivated/attenuated vaccine antigen.


Sin Nombre virus (SNV) and Andes virus (ANDV) are the two most prominent hantaviruses in North and South America, respectively (14). These viruses are the etiological agents of hantavirus pulmonary syndrome (HPS), also known as hantavirus cardiopulmonary syndrome, characterized by vascular leakage, resulting from infection and dysfunction of the endothelium, leading to tachypnea, shock, pulmonary edema, and cardiac failure (5, 6). The time from exposure to first symptoms is up to 2 weeks (7, 8), but the progression from first symptoms to severe disease is often measured in hours (6). The high case fatality rate (35 to 40%) associated with these emerging viruses highlights the need for medical countermeasures. However, at this time, there are no vaccines or drugs licensed, or in advanced development, to prevent or treat HPS (9).

Hantaviruses are trisegmented (S, M, L), negative-sense, RNA viruses (10). We have previously developed DNA vaccines to the M segment (glycoproteins; Gn and Gc) of ANDV and SNV and demonstrated immunogenicity in rabbits, ducks, and nonhuman primates (1114). Passive transfer of DNA vaccine–derived neutralizing antibodies from each of these species was efficacious in preventing infection or disease in hamster hantavirus models (11, 15, 16).

ANDV is unique among hantaviruses in its capacity to spread person to person (17, 18), resulting in clusters of cases, usually among close contacts. Recently, researchers in Chile found a medically significant beneficial effect of using convalescent plasma from HPS survivors to treat HPS (19). This suggests that a polyclonal antibody approach to treatment is feasible. However, the paucity of immune plasma and the necessity to match plasma to blood types, as well as other caveats associated with the use of human tissue, warrant the development of a neutralizing antibody–based product that can be manufactured without a need for infectious virus or human use.

SNV is not associated with person-to-person transmission; however, this virus is highly pathogenic (case fatality rate, 30 to 40%) and was responsible for the 2012 Yosemite National Park outbreak (20). The absence of vaccines, post-exposure prophylactics, or therapeutics to prevent HPS caused by SNV contributed to the angst experienced by about 270,000 Yosemite visitors who received notification of possible exposure (21).

Transchromosomal bovines (TcBs), when vaccinated, produce both fully human immunoglobulins (IgG) in addition to chimeric antibodies containing the human γ heavy chain and bovine κ light chain (22). Fully human polyclonal IgG can be purified from bovine serum or plasma, creating a highly potent source of neutralizing antibodies. A distinct advantage with the production of antibodies using TcB is that the IgG are fully human and no enzymatic treatment is needed to eliminate the risk of anaphylaxis and serum sickness associated with heterologous species IgG. Moreover, TcB human IgG have a longer half-life and also retain effector functions associated with the Fc that are removed by enzyme treatment.

Here, we demonstrate the possibility of using DNA vaccine technology to produce high-titer α-HPS neutralizing antibodies in TcB. The antibodies specifically target both ANDV and SNV, illustrating the capacity to use single TcB animal to produce antibodies aimed at multiple pathogens. Furthermore, we show that purified α-HPS TcB human IgG produced using this platform protect when used as a post-exposure prophylactic in two animal models of lethal HPS.


TcB vaccination responses

We have previously reported the construction and testing of first-generation ANDV DNA vaccines (14) and SNV DNA vaccines (13) that encode the full-length glycoproteins of ANDV and SNV, respectively. Here, our goal was to co-administer the ANDV and SNV DNA vaccine plasmids to TcB to produce human neutralizing antibodies against both ANDV and SNV. With the PharmaJet Stratis needle-free disposable syringe jet injection device, two TcBs carrying the HAC vector labeled as KcHACD in the triple KO background (23) were vaccinated four times, intramuscularly, with both the ANDV and SNV DNA vaccines at separate locations. To evaluate whether an adjuvant plus immune stimulator would enhance immune response to the vaccine, TcB #2 was administrated ISA 206 adjuvant plus the saponin-derived immune stimulant Quil A adjacent to each DNA vaccination site at the final booster [fourth vaccination (V4)]. As a control, TcB #1 was given a DNA vaccine without adjuvant for V4. Serum samples and large volumes of plasma were collected before the initial vaccination and after the second, third, and fourth vaccinations (Fig. 1A). Neutralizing antibody titers were determined using a pseudovirion neutralization assay (PsVNA) and a classical plaque reduction neutralization test (PRNT) (Fig. 1, B and C, and table S1). Both bovines developed neutralizing antibodies against ANDV and SNV. Titers decreased with time after the second vaccination, with higher responses after the third and fourth boost. The neutralizing activity against both ANDV and SNV was similar, and both the PsVNA and PRNT resulted in the same trends in neutralizing activity. α-ANDV titers from TcB sera eventually surpassed the α-ANDV titers in human convalescent fresh-frozen plasma (FFP) used in previous passive protection experiments (for example, TcB #1 PRNT80 = 10,240 versus human FFP PRNT80 = 7240) (11).

Fig. 1. Neutralizing antibody responses in TcBs vaccinated with hantavirus DNA vaccine plasmids.

Two (TcB #1 and TcB #2) were vaccinated with an HPS vaccine consisting of an ANDV DNA vaccine and an SNV DNA vaccine component. (A) Schedule of vaccinations V1 to V4 (black arrows) and blood collection (red arrows). ANDV and SNV neutralizing antibody assays were performed on sera collected before vaccination (week 0) and on the indicated weeks after the first vaccination. (B and C) PsVNA80 (B) and PRNT80 (C) titers for ANDV (red symbols) or SNV (blue symbols) for both animals. The cutoff for positive titers is 20. The α-ANDV titer of convalescent FFP from an HPS survivor is shown (red triangle).

Neutralizing activity of purified TcB human IgG

Fully human IgG (referred hitherto as TcB human IgG) were purified from plasma collected from one of the immunized animals (TcB #2). Human IgG and bovine IgM capture enzyme-linked immunosorbent assay (ELISA) and SDS–polyacrylamide gel electrophoresis (SDS-PAGE) indicated that about 97% of the immunoglobulin was human IgG and 3% was IgM (fig. S1). The human IgG isotype subclass distribution of IgG1, IgG2, IgG3, and IgG4 was 54, 44, 1.1, and 0.37%, respectively, comparable to in human IVIG (intravenous immunoglobulin): 67, 26, 2.6, and 2.5%, respectively. The purified TcB human immunoglobulin (8.42 mg/ml) was evaluated for α-SNV (Fig. 2A) and α-ANDV (Fig. 2B) neutralizing activity in the PsVNA and the PRNT. High levels of both α-ANDV and α-SNV neutralizing antibodies were detected. α-ANDV convalescent FFP was used as a positive control for α-ANDV activity. The purified TcB human IgG had higher α-ANDV activity than the FFP in both the PsVNA and the PRNT. Also, whereas the FFP had only low levels of cross-neutralizing activity against SNV, the purified TcB human IgG had potent neutralizing activity against both ANDV and SNV. To confirm that the human IgG was the major contributor to the neutralizing activity, we depleted IgG from the purified immunoglobulin (8.42 mg/ml) by incubating with protein G–labeled magnetic beads. This process removed IgG from the sample as visualized by SDS-PAGE (fig. S1). Pre- and post-depletion samples were evaluated for α-ANDV neutralizing activity in the PsVNA. The anti-ANDV neutralizing activity in the IgG-depleted sample was reduced from PsVNA80 titer of 4565 to 63, demonstrating that the IgM contribution to neutralizing activity was minimal (fig. S1).

Fig. 2. Neutralizing activity in purified IgG from TcB.

IgG antibodies were purified from plasma collected from TcB #2. α-HPS TcB human IgG (8.42 mg/ml) (solid bar), positive control α-ANDV convalescent FFP (stippled bar), and negative control normal TcB IgG (9.04 mg/ml) (empty bar) were evaluated for neutralizing antibody activity by PsVNA and PRNT. (A and B) α-SNV (A) and α-ANDV (B) PsVNA80 and PRNT80 titers. Samples with neutralizing antibody titers >1000 are considered to contain “high-titer” antibodies (dashed line). Samples with titers below 20 are denoted with a “<” symbol.

Bioavailability of TcB human IgG in Syrian hamsters

ANDV causes a disease in Syrian hamsters that closely resembles HPS in humans (23). Similarly, SNV causes HPS in Syrian hamsters that have been transiently immunosuppressed (15). Before testing the TcB human IgG for protection in the hamster models, we evaluated the bioavailability of this candidate product in hamsters. Groups of hamsters were injected subcutaneously with either a high dose {α-ANDV [64,000 neutralizing antibody units (NAU)/kg] equating to α-SNV human IgG (77,000 NAU/kg), or 52.63 mg/kg} or low dose [α-ANDV (12,000 NAU/kg) equating to α-SNV human IgG (14,000 NAU/kg), or 9.87 mg/kg] of purified fully human IgG from TcB #2. Serum samples were obtained on the indicated day after injection until neutralization activity was no longer detected. α-ANDV and α-SNV neutralizing antibody titers were determined by PsVNA (Fig. 3 and table S2). Hamsters administered a high dosage of α-ANDV and α-SNV human IgG had detectable titers against both viruses out to day 35, whereas hamsters administered a low dosage of fully human IgG only had detectable titers out to day 6. Half-lives were calculated to be 5.8 and 6.8 days for α-ANDV and α-SNV, respectively.

Fig. 3. Bioavailability of purified α-HPS TcB human IgG in Syrian hamsters.

Three hamsters each were injected subcutaneously with either a high dosage [α-ANDV (64,000 NAU/kg)/α-SNV (77,000 NAU/kg), solid lines] or low dosage [α-ANDV (12,000 NAU/kg)/α-SNV (14,000 NAU/kg), dashed lines] of purified α-HPS TcB human IgG. The dosage was based on α-ANDV and α-SNV neutralizing activity. Sera collected from hamsters on the indicated days after injection were analyzed for α-ANDV and α-SNV neutralizing titer by PsVNA. Mean titers ± SE are shown. The dashed line indicates limit of detection for the assay (PsVNA80 ≤20).

Protection from lethal ANDV challenge

To determine the protective efficacy of α-HPS TcB human IgG in a model of lethal HPS caused by ANDV, hamsters were challenged with 200 plaque-forming units (PFU) of ANDV. One group was then administered purified α-HPS TcB human IgG (16.45 mg/kg) (α-ANDV activity, 20,000 NAU/kg), subcutaneously, on days 5 and 8 after challenge. Negative control hamsters were injected with normal TcB human IgG (16.45 mg/kg) purified from the same, but prevaccinated, TcB. Positive control hamsters were administered a single dose (12,000 NAU/kg) of previously described α-ANDV rabbit sera produced using the ANDV DNA vaccine on day 5 (16). A fourth group of eight hamsters was exposed to virus but not treated with antibody.

Seven of eight hamsters treated with α-HPS TcB human IgG survived with no signs of disease (n = 8; P = 0.0025, Kaplan-Meier with log-rank test, when compared to normal TcB human IgG). Similarly, seven of eight hamsters receiving the positive control rabbit sera survived (n = 8; P = 0.0181, Kaplan-Meier with log-rank test, when compared to no antibody control). In contrast, all eight hamsters treated with the normal TcB human IgG developed HPS disease and succumbed 10 to 13 days after challenge. Only one of eight hamsters in the untreated group survived (Fig. 4A). All surviving hamsters had positive α-nucleocapsid (N) ELISA titers 28 days after challenge, indicating a productive ANDV infection (Fig. 4B). The results of this experiment demonstrated that the α-HPS TcB human IgG was capable of protecting against ANDV when used as a post-exposure prophylactic.

Fig. 4. Efficacy of α-HPS TcB human IgG to protect against lethal HPS caused by ANDV when administered after exposure.

Four groups of eight hamsters were challenged intramuscularly with ANDV. (A) On days 5 and 8 after exposure, one group of hamsters was injected subcutaneously with α-HPS TcB human IgG [containing α-ANDV (20,000 NAU/kg)]. A second group was injected with the purified normal TcB human IgG. A third group of hamsters was injected with α-ANDV (12,000 NAU/kg) produced in rabbits (Rb) on day 5 as published previously (16). A fourth group of eight hamsters was not treated (No antibody). P values were determined using Kaplan-Meier survival analysis with log-rank tests. (B) Sera collected from surviving hamsters on day 28 were tested by ELISA for evidence of ANDV infection. Symbols represent antibody titers of individual animals.

Protection from lethal SNV challenge

We recently described a lethal HPS animal model involving SNV infection of transiently immunosuppressed hamsters (15). Using this model, we determined whether the α-HPS TcB human IgG could protect against lethal disease caused by SNV. Groups of immunosuppressed hamsters were challenged with 2000 PFU of SNV. One group of hamsters was then administered purified α-HPS TcB human IgG (21.9 mg/kg) (α-SNV activity, 20,000 NAU/kg) on days 5 and 8 after challenge. Negative control hamsters were given normal TcB human IgG (21.9 mg/kg) purified from the same, but prevaccinated, TcB. Positive control hamsters were administered previously described α-SNV rabbit sera (20,000 NAU/kg) produced using the SNV DNA vaccine (13), and a fourth group of hamsters was not injected with antibody.

Five of eight hamsters receiving the α-HPS TcB human IgG survived (n = 8; P = 0.0036, Kaplan-Meier with log-rank test, when compared to normal TcB human IgG). The same survival rate was observed for the positive control group (n = 8; P = 0.0070, Kaplan-Meier with log-rank test, when compared to no antibody control). In contrast, seven of eight of the hamsters injected with either the normal TcB human IgG group or left untreated group succumbed. Moreover, α-HPS TcB human IgG treatment resulted in an increased mean time to death compared to normal TcB human IgG (17 to 18 days versus 12 to 15 days) (Fig. 5A). This is a statistically significant delay in death (n = 8; P = 0.0179, Kaplan-Meier with log-rank test, when compared to normal TcB human IgG; n = 8; P = 0.0157, Kaplan-Meier with log-rank test, when compared to untreated controls). The results of this experiment demonstrated that the α-HPS TcB human IgG was capable of protecting against SNV when used as a post-exposure prophylactic.

Fig. 5. Efficacy of α-HPS TcB human IgG to protect against lethal HPS caused by SNV when administered after exposure.

Four groups of eight hamsters each were immunosuppressed with a combination of dexamethasone and cyclophosphamide described previously (15) and challenged intramuscularly with SNV. (A) On days 5 and 8, one group of hamsters was injected subcutaneously with α-HPS TcB human IgG [containing α-SNV (20,000 NAU/kg)]. A second group was injected with purified normal TcB human IgG. A third group was injected on day 5 only with α-SNV (20,000 NAU/kg) produced in rabbits (13). A fourth group remained untreated (No antibody). P values were determined using Kaplan-Meier survival analysis with log-rank tests. (B) Lung tissue isolated on day 28 after infection was evaluated for viral genome by reverse transcription polymerase chain reaction (RT-PCR). Symbols represent viral genome detected in individual animals. P value was determined by Student’s t test.

Continuous cyclophosphamide treatment precludes the hamster from mounting an antibody response to SNV infection. Therefore, lung tissue collected from surviving hamsters on day 28 was used to evaluate the levels of SNV viral genome detected by RT-PCR (Fig. 5B). Lung tissue from hamsters receiving α-SNV rabbit sera contained similar levels of viral genome as seen previously (15). Hamsters receiving α-HPS TcB human IgG had significantly reduced viral genome detected (n = 5; P = 0.0079, t test).


Historically, polyclonal antibody therapy with immune serum from immunized animals has been used for the treatment of infectious diseases (24). However, with the emergence of antibiotic therapy and the relatively high toxicity associated with animal-derived antibody products, the development of anti-infective polyclonal antibody products in animals was largely abandoned. Recently, there has been a great deal of interest in the development of monoclonal antibody (mAb)–based drug products targeting inflammation, cancer, toxins, and infectious diseases (25, 26). One of the limitations of mAbs to treat infectious disease is the emergence of antibody escape variants and breakthrough disease. One way to reduce the chances of escape is to identify multiple mAbs targeting a particular virus, or viruses, and then combine those antibodies into a cocktail. These combinations can also show increased breadth and potency of activity. For example, a combination of three broadly neutralizing HIV antibodies delayed viral rebound after cessation of antiretroviral therapy (27). mAb cocktails, specifically ZMAb and MB-003, have also been evaluated for the treatment of lethal Ebola virus disease in both mouse and nonhuman primate models (2831). In those studies, the three-mAb cocktails were found to confer protection when used as post-exposure prophylactics and, in the case of MB-003, as a therapeutic.

With the recent successes of mAb cocktails, the question arises: can safe and potent polyclonal antibody therapies be developed? Furthermore, can safe polyclonal antibody products be developed without the need for human donors, despeciation protocols, or inactivated/attenuated virus vaccine antigen? Currently, there are several U.S. Food and Drug Administration (FDA)–approved immune globulin products that are animal-derived: whole polyclonal IgG derived from equine and rabbit species (3234), or purified F(ab)2 or Fab fragments fractionated from polyclonal IgG by either pepsin or papain digestion from equine or ovine species (3538). Products from human donors include plasma-derived immunoglobulins against vaccinia virus, cytomegalovirus, varicella virus, hepatitis B virus, hepatitis A virus, and measles virus. For these products, the source of human IgG is convalescent plasma, or plasma from humans vaccinated with a licensed vaccine (that is, vaccinia immune globulin). Whereas there are no antiviral polyclonal antibodies produced in vaccinated animals, there are several FDA-licensed polyclonal antibody products against bacteria, toxins, and venoms produced in horses or sheep. In general, polyclonal antibody products made in species other than human require a despeciation step to remove the Fc region of the heterologous antibody, thereby decreasing toxicity. However, removing the Fc region significantly reduces the half-life of antibody in vivo. In all scenarios involving animal-derived polyclonal antibodies, lot-to-lot consistency of the product is also a challenge. Nevertheless, the relative simplicity and low cost of polyclonal antibody–based product development is the reason that the only specific treatment for many rare afflictions (such as snake bites, scorpion stings, and toxin exposure) consists of animal plasma–derived polyclonal antibody–based products.

Here, we demonstrate proof of concept that it is possible to combine DNA vaccine technology with the TcB platform to produce polyclonal human IgG specifically targeting multiple viruses. In a relatively straightforward process, it was possible to synthesize the vaccine, vaccinate the TcB, collect large volumes of plasma, and purify human polyclonal antibodies with extraordinarily potent neutralizing activity against two different viruses. Humans who have survived HPS are not required for this product, and although it is made in an animal, there is no need for a despeciation step. Using a DNA vaccine approach resulted in consistent patterns of neutralizing antibody response against ANDV and SNV as measured by both the PsVNA and PRNT.

The vaccine used to produce the human IgG in TcB is purified plasmid DNA. This approach allowed us to specifically target the virus glycoproteins without the need to modify the proteins (such as remove transmembrane regions and produce His-tagged fusion proteins) or develop specific protocols for virus glycoprotein purification. Because we were not using a conventional killed or attenuated vaccine grown in mammalian cell culture, the possibility that the product will contain antibodies against cell culture contaminants was eliminated. The same DNA vaccine plasmids (pWRG/AND-M[opt2] and pWRG/SN-M[opt]) used in this process are advancing toward clinical trials as an active vaccine for HPS. The flexibility of the DNA vaccine/TcB platform allows for the rapid design and development of human polyclonal formulations against multiple viruses as long as the DNA vaccine elicits high-titer neutralizing antibodies.

The PharmaJet IM Stratis device was recently shown to effectively deliver hantavirus DNA vaccines to both rabbits and nonhuman primates (39). Here, DNA vaccination of the TcB using the Stratis device elicited high-titer neutralizing antibodies to two viruses after the second vaccination. This device is FDA 510(k)–cleared and can be used in the field without the need for an electric or compressed gas power source. Moreover, the vaccination procedure is relatively painless, and the animals did not need to be anesthetized before vaccination. It is possible that other means of DNA delivery, such as intramuscular electroporation, could also generate even higher levels of neutralizing antibody titers. Also, we used the ISA 206/Quil A adjuvant for a single vaccination in this study. The increased titer after that boost suggests that this veterinary-use adjuvant has the potential to significantly enhance the immune response to DNA vaccines.

Here, the TcB human immunoglobulin (8.4 mg/ml) was dosed at 20,000 NAU/kg delivered to hamsters on days 5 and 8 after exposure. This is equivalent to 16 mg/kg for a 70-kg human. Previous work with α-ANDV FFP in the hamster model indicated that 12,000 NAU/kg was a conservative protective dose (11). We can predict that if 12,000 NAU/kg is sufficient to confer clinical benefit, then a dose as low as 9.6 mg/kg might be effective. The TcBs used in this study produced 68% (TcB #1) and 79% (TcB #2) fully human IgG before purification. Purification increased the percent of fully human IgG to 97%. The manufacturing process for TcB-derived human IgG to be used in preclinical studies and future clinical studies includes a polishing step (Q Sepharose chromatography) to remove residual bovine IgM from the final Tc human IgG product. TcB produces fully human IgG (up to 15 g/liter), and 30 to 60 liters of plasma can be collected per animal per month. This platform can be scaled by the addition of animals depending on need. It is expected that optimization of vaccine dose, delivery, schedule, and adjuvant formulation will increase the potency of the product. Also, further manipulation of the TcB genome by additional gene transfer and/or the insertion of mutations in the Fc might increase not only levels of fully human IgG but also IgG half-life (40).

ANDV and SNV are considered NIAID (National Institute of Allergy and Infectious Diseases) category A agents, which means that they have characteristics (such as high lethality and low levels of immunity in the population) that make them potential biological weapons threats. Others have proposed that polyclonal antibody products could be used to provide “immediate immunity” and could serve as a relatively inexpensive component of a strategy to deter the nefarious use of these agents (41). In addition to the use of the α-HPS human IgG as a post-exposure prophylactic, it is possible that higher and/or more frequent dosing might confer clinical benefit after onset of disease. Therapeutic utility of the α-HPS human IgG might be realized by combined treatment with broad-spectrum antiviral drugs such as ribavirin or T-705. A path to licensure for a biologic to prevent or treat HPS could involve use of the FDA “Animal Rule.” The ANDV/hamster model, and more recently the SNV/immunosuppressed hamster model, closely mimics human disease and can be used to formally demonstrate efficacy of the α-HPS product. The recent discovery that SNV isolated directly from the natural reservoir (that is, deer mouse) causes lethal disease in macaques could provide an additional model to test, and possibly license, products targeting HPS (42). An alternative path to licensure could involve testing the candidate product for clinical benefit under an Investigational New Drug Application Expanded Access protocol in HPS patients in endemic regions of disease (such as Chile or the southwestern United States).

As with any antiviral treatment, it is important to control for nonspecific protection attributable to the innate immune response. Here, normal purified TcB human immunoglobulin was included as a control in both the ANDV and SNV challenge experiments. This material was collected, before vaccination, from the same animal used to produce the α-ANDV TcB human immunoglobulin. Ideally, the amount and isotype of immunoglobulin would be identical in the test article and control in a passive transfer experiments. However, because the amount of residual IgM in the control sample (0.84%) was less than that in the α-HPS TcB human immunoglobulin (3.07%), we cannot rule out the possibility that residual IgM could trigger a difference in the innate response to the treatment. There was no significant protection conferred by the control antibody in either the ANDV or SNV challenge experiments. The late time point (day 5) when α-HPS TcB human immunoglobulin was first administered makes it unlikely that the innate immune response contributed to the observed protection. Even strong inducers of innate immunity fail to protect hamsters against lethal HPS caused by ANDV if administered later than 3 days after challenge. For example, a VSV-vectored Ebola vaccine elicited a protective innate immune response that protected hamsters against lethal HPS caused by ANDV, but only if administered on day 3 or earlier. Conversely, administration of the vaccine on day 5 or later did not protect hamsters from lethal HPS (43). One limitation of this study was that the mechanism of protection against HPS in the hamster model was not addressed. Although we demonstrated that the purified TcB-derived IgG has potent ANDV and SNV neutralizing activity in vitro, we did not perform serial blood collections to measure the effects of antibody treatment on viremia, and we did not perform serial pathology experiments to monitor evidence of disease. The onset of viremia is dependent on the route of virus challenge. After intramuscular challenge, viremia is detected on day 6 and the mean day to death is day 11 (44), whereas after intranasal challenge, viremia is detected later and the mean day to death is day 17 (16). Consistent with differences in the onset of viremia, passive transfer of antibody protected against intramuscular challenge if antibody was administered starting on day 5 or earlier, whereas after intranasal challenge, passive transfer was successful when starting later (day ≤8) (11, 14, 16). On the basis of these findings, we hypothesize that the polyclonal neutralizing antibody administered before viremia confers protection by limiting or delaying widespread dissemination of virus to the endothelium. The most likely mechanism of in vivo neutralization is the binding of the antibodies to the virus envelope glycoproteins, preventing entry into target cells. For potential licensure under Animal Rule, determining the pathophysiological mechanism through serial pathology experiments will likely be required.

Although this proof-of-concept work targeted HPS, the implications are more wide-ranging. DNA vaccines eliciting protective antibodies to many infectious agents and toxins have been reported. Our work indicates that it is possible to combine this class of vaccine with the TcB platform to produce purified fully human IgG with functional activity (such as neutralizing antibodies), capable of conferring protection in lethal disease models when administered after exposure. Many of the polyclonal antibody–based products that are currently licensed target infectious diseases and toxins for which DNA vaccines have been described. The combination of these DNA vaccines and the TcB platform might allow the production of protective human IgG products without the need for human donors, heterologous animal species, or conventional yet more cumbersome vaccine methodology.


Study design

This study evaluated the potential of TcB vaccinated with ANDV and SNV DNA vaccines to express human IgG and the efficacy of these antibodies to prevent lethal hantavirus disease in two small-animal models. Purified TcB human IgG was evaluated by both classical plaque reduction and PsVNAs, and bioavailability of these antibodies was assessed in the Syrian hamster. Finally, the ability of TcB human IgG to prevent lethal disease caused by ANDV in immunocompetent hamsters and SNV in immunosuppressed hamsters was evaluated. Animals were assigned randomly at the beginning of each study. Sampling information is provided in figure legends.

Virus and cells

Twice plaque-purified SNV strain CC107 (44) and ANDV strain Chile-9717869 (23) were propagated in Vero E6 cells [Vero C1008; American Type Culture Collection (ATCC) CRL 1586]. The VSVΔG*rLuc pseudovirion (PsV) is a recombinant vesicular stomatitis virus (VSV) derived from a full-length complementary DNA clone of the VSV Indiana serotype in which the G protein gene has been replaced with the Renilla luciferase gene. ANDV and SNV PsV were produced in human embryonic kidney (HEK) 293 cells using DNA vaccine plasmids pWRG/AND-M[opt2] and pWRG/SN-M[opt], respectively, by methods described previously (45). Vero, HEK 293T, and Vero E6 cells were maintained in Eagle’s minimum essential medium with Earle’s salts containing 10% fetal bovine serum, 10 mM Hepes (pH 7.4), and penicillin-streptomycin (Invitrogen) at 1×, and gentamicin sulfate (50 μg/ml) (Vero E6 cells only) [complete Eagle’s minimum essential medium (cEMEM)] at 37°C in a 5% CO2 incubator.

Transchromosomal bovine

The two TcBs used in this study have homozygous triple knockout in endogenous bovine immunoglobulin genes (IGHM–/– IGHML1–/– IGL–/–) and contain human artificial chromosome (HAC) vector labeled as KcHACD (22, 23, 47). This HAC vector consists of human chromosome 14 fragment, which contains the entire human immunoglobulin heavy chain locus except that the IGHM constant region remains bovine; and human chromosome 2 fragment, which contains the entire human immunoglobulin κ light chain locus (22, 23, 47).

TcB vaccination

Two TcBs were vaccinated at 3- to 4-week intervals with both the ANDV DNA vaccine, pWRG/AND-M[opt2], and the SNV DNA vaccine, pWRG/SN-M[opt], via the PharmaJet IM Stratis injection device. Each vaccine was administered with two injections of 3 mg each behind the ear and on the hind leg (for 6 mg total per site). The ANDV DNA vaccine was administered on the left side of the animal; the SNV DNA vaccine was administered on the right side of the animal. To evaluate whether co-administration of DNA vaccine with adjuvant plus immune stimulator could enhance immune response, Tc animal #2 was administrated with Montanide ISA 206 adjuvant (Seppic) plus the saponin-derived immune stimulant Quil A (Accurate Chemicals) adjacent to each DNA vaccination site at the final booster (V4). The adjuvant formulation [Quil A (0.5 mg/ml) in a 50% solution of ISA 206] was administered as a 1-ml injection using needle and syringe. The adjuvant was administered adjacent (1 to 2 cm) to each DNA vaccination site (2 ml total). At V4, Tc animal #1 was administered the DNA vaccine without adjuvant.

Purification of fully human IgG

TcB plasma collected from day 10 after V4 of Tc animal #2 was the source material for purifying fully human IgG against hantaviruses using the method described below. Frozen Tc plasma was thawed at room temperature overnight, pH-adjusted to 4.80 with dropwise addition of 20% acetic acid (Fisher, catalog #A491), fractionated by caprylic acid (Amresco, catalog #E499) at a caprylic acid/total protein ratio of 1.0, and then clarified by centrifugation at 10,000g for 20 min at room temperature. The supernatant containing IgG was then neutralized to pH 7.50 with 1 M tris, 0.22 μm–filtered, and affinity-purified with an α-human IgG light chain–specific column, KappaSelect (GE Healthcare, catalog #17545804). Fully human IgG was further purified by passage over an α-bovine IgG heavy chain–specific affinity column (Capto HC15 from GE Healthcare, catalog #17-5457-03). The purified α-HPS human IgG has a protein concentration of 8.42 mg/ml in a sterile-filtered buffer consisting of 10 mM glutamic acid monosodium salt, 262 mM d-sorbitol, and Tween (0.05 mg/ml) (pH 5.5).

Pseudovirion neutralization assay

The PsVNA was performed as previously described (39). Briefly, an initial 1:10 dilution of heat-inactivated sera was made followed by fivefold serial dilutions (in triplicate) that were mixed with equal volume of cEMEM containing 4000 focus-forming units of PsV of interest with 10% human complement (Sigma) and then incubated overnight at 4°C. After this incubation, 50 μl was inoculated onto Vero cell monolayers in a clear bottom black-walled 96-well plate (Corning). Plates were incubated at 37°C for 18 to 24 hours. The medium was discarded, and cells were lysed according to the luciferase kit protocol (Promega #E2820). A Tecan M200 Pro was used to acquire luciferase data. The values were graphed using GraphPad Prism software (version 6) to calculate the percent neutralization. Data for each dilution series were fit to a four-parameter logistic curve using GraphPad Prism and then PsVNA80 neutralization titers were interpolated. Geometric mean titers from triplicates were reported.

Plaque reduction neutralization test

PRNT was performed using Vero E6 cells as previously described (47). The 80% PRNT titer (PRNT80 titer) is the reciprocal of the highest serum dilution reducing the number of plaques by 80% relative to the average number of plaques in control wells that received medium alone.

ANDV lethal disease model using Syrian hamsters

Female Syrian hamsters aged 6 to 8 weeks (Harlan) were anesthetized by inhalation of vaporized isoflurane using an IMPAC 6 veterinary anesthesia machine. Once anesthetized, hamsters were injected with 200 PFU of ANDV diluted in phosphate-buffered saline (PBS). Intramuscular (caudal thigh) injections consisted of 0.2 ml delivered using a 1-ml syringe with a 25-gauge, 5/8-inch needle. The mean day to death of a 200-PFU intramuscular challenge in this model is 11, with a range of 9 to 14 (11, 23).

SNV lethal disease model using transiently immunosuppressed Syrian hamsters

Female Syrian hamsters aged 6 to 8 weeks (Harlan) were used in this experiment. The hamsters were anesthetized by inhalation of vaporized isoflurane using an IMPAC 6 veterinary anesthesia machine for all injections described below. Hamsters were transiently immunosuppressed by daily injections of dexamethasone and cyclophosphamide administered by the intraperitoneal route starting on day –3 and ending on day 13 after virus challenge as previously reported (15). Hamsters were challenged by intramuscular injection of 0.2 ml of PBS containing 2000 PFU of SNV on day 0. The mean day to death of a 2000-PFU intramuscular challenge in this model is 13, with a range of 10 to 14 (15).

N-specific ELISA

The ELISA used to detect N-specific antibodies (N-ELISA) was described previously (47, 48). The endpoint titer was determined as the highest dilution that had an optical density (OD) greater than the mean OD for serum samples from negative control wells plus 3 SDs. The PUUV N antigen was used to detect ANDV and SNV N-specific antibodies as previously reported (23).

Isolation of RNA and real-time PCR

About 250 mg of lung tissue was homogenized in 1.0 ml of TRIzol reagent using gentleMACS M tubes and a gentleMACS dissociator on the RNA setting. Serum samples were added directly to TRIzol reagent. RNA was extracted from TRIzol samples as recommended by the manufacturer. The concentration of the extracted RNA was determined using a NanoDrop 8000 instrument and raised to a final concentration of 10 ng/μl. Real-time PCR was conducted on a Bio-Rad CFX thermal cycler using an Invitrogen Power SYBR Green RNA-to-Ct 1-Step kit according to the manufacturer’s protocols. Primer sequences are as follows: SNV S 26F, 5′-CTACGACTAAAGCTGGAATGAGC-3′; SNV S 96R, 5′-GAGTTGTTGTTCGTGGAGAGTG-3′ (49). Cycling conditions were 30 min at 48°C, 10 min at 95°C, followed by 40 cycles of 15 s at 95°C and 1 min at 60°C. Data acquisition occurs after the annealing step.

Statistical analysis

Survival analyses were done using Kaplan-Meier survival analysis with log-rank tests. P values of less than 0.05 were considered significant. Comparison of levels of viral genome was done using Student’s t test. Analyses were conducted using GraphPad Prism (version 6).

Ethics statement

All work involving the use of ANDV or SNV in animals was performed in United States Army Medical Research Institute of Infectious Diseases (USAMRIID) biosafety level 4 laboratories. Research was conducted under an Institutional Animal Care and Use Committee–approved protocol in compliance with the Animal Welfare Act, Public Health Service Policy, and other Federal statutes and regulations relating to animals and experiments involving animals. The facility where this research was conducted is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International and adheres to principles stated in the Guide for the Care and Use of Laboratory Animals, National Research Council, 2011 (50).


Fig. S1. Virus neutralizing activity in purified Ig from vaccinated TcB is associated with human IgG.

Table S1. Fig. 1 geometric mean titers.

Table S2. Fig. 3 geometric mean titers.


  1. Acknowledgments: We thank P. Vial (Universidad del Desarrollo, Santiago, Chile) for providing the anti-HPS convalescent plasma used in this study. We also thank M. Wisniewski, P. Maes, and L. Queen for technical contributions; S. Kern for statistical analyses; and the Veterinary Medicine Division staff at USAMRIID for assistance with these studies. Funding: This work was supported by internal SAB funding and by funding from the Military Infectious Disease Research Program, Program Area T. Author contributions: J.W.H., R.L.B., H.W., J.J., E.J.S., and H.M. wrote, designed, and performed the study; S.A.K., C.D.H., and M.D.J. performed the study; and M.R. and J.B. designed the study. Opinions, interpretations, conclusions, and recommendations are those of the authors and are not necessarily endorsed by the U.S. Army. Competing interests: J.W.H. and E.J.S. have patents pending related to this work (U.S. Provisional 62/031,540: Anti-hantavirus human IgG neutralizing antibody produced in transchromosomal bovines). M.R. was employed at PharmaJet Inc. during the course of this research. The other authors declare no other competing interests. Data and materials availability: The data for this study are included in the paper or in the Supplementary Materials.

Correction: The author has added a patent citation to the references.

View Abstract

Stay Connected to Science Translational Medicine

Navigate This Article