Maternally Derived Recombinant Human Anti-Hantavirus Monoclonal Antibodies Are Transferred to Mouse Offspring during Lactation and Neutralize Virus In Vitro

Abstract

Transgenic mice expressing a recombinant human monoclonal antibody (rHMAb) against hantavirus were generated. These mice could be used as models to explore the possibilities of producing rHMAbs for therapeutic purposes. The highest concentration of the rHMAb in the milk of the transgenic females was 6.6 mg/ml. The rHMAb was also detected in the sera of pups fed by the transgenic females. Both the rHMAbs in the milk of transgenic mice and those in the sera of suckling pups were found to be active against hantaviruses, although the light chain of the antibody absorbed by the pups was modified by N-linked glycosylation.

Hantaviruses compose a genus that belongs to the family Bunyaviridae but differ from the viruses of other genera in that they are enzootic (24). The natural hosts of hantaviruses are rodents, and transmission from their natural hosts to humans is not well known but it is generally accepted that inhalation of infected excreta causes infection (20, 29). Some species of hantaviruses cause severe infections in humans such as hemorrhagic fever with renal syndrome and hantavirus pulmonary syndrome (8, 22). The annually reported number of cases of hemorrhagic fever with renal syndrome caused by hantaviruses is about 150,000 worldwide, and two-thirds of the cases occur in China and are mainly associated with Hantaan (HTNV) and Seoul (SEOV) hantavirus infections (26).

To prevent hantavirus infection, vaccines consisting of inactivated viruses (formalin-inactivated, rodent brain-derived virus) (5, 28) have been developed. Recently, vaccines made by recombinant DNA technology (including recombinant vaccinia virus and naked DNA vaccines) have also been shown to be promising (6, 9, 14, 18). Protective immunity to hantavirus infections has previously been associated with neutralizing antibody responses directed against the viral G1 and G2 envelope glycoproteins (1, 2). High concentrations of neutralizing antibodies in serum efficiently block infection (12). However, production of sufficient quantities of monoclonal antibodies (MAbs) for therapy remains a major problem (12, 15). Production of MAbs in the milk of transgenic animals is one of the most attractive techniques for addressing this problem (10, 11).

In this study, the heavy-chain and light-chain genes of a human immunoglobulin G1 (IgG1) MAb against the HTNV G2 protein (12) were cloned into a commercial pBC1 vector and co-microinjected to create transgenic mice expressing a recombinant human MAb (rHMAb) in their milk. Altogether, 75 mice were produced through co-microinjection. PCR and Southern blotting identified seven (two females and five males) transgenic founders containing both the heavy- and light-chain genes (Fig. 1).

FIG. 1.

Detection of transgenic mice by Southern blotting. Lanes: P1, P2, P5, and P10 (P, plasmid), the amounts of plasmid corresponding to 1, 2, 5, and 10 copies of pBC1-hG2H (2.3 kb) and pBC1-hG2L (1.2 kb) integrated into the mouse genome, respectively; N, negative control; 5, F₀ hAHT5; 6, F₁ hAHT8-6; 9, F₁ hAHT8-9; 20, F₁ hAHT12-20; 28, F₁ hAHT61-28; 38, F₁ hAHT71-38; 64, F₁ hAHT71-64; 44, F₀ hAHT44. The estimated copy numbers are shown in Table 1.

High levels of rHMAbs directed against hantavirus were detected in the milk (but not the serum) of the founder (hAHT5) and the F₁ females (hAHT8-6, hAHT12-20, hAHT61-28, hAHT71-38, and hAHT71-64) (Fig. 2). Expression levels of the recombinant antibody in the milk of F₀ and F₁ transgenic (both heavy- and light-chain-containing) females were more than 1 mg/ml, and 6.6 mg/ml was the highest expression level found (Table 1). One F₁ mouse, hAHT8-9, showed no intact antibody in the milk when tested by enzyme-linked immunosorbent assay, despite a strong positive signal for both the heavy- and light-chain antibodies in its whey by Western blotting. Also, we found that the hAHT44 mouse, which had acquired only the heavy-chain gene, did not express detectable levels of the heavy-chain antibody.

FIG. 2.

Detection of anti-HTNV rHMAb expression in transgenic mice by Western blotting. (a) Detection of rHMAb expression in serum and milk samples from an F₀ hAHT5 transgenic mouse with heavy-chain-specific antibodies. Lanes: 1, human milk; 2, serum; 3, whey from a wild-type mouse; 4, whey from F₀ hAHT5; 5, whey from F₀ hAHT5 (2-mercaptoethanol untreated). (b) Detection of rHMAb expression in milk samples from transgenic females with both heavy- and light-chain anti-human antibodies. Lanes: 64, F₁ hAHT71-64; 38, F₁ hAHT71-38; 28, F₁ hAHT61-28; 20, F₁ hAHT12-20; 9, F₁ hAHT8-9; 6, F₁ hAHT8-6; N1 and N2, nontransgenic milk samples (negative controls); 5 and 44, milk samples from F₀ hAHT5 and F₀ hAHT44. All of the F₁ mice tested and hAHT5 contained both the heavy- and light-chain antibodies, whereas hAHT44 contained only the heavy-chain antibody. A, human IgG. 50kD, heavy chain; 25kD, light chain.

TABLE 1.

Analysis of transgenic mice^a

Transgenic line (sex)	Offspring	No. of copies		Level of rHMAbs (g/liter)	Neutralizing activity
		Heavy chain	Light chain
hAHT5 (F)		15	15	2.8	+
hAHT8 (M)	hAHT8-6	5	5	6.6	+
	hAHT8-9	3	2	0	−
hAHT12 (M)	hAHT12-20	12	4	4.3	+
hAHT44 (F)		1	0	0	−
hAHT61 (M)	hAHT61-28	3	2	2.7	+
hAHT71 (M)	hAHT71-38	12	1	3.6	+
	hAHT71-64	2	1	1.7	+

^aThese data are for the rHMAbs in the milk of transgenic female mice. F, female; M, male.

The activity of rHMAbs was determined by using immunofluorescent antibody (IFA) (Fig. 3). The results showed that expressed rHMAbs, exemplified by one of the transgenic whey samples and one of the serum samples collected from the offspring of transgenic females, could both bind specifically to the G2 antigen of HTNV. As the MAb against the HTNV G2 protein is able to bind both HTNV and SEOV, additional testing for the binding of rHMAbs to SEOV antigen slides was also performed. The results showed that expressed rHMAbs could bind to both HTNV and SEOV. The rHMAbs in the serum from the pups also showed binding to HTNV and SEOV antigens despite modifications (discussed below).

FIG. 3.

IFA analysis of F₀ (hAHT5) and F₁ (hAHT12-20 and hAHT61-28) transgenic mice. HTNV IFA antigen slides were exposed to a dilution of whey (1:1,000) from the transgenic mice or serum samples from transgenic pups (1:50). Positive signals (green) in slides show binding to whey samples from transgenic females (a) and to serum samples of pups suckling transgenic females (d), while there were no positive signals in the slides processed with negative whey (c) or negative serum controls (f). Panels b and e represent whey and serum samples derived from wild-type mice with addition of purified anti-hantavirus MAb, used as positive controls. Green fluorescence corresponds to the presence of the MAb.

To test if rHMAbs in the transgenic milk could be absorbed by pups and provide protection against HTNV, sera were collected from the 5-day-old offspring of two transgenic mice expressing high levels of rHMAbs against hantavirus (hAHT12-20 and hAHT61-28). Western blotting showed that the human IgG MAb was present in the sera of pups, regardless of whether they were transgenic, as long as they were suckled by transgenic mice (Fig. 4). As expected, rHMAbs were not found in the sera of the nontransgenic pups that were fed by nontransgenic mothers, indicating that rHMAbs in milk are absorbed directly by newborn pups.

FIG. 4.

Western blot analysis of pups' sera with horseradish peroxidase-conjugated anti-human IgG and horseradish peroxidase-conjugated anti-mouse IgG. Sera from the pups of hAHT12-20 and hAHT61-28 were analyzed by Western blotting with horseradish peroxidase-conjugated anti-human IgG (a and b) to validate whether the recombinant antibodies in the pups' sera were obtained through suckling or not. The samples in panel a were not treated with 2-mercaptoethanol, and most of the recombinant antibody is still intact, while the samples in panel b were treated with 2-mercaptoethanol, and most of the antibodies separated into heavy and light chains. (a and b) A, human IgG; B, human serum; N1 and N2, negative control sera from two wild-type mice; 1, 2, and 3, serum samples from three pups of hAHT12-20; 4, 5, and 6, serum samples from pups of hAHT61-28. Whereas lanes 1 and 4 contain transgenes, the remaining pups lack transgenes according to PCR results.

The activity of rHMAbs in the milk of the transgenic females and in the sera of pups suckling transgenic mothers was further determined based on their ability to neutralize hantaviruses (Tables 1 and 2). In the neutralization test, dilutions of transgenic whey or pups' sera were added to HTNV before challenging Vero-E6 cells with the virus. Neutralized HTNV cannot infect Vero-E6 cells, and dilutions were considered to have virus-neutralizing activity if less than 50% of the challenged cells were infected by HTNV as detected by IFA. The results (Tables 1 and 2), demonstrating the neutralizing activity of the rHMAbs toward hantaviruses, suggest that protection may be provided to the offspring of transgenic animals via milk.

TABLE 2.

Results of fluorescence reduction virus neutralization tests of the rHMAb titer^a

Source	Transgenic mouse	Neutralization dilution rate	Fluorescence reduction (%)
Whey	hAHT5	1:100	75
Whey	hAHT8-6	1:800	75
Whey	hAHT12-20	1:400	50
Whey	hAHT61-28	1:200	50
Whey	hAHT71-38	1:200	50
Whey	hAHT71-64	<1:100	25
Mixture of sera	hAHT12-20 pups	1:20	50
Mixture of sera	hAHT61-28 pups	1:20	50

^aAs the serum of each individual pup was not enough to perform the virus neutralization test, we mixed the sera from pups suckled by the same transgenic female to obtain enough serum to perform the neutralization test. The negative controls included (both whey and sera from wild-type mice) did not show neutralizing activity.

Unexpectedly, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis showed that the molecular weight of the rHMAb heavy-chain antibody in the sera of pups was lower than that found in the controls (rHMAb from the milk of the hAHT5 mouse and human IgG), while that of the light-chain antibody was higher (Fig. 5). These changes in molecular weight are likely to be due to glycosylations or other modifications. To validate this hypothesis, the pups' sera were digested with N-glycosidase and then analyzed by SDS-PAGE and Western blotting (Fig. 5). After enzymatic treatment, the molecular weight of the light chain antibodies in the pups' serum became lower while that of the heavy chain antibodies remained constant. Control proteins did not show any changes, suggesting that the light chain of rHMAb absorbed by the pups is modified by N-linked glycosylation. To test the change in molecular weight of the heavy-chain antibody, commercial human IgG was added to both the serum of the nontransgenic mice and the serum of the transgenic pups. This was then analyzed by SDS-PAGE and Western blotting. We found that the heavy-chain band of standard human IgG was consistent with the reduced size found for the recombinant antibody in the pups' sera, suggesting that the apparent changes in the heavy-chain antibody in sera from the pups are due to interference by other proteins and impurities.

FIG. 5.

Western blot analysis of pups' sera digested with N-glycosidase F. Pups' sera were digested with N-glycosidase F and analyzed by Western blotting to check the change in molecular weight in the heavy- and light-chain antibodies absorbed into the pups' sera. 5′, 1′, 2′, and A′ are the whey sample of hAHT5, the serum samples of two of hAHT12-20 pups, and purified human IgG, all processed with N-glycosidase F, respectively, while 5, 1, 2, and A are the corresponding samples that were not N-glycosidase F treated.

We have achieved expression levels of up to 6.6 mg/ml of recombinant antibody in the milk of transgenic females, which compares favorably with values (0.4 to 6 mg/ml) reported previously (4, 7, 13, 16, 21, 25, 27). In vivo protection studies should have been done, but a lack of containment and strict regulations regarding animal experiments dealing with disease-causing viruses after a severe acute respiratory syndrome outbreak in China have prevented us from challenging transgenic animals or their offspring with viruses. It has previously been shown that administration of neutralizing MAbs can protect animals from infection with hantaviruses (23). The neutralization activity of the rHMAb detected in the sera of pups implies that the maternally derived recombinant antibodies may protect newborns, which are usually more susceptible to viral infections than older animals are, from death, as it has been reported that hantavirus infections are age dependent (19). It is worth noting that hantavirus infection of humans appears to be mediated by inflammatory immune responses (17, 22) and antibody treatment may potentially strengthen the inflammation. IgM antibodies are generally proinflammatory since they are powerful activators of the complement system (3). However, anti-inflammatory properties of antibodies, particularly some IgGs, have also been observed in diverse models of infectious diseases (3). Future production of the rHMAb in the milk of large transgenic animals such as cows or goats may provide sufficient antibodies for therapeutic use in humans.