Mother-to-Child Transmission of Andes Virus through Breast Milk, Chile

Andes virus (ANDV) is the only hantavirus transmitted between humans through close contact. We detected the genome and proteins of ANDV in breast milk cells from an infected mother in Chile who transmitted the virus to her child, suggesting gastrointestinal infection through breast milk as a route of ANDV person-to-person transmission.


Mother-to-Child Transmission of Andes Virus through
ANDV RNA has previously been detected in bodily fluids other than blood, such as saliva, respiratory secretions, and urine (2). Therefore, close contact with such fluids may explain additional cases for which high-risk environmental and rodent exposure is absent or improbable. In our study of the infected newborn, we ruled out environmental exposure; the only possible source was close contact to her mother during the incubation period and initial clinical disease. The mother maintained breast-feeding until the baby was hospitalized and confirmed to be viremic.
To evaluate the presence of ANDV particles in breast milk, we performed a culture in Huh-7 cells mock-infected and incubated with a breast milk pellet and, as a positive control, ANDV at a multiplicity of infection of 1 (5). After infection, we identified viral nucleoprotein (N) and glycoprotein (Gc) through immunofluorescence assay (Appendix). We detected N protein in the cytoplasmic compartment of ANDVinfected cells and cells incubated with breast milk but not in mock-infected cells (Figure 2, panel A).
To verify detection specificity, we used 2 different ANDV N protein antibodies generated in mice and rabbits. Again, we identified N protein in ANDV-infected cells and incubated with breast milk but not in Breast milk contains a variety of blood cells (monocytes, T-cells, NK cells, B cells, and neutrophils) and hematopoietic stem cells (6). In this context, we know that ANDV can be present in buffy coat cells for up to 15 days before illness onset (7); is always present during the acute phase of the disease, including the febrile prodrome phase; and remains in a small proportion of cases during convalescence (7). Assuming that breast milk contained ANDV-infected cells, direct inoculation in Peyer's patches in the newborn may have resulted in Figure 1. Epidemiologic timeline for mother-to-child transmission of Andes virus through breast milk, Chile. A) Key epidemiologic events related to the mother (represented by M, red circles and lines) and the newborn (NB, blue circle and lines). Blue-and-red circle represents the birth of the newborn; light green rectangle represents the 15 days of close contact that included breastfeeding. We show details for the baby above the time bar and details for the mother below the time bar. B) Longest (black bar) and shortest (light gray bar) probable person-to-person incubation period.
virus entry (8). Another possible mode of transmission is close contact with respiratory secretions and saliva of the infected mother. However, because ANDV was present in breast milk and the newborn's exposure to breast milk was much greater than to other fluids, transmission by breast milk is very likely.
Another factor that may help explain oral infection in the neonatal period is the gastrointestinal characteristics in the first month of life, such as the adjustment of stomach pH, rapid gastric emptying time, and increased permeability of the intestine due to loosened intestinal intercellular spaces (9). Sin Nombre virus was detected in breast milk samples by RT-PCR, but the exposed child did not become infected (10). Similar results were found in 2 women infected with Puumala virus (11). Vertical transmission was excluded in 4 pregnant women infected with hantavirus species in Europe (12). Bellomo et al. reported a newborn infected with ANDV but did not report the presence of ANDV in breast milk (13). Our case provides further evidence for a gastrointestinal transmission of ANDV, which is consistent with previous reports of Puumala virus and ANDV infections in Syrian hamster models of hantavirus cardiopulmonary disease and in 1 newborn human case (13)(14)(15)).

Conclusions
We describe mother-to-child transmission of ANDV infection in Chile. Our analyses proved the presence of ANDV in breast milk, proposing breast-feeding as an additional mechanism of transmission. In this context, we recommend that ANDV-infected mothers refrain from breast-feeding until ANDV RNA is undetectable in blood and breast milk. In addition, we advise strict clinical and virologic surveillance of children potentially exposed to family members with ANDV infection for early diagnosis and hospitalization for adequate intensive care. Detection of ANDV N and Gc proteins from enucleated cells from breast milk from a mother in Chile. A) Detection of N protein and the cytoplasmic marker eIF3. B) Detection of N protein with 2 different primary antibodies. C) Detection of N and Gc proteins. Huh-7 cells were mock-infected (mock column), ANDV infected (ANDV column), or incubated with a pellet from breast milk (breast milk column). Coverslips were incubated with mouse and rabbit antibodies. Scale bars indicate 20 µm. Complete methods are described in the Appendix (https://wwwnc.cdc.gov/EID/article/26/8/20-0204-App1.pdf). ANDV, Andes virus; Gc, glycoprotein; N, nucleoprotein.