**1. Introduction**

Classical swine fever virus (CSFV) is the etiological agent of classical swine fever (CSF), which is a highly contagious disease of swine. The CSFV is an enveloped positive-stranded RNA virus belonging to the genus *Pestivirus* of the family *Flaviviridae* [1]. The genome of CSFV is approximately 12.3 kb in length and contains a single open reading frame encoding for a polyprotein of 3,898 amino acids, which is flanked by 5 and 3 non-translated regions (NTR). The translated polyprotein is processed by cellular and viral proteases to the mature viral proteins of four structural (C, Erns, E1, and E2) and eight nonstructural proteins (Npro, p7, NS2, NS3, NS4A, NS4B, NS5A, and NS5B) [2].

The CSFV strains are divided into three genotypes, each with three to four subtypes—1.1, 1.2, 1.3, and 1.4; 2.1, 2.2, and 2.3; and 3.1, 3.2, 3.3, and 3.4—by analyzing three genomic regions: 5- -UTR, E2, and NS5B [3,4]. Genotype 1 comprises most of the historical strains including vaccine strains. Genotype 2

contains the most currently globally prevalent strains over the last two decades. Genotype 3 contains most of the strains with restricted distribution [3,5–7]. In recent years, there has been a shift in CSFV genotypes in the field, from genotypes 1 and 3 to genotype 2, which was observed in Europe and Asia [8–11]. The mechanisms for this shift remain unclear. In Taiwan, two CSFV populations coexist, namely the historical strain of genotype 3.4 (94.4 strain) prior to the 1920s and the newly invading strain of genotype 2.1 (TD/96 strain) since 1994, the latter of which became dominant in 1996 [12,13]. This means that in the field, pigs could be infected by two different genotypes of CSFV. In pig infections, the dominance of one strain, such as TD/96, over another strain, such as 94.4, could cause potential problems in the diagnosis, pathogenesis and epidemiological studies and control of CSF in the field, if not given special attention. From a clinical point of view, this replication advantage of genotype 2.1 could likely mask the detection, isolation and more aspects of genotype 3.4. The situation is further complicated by the routine use of modified live vaccines (MLV) to prevent and control CSF, in which all MLV strains used nowadays in different countries belong to genotype 1, a third genotype [3,14]. The MLVs offer protection from field viruses of different genotypes [14–16]. However, several factors, including viral loads of vaccines, routes and ages of vaccination, and co-presence of other pathogens, can interfere with the vaccine's efficacy [14,16,17].

To deepen our understanding of the characteristics of diverse CSFV genotypes, it is important to elucidate the mechanism of the virus shift in the field. Previous study revealed that the newly invading genotype 2.1 replicated more efficiently than genotype 3.4 did both in vitro and in vivo [18]. To further the understanding of pathogenesis that occurs in dual infections of CSFV, this study analyzed the competitions of the viruses of the two genotypes in co-infected pigs with equal and different virus populations without neutralizing antibodies and also in co-infected pigs with maternally derived neutralizing antibodies induced by MLV of a third genotype, with the goal of experimentally simulating the natural situations in the field.
