To identify the species of origin of the Ao38 cell line, we amplified sequences from three marker genes that have been used to distinguish between and within lepidopteran species: a) mitochondrial cytochrome c oxidase subunit 1 (COI) [3-5], b) internal transcribed spacers (ITS) of the nuclear ribosomal 18S-5.8S-28S cistron [6-8], and c) a cadherin gene [9] (Figure ​(Figure1).1). After PCR amplification and sequencing of DNA from selected regions of the COI gene and ribosomal ITS region, we found identical alignment of sequences from Ao38 cells and High Five (T. ni) cells, and nearly identical alignment with sequences from T. ni larvae (Figure ​(Figure1A1A and ​and1B),1B), as well as with T. ni sequences in the BOLD database [3,4]. Control sequences amplified from Sf9 cells and an A. odorata adult moth showed substantial variation (Figure ​(Figure1A1A and ​and1B,1B, Sf9 and A. odorata). Thus, the T. ni origin of the Ao38 cells was established using these sensitive marker genes. To examine the origin of the Ao38 cell line within the species T. ni, we examined cadherin gene sequences. A fragment of the T. ni cadherin gene sequence that contains an intron, was previously found to be highly polymorphic in sequence [9]. Therefore, we amplified and compared those cadherin sequences from several sources of T. ni. The target cadherin sequence from Ao38 cells was identical to that from High Five cells but differed from that of other T. ni sources. Figure ​Figure1C1C shows 100% identity between High Five and Ao38 sequences over a 415 bp region of the cadherin sequence. In contrast, sequences derived from larvae of a laboratory strain of T. ni (Cornell strain [10]), and from a recently reported T. ni cell line (QB-9-4s) [11], showed app. 2.4% sequence divergence over the same region. Thus, the Ao38 line appears to be closely related to, and is likely derived from High Five cells. Based on these data, we decided to rename the Ao38 cell line as BTI-Tnao38.

The alphanodavirus TNCLV is readily detected by RT-PCR in growing High Five cells. We previously reported that we were unable to detect TNCLV in the Ao38 cell line by RT-PCR. We therefore performed a careful reexamination of the Ao38 cells for the presence of TNCLV RNA, and measured the sensitivity of our methods. For these studies, we amplified and cloned a 566 nt region of the TNCLV genome, then in vitro transcribed and purified an RNA representing that region of the TNCLV genome. The in vitro transcribed TNCLV RNA was subsequently used for spiking experiments with Sf9 cell RNA, to determine the sensitivity of TNCLV RNA detection in our RT-PCR studies (Figure ​(Figure2).2). Under these conditions we found that the sensitivity for detection of TNCLV RNA was approximately 53 molecules of TNCLV RNA in the background of total RNA from approximately 3000 Sf9 cells (Figure ​(Figure2,2, lanes 3-7). Under these conditions, TNCLV was not detected in RNA isolated from approximately 3000 cells of the Ao38 line (Figure ​(Figure2,2, lane 8). In a parallel analysis, Sf9 cells were also negative for TNCLV, as expected (lane 2). Thus, based on this level of sensitivity, we estimate that if TNCLV is present in Ao38 or Sf9 cells, it would be found at less than one TNCLV genome per 50 cells. In contrast to Ao38 and Sf9 cells, TNCLV RNA was detected at relatively high levels in High Five cells. Based on the sensitivity of detection in our TNCLV RNA spiking experiment, and the levels of TNCLV RNA detected in dilutions of High Five cell RNA (Figure ​(Figure2,2, lanes 9-11), we estimate that TNCLV RNA is present at levels of > 180-300 copies per cell in High Five cells. The marker genes and methods described above may be useful for confirmation or validation of cell lines, and the methods used for alphanodavirus detection may also be useful for laboratories examining T. ni or other cell lines suspected to harbor this alphanodavirus.