Contemporary seasonal human coronaviruses display differences in cellular tropism compared to laboratory-adapted reference strains

Abstract

Seasonal human coronaviruses (sHCoVs) cause 15%-30% of common colds. The reference strains used for research were isolated decades ago and have been passaged extensively, but contemporary sHCoVs have been challenging to study as they are notoriously difficult to grow in standard immortalized cell lines. Here, we addressed these issues by utilizing primary human nasal epithelial cells (HNECs) and immortalized human bronchial epithelial cells (BCi) differentiated at an air-liquid interface, as well as human embryonic stem cell-derived alveolar type II (AT2) cells to recover contemporary sHCoVs from human nasopharyngeal specimens. From 21 specimens, we recovered four HCoV-229e, three HCoV-NL63, and eight HCoV-OC43 viruses. All contemporary sHCoVs showed sequence differences from lab-adapted CoVs, particularly within the spike gene. Evidence of nucleotide changes in the receptor binding domains within HCoV-229e and detection of recombination for both HCoV-229e and HCoV-OC43 isolates was also observed. Importantly, we developed methods for the amplification of high-titer stocks of HCoV-NL63 and HCoV-229e that maintained sequence identity, and we established methods for the titration of contemporary sHCoV isolates. Comparison of lab-adapted and contemporary strains in immortalized cell lines and airway epithelial cells revealed differences in cell tropism, growth kinetics, and cytokine production between lab-adapted and contemporary sHCoV strains. These data confirm that contemporary sHCoVs differ from lab-adapted reference strains and, using the methods established here, should be used for the study of CoV biology and evaluation of medical countermeasures.IMPORTANCEZoonotic coronaviruses have caused significant public health emergencies. The occurrence of a similar spillover event in the future is likely, and efforts to further understand coronavirus biology should be a high priority. Several seasonal coronaviruses circulate within the human population. Efforts to study these viruses have been limited to reference strains isolated decades ago due to the difficulty in isolating clinical isolates. Here, we use human airway and alveolar epithelial cultures to recover contemporary isolates of human coronaviruses HCoV-NL63, HCoV-229e, and HCoV-OC43. We establish methods to make high-titer stocks and titrate HCoV-229e and HCoV-NL63 isolates. We show that contemporary isolates of HCoV-NL63 and HCoV-OC43 have a different tropism within the respiratory epithelium compared to lab-adapted strains. Although HCoV-229e clinical and lab-adapted strains similarly infect the respiratory epithelium, differences in host response and replication kinetics are observed. Using the methods developed here, future research should include contemporary isolates when studying coronavirus biology.

Keywords: AT2 cells; air-liquid interface airway cultures; seasonal coronavirus.

Fig 1

Isolation of contemporary sHCoVs. (A) Schematic of the procedure for recovering contemporary sHCoVs from human nasal swabs. Replication kinetics (Log₁₀ genome copies/mL from 0 to 7 dpi) of recovered HCoV-229e isolates in immortalized BCi cells grown at an ALI and H9-derived AT2 cells (B), HCoV-NL63 isolates in primary HNECs grown at an ALI (C), and HCoV-OC43 isolates in immortalized BCi cells grown at an ALI (D). For (B), (C), and (D), either the mean ± SD from four unpooled replicates or the mean from four pooled replicates is shown. Each graph represents data generated from a single attempt to recover the virus using a nasal swab specimen.

Fig 2

Genetic characterization of sHCoVs. Time-resolved maximum likelihood phylogenetic analysis of full genome nucleotide sequences of contemporary HCoV-229e (A), HCoV-NL63 (B), and HCoV-OC43 (C) isolates compared to publicly available sequences. Sequences from viruses isolated in this study are shown in red text. Colored symbols indicate different countries of origin.

Fig 3

Sequence analysis of sHCoVs. Simplot analysis of full genome sequences of contemporary HCoV-229e (A), HCoV-NL63 (B), and HCoV-OC43 (C) isolates showing the percent nucleotide similarity across nucleotide positions compared to reference strains (HCoV-HCoV-229e/USA/Inf1/1962, HCoV-HCoV-NL63/NED/Amsterdam1/2004, and HCoV-HCoV-OC43/USA/VR-1558/1967, respectively). Analysis of the HCoV-229e receptor binding domain (D), HCoV-NL63 receptor binding domain (E), and HCoV-OC43 sialic acid binding domain (F) from contemporary isolates compared to the reference strain. RBL, receptor binding loop; SBL, sialic acid binding loop. Insertions upstream of SBL1 are highlighted.

Fig 4

Alterations in sHCoV spike sequence. Time-resolved maximum likelihood phylogenetic analysis of spike nucleotide sequences of contemporary HCoV-229e (A), HCoV-NL63 (B), and HCoV-OC43 (C) isolates compared to publicly available sequences. Sequences from viruses isolated in this study are shown in red text. Colored symbols indicate different countries of origin.

Fig 5

Comparison of virus growth in immortalized cell lines. Sequence alignments of (A) HCoV-NL63 and (B) HCoV-229e virus stocks isolated in HNEC (for HCoV-NL63) and BCi (for HCoV-229e) compared to passage 1 (P1) and passage 2 (P2) nucleotide sequences of the same virus produced in either LLC-AT (HCoV-NL63) or Huh7 (HCoV-229e) cells. Black lines indicate nucleotide changes, and red triangles indicate amino acid changes. Amino acid changes were only observed for the HCoV-NL63 P2 virus. Infectious virus titers and genome copies in LLC-MK2 (C) and LLC-AT (D) cells infected with lab-adapted (Am-1) and contemporary HCoV-NL63 isolates (18091206, 18111908, 19071116). Infectious virus titers and genome copies in Huh7 (E) and Huh7-T2 (F) cells infected with lab-adapted (VR740 [Inf1]) and contemporary HCoV-229e isolates (21021519, 21042820, 22050721). Data is a representative of at least two experimental repeats, each with three replicates. Mean ± SD is shown. Data were analyzed using a two-way analysis of variance (ANOVA) with a Dunnett’s post-test. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Fig 6

Comparison of virus growth across the human respiratory tract. Infectious virus titers (A) and genome copies (B) in HNECs, BCi, and AT2 cells infected with lab-adapted HCoV-NL63 (Am-1) or contemporary HCoV-NL63 18111908 (1908). Infectious virus titers (C) and genome copies (D) in HNECs, BCi, and AT2 cells infected with lab-adapted HCoV-229e (VR740-Inf1) or contemporary HCoV-229e 22050721 (0721). Data are representative of two experimental repeats, each with three replicates. Mean ± SD is shown. Data were analyzed using a student’s t test. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Fig 7

Comparison of the host response to sHCoVs. IP-10, IFN-λ1, IFN-λ2/3, and IFN-β1 levels in the supernatant of HNECs infected with HCoV-NL63 18111908 or Am-1 (A), HNECs infected with HCoV-229e VR740 or 22050721 (B), BCi infected with HCoV-229e VR740 or 22050721 (C), and AT2 cells infected with HCoV-229e VR740 or 22050721 (D). Data is a representative of two independent experiments. Individual replicates and mean ± SD are shown. Data were analyzed using a two-way ANOVA with Tukey’s post-test. *P < 0.05, **P < 0.01, and ***P < 0.001.