Rapid and sensitive identification of omicron by variant-specific PCR and nanopore sequencing: paradigm for diagnostics of emerging SARS-CoV-2 variants

Already in the last weeks of December 2021, omicron has become the dominant variant of concern (VoC) in many countries, including Great Britain, France, Denmark, Greece, Spain, Portugal, and the US [1] with numbers of new infections reaching all-time highs. Earlier VoCs were characterized either by an increased ability for transmission in the population [VoCs alpha (B.1.1.7) and delta (B.1.617.2)] or a partial immune escape with variable effects on neutralization by polyclonal serum antibodies [VoCs beta (B.1.351), gamma (P.1/B.1.1.28) and delta] [2‐7]. A striking characteristic of the apparently independently evolved omicron VoC is a large number of amino acid substitutions, insertions and deletions in the spike protein, i.e. 32 changes compared to the original Wuhan-hu-1 virus [8], that likely contribute to its extraordinarily rapid spread in the population. Moreover, the number of epitopes in the spike protein relevant for neutralization are an important determinant of the genetic barrier to viral escape from humoral immunity [6, 9]. Consequently, physician-scientists have been alerted since its discovery based on the genetic information alone by omicron’s potential for a pronounced immune escape. Recent laboratory studies and epidemiological reports have now confirmed this exceptional combination of high contagiousity and drastic escape from neutralizing antibodies [10‐12]. The resistance to neutralization is most likely due to mutations K417N, N440K, G446S, S477N, T478K, E484A, Q493R, G496S, Q498R, N501Y and Y505H, which are located within or close to the epitopes bound by these antibodies [13].

Here, we report the laboratory-based diagnosis of the first four COVID-19 cases with an omicron infection reported in Germany based on a combination of variant-specific PCRs and next-generation-sequencing using an established nanopore platform.

In four individuals returning from South Africa on November 24 or 26, 2021 to Munich, Germany, SARS-CoV-2 was detected in respiratory swabs by PCR on November 27, 2021 with cycle-threshold (ct) values ranging from 23.2 to 31.2 (Table 1). Spike-gene dropouts (or spike-gene target failure, SGTF) in conventional PCR systems, which can present either as a markedly delayed Ct value for or an entirely absent detection of the SARS-CoV-2 spike gene, have been reported for B.1.1.529 over the last weeks [22‐24] as indirect diagnostic evidence for its identification. Typically, non-spike gene-targeted sequences, e.g. in the nucleocapsid gene, envelope gene or RNA-dependent RNA polymerase gene [25], performed in these omicron-containing samples in parallel, can be readily amplified.

High-resolution probe-based melting curve assays [26] and multiplex-based PCR assays [27, 28] have allowed simultaneous detection of discriminatory mutation sites for identification and differentiation of VOCs alpha [B.1.1.7], beta [B.1.351], gamma [P.1], and delta [B.1.617.2 and AY]. These tests’ applicability to the identification of omicron [B.1.1.529] had not been reported at the time of this study in late November 2021.

Based on the patients’ recent travel history to South Africa and available sequences for B.1.1.529 from GISAID [29], we employed SARS-CoV-2 variant-specific PCR systems (Assays 1 and 2) from two globally operating commercial vendors (TIB MOLBIOL, Berlin, Germany, and Seegene, Germany, Düsseldorf, Germany) [26‐28], which had at the time either not been officially launched or only recently  been launched, on November 27, 2021: RT-PCR products of samples from patients 1–3 (Table 1) showed a peak for del 69/70 (like alpha; unlike beta, gamma and delta) and a shifted and dampened peak for 484 compared to the other VOCs (Fig. 1a (top panel), Tables 2, and 3). The peak for 501 was lost, likely due to surrounding mutations (G496S, Q498R, Y505H). A peak was detected for 417N (like beta; unlike alpha, gamma and delta), and a shifted and dampened peak for 681H (Fig. 1a (middle panel), Tables 2, and 3) (unlike all other VOCs), the latter likely due to omicron’s N679K mutation. Also, a distinct melting curve for 371L/373P compared to alpha, beta and delta was observed (Fig. 1a (lower panel), Tables 2, and 3) [26]. Results for the sample from patient 4 were indeterminate in both assays (Fig. 1b), likely due to the very low virus load in this respiratory sample (ct value of 31.3 or negative; Table 1). Figure 1b summarizes the results for the VOC-specific assays for omicron (see Table 3 for primary output data for Assay 2 for all VOCs [27, 28]).

Next, RNA extracts of patients’ samples were studied by Nanopore sequencing using the ARTIC protocol version 3 [15, 16], which contains 98 primer pairs designed to cover the entire SARS-CoV-2 genome with overlapping 400-bp amplicons. In less than 12 h, sequencing results and SARS-CoV-2 Pangolin lineage calling were completed (GISAID accession ID in Table 1). Some mutations in the spike gene of the omicron VOC result in mismatches for primers used in amplicon generation. For example, amplicon number 76 of the Artic protocol v3/v4 is affected by spike mutations A23040G, G23048A and A23055G, which results in reduced coverage of the spike region including the amino acid mutations K417N, N440K and G446S. Modified protocols that include additional primers matching these polymorphic sites have been developed and will be implemented in future versions [30]. Of particular note, also the sample of patient 4 with very low/negative virus load as assessed by PCR could now be identified as B.1.1.529 with a genome coverage of 95.04%. This indicates that nanopore sequencing may allow a rapid, highly sensitive determination of the SARS-CoV-2 lineage, including omicron. Compared to established short-read sequencing technologies such as Illumina, the long-read nanopore technology allows more cost-effective processing of smaller batches of samples and offers shorter turnaround times. However, the analytical accuracy of nanopore sequencing is limited for the detection of short indels and variants at low read-count frequencies [31]. For larger batches of samples, where turnaround time is not critical, other sequencing technologies could be more cost-effective.

In conclusion, we report the combination of PCRs specific for heavily mutated sites in the spike gene and nanopore-based full-length genome sequencing for the rapid and sensitive identification of the first four COVID-19 patients diagnosed in Germany to be infected with omicron. These findings will assist unambiguous laboratory-based clinical diagnosis and global surveillance for this highly contagious VOC, which carries spike mutations conferring reduced COVID-19 vaccine efficacy. Moreover, specialized diagnostic laboratories should continuously update their assays for variant-specific PCRs in the spike gene of SARS-CoV-2 to readily detect and diagnose emerging variants of interest and VoCs. The combination with established nanopore sequencing procedures allows both the rapid confirmation by whole genome sequencing as well as the sensitive identification of newly emerging variants of this pandemic β-coronavirus in years to come.