The cDNA of SARS-CoV strain Hanoi provided by Dr. Kariwa of Hokkaido University was used as a template for PCR by Accupower premix (Bioneer, Korea). The full length of the SARS-CoV N gene was inserted into vector pET21a (Novagen, German). After transformation, E. coli strain Origami B (DE3) pLysS competent cells were induced using 1 mM IPTG (Invitrogen, USA) at 37℃ for 4 h. The expressed proteins were purified using the His-Bind kits (Novagen, German). Production of polyclonal and monoclonal Abs was performed as described by Shang et al. [7]. Human coronaviruses, HCoV 229E (ATCC, VR-730) and HCoV OC43 (ATCC, VR-1558), were infected to MRC-5 cells (Korean Cell Line Bank). Replication of these viruses was confirmed by RT-PCR of cell lysates infected with HCoV 229E and HCoV OC43. RT-PCR was performed as described in a previous study [10]. To examine whether Abs against recombinant N protein of the SARS virus react with other HCoVs, we performed Western blotting on recombinant SARS N protein or cell lysates infected with HCoV 229E and HCoV OC43.

In our previous study, the antigenicity of recombinant of SARS-CoV N protein was checked with a mouse anti-SARS-CoV N protein monoclonal IgG2a (Zymed, USA), and convalescent SARS serum provided by the National Institute of Hygiene and Epidemiology in Vietnam [4]. We selected polyclonal and monoclonal Abs that showed the highest reactivity with the N protein in an ELISA. With these Abs, we determined cross-reactivity against cell lysates infected with HCoV 229E and HCoV OC43 by Western blotting. The viruses in these cells lysates were confirmed by RT-PCR (Fig. 1). Abs reacted with recombinant N protein, but did not react with HCoVs in cell lysates (Fig. 2). To determine the specificity of these Abs, cross reactivity with porcine epidemic diarrhea virus (coronavirus group I) and mouse hepatitis virus (coronavirus group II) were analyzed by Western blotting but showed no reaction (data not shown).

Coronaviruses are a group of large, enveloped, positive-sense, single-stranded RNA viruses that are known to associate with respiratory, enteric and neurological diseases in humans and domestic animals [2]. Many researchers have reported cross-reactivity with other HCoV when the diagnostic systems are based on SARS N protein [9,14]. It is therefore important to explore the possibility of developing a diagnostic test for SARS-CoV that does not show this cross-reactivity with the other HCoVs. Only two coronaviruses, HCoV 229E (Group I) and HCoV OC43 (Group II), have previously been known to cause illness in humans [2]. These coronaviruses are responsible for 10~35% of upper respiratory tract infections [9]. Another human coronaviruses, HCoV NL63 and Co HKU1, were reported in 2004 and 2005 [11,13]. As such, a SARS diagnostic system that cross-reacts with HCoVs could easily result in false-positive reactions. Previous researchers have tried to develop a monoclonal Ab against SARS N protein based ELISA. Some checked cross-reactivity with chicken CoV [3], HCoV OC43 [8] and various CoVs [1]. We need more Ab candidates for the diagnosis of SARS. In this study, we checked Abs cross-reactivity against SARS virus with HCoVs 229E and OC43, before developing a diagnostic system. Because the polyclonal and monoclonal Abs produced in this study did not react with HCoV 229E or HCoV OC43 in Western blotting, it could be possible to develop a specific diagnostic system to detect SARS-CoV in infected patients with theses Abs. Cross-reactivity with HCoV NL63 and Co HKU1, newly arising HCoVs, should be confirmed to strengthen the specificity of our Abs against SARS-CoV.