SARS‐CoV2 virions in PM₁₀ pollutants. May further research reject this thesis?

Dear Editor,

The animated debate about the possibility that infecting SARS‐CoV2 are widespread in open‐air via particulate matter (PM), should need a thorough reappraisal. So far, purported SARS‐CoV2 genomes were retrieved in PM10₁₀ airborne pollutants in Italy, despite the consideration that these particles were never cell‐cultured to ascertain an in vitro viral growth. ¹ Actually, some authors are describing the detection of possible SARS‐CoV2 RNAs in PM₁₀, rather than complete infectious viral particles. ¹ If confirmed, the evidence will raise great concerns for COVID‐19 epidemiology.

However, from a technical point of view, the ability to be particularly ensured about the existence of SARS‐CoV2 RNAs, should account on the contemporary positive detection of three separated gene markers in a quantitative reverse‐transcription polymerase chain reaction (RT‐qPCR), as recommended by the World Health Organization (WHO) guidelines for the many challenged countries from pandemic, that is, genes E, N, and RdRp, as reported. ² Positive outcomes on one single gene marker, such as the E gene, usually linked to the lipidic envelope, which is frail in nonbiological microenvironments, may be biased, particularly if RT‐PCR Co_t occurs beyond the 34th amplification cycle, suggesting a false‐positive result. ³ The direct association of SARS‐CoV2 with PM₁₀, that is, viral particles borne from PM₁₀ pollutants in urban areas, is a real concern for public health, therefore any researcher engaged with this study issue must be particularly ensured about its reliability. ⁴

Particulate matter, as airborne pollutants from urban and industrial wastes, may contain a huge deal of micro‐organisms, including viruses and bacteria. Actually, a huge amount of unclassified bacterial, viral, and further micro‐organisms can be found in industrial urbanized areas, usually highly polluted by PM₁₀, when pollutant material is collected on quartz fiber filters. ⁵ The majority of these organisms are still unidentified, so, their RNAs, are possible contaminants in a poorly stringent RT‐qPCR approach. Table 1 summarizes some of the major microbic species found in particulate matter PM₁₀, here exposed only as a few exemplificative cohort of airborne pollutant micro‐organisms, matching a sequence similarity with E primers sequences used in RT‐qPCR for SARS‐CoV2. ⁶ The choice focused onto a ≥70% sequence similarity to detect a minimal, biased primer/probe aspecific cycling. ⁷ Highly specific forward primers for detecting and amplifying the SARS‐CoV2 E gene and endowing researchers in diagnosing the presence of SARS‐CoV2 viral particles, should give amplified products in a RT‐qPCR around the 15–20th cycle, ⁶ , ⁷ a way to be ensured that the sequence similarity is particularly high and close to 100%. Table 1 shows that many further coronaviruses can share a great sequence homology, in some genome regions, with primers usually employed in diagnostics. ⁶ The same SARS‐CoV1, for example, the strain Frankfurt 1 (GeneBank: AY291315.1), shares 100% similarity (p = 2.2 × 10⁻⁸), with maximum likelihood estimation (MLE statistics) giving λ = 0.2558, K = 7.855 (with 500 sequences) and a Waterman‐Eggert score ⁸  = 130, showing complete overlap (26 nt) at 26141–26166. This level is perfectly comparable with the Italian isolate SARS‐CoV2 (INMI1/2020): similarity 100% (p = 4 × 10⁻⁸), MLE λ = 0.2505, K = 7.235, Waterman‐Eggert score = 130, complete overlap = 26213–26238). The stringency in primer‐template matching is only one of the crucial steps to warrant for a correct identification and to claim the existence of a specified coronavirus strain in a sampled matrix. A high overlap, as occurring for some RNA molecule in Kineococcus radiotolerans SRS30216, for example, the 23S rRNA (Table 1), does not mean high genetic similarity but a significant percentage of aspecific PCR products carry over. ⁶ , ⁷ Due to the wide diffusion of this bacterial species in PM₁₀ pollutants, ⁵ it cannot be excluded that a Co_t positivity over the 34th cycle using some recommended SARS‐CoV2 E gene forward primers, may hide a false positivity bias. In this specific case, the use of three markers, that is, E, RdRp, and N, notably reduces similarity (from 68% with 25 nt overlap to 58.2% with 62 nt overlap).

Table 1.

FASTA searching for possible probing sequences of environmental PM₁₀‐associated microbes with SARS‐CoV2 E sequence gene (1)

N	Identification	Genebank	Similarity	Overlap	Nucleotide sequence	Score
	Viruses
	Bat SARS CoV Rs672/2006, complete genome	FJ588686.1	100%	26 nt overlap	1–26:25502–25527	Waterman‐Eggert score: 130; 44.8 bits; E(1) < 2.5 × 10−8
	Bat coronavirus 16B0133, complete genome	KY938558.1	100%	26 nt overlap	1–26:25493–25518	Waterman‐Eggert score: 130; 45.0 bits; E(1) < 2.2 × 10−8
	Bat coronavirus isolate Anlong‐103 complete genome	KY770858.1	100%	26 nt overlap	1–26:26065–26090	Waterman‐Eggert score: 130; 45.7 bits; E(1) < 1.4 × 10−8
	Bat SARS coronavirus Rm1, complete genome	DQ412043.1	100%	26 nt overlap	1–26:26114–26139	Waterman‐Eggert score: 130; 45.4 bits; E(1) < 1.6 × 10−8
	Sparrow coronavirus HKU17, complete genome	NC_016992.1	87.5%	16 nt overlap	4–19:5395–5410	Waterman‐Eggert score: 62; 20.3 bits; E(1) < 0.4
	Human adenovirus C, complete genome	NC_001405.1	85%	20 nt overlap	26–8:2276–2295	Waterman‐Eggert score: 63; 22.2 bits; E(1) < 0.17
	Pigeon coronavirus PSH050513 complete (S protein sequence)	DQ160004.1	77.8%	18 nt overlap	9–26:1989–2006	Waterman‐Eggert score: 54; 16.5 bits; E(1) < 0.64
	Influenza A virus A/NewYork/PV01553/2018(H1N1)) segment 2	MH798340.1	76.5%	17 nt overlap	2–18:491–507	Waterman‐Eggert score: 49; 15.5 bits; E(1) < 0.73
	Enterobacteria phage P1, complete genome	NC_005856.1	76.2%	21 nt overlap	Waterman‐Eggert score: 60; 19.9 bits; E(1) < 0.92	6–26:24135–24155
	Pseudomonas phage F116 complete genome	NC_006552.1	69.2%	26 nt overlap	25–2:6852–6877	Waterman‐Eggert score: 50; 18.9 bits; E(1) < 0.97
	Bacteria and other micro‐organisms
	Mycobacterium gilvum (2)	CP000658.1	78.9%	19 nt overlap	7–25:18679–18697	Waterman‐Eggert score: 59; 22.8 bits; E(1) < 0.087
	Geodermatophilus obscurus DSM 43160 23 S rRNA	NR_076134.1	78.6%	14 nt overlap	15–2:1521–1534	Waterman‐Eggert score: 43; 15.5 bits; E(1) < 0.82
	Blastococcus saxobsidens DD2 23 S rRNA	NR_102997.1	78.6%	14 nt overlap	15–2:1525–1538	Waterman‐Eggert score: 43; 15.6 bits; E(1) < 0.81
	Kineococcus radiotolerans SRS30216 23 S rRNA	NR_076522.1	68%	25 nt overlap	2–26:1621–1643	Waterman‐Eggert score: 45; 16.1 bits; E(1) < 0.7
	Modestobacter marinus 42H12‐1 16 S rRNA (partial)	NR_116228.1	66.7%	18 nt overlap	2–19:492–509	Waterman‐Eggert score: 36; 13.4 bits; E(1) < 0.97
	Blastococcus saxobsidens DSM 44509 T 16 S rRNA	FN600641.1	66.7%	18 nt overlap	2–19:488–505	Waterman‐Eggert score: 36; 13.3 bits; E(1) < 0.98

(1) From Corman et al. ⁶   forward primer 5′‐ACAGGTACGTTAATAGTTAATAGCGT‐3′ (2) PYR‐GCK plasmid pMFLV02, complete sequence (Software Lalign Expasy version 2.10).

Despite these issues could be highly questionable, on the technical perspective and from a clinical point of view, Table 1 exemplifies the possibility that further micro‐organisms' derived RNAs may be biased for SARS‐CoV2 genomes, due to a similarity higher than 95% and close to 100% with long strings of primer sequences. As indicated, one possible “cut off” for the achievement of a proper decision‐making is to cultivate PM₁₀‐associated biological matrices onto living substrates, such as Vero or LLC‐MK2 cell lines. ⁹

The evidence that PM₁₀ can bear SARS‐CoV2 RNAs ¹ has been erroneously translated by nonscientific and popular information as the ability of infectious SARS‐CoV2 particles to colonize PM₁₀ and to widespread wherever people lives. Naked RNAs from SARS‐CoV2 are very difficult to be able to infect humans but, furthermore, more insightful clues are needed to be sure that PM₁₀ really bear COVID‐10 causative agents.

CONFLICT OF INTERESTS

The authors declare that there are no conflict of interests.