Category Technology

A group from Sejong University, Seoul, etc. has reported on the accuracy of COVID-19 diagnosis in chest CT images with applying Deep Learning.
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0249450

The neural network using in Deep Learning was composed of 20 stages, and convolution and pooling functions were incorporated. The resolution of the input images were (224 x 224, and the sizes of convolution were (3 x 3）and（5 x 5). The obtained overall accuracy was 99.83%（sensitivity＝0.9286, specificity＝0.99). In the future, it will be accelerate to adopt Deep Learning in diagnostic applications.

A group from Scripps Research Institute has reported that a broadly neutralizing antibody strong to SARS-CoV-2 variants could be induced by using SARS-CoV-2 nanoparticles as a vaccine.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8010731/

The main stream of the current vaccines is using platforms which deliver the recombinant SARS-CoV-2 spike as an antigen such as mRNA-encapsulated liposomes (e.g., BNT162b2 and mRNA-1273), adenovirus vectors (e.g., ChAdOx1 nCoV-19 [AZD1222], CTII-nCoV, Sputnik V, and Ad26.COV2.S). There vaccines are compatible with B.1.1.7 variant, but a notable loss of efficacy was reported for B.1.351 and P.1 variants. Therefore, it is required to develop vaccines that can elicit a broadly neutralizing antibody (bNAb) response to SARS-CoV-2 variants. For this purpose, those vaccines have to induce long-lived germinal center (GC) reactions to activate precursor B cells, stimulate affinity maturation, and form long-term immune memory.

The SARS-CoV-2 spike protein is a trimer of S1-S2 heterodimers. The S1 subunit contains a receptor-binding domain (RBD) to initiate infection. The S2 subunit consists of a fusion peptide (FP) and heptad repeat regions 1 and 2 (HR1 and HR2). Authors designed an HR2-deleted glycine-capped spike (S2GΔHR2) to increase the spike stability, and assembled those spikes on a nanoparticle platform called SApNP using I3–01v9 60-mers as a linker (S2GΔHR2-10GS-I3-01v9-LD7-PADRE (I3-01v9-L7P)). The I3–01v9-LP7 presents 20 stabilized spikes.

Using S2GΔHR2-10GS-I3-01v9-L7P as a vaccine, a bNAb were induced showing compatible titers for B.1.1.7, B.1.351, and P.1 variants. Compared with the soluble spike, this nanoparticle showed 6-fold longer retention, 4-fold greater presentation on follicular dendritic cell dendrites, and 5-fold higher germinal center reactions in lymph node follicles. The reason of this effect is not clear yet, but would be related to the size effect.

A group from National Health Research Institutes, Miaoli, Taiwan, etc. has established a transgenic Aedes aegypti line that expresses human CLEC18A, and has confirmed suppression of dengue virus (DENV) infection and proliferation in Aedes aegypti.
https://www.frontiersin.org/articles/10.3389/fimmu.2021.640367/full

A human C-type lectin protein CLEC18A binds to glycoprotein of dengue virus, induces type-I interferon secretion, making CLEC18A one of players in innate immune responses to DENV infection (detailed glycan binding specificity of CLEC18A is not clear, but usually C-type lectins bind to high mannose or Gal/GalNAc). This study suggests a possibility that could suppress the spread of DENV infection and proliferation by using transgenic mosquitoes.

A group from University of Milan, etc. has reported on inactivation of SARS-CoV-2 with UV-C irradiation.
https://www.nature.com/articles/s41598-021-85425-w

Authors irradiated UV-C (254nm) on V6 cells with different Multipulicity of Infection (MOI). 0.05 MOI, 5 MOI, and 1000MOI. 0.05 MOI is equivalent to the low-level contamination observed in closed environments, 5 MOI corresponds to the average concentration found in the sputum of COVID-19 infected patients, and 1000 MOI is a very large concentration, corresponding to that observed in terminally diseased COVID-19 patients. In a range from 0.05 MOI to 5 MOI, a very small dose of 4 mJ/cm² was enough to achieve full inactivation of the virus. Even at the highest value, 1000 MOI, virus was totally inactivated with a dose of 16.9 mJ/cm².

A group from University of Pennsylvania Perelman School of Medicine has reported longitudinal analysis results of IgG levels and memory B cells following Pfizer (BNT162b2)/Moderna (mRNA-1273) vaccination.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7941668/

1. Vaccination boosts antibody levels for both SARS-CoV-2 recovered people and SARS-CoV-2 inexperienced people
2. but, no additional boosting is observed after the second vaccine dose for SARS-CoV-2 recovered people
3. Memory B cell responses continue to improve after the second dose for SARS-CoV-2 inexperienced people
4. All ages benefit from vaccination, but vaccine induction of memory B cells declines with age
5. No relationship between post-vaccine IgG levels and memory B cells in SARS-CoV-2 inexperienced people
6. Memory B cells strongly correlate with post-vaccination IgG levels in SARS-CoV2 recovered people
7. The importance of measuring memory B cells induced by the vaccination is emphasized

So, this means that SARS-CoV-2 recovered people do not need vaccination twice, but SARS-CoV-2 inexperienced people do need vaccination twice definitely.

A group from Fred Hutchinson Cancer Research Center, etc. has investigated the effect of genomic mutations onto two types of monochronal antibodies (LY-CoV555 and LY-CoV016) against RBD of SARS-CoV-2.
https://www.biorxiv.org/content/10.1101/2021.02.17.431683v1

It was found that LY-CoV016 is very weak for K417N mutation, and LY-CoV555 is so weak for E484K mutation. In both cases, IC₅₀ increased more than 1,000 times with these mutations. Therefore, a strategy using antibody cocktails with different monochronal antibodies with different target epitopes would be quite effective in developing therapeutic drugs for rapidly mutating RNA viruses such as SARS-CoV-2.

A group from Massachusetts General Hospital, etc. has reported on effectiveness of two typical vaccines, BNT162b2 (Pfizer) and mRNA-1273 (Moderna), for various SARS-CoV-2 variants occuring in the world.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7899476/

Distribution of typical SARS-CoV-2 variants in the world.

Effectiveness of Pfizer and Moderna vaccines is shown by the neutralization response. There seems to be almost no change for B.1.1.7 variant, but the neutralization activity of these vaccines are significantly reduced in B.1.351 v2 and P.1 variants. Two common mutations, K417N and E484K, exist in B.1.351 and P.1 variants, and these mutations do not exist in B.1.1.7. 

A group from Memorial Sloan Kettering Cancer Center, New York etc. has reported on the development of therapeutic agents targeting Galectin-3 for breast and ovarian cancers.
https://www.nature.com/articles/s41598-021-82686-3

In ovarian and breast cancers, mucin named MUC16 with CA125 epitope is highly expressed. Mucins are strongly glycosylated with O-glycans, and the glycan chains are sometimes stretched by adding a poly LacNAc chain to the terminal. Galectin-3 (Gal-3) has affinity to the poly LacNAc, hence Gal-3 could bind to MUC16 via such glycans. Galectin family has various functions, and MUC16 is closely related to cancer cells and is considered to be related to cancer growth and metastasis.

Using a monoclonal antibody (14D11) against Gal-3, the authors evaluated the inhibitory effects of Gal-3 on ovarian and breast cancers in vitro and in vivo. The Kd value for Gal-3 and LacNAc binding was ~0.2mM, and the Kd value for Gal-3 and 14D11 was ~14.6nM. Therefore, it turns out that 14D11 has stronger affinity about 13,000 times.

Two ovarian cancer cell lines (A2780, SKOV3) with high expression of MUC16 were transplanted into mice and the effects of 14D11 administration were compared with survival rates.

In addition, breast cancer cells (MDA-MB-231) were transplanted into mice and the effects of 14D11 administration were compared as below. These experimental results show that inhibition of Gal-3 can reduce cancer growth significantly.

A group from UC Davis had proposed the idea of using ACE2-Fc fusion protein as a therapeutic agent for COVID-19 as follows,
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0237295

A group from Chulalongkorn University, Bangkok, Thailand etc. has  developed ACE2-Fc fusion proteins expressed in N. benthamiana leaf, and demonstrated the inhibitory effect of SARS-CoV-2 in vitro. Vero cells were infected with SARS-CoV-2, and then ACE2-Fc fusion protein was applied, 0.84 μg/ml (IC50) was obtained as the inhibitory effect. Why did authors use plants? The advantages of using plants are low cost, scalability of production, and no risks from pathogens of animal or human origin.
https://www.frontiersin.org/articles/10.3389/fpls.2020.604663/full

A group from Univ. of Rochester’s group has developed and used sensor chips with various glycans immobilized on the Arrayed Imaging Reflectometry (AIR) platform as a simple way to detect the hemagglutinin subtypes of influenza viruses and the subtypes of neuraminidase.
https://pubs.acs.org/doi/10.1021/acs.bioconjchem.0c00718

What is more curious is what is the Arrayed Imaging Reflectometry (AIR) platform rather than the application example itself. This platform seemed to be developed in Benjamin Miller Lab., Univ. of Rochester. The principle is physically very simple, growing a thin oxide on a Si substrate with the mirror surface, and using the interference effect of light reflected on the upper and lower surfaces of SiO2 (i.e., at the SiO2/Si interface). There could be a reflection condition that the surface becomes non-reflective as a result of the interference of the diagonally incident light onto the substrate, and molecular interaction between probes pre-fixed on the substrate and applied ligands happens, reflected light appears depending on degree of disruption of the light interference. Therefore, like SPR, AIR does not need fluorescence labeling to the sample (i.e., label-free). The question about this technology would be the sensitivity. However, the above example explained that it provides comparable or better sensitivity than SPR.
https://www.urmc.rochester.edu/labs/benjamin-miller/projects/arrayed-imaging-reflectometry.aspx