Detection of clade 2.3.4.4b highly pathogenic H5N1 influenza virus in New York City

ABSTRACT Highly pathogenic avian influenza viruses of the H5N1 clade 2.3.4.4b were detected in North America in the winter of 2021/2022. These viruses have spread across the Americas, causing morbidity and mortality in both wild and domestic birds as well as some mammalian species, including cattle. Many surveillance programs for wildlife as well as commercial poultry operations have detected these viruses. In this study, we conducted surveillance of avian species in the urban environment in New York City. We detected highly pathogenic H5N1 viruses in six samples from four different bird species and performed whole-genome sequencing. Sequencing analysis showed the presence of multiple different genotypes. Our work highlights that the interface between animals and humans that may give rise to zoonotic infections or even pandemics is not limited to rural environments and commercial poultry operations but extends into the heart of our urban centers. IMPORTANCE While surveillance programs for avian influenza viruses are often focused on migratory routes and their associated stop-over locations or commercial poultry operations, many bird species—including migratory birds—frequent or live in urban green spaces and wetlands. This brings them into contact with a highly dense population of humans and pets, providing an extensive urban animal–human interface in which the general public may have little awareness of circulating infectious diseases. This study focuses on virus surveillance of this interface, combined with culturally responsive science education and community outreach.

Americas and have heavily impacted wild bird populations and have affected the poultry industry (16)(17)(18)(19).In addition, infections in mammals-often leading to neurological symptoms and fatal outcomes-have been reported.This includes predatory animals and scavengers feeding on sick or dead birds (20)(21)(22).These are mostly seen as deadend hosts.Marine mammals have also been affected, especially in South America, and mammal-to-mammal transmission is suspected in some of these outbreaks (23)(24)(25).Furthermore, clade 2.3.4.4bH5N1 seems to have caused outbreaks in fur farms in Europe in mink and foxes with potential mammal-to-mammal transmission (26)(27)(28), and recently reported cases in dairy cattle are also raising concerns.Human cases caused by clade 2.3.4.4bH5N1 so far have been rare, and only two severe infections are known in the Americas (with a low number of additional ones in Asia), which is remarkable given the extent of the spread of this virus and the potential exposure to humans (29)(30)(31).
Nevertheless, it is very important to track the spread of this virus to determine potential risk to humans.There is a need for viral surveillance in urban areas that often have plenty of green spaces and wetlands for both resident and migratory birds.This, in combination with high human population densities, creates an extensive urban animal-human interface.In this interface, pets can also be impacted, as shown by infections with H5N1 in cats and dogs (32)(33)(34)(35).Communicating this risk to urban populations is critical.Here, we set out to detect HPAI H5N1 viruses in New York City using surveillance in wildlife rehabilitation centers as well as sampling bird feces from the environment.Our approach is based on a collaboration between research institutions, science outreach organizations, wildlife welfare nonprofit organizations, and community scientists.Community scientists working with our research team have previously reported the first detection of avian paramyxovirus 1 in New York City's pigeon population (36).The growing interest in biodiversity protection and citizen science has resulted in initiatives that collect a massive quantity of data about birds (37,38).However, these approaches are frequently limited to participatory data collection (38,39).In the collaborative New York City Virus Hunters initiative described in this study, we aim to engage the community in every step of the research process.Mentored research for high school students who self-identify as members of racial or ethnic minoritized groups in science is a core element.The students work alongside expert mentors and actively engage in overall study design before safely participating in sample collection, processing, data analysis, dissemination of results, and community outreach.The outputs of this program benefit all participants (40).

Surveillance strategy and virus detection
Our sampling strategy prioritized samples collected from birds known to contract HPAI H5N1-principally wild aquatic avian species of Anseriformes (ducks, geese, and swans), Charadriiformes (gulls, terns, auks and other shorebirds) and raptors such as Accipitri formes (hawks, ospreys, and other birds of prey) and Falconidae (falcons and kestrels).Samples for this study were collected from January 2022 to November 2023.In total, 1,927 samples were collected and processed for this study.We used two sampling streams.First, 125 environmental fecal samples were collected from New York City parks and green spaces using proper personal protective equipment (masks and gloves).In addition, professional animal rehabilitators at the Wild Bird Fund (WBF) and veterinarians of the Animal Care Centers (ACC) of New York City provided four water samples (3 mL each) and 1,798 cloacal (CS), oropharyngeal (OS), and fecal swabs from urban wild and domestic birds submitted to them.From these 1,798 samples (237 fecal samples, 783 CS and 764 OS samples, and 14 samples where CS or OP was nonspecified; from 895 birds), six were found positive for HPAI H5N1.While for environmental fecal samples collected in urban parks and green spaces, the avian species is hard to determine by the appearance of the sample, CS, OS, and fecal samples provided by wildlife rehabilitation centers were documented to be from 80 different species (see     1. RNA was extracted from 1,927 samples, and reverse transcription was used to generate cDNA.We then screened the cDNA preparations via a multiplex PCR using primers for the matrix (M) genomic segment, the nucleoprotein (NP) genomic segment, and for the hemagglutinin (HA) genomic segment.Primers for the HA segment were H5 HA-specific, while M and NP primers were designed to detect all known influenza A viruses.Gene products were sent for Sanger sequencing.If Sanger sequencing indicated the presence of influenza A virus, whole-genome sequencing was performed.Samples from six birds were found to be positive for HPAI H5N1, and whole genomes could be obtained.No environmental fecal samples were found positive for HPAI H5N1.No other avian influenza viruses were detected.Of these positive samples, the first [A/Canada goose/New York/NYCVH 22-6038/2022 (H5N1)] was collected from a Canada goose (Branta canadensis).This animal was initially found in Hutchinson River Parkway, in the Bronx, and died before the intake exam in August 2022.The next positive sample [A/red-tailed hawk/New York/NYCVH 22-8477/2022 (H5N1)] was derived in October 2022 from a red-tailed hawk (Buteo jamaicensis) that was found in close proximity to a major highway in Queens.The bird displayed neurological symptoms in the clinic.In December 2022, two birds found in   2).No additional positive samples/birds were detected from April 2023 to November 2023.
To further analyze our detected viruses, we performed a multiple sequence alignment of their amino acid sequences and mapped amino acid (AA) changes from the HPAI H5N1 strain A/bald eagle/FL/W22-134-OP/2022 (accession number UWI70064) (41) onto an HA structure from A/chicken/Vietnam/4/2003 (42) (Fig. 2).The detected AA differences mainly fell outside the receptor-binding site and antigenic sites of H5N1 (43,44), except for T71I.Most differences were only found in one of our NYCVH strains, except for T71I, which was present in all NYCVH strains.It should be noted that isoleucine (I) was present at this position in all 50 strains used to construct our phylogenetic tree, and it is atypical for A/bald eagle/FL/W22-134-OP/2022 to have a threonine (T) at this position.To our knowledge, none of the amino acid changes relative to A/bald eagle/FL/W22-134-OP/ 2022 have specifically been implicated with increases in pathogenicity or mammalian adaptation.
Upon confirmation, detections were reported to the United States Department of Agriculture (USDA), and the associated original samples were transferred to Mount Sinai's BSL3 +select agent facility (Emerging Pathogens Facility (EPF)/BSL-3 Biocontain ment CoRE) for storage.Results were also discussed with the New York City Depart ment of Health and Mental Hygiene as well as the Wild Bird Fund and the Animal Care Centers of New York City, following a previously developed internal and external communication strategy (45,46).In brief, the strategy aimed to ensure prompt and informed decisions and that all participants, collaborators, and stakeholders are kept fully informed.Successful communication of science and public health messages is complex, and reaching potentially vulnerable audiences remains an important challenge.
Our communication aimed to calm potential anxiety by providing information and instilling confidence and trust by addressing all questions to the best of our abilities.It has been noted that communications around emerging infectious disease can be improved when it comes from individuals inside the same community as those receiving the information, simply for the fact that they often share the same language, values, and beliefs (47).Therefore, it is incredibly important to ensure researchers involved in pandemic preparedness are committed to bidirectional communication, listening and serving the needs of the community.To reach the scientific community and general public alike, involved students shared their results in multiple languages and through multiple channels.These range from live virtual events and talks at community boards to in-person symposia and presentations at scientific conferences.Results were also presented to the public at three student research symposia, including the New York City Virus Hunters Symposium on 31 May 2023.

Phylogenetic analysis of detected HPAI H5N1 viral genomes
Phylogenetic analysis of the six viral genomes and genotype assignment was performed.The H5 and N1 genes of all six viruses were all typical of the currently circulating 2.3.4.4b clade in the Americas.HA sequences of A/Canada goose/New York/NYCVH 22-6038/2022, A/red-tailed hawk/New York/NYCVH 22-8477/2022, A/Canada goose/New York/NYCVH 22-9190/2022, and A/peregrine falcon/New York/NYCVH 160820/2022 were clustered closely together in a tree constructed of 50 HA sequences randomly selec ted from a list of all available H5N1 strains collected since 1 January 2020 on NCBI's influenza virus database, downloaded on 26 October 2023 (Fig. 3).They also cluster with contemporary H5 sequences from 2022 from Ohio, North Carolina, and also Colombia.Similarly, their NA sequences cluster together next to the NA sequences of the Ohio and Colombia isolates for which the HAs cluster as well.The two 2023 sequences A/chicken/New York/NYCVH 168127/2023 and A/Canada goose/New York/NYCVH 23-453/2023 are clustering together as well and form their own branch close to a cluster of sequences from 2022 and 2023 North and South American isolates.The NA sequences of these two viruses cluster together but are also located closely to many different isolates from both North and South America.
To identify the genotypes of internal genes, we used a script provided by Youk et al. (16) that allows for classification of segments into lineages and determines a genotype based on the genomic segment composition of a virus.We compared our virus sequences with available full-length genome sequences from the New York State, New Jersey, and Connecticut areas surrounding New York City where many infections were detected (Fig. S1).All detected viruses were re-assortant viruses between the Eurasian (EA) and American (AM) lineages.All HA and NA segments were of course from the EA lineage, but segments encoding for internal proteins differed.A/Canada goose/New York/NYCVH 22-6038/2022, A/red-tailed hawk/New York/NYCVH 22-8477/2022, A/Canada goose/New York/NYCVH 22-9190/2022, and A/peregrine falcon/New York/NYCVH 160820/2022 were all determined to be genotype B1.3 with AM lineage polymerase and NP segments, and all other segments were from the EA lineage (Table 3).B1.3 lineage viruses were also found in New York State and neighboring states (New Jersey and Connecticut) during our observation period (Table 4).The more recent A/chicken/New York/NYCVH 168127/2023 and A/Canada goose/New York/NYCVH 23-453/2023 viruses belonged to lineage B3.3, a lineage also detected in New York State in a turkey vulture in April 2023.This lineage features PB2, PB1, NP, and NS segments from the AM lineage, while PA, HA, NA, and M segments are derived from the EA lineage.

DISCUSSION
The recent spread of the panzootic clade 2.3.4.4bH5N1 across the globe has caused significant damage to wild bird populations and to the poultry industry (16,41,49,50).Spillovers into mammals (including cattle) have caused concerns about mammalian adaptation of this clade.However, despite the extensive spread of the clade 2.3.4.4bH5N1 virus and likely significant exposure of humans to it (hunters, poultry farmers, etc.), human infections have so far been rare, with only two known severe cases in the Americas (29,30) and a small number of cases in Asia (31).Avian influenza virus surveillance is often carried out in wild birds in rural areas, through hunter programs as well as domestic poultry operations.However, surveillance systems to detect the virus in urban wild birds are often absent.Despite that, many bird species inhabit or temporarily visit urban areas, which in many cases have ample green space as well as aquatic habitats for waterfowl.This is exemplified by the long list of species sampled in this study.Our study focused on this urban space using two sample streams, namely, samples from animal rehabilitation centers (Wild Bird Fund and Animal Care Centers of New York City) and environmental fecal samples sourced via a citizen/community science project (New York City Virus Hunters).Including the community in viral surveillance in a safe way generates interest and an understanding of the topic in the population, which is important given the science skepticism that has come to light through the coronavirus disease 2019 (COVID-19) pandemic (51,52).
Our work identified six HPAI H5N1 viruses in 1,927 samples (corresponding to at least 895 birds).These viruses were found in species known to be susceptible for H5N1    infection.Based on infection patterns in our area, we did expect to find HPAI H5N1 virus in Canada geese (which are highly susceptible to H5N1 infections (53,54)) as well as in raptors (peregrine falcon and red-tailed hawk), which often get infected when feeding on infected prey or carcasses.While H5N1 is known to infect chickens, it was somewhat unexpected to receive samples from a chicken found in Marcus Garvey Park in Manhat tan.Almost all our other samples from chickens were from birds in captivity.41).The genotypes of our NYCVH-detected HPAI H5N1 viruses have also been detected in the region (defined as the states of New York, New Jersey, and Connecticut) during the same time period.Of note, while many infections in mammals have been reported in the Americas with severe (and often neurological) symptoms and outcomes, most have been "dead end" infections in scavengers or predatory animals that presumably fed on infected birds or bird carcasses (20)(21)(22).However, mammal-to-mammal transmission is suspected in several outbreaks in fur farms in Europe and in marine mammals in South America (26)(27)(28), and recent cases in cows in several US states and in a goat have raised concerns.
Our study shows that clade 2.3.4.4bH5N1 highly pathogenic avian influenza virus can be present in birds that migrate through or live in urban centers.This highlights the importance of viral surveillance at the urban animal-human interface in which wild animals may potentially interact with a high-density population of humans and their pets.Humans may interact with infected birds directly (handling an injured bird) or indirectly (e.g., by coming in contact with feces or contaminated water in parks).Pets including cats and dogs are susceptible to HPAI H5N1, and transmissions from birds to both pet species have happened via contact with infected birds or bird carcassesscenarios which could occur in urban green spaces where pets are frequently taken (32)(33)(34)(35)55).Our study highlights these risks.However, it needs to be emphasized that a very small number of birds were found positive.Of note, the low percentage of positive animals could also be due to the sensitivity of the screening pipeline used, and other assays or tests may produce a higher number of positives.
An important aspect of our work is to involve the population and all stakeholders in surveillance efforts and communicate findings and risks efficiently.For this, we have shared and discussed our results with the New York City Department of Health and Mental Hygiene, and we have worked out a communication strategy.Junior scientists from the New York City Virus Hunters Program have also shared the results of our study with the public during our annual symposium in May 2023.We believe it is important for the public to understand that HPAI H5N1 may be present in birds, as well as their feces and other secretions in urban spaces, that sick or weirdly behaving birds (or other wildlife) should be reported to the authorities and only be handled by professionals in proper personal protective equipment, and that pets should be kept away from urban wildlife.Furthermore, it is important for physicians in urban centers to know about the potential presence of HPAI H5N1 and be aware that atypical influenza cases in humans may be caused by avian influenza viruses.So far, studies suggest that North American clade 2.3.4.4b viruses are susceptible to all classes of influenza drugs that are available as treatment options (41).
In summary, through a science outreach and community science project, we found clade 2.3.4.4bH5N1 highly pathogenic avian influenza viruses in New York City birds.
The presence of the virus poses a low but non-zero risk for humans and pets, and more awareness about the presence of this virus at the urban animal-human interface is needed.

Sample collection
No birds were killed for the purposes of this study.Work at Mount Sinai was approved by the Icahn School of Medicine at Mount Sinai Institutional Animal Care and Use Committee (IPROTO202300000038). Sampling in New York City parks was permitted by the New York City Parks and Recreation.The WBF operates under a Department of Environmental Conservation license, and both WBF and ACC operate under a U.S. Fish & Wildlife Service Migratory Bird permit.Fecal and swab samples from live, sick, and recently deceased or euthanized birds were provided by the WBF and ACC and were collected by veterinarians or licensed veterinary technicians as part of standard veterinary care of the birds.Water samples (3 mL each) were collected at the WBF's indoor waterfowl rehabilitation pool, using sterile pipettes and stored individually in cryotubes.Sampling focused on aquatic birds, particularly Anseriformes (including Anas ducks, geese, and swans), Ciconiiformes (including gulls, cormorants, and shorebirds), and raptors (including hawks, eagles, and falcons).All live bird sampling procedures were performed by New York State licensed wildlife rehabilitators employed by the WBF or ACC.Fecal, oropharyngeal, and cloacal swabs were collected from each bird using sterile flocked nylon-tipped swabs and stored individually in cryovials containing either MicroTest viral transport medium (Thermo Scientific, USA) or a medium containing 50% phosphate-buffered saline and 50% glycerol, supplemented with 1% antibiotic-antimy cotic 100X (Gibco, Thermo Scientific, USA).Samples were kept at 4°C for up to 4 hours, stored at −20°C for up to 7 days, and then stored at −80°C .The cold chain was main tained throughout delivery of samples to the laboratory.Environmental fecal samples that appeared fresh (still moist) were collected opportunistically in urban parks and greenspaces where birds were observed congregating, sacrificing the specific identity of the birds being sampled.Environmental fecal samples included in this study were collected on the following locations and dates (12 sampling field trips total): Manhattan, New York: Central Park (May 2022 and January and November 2023), Riverside Park (April and June 2023), Tompkins Square Park ( April 22 and May 2023), and South Bronx, New York: Saint Mary's Park (October 2023).For each location, samples were obtained over a wide area of interest and not a single point.To avoid sampling of the same bird more than once, samples were collected with a minimum distance of 20 cm between each other.A transect sampling strategy was employed when sampling around bodies of water, like city ponds.All samples were collected and preserved in the same manner as those collected from live birds.When possible, samples were collected avoiding visible uric acid and soil to prevent potential contamination with PCR inhibitors.

RNA extraction and RT-PCR
Fecal samples were diluted in phosphate-buffered saline, pH 7.4 (1X, Thermo Scien tific, USA) for processing.Suspended fecal samples, oropharyngeal swabs and cloacal swabs, were centrifuged at 4,000 × g for 15 min, and viral RNA was extracted from each supernatant using the QIAamp Viral RNA minikit (Qiagen, USA) according to the manufacturer's instructions.The Stool Total RNA purification kit (Norgen Biotek Corporation, Canada) was also used to extract RNA from fecal samples.Samples collected from the same bird were not pooled.The conventional two-step reverse transcriptase polymerase chain reaction (RT-PCR) was employed using the Invitrogen SuperScript IV first-strand synthesis system (Thermo Scientific, USA) for cDNA synthesis and Dream Taq Green PCR Master Mix (2X) (Thermo Scientific, USA) for RT-PCR.First, cDNA was synthesized using a minimum of 250 ng of RNA at 55°C for 10 min using a previously described universal primer [Uni12, AGCAAAAGCAGG (56)].Then, cDNA was amplified using previously described primers for HPAI H5N1 surveillance that target the nucleo protein [NP, NP1200 forward, CAGRTACTGGGCHATAAGRAC and NP1529 reverse, GCATT GTCTCCGAAGAAATAAG (57)], matrix [M, M52C forward, CTTCTAACCGAGGTCGAAACG and M253R reverse, AGGGCATTTTGGACAAAKCGTCTA (57)], and hemagglutinin [H5, H5.2344-1673 forward, TACCAAATAYTGTCAATTTATTCAAC and H5.2344-1749 reverse, GTAAYGACCCR TTRGARCACATCC (58)] genes.Primers for HA (H5) were included for prompt identification of HPAI H5N1 viruses, facilitating quick notification of partner organizations as neces sary to handle infected birds.Cycling conditions for the multiplex PCR consisted of a pre-denaturation step at 95°C for 1 min, followed by 30 cycles of denaturation at 95°C for 1 min, annealing at 45°C for 30 s, and extension at 72°C for 30 s, with a final extension step at 72°C for 5 min.PCR amplicons were visualized with SYBR Safe DNA Gel Stain in 2% Ultra-Pure Agarose (Thermo Scientific, USA).DNA bands were excised and purified using the QIAquick Gel Extraction Kit (Qiagen, USA) and sent for commercial Sanger sequencing (through Genewiz, New Jersey facility) to confirm the identity of samples that screened positive.Samples that screened positive for H5 HA and also were identified as H5 by Sanger sequencing (to exclude false positives) were reported to the USDA.The remaining sample material was moved to Mount Sinai's select agent facility for storage.

Next-generation sequencing
Samples which tested positive for H5, NP, and M by PCR were then used for next-gen eration sequencing.RT-PCR products were quantified on a Qubit 4 Fluorometer using HS DNA reagents (Invitrogen).A volume of 3.5 µL of the cDNA product was used in a 50 µL PCR reaction with Phusion™ High-Fidelity DNA Polymerase (2 U/µL) (Thermo Fisher).Three universal influenza A primers at 0.20 µM concentration were used in the PCR.Commonuni13 (GCCGGAGCTCTGCAGATATCAGTAGAAACAAGG), Commonuni12G (G CCGGAGCTCTGCAGATATCAGCGAAAGCAGG), and Commonuni12A (GCCGGAGCTCTGCA GATATCAGCA AAAGCAGG), 0.2 µM dNTPs, and 1X HF buffer were also components of the reaction.Single-reaction multiplex PCR was performed for sample amplification.Amplification occurred under the following cycling parameters: samples were initially denatured at 94°C for 2 minutes and then underwent five cycles of 94°C for 30 seconds, 45°C for 30 seconds, and 60°C for 3 minutes, followed by cycling parameters 31 cycles of 94°C for 30 seconds, 57°C for 30 seconds, and 68°C for 3 minutes and then a 4°C hold.PCR products underwent DNA purification and size selection using AMpure XP beads.Library preparation was performed using the Nextera XT DNA Library Preparation Kit (Illumnia, CA, USA) following the manufacturer's protocol to generate multiplex paired-end sequencing libraries.Post-fragmentation automated electrophoresis was performed using the Tape Station 1450 (Agilent Technologies).Sample libraries were quantified using a Qubit 4 Fluorometer (Invitrogen).Sample molarity was determined according to the following formula: Sample concentration in ng/μl Average number of bp x 660 g/mols x 10 6 Samples were pooled using Illumina's pooling calculator in equimolar amounts, creating a paired-end fragmented library pool in which each sample is represented with unique indices.Upon pooling and diluting samples to 4 nM using 8.5 pH 10 mM tris(hydroxymethyl)aminomethane (Tris)-HCl with 0.1% Tween 20 following "Protocol A: Standard Normalization method" from the "illlumina MiSeq System: Denature and Dilute Libraries Guide".Samples were sequenced using a MiSeq device (Illumnia, CA, USA) at a final concentration of 12 pM with a 15% PhiX spike in.After the 2 × 300 bp MiSeq paired-end sequencing run, the instrument performed base-calling on the data and collected reads with matching indices to generate paired-end forward and reverse read Illumina FASTQ files for each sample.
Illumina reads in the form of FASTQ files are produced by the MiSeq with adapter sequences on the 5' and 3' ends for sample identification; therefore, the first step in analysis was to remove the non-viral DNA.FASTQ files were uploaded to the Galaxy web platform, and several tools available through the usegalaxy.orgpublic server (59) were used for genome assembly.Cutadapt (Martin) was used to remove adapter sequences and to generate FASTA files for analysis.Metagenome de novo assembly was performed using the metaSPAdes approach.Nucleotide to protein BLAST (NCBI) was performed on generated sequences to confirm the identity and relevance (60).Influenza A virus segments were identified by NCBI Flu annotator (61).Nucleotide to nucleotide BLAST was run to determine reference sequences.FASTA files were then indexed and map ped to the reference genome using BWA-MEM2 (62) and Simple Illumina parameters.BWA-MEM2 produces sorted and indexed BAM files.To account for potential laboratoryderived influenza virus contaminants that could alter the final consensus sequences, we performed the same analysis on a filtered read set, generated by taking the unaligned reads left over after aligning the data to the A/Puerto Rico/8/1934 reference genome (NCBI Taxonomy ID id183764) using Bowtie2 (63).To ensure only reads definitively derived from known contaminants were excluded and no reads derived from influenza virus from the sample were discarded, we adjusted the parameters of the alignment algorithm.To increase strictness, the alignment was run in the end-to-end mode; the mismatch penalties, gap opening, and extension penalties were increased; and the maximum seed substring length allowing for zero sequence mismatches in the seed alignment during multiseed alignment was used.Final analysis and consensus sequences were produced using the indexed BAM file and iVar Consensus (64) and Geneious.The antigenic genotypes of the consensuses sequences were confirmed using the NCBI Flu annotator.Nucleotide to protein BLAST (NCBI) was performed on generated sequences to confirm identity and relevance.Sequence relevance is determined by the location of isolation and date of isolation in terms of the closest reference sequence, as well as its similarity to existing reference sequences for HPAI H5N1 viruses from 2022 onward.

FIG 1
FIG 1 Map of location of birds that tested positive for highly pathogenic avian influenza H5N1 virus (HPAI H5N1) in New York City.The approximate locations are plotted based on geocoded addresses (latitude and longitude), providing a visual representation of affected areas.Major parks and natural areas are highlighted in green and labeled for context.The map was created using the leaflet package for mapping visualizations, with additional spatial data handling and esthetic enhancements performed using the sf, ggplot2, and dplyr packages in RStudio/Posit (Version 2023.09.1+494).The basemap was provided by CARTO, with data sourced from OpenStreetMap under the Open Data Commons Open Database License (ODbL) by the OpenStreetMap Foundation (OSMF).

FIG 2
FIG 2 Amino acid sequence analysis of detected HPAI H5N1 strains.A multiple sequence alignment was performed using the HA sequences of NYCVH-detected strains and A/bald eagle/FL/W22-134-OP/2022 in Clustal Omega.Only areas of the alignment with amino acid differences are shown.Residues that were different in NYCVH strains compared to A/bald eagle/FL/W22-134-OP/2022 were highlighted in red on an H5N1 structure based on A/chicken/Vietnam/4/2003 (PDB #6VMZ) (42), visualized with UCSF ChimeraX.V510I and A522V are changes to positions not present in the PDB #6VMZ ectodomain structure.The receptor-binding site and antigenic sites are indicated by blue highlighting and red outline, respectively.

TABLE 1
Details on samples for virological analysis collected from different avian species present at Wild Bird Fund or Animal Care Centers of New York City

TABLE 1
Details on samples for virological analysis collected from different avian species present at Wild Bird Fund or Animal Care Centers of New York City (Continued)

TABLE 2
Clinical and sampling information for wild birds positive on RT-PCR for highly pathogenic avian influenza virus (H5N1) in New York City from January 2022 to November 2023 June 2024 Volume 98 Issue 6 10.1128/jvi.00626-246

TABLE 4
Genotypes of strains detected in New York, New Jersey, and Connecticut from August 2022-April 2023 a It remains unclear if the chicken in Marcus Garvey Park was intentionally released or escaped from captivity elsewhere, as does the context in which they became infected (in captivity or after release).It is important to state that all six positive samples came from either the Wild Bird Fund or the Animal Care Centers of New York City, stressing the important role that urban wildlife rehabilitation centers can play in urban viral surveillance efforts.The detected HA and NA sequences clustered with other H5 and N1 sequences from North and South American clade 2.3.4.4bH5N1 viruses circulating at approximately the same time, and they belonged to two different genotypes, which are both reassortants between the Eurasian 2.3.4.4bH5N1 and American avian influenza viruses.It has recently been shown that these reassortants can have increased pathogenicity in mammals as compared to the full Eurasian genotype of 2.3.4.4bH5N1 (