All four cases and family members were interviewed using a standard questionnaire to collect basic demographic information, information pertaining to exposure to domestic ruminant animals (e.g., sheep, goats, dairy cattle), movement of animals on and off the farm and introduction of new animals, daily farm activities, number of years of animal exposure, glove use during animal care, and observations of parapoxvirus infections in their animals. In addition, individuals with a history of parapoxvirus infection (including the four primary cases) were questioned regarding the symptoms associated with their illness, health care seeking behavior, and treatments received. Samples were collected with dry swabs from both symptomatic domestic ruminants and bystander animals, and their pens.

Farms neighboring case farms were selected on a referral basis (i.e., referred by case farm owner) to serve as points of comparison. Owners, family members and employees were interviewed using a standard collection tool similar in nature to the one described above. This survey was supplemented with photographic aids of classic presentations of parapoxvirus infections in both humans and animals. If interviewees themselves had evidence of recent parapoxvirus infection (i.e., active or resolving lesions) they were offered free testing to include swab of lesion (if lesion was active) or a blood draw for parapoxvirus serology (if lesion was resolving). In addition, such individuals were questioned regarding their symptoms, health care seeking behavior, and treatments received. Whenever possible, domestic ruminant animals were visually examined and swabs of their oral mucosa and the animal environment (e.g., pens, bedding, discarded wool and artificial milk nipples, etc.) were obtained. We utilized SAS V9.1 (Cary, NC, USA) for all statistical analyses. Odds ratios (ORs) with 95% confidence intervals (CIs) were calculated using logistic regression analysis.

A self-administered survey was mailed to veterinarians from the three counties in which human parapoxvirus infections had been confirmed and one adjacent county (Lincoln), as well as to state/federal veterinarians from Missouri Department of Agriculture and USDA (assigned to Missouri). This brief survey requested information regarding (1) frequency of consultation for parapoxvirus infections among animals, (2) frequency of consultation for parapoxvirus infections among humans (i.e., herders), (3) knowledge of the use of the orf vaccine among local herders.

In order to obtain a broader sample of Missouri livestock handlers, animal exhibitors of target species (goat, sheep, cattle) at 3 county fairs and the 2006 Missouri state fair were asked to respond to the ‘herder survey’ (see above). In addition, sheep from exhibits at the State Fair were examined by one investigator (H.P.) and oral swabs were obtained. The owners of selected sheep were queried regarding previous orf in their flock in the previous year.

Swab and scab specimens were hydrated with 300 µL of phosphate buffered saline and centrifuged for 2 minutes in a swab extraction tube system (S.E.T.S., Roche Diagnostics Corporation, Indianapolis, IN, USA) per the manufacturer’s guidelines. DNA was extracted from 100 µL of elutant with a standard tissue protocol for the BioRobot EZ1 (Qiagen^®) and the product was analyzed by real-time polymerase chain reaction (PCR) with parapoxvirus generic and orf specific assays.

The primer/probe sequences were selected from the multiple DNA sequences alignment of DNA polymerase gene (GenBank AY386263, NC_005336, NC_005337) with Primer Express (version 1.5; Applied Biosystems). These included: (1)

Parapoxvirus-generic assay:

Parapox-generic forward primer (5'-GTA_CCG_CGC_CAT_GTC_CA-3')

Primers and probe were synthesized in the Biotechnology Core Facility (CDC, Atlanta, GA, USA), utilizing standard phosphoramidite chemistry. Primers and probe were synthesized in the Biotechnology Core Facility (CDC, Atlanta, GA, USA), utilizing standard phosphoramidite chemistry. The detection probe contained a 5' reporter molecule (FAM) and a 3' aminomodifier (both from Glen Research, Sterling, VA, USA). A 3' Blackhole quencher molecule (Molecular Probes, Eugene, OR, USA) was conjugated to the 3' amino group after synthesis.

Real time-PCR assay conditions were optimized according to standard protocols (protocol 04304449, Applied Biosystems, Foster City, CA, USA) by adjusting primer and probe concentrations, and thermal cycling temperatures/duration. Each reaction (30 µL) contained 2× Taqman universal PCR Master Mix (Applied Biosystems, Foster City, CA, USA), 25 µmol/L of each primer and probe, and 1 µL of template DNA. Thermal cycling conditions for the ABI PRISM 7900 (Applied Biosystems, Foster City, CA, USA): one cycle of 95 °C for 6 min; followed by 45 cycles of 95 °C for 5 s, and 60 °C for 25 s. PCR amplification is based on fluorescent emission after annealing/elongation (60 °C).

Serum specimens were evaluated by the following immunologic technique. Antibody to orf virus was detected by an indirect immunofluorescence assay [12] using pseudocowpox virus as the antigen. For IgG detection, serial two fold dilutions of sera, starting at 1:8 were added to pseudocowpox virus infected acetone fixed cell spots and incubated at 37 °C for 30 min. Slides were then washed in phosphate-buffered saline (PBS) and anti-human IgG (Panbio, Columbia, MD, USA) added and allowed to incubate for 30 min at 37 °C. The slides were then washed with PBS, mounted with cover slips in polyvinyl alcohol (PVA) and read with a fluorescence microscope. Results were reported as the reciprocal of the highest dilution at which specific fluorescence was observed. Titers of 1:8 or greater were considered positive. For IgM detection, serial two fold dilutions of MarSorb™ G (MarDx Diagnostics, Inc., Carlsbad, CA, USA) treated sera starting at 1:10 were tested with a primary incubation of 45 min at 37 °C, followed by a 30 min incubation with anti-human IgM conjugate (Panbio, Columbia, MD, USA) at 37 °C. Slides were washed, mounted and read using a fluorescence microscope. Titers of 1:10 or greater were reported as positive.

PCR amplified specimens were cleaned and concentrated according to standard protocols (short protocol, DNA Clean & Concentrator Kit, Zymo Research, Orange County, CA, USA) in order to remove unincorporated primers and dNTPs. The specimens were then sequenced according to standard protocols (BigDye Terminator v3.1 Cycle Sequencing Kit, applied Biosystems, Foster City, CA, USA). Each reaction (10 μL) contained 4 μL Terminator Ready Reaction Mix, 3 μL sterile dH₂0 (to adjust total volume to 10 μL ), 1 μL of each sequencing primer (concentration 1.6 μM), and 2 μL of template DNA. Thermal cycling conditions: an initial denaturation at 96 °C for 1 min, followed by 96 °C for 10 s, 50 °C for 5 s, 60 °C for 4 min for 30 cycles followed by holding at 4 °C.

The samples were then purified according to standard protocols (DyeEx 2.0 Spin Protocol for Dye-Terminator Removal, Qiagen, Valencia, CA, USA).

DNA sequencing primers targeting a virion core protein gene (OV_SA00_027, GenBank NC_005336) and a virion morphogenesis gene (OV_SA00 108, GenBank NC_005336) were synthesized in the Biotechnology Core Facility (CDC, Atlanta, GA, USA), utilizing standard phosphoramidite chemistry:

Primers orf_027 for: 5' CGT CTA GAG CAT GCC CTC GTA and orf_027 rev: 5' CGA TCC TCG AGT CGC TTC TG generated a 440 bp DNA fragment and primers orf_108 for (AAG GCG AAC TCG TGC TTG A) and orf_108 rev (CCT CAT CAT TCT CGA CGA TAT G) generated a 870 bp DNA fragment.

The sequences of a virion core protein gene and a virion morphogenesis gene from clinical sample ORFV 031, 036, and 044 have been deposited in GenBank under accession no. FJ442820 and FJ460507, FJ460505 and FJ460508, FJ460506 and FJ460509, respectively. The sequences were concatenated and aligned with BioEdit [13] and Clustal alignment programs [14]. Phylogenetic analysis was performed using the Bayesian analysis software package (v1.7, http://beast.bio.ed.ac.uk), BEAST, BEAUti, and Tracer [15]. The analyses run MCMC chain length of 10,000,000 with a GTR + Gamma (category 4) + Invariant sites nucleotide rate substitution model, strict molecular settings and sampling of every 1,000 states.

Characteristics of the four confirmed parapoxvirus cases reported to CDC by Missouri Department of Health and Senior Services are represented in Table 1. Three of four individuals who developed these infections were directly involved in the manual feeding of ill juvenile animals; cases did not use gloves while handling the ill animals. Two of the three operations (farms) were closed (did not obtain new animals in the past year). None of the case farms reported using the orf vaccine. We did not identify any additional recent cases (in the preceding year) among family members or employees.

Five additional farms in proximity to three of the four cases were evaluated with employee questionnaires and animal assessment/sampling; the fourth case (orf virus) was excluded because of lack of proximal farms and limitations in time for the investigation. Three of these five farms had overtly infected animals (sheep and goats). The kids examined were younger than the infected lambs (3–5 weeks old vs. 12–16 weeks old) and had more exuberant lesions, especially on their gingiva (Figure 1). Two additional cases of employees with orf lesions were identified during the evaluation of one neighboring dairy farm. As the lesions were healing and not amenable to confirmatory diagnosis using PCR, serology was performed to assess infection etiology as described above. Both employees had IgM and IgG titers to parapoxvirus ≥1:128. Both employees were young men from an urban area who had arrived several months before as part of a youth rehabilitation program; neither had prior farm animal exposure before being employed on this dairy. They both cared for kids (including bottle feeding) on a daily basis, several of which were actively infected at the time of our visit (infection status was confirmed by real-time PCR).

In addition, questionnaires were administered to livestock owners at county fairs. A total of 97 questionnaires from individuals on community farms and attending Missouri county fairs were completed. Seventeen percent of respondents (n = 16) reported having at least one perceived parapoxvirus infection in the past. Thirty-five respondents reported the age at which they began farming and their current age, allowing us to compute the number of farm years for this subset (i.e., potential for contact with parapoxvirus-infected livestock). These respondents had collectively 1,138 farm years and 3 infections yielding a rate of 26 infections/10,000 farm years. Table 2, Table 3 summarize the comparison of demographic, knowledge and risk factors evaluated for association with parapoxvirus infections. In the univariate analysis, age was not a significant risk factor, but male sex and having knowledge of parapoxvirus-associated conditions prior to the survey were significant risk factors;having an occupation listed as “other” (not a herder, student or homemaker) was protective. Only male sex (OR 5.1 [1.5–17.2]) remained an independent risk factor (p < 0.05) for parapoxvirus infection in the multivariate model (data not shown). Frequency and type of healthcare utilization as well as treatment for individuals with a history of parapoxvirus infections is described in Table 4. The majority of cases (13/21, 61.9%) sought healthcare during their presumptive infection, most (10/13, 76.9%) from a general practitioner. Among those evaluated by a health care provider, antibiotic prescription was common (11/13, 84.6%) and if a topical antiviral treatment were available in the future 19/22 (86.4%) of cases thought it would be worthwhile seeking healthcare to obtain a prescription.

Of the 52 veterinarians invited to participate, 33 surveys were returned. Five surveys were discarded (three were retired, one had no forwarding address, one was incomplete) leaving data from 28 respondents (54%) available for analysis. Sixty-four percent of respondents had been consulted for orf in sheep and goats at some point in their career; 32% reported having been consulted at least once a year. Only 29% percent of respondents reported ever having been consulted for pseudocowpox virus infection (in dairy cattle) during the course of their careers. Thirty-six percent surveyed were aware of the use of the live (non-attenuated) orf virus vaccine by local herders. Finally, 46% of veterinarians surveyed had been consulted for suspected human orf virus infection at some point; 18% reported being consulted at least once a year. Twelve percent of survey respondents had been consulted for human pseudocowpox virus infection at least once in their career.

No overtly infected animals were found on any of the case farms. Swabs collected from previously infected, randomly selected animals and animal environments (pen walls, soil and bedding) were all negative for parapoxvirus DNA signatures (n = 12). Of the oral swabs obtained from domestic ruminants and/or environmental swabs from community farms 100% (7/7) sheep and goats with symptomatic disease as well as 21% (3/14) asymptomatic sheep and goats were positive for orf virus by real-time PCR. The only PCR-positive environmental sample was a swab of an artificial nipple used to feed kids at a commercial goat dairy. All sheep visually examined at the State Fair were free of overt signs of orf infection yet 20% (5/25) randomly sampled were positive for orf virus DNA via PCR, including two from a farm without report of any orf disease in the past year.

Sufficient orf virus DNA was obtained from 2 animal samples (kid and lamb on community farms) to perform DNA sequencing. A virion core protein gene on the left end and an ATPase gene on the right end of the genome were partially sequenced, and the sequences were concatenated for the phylogenetic analyses to determine the relationships between the isolates collected and those of known strains of orf virus previously sequenced.

When gene sequences of the virus strains obtained from this investigation were compared to those from published sources and an additional virus archived at CDC, sheep and goat isolates tended to cluster in species-preferential fashion (Figure 2). This clustering occurred despite wide geographic distances and decades of time between the collection of isolates.