By Carol A. Kemper, MD, FACP

The Origins of Smallpox Vaccine Virus

The origins of vaccinia virus, used for smallpox vaccination, have long been debated. For years, many believed that cowpox virus was the source for vaccinia virus, ever since British surgeon Edward Jenner from Berkeley, Gloucestershire, first published accounts in 1798 and 1799 regarding the use of cowpox virus for scarification to generate protective immunity (ostensibly following observations that milkmaids were less likely to succumb to smallpox). However, it has been recognized for years that vaccinia virus differs from cowpox virus (CPXV), and the exact origins for the vaccine virus have remained a mystery. Complicating matters is the fact that various smallpox vaccines were manufactured in different countries and remained in use for more than a century before the World Health Organization (WHO) standardized the vaccine in 1967 using four different strains of vaccinia virus.

A sample of 1902 smallpox vaccine manufactured by H.K. Mulford Co., an American company in Philadelphia, was recently discovered. DNA extracted from the vaccine was submitted to whole genome amplification and compared with 65 other published genome orthopoxvirus sequences.1 Contrary to the cowpox theory, the 1902 vaccinia virus most closely resembles horsepox virus (99.7%), using different phylogenetic algorithms. Interestingly, deletions found at each end of the vaccine strain were not observed in either natural cowpox or horsepox viruses, but were similar to current vaccinia strains.

This suggests that horsepox, and not cowpox, may have served as the progenitor virus for vaccinia virus. In actuality, orthopox viruses are known to infect many different animals, and horsepox virus and CPXV are very similar, and probably originally derived from rodent poxvirus, which serves as a reservoir for cowpox infection. Further, an earlier study observed that CPXV is not a single species but a composite of different strains of virus (up to five) that can infect cows, other animals, and (accidentally) humans.

Carroll and colleagues examined whole genome sequences of CPXV in 2011, creating a phylogenetic tree, with comparisons to other animal orthopoxviruses.2 CPXV fell into two major clades, one of which contained strains of virus from the United Kingdom and Germany. The other clade contained a group of viruses including buffalopox (from India), horsepox, rabbitpox, and CPXV strains found in Finland and Austria. CPXV isolates from Germany, most of which came from accidental infection of humans, showed more genetic variability than those from the United Kingdom, suggesting independent evolution and, possibly, different rodent reservoirs. In reviewing that phylogenetic tree, horsepox virus certainly appears most closely related to three different strains of vaccine-derived vaccinia virus.

Horsepox virus apparently no longer exists in nature, although samples of horsepox virus are maintained at the CDC. In 2006, scientists in New York published the sequence of horsepox virus, derived from a wild strain recovered 40 years earlier from horses in Mongolia.

Orthopox viruses have fairly sizeable genomes, and both CPXV (averaging 230 kb pairs) and horsepox virus (~212 kb pairs) are about 30 kb pairs larger than vaccinia virus — and 30 times larger than poliovirus. Discord occurred in 2017 when a Canadian researcher, using DNA sequences purchased from a German company, was able to recreate horsepox virus.3 Apparently, he had applied to the CDC for samples of horsepox virus, but was declined — so decided to make his own. Both the United States and the WHO prohibit the production of variola (smallpox) virus, and researchers are not allowed to recreate more than 20% of the variola genome. It has been hoped that no one would try or that it would not be so readily possible. The United States maintains a list of 15 “dual-use” agents, including smallpox virus, for which research and production are restricted, but horsepox is not included on this list. Nonetheless, the implication of the Canadian work is obvious. As one researcher put it, “No question, if it’s possible with horsepox, it’s possible with smallpox.”

Meanwhile, the exact origins of Jenner’s “vaccine” are becoming somewhat less of a mystery. All of this genomic work strengthens the likelihood that horsepox may be the progenitor for the original vaccinia virus. Earlier WHO reports of the origins of vaccinia virus in 1988 suggested that many physicians, including Jenner, likely turned to horsepox, rather than cowpox, for scarification. For one, cowpox virus was available only sporadically — and physicians at the time may have substituted horsepox as needed. They even may have thought they were the same virus. But one of the best parts of researching this article was the term used for inoculation of horsepox virus “equination” as in “equinate your kids,” (although that equination has taken on new meaning with online horse racing).


  1. Schrick L, Tausch SH, Dabrowski PW, et al. An early American smallpox vaccine based on Horsepox. N Engl J Med 2017;377:1491-1492.
  2. Carroll DS, Emerson GL, Li Y, et al. Chasing Jenner’s vaccine: Revisiting cowpox virus classification. PLoS One 2011;6:e23086.
  3. Kupferschmidt K. How Canadian researchers reconstituted an extinct poxvirus for $100,000 using mail-order DNA. Science 2017; July 6.

No More Fun Helping Mommy Bake

SOURCE: Crowe SJ, Bottichio L, Shade LN, et al. Shiga toxin-producing E. coli infections associated with flour. N Engl J Med 2017;377:2036-2043.

Being a kid just isn’t as much fun as it used to be. One of my favorite things growing up was getting to lick the bowl whenever mom made a cake. And, my mother and my grandmother used to set aside time every few months to make 50-60 pie crusts for the freezer, yielding plenty of extra pie dough to play with and nibble on. Food safety experts now warn against this practice.

While some home cooks may recognize raw eggs as a potential source for Salmonella and Campylobacter infection, few would suspect flour as a source for serious infection. It sits on the shelf for months, it looks clean, white, and innocuous — and it’s dry. Not the usual medium one would imagine for a food-borne illness.

In 2016, a multi-state outbreak of Shiga toxin-producing Escherichia coli (STEC) serogroups O121 and O26 occurred, resulting in a total of 63 infections in 24 states within the United States and Canada. Seventeen people required hospitalization. Many of those affected clustered within families, suggesting a common source for infection. STEC typically causes a gastroenteritis with abdominal cramping and bloody diarrhea. A minority of those infected, especially children younger than 5 years of age, may develop the more serious complication of hemolytic-uremic syndrome.

As soon as the outbreak was recognized, a multijurisdictional investigation was launched. Conditional logistic-regression analysis suggested that the infection was related to the use of a single brand of flour. The investigation quickly pointed to a single manufacturer of flour from one manufacturing facility in Kansas City. The organism was isolated from flour samples, and whole genome sequencing of clinical isolates and strains isolated from food samples proved they were closely related. Tasting or handling unbaked cookie dough or uncooked batter proved to be the culprit activity. Multiple step-wise recalls of flour ensued, resulting in the eventual destruction of 10 million pounds of flour, including unbleached, all-purpose, and self-rising flours, from this single facility.

The CDC and the FDA offer several tips for avoiding food-borne illness:

  • Do not eat raw cookie dough, cake mix, batter, or any other dough or mix that should be baked or cooked.
  • Keep raw food separate from other foods while preparing them and keep in mind that dry flour can spread easily when sifting or mixing.
  • After contact with raw foods, wash hands, work surfaces, and utensils well with good soap and water.
  • Restaurants and preschools should not allow children to play with raw dough.

Annual Influenza Vaccination of Physicians

SOURCE: California Department of Public Health. Healthcare-Associated Infections Program. Healthcare Personnel Influenza Vaccination in California Hospitals. Nov. 14, 2017. Available at: Accessed Dec. 11, 2017.

To help achieve the CDC Healthy People 2020 goal and reduce healthcare-associated illness, many counties within California launched a goal to achieve 90% or better vaccination of healthcare workers beginning in 2011-2012. Thirty-five counties in the state, including our own, have issued a public health order requiring unvaccinated healthcare providers to wear a mask whenever working in clinical areas during flu season. Since the introduction of these county mandates, influenza vaccination of healthcare personnel has increased steadily from 63% in 2010-2011 to 83% in 2016-2017. The highest rates of compliance (87%) were observed in paid employees of hospitals, while the lowest rates (67%) were observed among licensed independent practitioners, including physicians and physician assistants, not directly paid by a hospital. Influenza vaccination of licensed independent practitioners in 2016-2017 plateaued and even slightly fell compared with earlier influenza seasons.

Almost one-third of California counties have achieved a > 90% vaccination rate across all healthcare provider groups. Reported influenza vaccine rates for healthcare personnel working within the 35 California counties with a county mandate (84%) were higher than those working in one of the 20 counties without a mandate (81%).

Hospitals throughout the United States now are required to collect data for annual influenza vaccination for all healthcare personnel working within their facility, paid or otherwise, and report these data to the National Healthcare Safety Network using their secure, web-based system. To achieve the 90% goal, developing additional strategies to improve influenza vaccination rates is important.

For one thing, California state law requires acute care hospitals to offer influenza vaccination at no cost to all healthcare personnel. For another, California hospitals are required to provide an attestation/declination form to all healthcare personnel working within their facility. Our community-based hospital offers a badge decal to physicians upon proof of vaccination. But how do you best monitor and enforce this process? What “teeth” do community hospitals have, especially in counties or states without such a mandate? And how do you get community-based physicians to submit proof of their voluntary vaccination, short of threatening resumption of their hospital staff privileges?

While the California Department of Public Health report suggests that independent physicians vaccine rates lag behind paid employees, I suspect this is largely due to a gap in voluntary physician reporting. For example, in our hospital, thus far only 17% of physicians have returned their attestation form for the current influenza season (and which physician in their right mind would voluntarily provide a declination?). My guess is that most have been vaccinated.

Hospitals are already squeezed to improve patient safety in myriad ways and financially penalized if they fall short. For a community-based hospital to find the FTE to track down forms for 1,600 physicians is no small task. It’s easy for authorities to draft laws and issue mandates, but then adequate bonus funding should be provided for hospitals to perform this task successfully.