The distribution of eastern equine encephalitis virus (EEEV) activity in Maine is currently not well understood. Before 2005 EEEV had only been detected in an American goldfinch (Carduelis tristis) in York County (Lubelcyck et al. 2013). Between 2005 and 2007 EEEV was detected in birds, horses, and mosquito pools in York and Cumberland counties in southern Maine and at that time it was thought that southern Maine represented the northern limits of EEEV activity in the North America (Lubelcyck et al. 2014). However, in 2009 an epizootic of EEE occurred in 5 counties in southern (York and Cumberland Counties) and central Maine (Kennebec, Waldo and Penobscot Counties) and it involved fatalities in at least 15 horses, a llama and 3 flocks of pheasants (Gibney et al. 2011, Lubelcyck et al. 2013). This suggested a much broader distribution of EEEV activity in the state. Following the outbreak in 2009, we conducted a pilot study and found high EEEV antibodies levels in free-ranging deer (Odocoileus virginianus) sera in central Maine which suggested that EEEV is endemic in this region (Mutebi et al. 2011). Since then deer serosurveys have been utilized to study the distribution of EEEV activity and other arboviruses in other states in Northeastern United States (Berl et al. 2013, Nofchissey et al. 2013).

In 2010, we expanded the deer serosurvey studies to include most of the counties in Maine to get a more complete picture of the distribution of EEEV activity in the state. In the course of these studies we noticed that the number of deer tagged was significantly reduced in northern Maine corresponding to lower densities in that region of the state (Lavigne 1997). We also noticed that the numbers of moose (Alces alces) tagged increased in northern Maine suggesting increased moose population size in that part of the state (Wattles and DeStefano 2011, Lubelcyck et al. 2014). We screened the moose sera and detected EEEV antibodies in approximately 11% of the samples which showed that moose were exposed to EEEV infections in northern Maine (Lubelczyk et al. 2014). Previously we had observed that 29% of the moose sera collected in northern Vermont were positive for EEEV antibodies (Mutebi et al. 2012). Taken together these observations suggest that similar to deer, moose can be used as sentinels to detect EEEV activity. In the present study, EEEV antibody positive moose sera were detected near Fort Kent very close to the Canadian border suggesting EEEV activity throughout the state of Maine (Lubelczyk et al. 2014). However there are large areas in the state of Maine where EEEV activity is unknown. Since the distribution of EEEV activity in North America is patchy and not uniform throughout the distribution range (Morris 1988) it is essential to investigate all areas of the state to obtain an accurate picture of EEEV activity. In this manuscript we present and discuss our observations of the expanded deer serosurvey in Maine in 2010.

Three hundred and thirty-two (332) deer serum samples were collected from 15 of the 16 (93.8%) ME counties and 47 (14.2%) were positive for EEEV antibodies by PRNT (Tables 1 and 2). This shows a much wider distribution of EEEV activity in Maine than previously reported by Mutebi et al. (2011) and Lubelczyk et al. (2014). The percentage of EEEV antibody positive deer sera was ≥10% in six counties Piscataquis (100%), Somerset (28.6%), Waldo (22.2%), Penobscot (21.7%), Kennebec (13.7%) and Sagadahoc (10%) (Table 1). Although the highest percentage of EEEV positive sera (100%) was detected in Piscataquis County, only 2 deer samples were collected in that County and therefore the high percentage may be attributed to the extremely low sample size. In counties where more than 20 samples were collected the highest percentage of EEEV antibody sera was detected in Somerset County (28.6%) followed by Waldo (22.2%), Penobscot (21.7%), Kennebec (13.7%) and Cumberland (7.3%). Positive sera were detected in all the 6 counties (Somerset, Waldo, Penobscot, Kennebec, Cumberland and York) that were positive in 2009 (Mutebi et al. 2011) which suggest endemic EEEV activity in these counties. The high percentage of positive sera detected in Somerset County in both 2009 (19%) and 2010 (28.6%) suggest consistently high EEEV activity in this County. Similarly the percentages of positive deer sera detected in Cumberland, York and Penobscot Counties (9.4%, 5% and 12.5% respectively) were very similar to those observed in 2009 (7.3%, 2.9% and 21.7% respectively) (Mutebi et al. 2011) suggesting consistent EEEV activity in this county as well as well. However, in Kennebec and Waldo counties, the percentage of EEEV positive sera detected was substantially lower in 2010 (3.8% and 7.9% respectively) when compared to what was detected in 2009 (13.7% and 22.2% respectively) (Mutebi et al. 2011) which suggests that EEEV activity may be a recent expansion into these counties. As epizootic activity of the virus was also higher in 2009 than 2010 in this region of Maine (Gibney 2011), the parallels seen in cervids may also speak to the usefulness of this surveillance method.

Although positive sera has been detected in York and Cumberland in both 2009 and 2010 and EEEV activity has historically been associated with these two southern counties, the percentages of positive sera detected in these two counties were substantially lower than those detected in some counties in central Maine such as Somerset, Waldo and Penobscot (Table 1) (Mutebi et al. 2011). This suggests that the highest EEEV activity in the state may not be in southern Maine, despite historical viral activity in veterinary cases (Lubelczyk 2013). Most of the positive sera were detected at the junction where five counties, Somerset, Piscataquis, Penobscot Waldo and Kennebec come together in central Maine suggesting that this may be a focus of EEEV activity (Fig. 1).

EEEV antibodies were not detected in sera collected in five counties: Franklin, Knox, Lincoln, Oxford and Washington (Table 1, Fig. 1). However, with the exception of Washington County where 20 serum samples were collected, the sample sizes from the other four counties were limited to one or 2 samples (Table 1) and the inability to detect positive samples in these counties may be attributed to the small sample sizes. Although there was a sizable number of samples from Washington County (20) (Table 1), all these samples were collected in the coastal areas in the south and none from the central or northern areas of this county (Fig. 1). The absence of positive deer sera in southern Washington County is intriguing especially because the coastal areas of this county have numerous ideal habitats for Cs. melanura and other EEEV vectors.

Only 17 (5.1%) deer sera were collected from the northern part of the state; 15 (4.5%) from Aroostook County, and 2 of the 23 samples from Penobscot County (Table 1, Fig. 1). Our observations based on deer tagging data suggest that deer populations are not uniformly distributed throughout Maine but rather more restricted to the southern, costal and central parts of the state (Rand et al. 2003). These observations are consistent with the conclusions by Krom (http://www.umaine.edu/cfru/Events/Munsungan_DWA_12.07/Krohn_HistoricalEcology.pdf). Krom compiled available information on the distribution of white-tailed deer in Maine, from the time of European settlement to the early 2000s and found significant distribution variations of deer populations over time; currently deer populations are abundant in the south, central and coastal areas of Maine and very low in the North.