Phylogeographic Analysis of Tick-borne Pathogen - Recent Advances and Issues in Environmental Science - page 83

(12.8%) were reported solely in ecoregion cluster 2 with 64.5% identical amino acids.
Twelve R1 unique sequences (30.7%) were found only in ecoregion cluster 3, sharing
64.5% of their amino acids. Only three R1 sequences were exclusively associated with
ecoregion cluster 4, with 77.4% identical amino acids. All of the A. marginale MSP1a
R1 sequences within each ecoregion cluster appeared to be under positive selection
as shown by d N /d S indexes of 1.83, 1.61, 1.54, and 1.21 for ecoregion clusters 1-4,
respectively.

Table 2. MSP1a R1 and RL repeat sequences of unique sequences unambiguously associated to only
one ecoregion cluster, including the consensus sequence of the isolates of that cluster.

MSP I a Repeat Unique sequences Consensus sequence Other strains and number of

substitutions

R I /Ecoregion I 4, 8, 16, 56, 60, 64, 67,

gamma, pi, tau

***SSA***QQ*SSV*S*S**
AS*SSQ*G--

A(O), B(O), D( I), T(O), 13( I),
23( I), alfa(O)

R I /Ecoregion 2 28, 48, 53, E, F,

epsilon

***SS**GQQQESSV***S*-
ASTSSQLG--

A(O), B(O), L( I), T(7), 13( I),
23( I), alfa(O)

R I /Ecoregion 3 I, 3, 5, 6, 27, 33, 34, 39,

M, 0, Q, U

**SSSA*GQQQESSV-*
QA*TSSQLG--

A(O), D(O)

R I /Ecoregion 4 I, J, K *D*S*A*GQQQESSVSSQS

*QASTSSQLG--

A(O), B(O), L(O), alfa(O)

RUEcoregion 1 8, 9, 12, 15, 59, 61 , 66 ***SSSA**QQQES*V*SQS**

ASTSSQ*G--

B(O), C(O), M(O), 18(0),
27(0), gamma(O)

RUEcoregion 2 10, 31 , 52, pi, beta *DSSSA**QQQ*S*V*S*S*-

ASTSSQLG--

F(O), H(O), M(O), 27(0),
gamma(O)

RUEcoregion 3 B( I), F(O), H, 18(0),

gamma(O)
RUEcoregion 4 none B, C, H
Other strains recorded from more than one ecoregion cluster are included in the last column, stating the number of
substitutions in their amino
acid sequences as compared with the consensus sequence in that ecoregion cluster (in parent heses).

3, 7, 35, 37, 38, 44, E, N,
P, Q, U, ro

*DSSSAS*QQQESS**S*S
*QA**S*Q*G--

Differences were found among the R1 sequences at each ecoregion cluster. Com-
parison of consensus sequences between clusters 1 and 2 revealed 22.5% of amino
acid differences. Ecoregion clusters 1 and 3 differed by 25% of their consensus se-
quences while ecoregion clusters 2 and 3 had only 19% of different amino acids in
their R1 sequences. Five R1 sequences, T, 13, 23, D, and L were found simultaneously
in two of the ecoregion clusters (Figure 2, Table 2). Details of their similitude with
the consensus sequences of each ecoregion cluster are included in Table 2. Three R1
sequences, A, B, and alpha appeared associated with different ecoregions. Pairwise
comparisons between A. marginale R1 sequences demonstrated that genetic distances
were lower within than between ecoregion clusters (Table 3). However, when strains
were compared as clustered geographically, intra-cluster distances were much higher,
revealing a clear heterogeneity between geographic strains and rejecting the hypoth-
esis of a pure geographical association. Mantel's test on R1 sequences was 0.82 (P <
0.001) when applied to ecoregion clusters using only unique sequences. Mantel's test
on R1 sequences dropped to 0.72 (P < 0.005) when every sequence in the ecoregion
clusters was included. The same test provided a value of 0.31 (P = 0.145) for the dis-
tances matrix based on geographical association of strains.