Pinus sylvestris Breeding for Resistance against Natural Infection of the Fungus Heterobasidion annosum

Abstract

Increasing resistance against biotic and abiotic factors is an important goal of forest tree breeding. The aim of the present study was to develop a root rot resistance index for Scots pine breeding and evaluate its effectiveness. The productivity, branch diameter, branchiness, stem straightness, spike knots, and damage from natural infection of root rot in 154 Scots pine open-pollinated families from Latvia were evaluated through a progeny field trial at the age of 38 years. Trees with decline symptoms were sampled for fungal isolations. Based on this information and kriging estimates of root rot, 35 affected areas (average size: 108 m²; total 28% from the 1.5 ha trial) were delineated. Resistance index of a single tree was formed based on family adjusted proportion of live to infected trees and distance to the center of affected area. Heritability for resistance to root rot based on the value of this index, was high (0.37) and comparable to indices of growth traits. Correlations of family breeding estimates between resistance to root rot and the other traits were not significant, except for a weak, yet significant, positive correlation with diameter at breast height and branch diameter. Selection index including only growth traits (height and stem volume) had a negligible effect on damage by root rot. We detected a maximum genetic gain in resistance index of 33.7% when incorporating it into the selection index with positive gains for growth traits (6.5–11.0%). Two-stage selection with prior selection of the most resistant families was not superior to the use of selection index with only rot resistance included. Overall; rot resistance index appeared to be an effective tool in tree breeding for the selection of more resistant families, using the existing trials with natural (uncontrolled) infection

Full text

Article
Pinus sylvestris Breeding for Resistance against
Natural Infection of the Fungus
Heterobasidion annosum
Raitis Rieksts-Riekstin
,š1, Pauls Zeltin
,š1, * , Virgilijus Baliuckas 2, Lauma Br¯
una 1,
Astra Zal
,uma 1and Rolands K¯
apostin
,š1
1Latvian State Forest Research Institute Silava, 111 Rigas street, LV-2169 Salaspils, Latvia;
raitis.riekstins@silava.lv (R.R.-R.); lauma.bruna@silava.lv (L.B.); astra.zaluma@silava.lv (A.Z.);
rolands.kapostins@silava.lv (R.K.)
2Forest Institute, Lithuanian Centre for Agriculture and Forestry, Department of Forest Tree Genetics and
Breeding, Liepu St. 1, Girionys, LT-53101 Kaunas distr., Lithuania; virgilijus.baliuckas@mi.lt
*Correspondence: pauls.zeltins@silava.lv
Received: 28 November 2019; Accepted: 19 December 2019; Published: 22 December 2019


Abstract:
Increasing resistance against biotic and abiotic factors is an important goal of forest tree
breeding. The aim of the present study was to develop a root rot resistance index for Scots pine breeding
and evaluate its effectiveness. The productivity, branch diameter, branchiness, stem straightness,
spike knots, and damage from natural infection of root rot in 154 Scots pine open-pollinated families
from Latvia were evaluated through a progeny field trial at the age of 38 years. Trees with decline
symptoms were sampled for fungal isolations. Based on this information and kriging estimates of root
rot, 35 affected areas (average size: 108 m
2
; total 28% from the 1.5 ha trial) were delineated. Resistance
index of a single tree was formed based on family adjusted proportion of live to infected trees and
distance to the center of affected area. Heritability for resistance to root rot based on the value of this
index, was high (0.37) and comparable to indices of growth traits. Correlations of family breeding
estimates between resistance to root rot and the other traits were not significant, except for a weak,
yet significant, positive correlation with diameter at breast height and branch diameter. Selection
index including only growth traits (height and stem volume) had a negligible effect on damage by
root rot. We detected a maximum genetic gain in resistance index of 33.7% when incorporating it into
the selection index with positive gains for growth traits (6.5–11.0%). Two-stage selection with prior
selection of the most resistant families was not superior to the use of selection index with only rot
resistance included. Overall; rot resistance index appeared to be an effective tool in tree breeding for
the selection of more resistant families, using the existing trials with natural (uncontrolled) infection
Keywords: selection index; root rot; growth; genetic gain; two-stage selection; heritability
1. Introduction
Scots pine (Pinus sylvestris L.) is one of the most economically important tree species in the Baltic
States and Scandinavia. In this region, it is regenerated primarily by planting, and most of all the plant
material is a result of tree breeding; seeds for plant production are collected from seed orchards. Thus,
it is both relatively straightforward as well as important to improve a particular trait in significant
portion of Scots pine stands. Scots pine breeding primarily had been focused on achieving gain in traits
related to productivity (height, diameter, volume production), and frost hardiness in the northern part
of the region. In these traits, considerable gain has been achieved [
1
]. However, little has been done
in resistance breeding at practically applicable scale; the most prominent activities include selection
Forests 2020,11, 23; doi:10.3390/f11010023 www.mdpi.com/journal/forests

Forests 2020,11, 23 2 of 10
of resistance against needle cast (Lophodermium spp.) across the region and against pine blister rust
(Cronartium flaccidum and Peridermium pini) in northern Sweden [2–4].
Two species of the fungus genus Heterobasidion are found in Northern Europe: H. annosum (Fr.)
Bref. and H. parviporum Niemelä and Korhonen. Scots pine is highly susceptible to root rot caused
by H. annosum [
5
]. The biology of the pathogen has been studied intensively [
6
], nevertheless host
genotypes with total resistance to root rot have not been observed for Scots pine or Norway spruce [
7
].
Considerable effort has been made to study genetic control of root-rot resistance in Norway spruce [
8
,
9
].
However, in context of changing climate, Scots pine must also be given a high priority. Genetic control
of resistance has been previously detected in P. sylvestris [
10
], and clones with apparently varying
resistance have been detected in seed orchards [
11
] and among seedlings [
12
]. At the genetic level,
a recent study about terpenoids associated with P. sylvestris defensive mechanisms against H. annosum
has indicated higher expression of genes encoding certain terpenoid compounds, thus showing potential
to identify susceptibility of trees based on inherent genetic and chemical properties [
13
]. Moreover,
copy number variation polymorphism had been detected for a gene encoding a thaumatin-like protein,
involved in antimicrobial activity against 12 fungal species. As increased gene copy number can lead
to increased gene product amounts in cells, candidate of more resistant genotypes can be detected [
14
].
Previous studies have recommended planting the most resistant genotypes, though it remains
difficult to perform the mass testing of these genotypes for their selection and use in practical breeding.
One of the methods—genetic screening—can be biased, while controlled inoculation is expensive
at large scale. Besides reported results from inoculation experiments allow to evaluate resistance
against pathogen spread from already infected trees [
15
–
17
], but not the ability of trees to avoid
infection. It would be beneficial (saving time and resources) to use existing trials with naturally
occurring infection. Heterobasidion spp. has a mixed infection biology: (i) Primary infections via
airborne spores can originate form freshly exposed sapwood, (ii) secondary infection via vegetative
growth of mycelium to neighboring trees. Other ways—e.g., via browsing damage—are very unlike
for Scots pine [5].
Regarding substantial economic losses due to dieback caused by the pathogens, inclusion of
resistance to H. annosum in Scots pine breeding strategies should be considered. Resistance to disease
commonly exhibits strong gains and lacks undesirable correlations with other priority traits, hence
the common integration of this strategy in tree breeding programs [
18
]. Incorporation of root rot
resistance into the selection criteria—e.g., index selection [
19
]—alongside genetic gains of economically
important traits, such as tree height and stem volume, may represent a viable strategy to increase net
genetic improvement [
20
]. For Norway spruce, early selection for resistance to H. parviporum observed
not to have negative effect on growth and wood quality later [
9
]. However, an additional trait in the
selection criteria reduces the gains on already included traits [
18
]. Thus, a reasonable measurement
of resistance against root rot—one that preferably possesses substantial genetic variance and would
provide practically important genetic gains—is required. As such, the aim of the present study is to
construct a root rot resistance index for Scots pine and evaluate its effectiveness.
2. Materials and Methods
2.1. Studied Trial and Measurements
The assessed Scots pine progeny trial consisted of 154 open-pollinated families from 13 Latvian
populations. The trial site was located in eastern Latvia (56
◦
40
0
N, 25
◦
57
0
E), in flat terrain, ca. 110 m
above sea level, and was characterized by a rather continental climate [
21
]. Mean annual temperature
was +6.0
◦
C, and mean monthly temperature ranged from
−
6.4
◦
C to +17.1
◦
C in February and July,
respectively. Mean annual precipitation was ca. 700 mm [22]. The trial was established in 1979 using
one-year-old bare-rooted seedlings with an initial spacing of 2
×
1 m. The site corresponded to the
Cladinoso-callunosa forest type according to Latvian forest typology [
23
]. A randomized block design
was used with eight blocks and eight trees row-plot. In total, 4695 trees were available for analysis.

Forests 2020,11, 23 3 of 10
No thinning or other silvicultural measures had been carried out during the trial, thus the only source
of natural infection of H. annosum had been old Scots pine stumps remaining from previous stand
(harvested at mature age).
Evaluations were performed after 38 years of growth, and height (H), diameter at breast height
(DBH), and diameter of the thickest branch until the stem height of 2 m (BrD) was measured for each
tree. Stem volume (StVol) was calculated according to Liepa [
24
], and the occurrence of spike knots
(SpKn) was recorded as a binary variable (1 =present, 0 =absent). Arbitrary scores using a three-point
scale of stem straightness (StStr) and branchiness (Branch) were assessed (1 =straight stem/relatively
thin branches, 3 =crooked stem/relatively thick branches).
Regarding their potential resistance to H. annosum, trees were allocated in three categories:
(1) Asymptomatic trees; (2) symptomatic trees based on crown conditions and fruit bodies on root
collar; and (3) dead trees. To evaluate the proportion of actually infected trees, all trees from the
Category 2 and 3 (altogether 421) were cut. Wood samples (cross sections) were collected from those
stumps. Asexual sporulation (conidiophores) was used to confirm the presence of the pathogen.
Isolation and identification of Heterobasidion from all samples was attempted according to previous
work [17].
Provided data were analyzed using the procedures VARIOGRAM and KRIGE2D in SAS 9.3
software (SAS Institute, Cary, NC, USA) [
25
] to define affected areas of root rot. Kriging was used
as a geostatistical interpolation method to estimate unknown value for trees considering both the
distance and the degree of variation between known data points when estimating values in unknown
areas. A kriged estimate of root rot infection for a tree was a weighted linear combination of the
known sample values (infected trees with registered H. annosum) around the point to be estimated as
an optimal and unbiased estimate. A semivariogram of the data was constructed to weight nearby
sample points when interpolating [
26
]. Based on kriged estimates larger than zero, presumably infected
plots were delineated.
2.2. Data Analysis
Within the established affected area, root rot resistance index (RotResist) was formed:
RotResist =IndDist +IndRot +IndRotFam (1)
where IndDist is inverted distance from the tree to the plot center (average plot radius minus tree
distance to the plot center), IndRot is the ratio of number of trees in the plot to number of killed by
H. annosum trees in the plot, and IndRotFam is the ratio of number of certain families’ trees in the parcel
to number of trees in the certain parcel that are included into a delineated root-rot plot. The index
IndDist is inverted distance, because the larger value should indicate higher resistance for a living tree
closer to the plot center. Similarly, IndRot and IndRotFam are inverted proportions of killed trees by
root rot in the plot and trees per family parcel that is included in the plot, respectively, by this means
securing larger resistance index values for alive trees under higher risk conditions.
Variance components for variables were estimated using the SAS Mixed (PROC MIX) and
Generalized Linear Mixed (PROC GLIMMIX) procedures with the restricted maximum likelihood
approach [
27
,
28
]; and standard error values were calculated using Dickerson’s approximation [
29
].
Root rot variance components for heritability calculations were estimated using data only from affected
areas. The following model was used:
yijk =µ+Bi+Fj+BFij +εi jk,(2)
where y
ijk
is the observation on the kth tree from the jth family in the ith block;
µ
is the overall mean;
B
i
is the fixed effect of the ith block; F
j
and BF
ij
are the random effects of the jth family and interaction
of the ith block and the jth family, respectively, and
εijk
is the random residual effect. For estimating
the variance components of the root rot resistance index, block effect was substituted in the model

Forests 2020,11, 23 4 of 10
by plot effect since resistance index was calculated only in established affected areas that represented
more homogenous growing conditions than blocks.
Estimates of narrow-sense heritability (h2) were calculated as follows:
h2=
4׈
σ2
f
ˆ
σ2
f+ˆ
σ2
f b+ˆ
σ2
ε
, (3)
where h
2
is narrow-sense heritability and
ˆ
σ2
f
,
ˆ
σ2
f b
, and
ˆ
σ2
ε
are the estimated variance components of the
family, family ×block interaction, and the residual, respectively.
Family breeding values were estimated as double general combining ability values [
19
] using a best
linear unbiased predictors (BLUP) procedure in SAS [
25
]. For binary traits, predicted probabilities were
estimated by applying the inverse of the link function [
27
]. We applied different alternative criteria for
the selection. However, all of the alternative strategies had independent culling levels [
18
] set as a first
step. Namely, trees with breeding values for studied stem quality traits no more than 10% lower than
the trial mean were selected. In combination with the culling, further criteria were applied (Figure 1).
Forests 2019, 10, x FOR PEER REVIEW 4 of 10
ℎ=4𝜎

𝜎
+𝜎
+𝜎
, (3)
where h
2
is narrow-sense heritability and 𝜎
, 𝜎
, and 𝜎
 are the estimated variance
components of the family, family × block interaction, and the residual, respectively.
Family breeding values were estimated as double general combining ability values [19] using a
best linear unbiased predictors (BLUP) procedure in SAS [25]. For binary traits, predicted
probabilities were estimated by applying the inverse of the link function [27]. We applied different
alternative criteria for the selection. However, all of the alternative strategies had independent culling
levels [18] set as a first step. Namely, trees with breeding values for studied stem quality traits no
more than 10% lower than the trial mean were selected. In combination with the culling, further
criteria were applied (Figure 1).
Figure 1. Scheme of the different selection criteria applied. Abbreviations: H—tree height; StVol—
stem volume; RotResist—root rot resistance index.
First, standard Scots pine breeding methodology in Latvia was applied for a selection of the best
performing families and trees [30]. Tree height and volume were used (selection index = 0.5 × (height
breeding estimate) + 1.0 × (stem volume breeding estimate)). We selected the 25 best trees within the
25 top-ranked families (one from each selected family) and weighted each tree by the number
(proportion) of ramets for a seed orchard. None of the families was represented by less than 10–15
progeny nor were any of the best families’ clones allowed to comprise more than 10% of the total
number of ramets. Second, root rot resistance index was included separately to the selection index
formula by weighting it equally to the productivity traits (Selection index = 0.5 × (height breeding
estimate) + 1.0 × (stem volume breeding estimate) + 1.5 × (root rot resistance index breeding estimate)).
The third and the fourth approaches were similar to the aforementioned approaches but involved
applying two-stage selection (tandem selection in the same breeding cycle) [18] with an initial
selection of the 30% most resistant families. Thereafter, the four selection criteria were labelled as S I,
S II, S III, and S IV (Figure 1).
3. Results
For 149 trees, 28 different H. annosum genotypes were registered over a total area of 1.55 ha, and
35 affected areas of root rot obtained. The mean affected area was 108 m
2
. Distance between areas
ranged from 2 to 74 m, and in total they covered 27.5% of the trial area.
Narrow-sense heritability for RotResist was high (0.37, cf. Falconer and Mackay [19]) and
comparable to the heritability of productivity traits, which were moderate to high (0.25–0.45). Stem
Figure 1.
Scheme of the different selection criteria applied. Abbreviations: H—tree height; StVol—stem
volume; RotResist—root rot resistance index.
First, standard Scots pine breeding methodology in Latvia was applied for a selection of the best
performing families and trees [
30
]. Tree height and volume were used (selection index =0.5
×
(height
breeding estimate) +1.0
×
(stem volume breeding estimate)). We selected the 25 best trees within the 25
top-ranked families (one from each selected family) and weighted each tree by the number (proportion)
of ramets for a seed orchard. None of the families was represented by less than 10–15 progeny nor
were any of the best families’ clones allowed to comprise more than 10% of the total number of ramets.
Second, root rot resistance index was included separately to the selection index formula by weighting
it equally to the productivity traits (Selection index =0.5
×
(height breeding estimate) +1.0
×
(stem
volume breeding estimate) +1.5
×
(root rot resistance index breeding estimate)). The third and the
fourth approaches were similar to the aforementioned approaches but involved applying two-stage
selection (tandem selection in the same breeding cycle) [
18
] with an initial selection of the 30% most
resistant families. Thereafter, the four selection criteria were labelled as S I, S II, S III, and S IV (Figure 1).
3. Results
For 149 trees, 28 different H. annosum genotypes were registered over a total area of 1.55 ha, and
35 affected areas of root rot obtained. The mean affected area was 108 m
2
. Distance between areas
ranged from 2 to 74 m, and in total they covered 27.5% of the trial area.

Forests 2020,11, 23 5 of 10
Narrow-sense heritability for RotResist was high (0.37, cf. Falconer and Mackay [
19
]) and
comparable to the heritability of productivity traits, which were moderate to high (0.25–0.45). Stem
quality traits, such as stem straightness, branchiness, branch diameter, and the occurrence of spike
knots, had low (0.02–0.13) heritability estimates (Table 1). Correlations of family breeding values of
RotResist and the remaining analyzed variables were not significant, with the exception of a weak
(cf. Evans [
31
]), positive correlation with DBH (r=0.16, p=0.05), StVol (r=0.19, p=0.02), and BrD
(r=0.28, p<0.01). Productivity traits (H, DBH and StVol) showed very high correlation (r
≥
0.88,
p<0.01) among themselves and moderate to high (0.51 <r<0.65, p<0.01) correlation with BrD
(Table 2).
Table 1.
Mean, minimum (Min), and maximum (Max) values and narrow-sense heritability indices of
the studied traits.
Trait Mean ±Standard
Deviation Min Max Narrow-Sense Heritability
h2±standard Error
Height (m) 9.7 ±2.42 1.9 16.1 0.45 ±0.094
Diameter at breast height (cm)
10.3 ±4.06 1.0 23.0 0.25 ±0.069
Stem volume (dm3)55.6 ±43.39 0.3 274.9 0.28 ±0.072
Branch diameter (cm) 1.5 ±0.43 0.3 4.5 0.13 ±0.053
Branchiness (score) 1.4 1.0 3.0 0.10 ±0.048
Spike knots (% of trees) 21.8 - - 0.03 ±0.077
Stem straightness (score) 1.2 1.0 3.0 0.02 ±0.042
Root rot resistance index 0.78 ±1.34 −1.9 8.6 0.37 ±0.220
Table 2.
Family mean breeding value correlations (significant correlations with p
≤
0.05 in bold) in
the upper diagonal part and their p-values in the lower diagonal part. Abbreviations: H—height;
DBH—diameter at breast height; StVol—stem volume; BrD—branch diameter; Branch—branchiness;
StStr—stem straightness; SpKn—spike knots; RotResist—root rot resistance index.
H DBH StVol BrD Branch StStr SpKn RotResist
H 1 0.89 0.88 0.51 −0.10 0.39 −0.27 0.15
DBH <0.01 1 0.98 0.65 −0.34 0.31 −0.25 0.16
Vol <0.01 <0.01 1 0.62 −0.34 0.26 −0.25 0.19
BrD <0.01 <0.01 <0.01 1 −0.58 0.04 −0.13 0.28
Branch 0.19 <0.01 <0.01 <0.01 1 0.15 0.02 −0.15
StStr <0.01 <0.01 <0.01 0.59 0.06 1 −0.24 −0.03
SpKn <0.01 <0.01 <0.01 0.11 0.76 <0.01 1 0.07
RotResist 0.06 0.05 0.02 <0.01 0.06 0.71 0.38 1
The applied different selection criteria highlighted a varying response to selection (Table 3).
Selection index considering only H and StVol (S I) would not notably alter damage by root rot via
selection of the 25 best individuals from each family (genetic gain 2.3%) or from the 25 top families
(genetic gain 4.9%). If the tree root rot resistance index was included to the selection index formula
with equal weighting-to-productivity traits (S II), then the observed increase in root rot resistance
comprised 30.0% of cases where individuals were selected and 33.7% when top families were selected.
Genetic gain estimates for H, DBH and StVol were reduced considerably, but the other traits were quite
robust with the inclusion of RotResist to the selection index. The two-stage (tandem) selection process
allowed increased genetic gain for root resistance index when applying both S III and S IV. Initial
selection of the top 30% of resistant families resulted in an improved resistance index among 15.4%
(the best individuals) and 27.0% (the best families) when applying S III. The inclusion of resistance
in the selection index in S IV increased the genetic gain for rot resistance to 27.7 and 35.2% for the
individuals and the families, respectively. Generally, when selecting the best 25 families instead of
the best 25 individuals from each family, the estimated genetic gains for H, DBH, and StVol were less
sensitive to the inclusion of resistance in the selection index (Table 3).

Forests 2020,11, 23 6 of 10
Table 3.
Estimated genetic gain for the studied variables using different selection criteria: S I—selection index with only height (H) and stem volume (StVol) included;
S II—selection index with root rot resistance index (RotResist) included; S III—two-stage selection with the selection of 30% of individuals from families with the
greatest resistance prior to the application of conventional selection index; S IV—two-stage selection with the selection of 30% of families with the greatest resistance
prior to the application of selection index with RotResist included. For all criteria, the selection of the 25 best individuals within the best families or selection of the
25 best families regarding the selection index are applied.
Selection Criterion Genetic Gain (%)
Height Diameter at
Breast Height
Branch
Diameter Branchiness Stem
Straightness
Spike
Knots
Stem
Volume
Root Rot
Resistance Index
S I (0.5 ×H+StVol) 25 best individuals 16.5 13.2 3.0 −0.6 0.1 −1.1 44.3 2.3
25 best families 14.5 17.8 6.3 −2.4 0.2 −15.4 50.6 4.9
S II (0.5 ×H+StVol +1.5*RotResist)25 best individuals 6.5 4.7 1.6 0.3 0.1 −1.6 10.9 30.0
25 best families 10.4 13.5 6.0 −2.4 0.0 −0.5 38.1 33.7
S III (Two-stage selection/0.5 ×H+StVol) 25 best individuals 16.4 12.5 2.8 −0.2 0.0 −1.0 41.0 15.4
25 best families 11.5 14.3 4.8 −2.4 0.1 −2.3 39.7 27.0
S IV (Two-stage selection/0.5
×
H+StVol +
1.5 ×RotResist)
25 best individuals 6.5 4.6 1.7 0.2 0.1 −1.4 10.7 27.7
25 best families 9.5 12.8 5.5 −2.3 0.0 0.6 35.7 35.2

Forests 2020,11, 23 7 of 10
4. Discussion
4.1. Genetic Parameters
Although there are no conifer species with genetically determined total resistance against the
pathogen Heterobasidion spp. [
6
,
7
], estimated high heritability of resistance index (Table 1) suggests
existing variation in the susceptibility of Scots pine to this fungus being strongly genetically controlled.
Genetic control has been commonly reported in the resistance of Norway spruce P. abies to Heterobasidion
spp. [
8
,
15
,
32
–
37
], moreover the genetic component in the resistance of Scots pine has also been
previously detected [
10
]. The narrow-sense heritability (h
2
=0.37) for RotResist was similar or higher
to the indices estimated for H. parviporum lesion length in the phloem (0.17 <h
2
<0.33) and fungal
growth in sapwood (h
2
=0.42) in Norway spruce open-pollinated progenies [
9
,
37
]. Also, for Norway
spruce clones, moderate broad-sense heritability (H
2
=0.21) has been estimated for lesion length [
15
].
Swedjemark and Karlsson [
33
] reported moderate broad-sense heritability (H
2
=0.28) for the frequency
of stems infected with H. annosum in a progeny trial of 44-year-old Norway spruce full-sib families.
For Scots pine, the estimated h
2
in open-pollinated families for RotResist was similar to the observed
resistance in full-sib families to Dothistroma needle blight (h2=0.38) [38].
Notably, we detected only very weak undesirable relationships between root rot resistance index
and other valuable traits (Table 2), implying potential simultaneous improvement. Besides, the
highest estimated correlation (0.28) with BrD might be somewhat exaggerated because of the RotReist
construction characteristics. Namely, trees with higher RotResist in the root-rot affected areas tended
to face higher mortality around them, thus promoting development of thicker branches because of
improved light conditions.
Overall, fungal growth and lesion length in the phloem were not found to correlate with growth
and wood quality traits in a study evaluating the incorporation of resistance to H. parviporum in the
Norway Spruce Breeding Program [
9
]. The rather weak, yet significant, positive correlations between
family breeding values of RotResist and StVol (Table 2) might be explained by reduced growth among
more infected families. In Southern Sweden, assessing relative small sample size, significantly reduced
the growth nine years after the first thinning had been detected for trees with even a small proportion
of roots being infected; general decline in volume increment was detected as early as three years after
the first thinning [39].
4.2. Selection Methods
Root rot resistance can be considered a high priority trait alongside H and StVol for incorporation
into the selection index since it shows notable genetic variance, has high economic impact on timber
production value, and does not show an undesirable relationship with growth [
18
], as also observed
for Norway spruce [
37
]. Unsurprisingly, the reduction of gains in H and StVol were observed (Table 3)
with the inclusion of RotResist as an additional trait into the selection index [
18
]. Gains in productivity
remained positive (6.5–10.4%), while genetic gains for RotResist increased from ca. 2%–5% to 30%–33.7%
(Table 3). Considerable heritability for RotResist (Table 1) might have determined high genetic gain in
our case, when assigning equal weight to productivity traits.
Still, higher gains for RotResist and smaller reductions in gains in growth could be achieved by
selecting the best (based on selection index) families (backward selection) instead of the best individuals
within the best families (Table 3), indicating a higher effectiveness in family selection likely due to
sufficiently large environmental deviations affecting the phenotypic variance of the traits [19].
The two-stage selection (S III and S IV) appears to be as effective as single-stage selection (S I
and S II), having a small impact on growth and quality traits (Table 3). Under two-stage selection,
the aforementioned effectiveness of family selection seems to be retained, providing higher gains
for growth traits. However, regarding similar improvement in RotResist in the index and two-stage
selection processes, S II should be preferred because little yet improved gains in growth traits were
observed for S II compared to S IV (Table 3).

Forests 2020,11, 23 8 of 10
Overall, the monetary value of forest products must reflect genetic gains for resistance index, and
evaluation is necessary regarding whether the added value of RotResist outweighs selection for pure
productivity [
37
]. In our study, we applied equal weights for growth traits and RotResist. Therefore,
potential further research should address the optimal economic weight for each trait in the selection
index to achieve the greatest economic gain; however, uncertain future economic conditions makes
this a difficult task [
18
]. The loss of merchantable timber because of pathogens is hardly predictable
because of numerous site factors affecting the spread of infections [
6
,
40
] and reduced resistance to
wind damage [6].
Although inclusion of RotResist into the selection criteria indicates substantial improvements in
genetic gain, some practical issues may arise regarding the application of the index. In the present
study, RotResist was calculated for trees in the sample plots derived from the kriging estimates, which
were based on detected rot in cross-sectional wood samples. Such a destructive method involves
laborious extra work to include in breeding activities, thereby increasing breeding costs. However,
H. annosum may infect Scots pine root systems without any obvious crown symptoms; thus, visual
assessment alone may result in the underestimation of disease distribution [
7
,
39
,
40
]. As only Category
2 and 3 trees have been sampled, calculations are based on the minimum size of the pathogen genets.
To estimate the exact borders of Heterobasidion genets more accurate sampling would be required
(excavated root systems). Still, the observed effectiveness of RotResist may be successfully applied in
combination with other, less resource-consuming root rot detection methods at an earlier age, such as
non-destructive resistography [41] or assessment during inoculation experiments [15].
5. Conclusions
Root rot resistance index appeared to be an effective measure for inclusion in the selection criteria
because of its notable genetic variance and lack of undesirable relationship to other growth and
quality traits. We suggest the incorporation of the root rot resistance into the conventional selection
index of growth traits for family selection, which would considerably improve resistance to root
rot causing dieback while maintaining reasonable gains in growth traits. Notwithstanding, further
analysis is necessary to adjust economic weights for the traits in the selection index to obtain optimal
index performance.
Author Contributions:
Conceptualization, R.R.-R. and R.K.; methodology, L.B. and R.R.-R.; software, P.Z. and
V.B.; formal analysis, P.Z.; investigation, R.R.-R., R.K., and A.Z.; data curation, V.B.; writing—original draft
preparation, R.R.-R. and P.Z.; writing—review and editing, A.Z. and R.K. All authors have read and agreed to the
published version of the manuscript.
Funding: This research received no external funding.
Acknowledgments:
Study was carried out in Forest competence center (ERDF) project “Technologies for efficient
transfer of genetic gain in plant production and forestry” and Latvian state forests (LVM) projects “Tree breeding
for selection of superior forest reproductive material”, “Investigation of the factors limiting the spread of root rot”.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
Jansson, G.; Hansen, J.K.; Haapanen, M.; Kvaalen, H.; Steffenrem, A. The genetic and economic gains from
forest tree breeding programmes in Scandinavia and Finland. Scand. J. For. Res.
2017
,32, 273–286. [CrossRef]
2.
Persson, T. Genotype by environment interaction for survival, growth and Cronartium resistance in northern
Scots pine. In Genetic Aspects of Adaptation and Mitigation: Forest Health, Wood Quality and Biomass Production,
Proceedings of the Book of Abstracts of International Scientific Conference, Riga, Latvia, 3–5 October 2012; Jansons,
¯
A.,
Konstantinova, I., Eds.; LSFRI Silava: Salaspils, Latvia, 2012; p. 150.
3.
Krakau, U.-K.; Liesebach, M.; Aronen, T.; Lelu-Walter, M.-A.; Schneck, V. Scots Pine (Pinus sylvestris L.);
Springer: Dordrecht, Germany, 2013; pp. 267–323.

Forests 2020,11, 23 9 of 10
4.
Neimane, U.; Polmanis, K.; Zaluma, A.; Klavina, D.; Gaitnieks, T.; Jansons, A. Damage caused by Lophodermium
needle cast in open-pollinated and control-crossed progeny trials of Scots pine (Pinus sylvestris L.). For. Chron.
2018,94, 155–161. [CrossRef]
5.
Gonthier, P.; Thor, M. Annosus root and butt rots. In Infectious Forest Diseases; Gonthier, P., Nicolotti, G., Eds.;
CAB International: Wallingford, UK, 2013; pp. 128–158.
6.
Garbelotto, M.; Gonthier, P. Biology, Epidemiology, and Control of Heterobasidion Species Worldwide.
Annu. Rev. Phytopathol. 2013,51, 39–59. [CrossRef] [PubMed]
7.
Asiegbu, F.O.; Adomas, A.; Stenlid, J. Conifer root and butt rot caused by Heterobasidion annosum (Fr.) Bref.
s.l. Mol. Plant Pathol. 2005,6, 395–409. [CrossRef] [PubMed]
8.
Swedjemark, G.; Stenlid, J.; Karlsson, B. Genetic variation among clones of Picea abies in resistance to growth
of Heterobasidion annosum.Silvae Genet. 1998,46, 369–374.
9.
Chen, Z.-Q.; Lund
é
n, K.; Karlsson, B.; Vos, I.; Olson, Å.; Lundqvist, S.-O.; Stenlid, J.; Wu, H.X.; Garc
í
a
Gil, M.R.; Elfstrand, M. Early selection for resistance to Heterobasidion parviporum in Norway spruce is not
likely to adversely affect growth and wood quality traits in late-age performance. Eur. J. For. Res.
2018
,137,
517–525. [CrossRef]
10.
Korshikov, I.I.; Demkovich, A.E. Genetic features of root fungus-resistant scotch pine trees in artificial stands
in the steppe zone of Ukraine. Cytol. Genet. 2008,42, 323–328. [CrossRef]
11.
Vasiliauskas, A. Šaknin
˙
es pinties (Heterobasidion annosum (Fr.) Bref.) paplitimas pušynuose,
ı˛
veistuose žem
˙
es
¯
ukio naudmenuose, kovos su ja priemon
˙
es ir kovos rezultatai [Root rot caused by Heterobasidion annosum in
Pinus sylvestris plantations on former agricultural land. Miskininkyste 2001,2, 42–59.
12.
Zaluma, A.; Gailis, A.; Bur
n
,
eviˇca, N.; Korhonen, K.; Gaitnieks, T. Susceptibility of Picea abies and Pinus
sylvestris seedlings of various origins to Heterobasidion annosum and H. parviporum.Proc. Latv. Acad. Sci.
2016,70, 29–33.
13.
Mukrimin, M.; Kovalchuk, A.; Ghimire, R.P.; Kivimäenpää, M.; Sun, H.; Holopainen, J.K.; Asiegbu, F.O.
Evaluation of potential genetic and chemical markers for Scots pine tolerance against Heterobasidion annosum
infection. Planta 2019,250, 1881–1895. [CrossRef]
14.
Š
n
,
epste, I.; Š
k
,
ipars, V.; Krivmane, B.; Br
¯
una, L.; Ru
n
,‘
g
is, D. Characterization of a Pinus sylvestris thaumatin-like
protein gene and determination of antimicrobial activity of the
in vitro
expressed protein. Tree Genet. Genomes
2018,14, 58. [CrossRef]
15.
Skrøppa, T.; Solheim, H.; Steffenrem, A. Genetic variation, inheritance patterns and parent–offspring
relationships after artificial inoculations with Heterobasidion parviporum and Ceratosystis polonica in Norway
spruce seed orchards and progeny tests. Silva Fenn. 2015,49, 1191. [CrossRef]
16.
Marˇciulynas, A.; Sirgedait
˙
e-Š
˙
ežien
˙
e, V.; Žemaitis, P.; Baliuckas, V. The Resistance of Scots Pine (Pinus sylvestris
L.) Half-sib Families to Heterobasidion annosum.Forests 2019,10, 287. [CrossRef]
17.
Za
l
,
uma, A.; Muižnieks, I.; Gaitnieks, T.; Bur
n
,
eviˇca, N.; Jansons,
¯
A.; Jansons, J.; Stenlid, J.; Vasaitis, R. Infection
and spread of root rot caused by Heterobasidion spp. in Pinus contorta plantations in Northern Europe: Three
case studies. Can. J. For. Res. 2019,49, 969–977. [CrossRef]
18.
White, T.L.; Adams, W.T.; Neale, D.B. Forest Genetics; CABI Publishing: Wallingford, UK, 2007; ISBN
9780851990835.
19.
Falconer, D.S.; Mackay, T.F. Introduction to Quantitative Genetics, 4th ed.; Longman Group Ltd.: London, UK,
1996.
20. Hazel, L.N. The Genetic Basis for Constructing Selection Indexes. Genetics 1943,28, 476–490.
21.
Laivi
n
,
š, M.; Melecis, V. Bio-geographical interpretation of climate data in Latvia: Multidimensional analysis.
Acta Univ. Latv. 2003,654, 7–22.
22.
Harris, I.; Jones, P.D.; Osborn, T.J.; Lister, D.H. Updated high-resolution grids of monthly climatic
observations—The CRU TS3.10 Dataset. Int. J. Climatol. 2014,34, 623–642. [CrossRef]
23. Bušs, K. Meža Ekolo ‘
gija un Tipolo‘
gija [Forest ecology and typology]; Zin¯
atne: R¯
ıga, Latvia, 1981.
24. Liepa, I. Pieauguma M¯
ac¯
ıba [Increment Science]; LLU: Jelgava, Latvia, 1996.
25. SAS. SAS/IML 9.3 User’s Guide; Sas Institute: Cary, NC, USA, 2011; ISBN 1607649136.
26.
Oliver, M.A.; Webster, R. Kriging: A method of interpolation for geographical information systems. Int. J.
Geogr. Inf. Syst. 1990,4, 313–332. [CrossRef]
27.
Littell, R.C.; Milliken, G.A.; Stroup, W.W.; Wolfinger, R.D. SAS System for Mixed Models; Sas Institute: Cary,
NC, USA, 1996.

Forests 2020,11, 23 10 of 10
28.
Littel, R.C.; Milliken, G.A.; Stroup, W.W.; Wolfinger, R.D.; Schabenberger, O. SAS for Mixed Models, 2nd ed.;
Sas Institute: Cary, NC, USA, 2006; Volume 816.
29.
Dickerson, G.E. Techniques for research in quantitative animal genetics. In Techniques and Procedures in
Animal Science Research; American Society of Animal Production: Albany, NY, USA, 1969; pp. 36–79.
30.
Baumanis, I.; Jansons, A.; Neimane, U. Priede: Selekcija,
G
,
en
¯
etika un S
¯
eklkop
¯
ıba Latvij
¯
a [Scots Pine: Breeding,
Genetics and Seed Orchard Management]; Daugavpils Universit
¯
ates Akad
¯
emiskais apg
¯
ads “Saule”: Daugavpils,
Latvia, 2014.
31.
Evans, J.D. Straightforward Statistics for the Behavioral Sciences; Thomson Brooks/Cole Publishing Co.: Belmont,
CA, USA, 1996.
32.
Swedjemark, G.; Karlsson, B. Genotypic variation in susceptibility following artificial Heterobasidion annosum
inoculation of Picea abies clones in a 17-year-old field test. Scand. J. For. Res. 2004,19, 103–111. [CrossRef]
33.
Swedjemark, G.; Karlsson, B. Variation in incidence and genetic impact on natural infection of Heterobasidion
annosum in Picea abies (L.) Karst. in genetic trials in south Sweden. For. Ecol. Manag.
2004
,203, 135–145.
[CrossRef]
34.
Wellendorf, H.; Thomsen, I.M. Genetic variation in resistance against Heterobasidion annosum (Fr.) bref. in
Picea abies (L.) karst. expressed after inoculation of neighboring stumps. Silvae Genet.
2008
,57, 312–324.
[CrossRef]
35.
Arnerup, J.; Swedjemark, G.; Elfstrand, M.; Karlsson, B.; Stenlid, J. Variation in growth of Heterobasidion
parviporum in a full-sib family of Picea abies.Scand. J. For. Res. 2010,25, 106–110. [CrossRef]
36.
Skrøppa, T.; Solheim, H.; Hietala, A. Variation in phloem resistance of Norway spruce clones and families
to Heterobasidion parviporum and Ceratocystis polonica and its relationship to phenology and growth traits.
Scand. J. For. Res. 2015,30, 103–111. [CrossRef]
37.
Steffenrem, A.; Solheim, H.; Skrøppa, T. Genetic parameters for wood quality traits and resistance to the
pathogens Heterobasidion parviporum and Endoconidiophora polonica in a Norway spruce breeding population.
Eur. J. For. Res. 2016,135, 815–825. [CrossRef]
38.
Perry, A.; Wachowiak, W.; Brown, A.V.; Ennos, R.A.; Cottrell, J.E.; Cavers, S. Substantial heritable variation
for susceptibility to Dothistroma septosporum within populations of native British Scots pine (Pinus sylvestris).
Plant Pathol. 2016,65, 987–996. [CrossRef]
39.
Wang, L.; Zhang, J.; Drobyshev, I.; Cleary, M.; Rönnberg, J. Incidence and impact of root infection by
Heterobasidion spp., and the justification for preventative silvicultural measures on Scots pine trees: A case
study in southern Sweden. For. Ecol. Manag. 2014,315, 153–159. [CrossRef]
40.
Rönnberg, J.; Petrylaite, E.; Nilsson, G.; Pratt, J. Two studies to assess the risk to Pinus sylvestris from
Heterobasidion spp. in southern Sweden. Scand. J. For. Res. 2006,21, 405–413. [CrossRef]
41.
Allikmäe, E.; Laarmann, D.; Korjus, H. Vitality assessment of visually healthy trees in Estonia. Forests
2017
,
8, 223. [CrossRef]
©
2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).