Table 5.

Variables that defined each cluster (p < 0.05) using a principal component analysis (PCA) and hierarchical clustering on principal components (HCPC).

Cluster	Treatment		Variables
1	Control	+	Fe_FT, Cr_FT, Co_FT
		−	P_uptR, Mg_R, Ca_uptR, K_uptT, P_R, Ca_uptT, BA, Mg_uptR, P_uptT, Mg_uptT
2	Paraburkholderia ultramafica STM10279T	+	Mg_R, Na_R, Na_upT, Na_T
		−
2	EPS of P. ultramafica STM10279T	+	Ca_uptT, K_uptT, Ca_uptR, Mg_uptT, Mg_uptT, P_uptR, K_uptR, K_R,P_uptT
		−

PCA and HCPC were performed with 29 variables: biomass [dry weight tissue of shoots (BA) and roots (BR)], mineral and metal content in shoots (i.e., Ca_T) and roots (i.e., Ca._R) (Ca, Mg, K, Na, and P), metal translocation factor (i.e., Co_FT) (Co, Cr, Fe, Mn, and Ni), mineral element uptake in shoots (i.e., Ca_uptT) and roots (i.e., Ca_uptR) (Ca, Mg, Na, K, and P), and Ca/Mg ratio in shoots (CaMg_T) and roots (CaMg_R). Analyses were carried out using R software with the “FactoMineR” package. HCPC used Euclidean distances for calculating dissimilarities between observations and average method to define clusters.

EPS, exopolysaccharide.