There is accumulating evidence that root system collapse is a primary symptom associated with Huanglongbing (HLB)-induced tree decline, especially for commercial sweet orange and grapefruit trees on Swingle and Carrizo rootstocks. Maintaining root health is imperative to keep trees productive in an HLB-endemic environment. Preliminary greenhouse and field studies have shown that HLB-impacted trees had secondary and micronutrient deficiencies that were much greater in the roots than in the leaves, and that treatments containing three-times the recommended dose of manganese (Mn) improved tree health and growth and increased feeder root density in greenhouse trees. These results suggested that trees in an HLB-endemic environment have higher specific micronutrient requirements than those currently recommended. To test this hypothesis, established Vernia sweet orange grafted onto rough lemon rootstock trees were divided into eight supplemental CRF nutrition treatments (including two-times and four-times the recommended doses of Mn and boron) using a randomized complete block design in a commercial grove in St. Cloud, FL. The following supplemental nutrition treatments were used: no extra nutrition (control); Harrell’s–St. Helena mix 0.9 kg per tree; Harrell’s with 32 g of Florikan polycoated sodium borate (PSB) per tree; Harrell’s with 90 g of TigerSul® Mn sulfate (MS) per tree; Harrell’s with 32 g of PSB and 90 g of MS per tree; 180 g of MS per tree; 64 g of PSB per tree; and 180 g of MS plus 64 g of PSB per tree applied every 6 months since Fall 2015. Leaf and soil nutritional analyses were performed in Mar. 2017, Sept. 2017, and May 2018; a quantitative polymerase chain reaction was performed for Candidatus Liberibacter asiaticus (CLas) titer estimation in Nov. 2017. Significantly higher cycle threshold (Ct) values indicating reduced CLas bacterial populations were observed in trees that received the higher doses of Mn, especially those receiving four-times the recommended dosage of Mn (180 g Mn). Many trees exhibited Ct values of 32 or more, indicating a nonactive infection. Fruit yields of these trees were also increased. No significant differences in juice characteristics, canopy volume, and trunk section area were found between control plants and plants treated with 180 g Mn. Soil and leaf nutrients B, K, Mn, and Zn were significantly different among treatments at various times during the study. Our results strongly suggest that overdoses of Mn can suppress CLas bacterial titers in sweet orange trees on rough lemon rootstock, thus providing a therapeutic effect that can help restore tree health and fruit yields. This response was not observed when Mn and B were combined in the overdose, suggesting an antagonistic effect from B on Mn metabolism. When an overdose of Mn is used, biological functions and tree tolerance lost due to nutritional imbalances caused by HLB might be restored. Further studies are needed to elucidate which metabolic pathways are altered by comparing overdosed and conventionally fertilized HLB-impacted trees and to determine if the observed therapeutic effects can be achieved in trees grafted to other important commercial rootstocks.