Arm Control and its Recovery after Selective Lesions of Sensorimotor Cortex and the Red Nucleus: A Kinematic Study in Non-Human Primates

Reviewer #1 (Public review):

Summary:

This is a very interesting and well-done study of the effects of selective lesions to the sensorimotor cortex and the red nucleus on control of upper limb movements. The findings that the red nucleus may subserve recovery of upper limb motor function after cortical lesions in macaques and the different motor functions of different cortical sensorimotor areas are significant findings of considerable interest to sensorimotor neuroscientists, neurologists and neurosurgeons. The methods are mostly excellent, but there are some questions about the use of endothelin lesions in cortical areas and the use of trajectory variability as a marker of movement quality and fine motor control. Furthermore, it is questionable that increased trajectory variability in reaching a target reflects reduced movement quality, reduced ability to independently control muscles, and is a proximal analog of reduced dexterity.

Strengths:

The rationale that rubrospinal projections onto spinal neurons may subserve the good recovery of upper limb movements observed after lesions of sensorimotor cortex is compelling. The methods involving complete lesions of the red nucleus followed by recovery prior to lesions affecting various sensorimotor cortical areas are a strength. The excellent interpretations offered in the Discussion section are also a strength.

Weaknesses:

There are weaknesses in the Methods, including:

(1) no information on dimensions of the cup containing the food reward or types of food rewards,

(2) recording 3D hand movements with a single camera,

(3) cortical endothelin lesions were not very precise,

(4) the use of trajectory variability as a measure of movement quality and reduced ability to independently control muscles.

Some interpretations presented in the Discussion are not well supported. The discussion related to movement quality should be modified to focus on trajectory variability. The suggestion that rubrospinal projections onto motor neurons are apparently irreplaceable is not well justified because one monkey receiving a complete red nucleus lesion showed nearly full recovery of maximum movement speed, while the other monkey did not. The nearly full recovery of one monkey was probably due to new corticospinal connections onto motor neurons, whereas it is possible that the other monkey would have recovered better given more time before the 2nd lesion to cortical areas.

Reviewer #2 (Public review):

Summary:

This study made selective lesions in motor cortical subregions and the magnocellular red nucleus in nine macaque monkeys, and evaluated reaching and grasping movements using maximum speed and trajectory variability. The results suggest that damage to the posterior old primary motor cortex (M1) was mainly associated with reduced maximum speed, whereas damage to the new M1 was mainly associated with increased trajectory variability. Damage to the anterior old M1 did not clearly add further impairment. Lesions of the magnocellular red nucleus (RNm) alone mainly reduced reaching speed, but recovery after subsequent cortical lesions was worse than after cortical lesions alone, suggesting that the rubrospinal pathway may be important for compensation after cortical damage in monkeys. Overall, this is a valuable study that examines differences among M1 subregions using selective lesions in macaques.

Strengths:

(1) This study tackles an important question. It attempts to decompose the diverse upper-limb impairments after stroke into the effects of different primary motor cortex subregions.

(2) Another strength is that the lesions and behavioral impairments were evaluated quantitatively. The use of nine macaque monkeys with different lesion patterns, together with quantitative behavioral evaluation, provides a rare and valuable dataset. The authors also followed recovery using quantitative behavioral measures such as maximum speed and trajectory variability.

(3) The inclusion of RNm lesions is also valuable, as it revisits the classic question raised by Lawrence and Kuypers (1968) of how brainstem descending pathways contribute to recovery after cortical motor damage.

Weaknesses:

(1) The main limitation is that the contribution of each cortical lesion is sometimes interpreted from largely qualitative comparisons. Because the lesion extent was not always limited to the intended region, it is difficult to fully separate the independent contribution of each subregion. Some conclusions are also based on comparisons between a small number of animals. The dataset itself is valuable, and the manuscript would be strengthened by presenting these conclusions more cautiously and explicitly acknowledging this limitation.

(2) Because the behavioral evaluation is quantitative, it would be helpful to show the relationship between lesion size and behavioral impairment more quantitatively. For example, rank correlations between the lesion size of each cortical region and behavioral measures could help readers evaluate whether the type and size of lesion are related to behavioral impairment.

(3) The discussion of area 4s could be further developed. The authors suggest that this region may have a different role, but the specific hypothesis is not fully clear. There has also been skepticism in the previous literature about area 4s, for example, Meyers et al. (1954), and this broader background could be discussed. (Meyers R, Knott JR, Skultety FM, Imler R (1954) On the Question as to the Existence of a "4s" Suppressor Mechanism. Journal of Neurosurgery 11:7-23.)

Reviewer #3 (Public review):

Summary:

In this article, the authors performed targeted lesions in cortical areas involved in forelimb control of rhesus macaques. Using a reaching task with kinematic tracking, they compared kinematic variability (as a proxy for dexterity) and reaching speed (as a proxy for strength) before and after cortical (n=7) and magnocellular red nucleus (RNm) (n=2) lesion. Changes in these movement metrics were related to the location and extent of the lesions, reconstructed from histology. The authors report that lesions with a large component in New M1 had a pronounced effect on kinematic variability, whereas lesions with a large component in posterior Old M1 primarily affected reaching speed. Lesions of the RNm were performed in two animals approximately seven to eight weeks before the cortical lesion. By themselves, RNm lesions produced a significant but small reduction in reach speed. They also magnified the effect of the subsequent cortical lesions.

Strengths:

(1) For non-human primate (NHP) research, this is a large cohort.

(2) The behavioural analyses are clear and precise.

(3) The additional red nucleus lesions in two monkeys provide unique complementary information.

Weaknesses:

(1) Description of injuries. As described and reported in the current result section and figures, readers do not have a clear understanding of the lesion extent and location.

(2) Lack of formal correlative analyses between lesion characteristics and behaviour. Currently, it seems that the conclusions are based on impressions between some aspects of the lesions and precise kinematic measures.

(3) The data will be of interest to a large community of researchers working on brain injury and stroke recovery, as well as cortical motor control. There are, however, some major methodological issues in the current version of the manuscript that prevent a clear evaluation of the findings and potential contribution of the work in the field. These issues need to be addressed to support the conclusions.