Sex Matters: Robust Sex Differences in Signal Detection in the HIV-1 Transgenic Rat

Abstract

Sex differences in human immunodeficiency virus type-1 (HIV-1) have been repeatedly suggested. Females, who account for 51% of HIV-1 seropositive individuals, are inadequately represented in clinical and preclinical studies, as well as in the description of HIV-1 associated neurocognitive disorders (HAND). Direct comparisons of neurocognitive decline in women and men must be made to address this underrepresentation. The effect of biological sex (i.e., the biological factors, including chromosomes and hormones, determining male or female characteristics; WHO, 2017) on sustained attention, which is commonly impaired in HIV-1 seropositive individuals, was investigated in intact HIV-1 transgenic (Tg) and control animals using a signal detection operant task. Analyses revealed a robust sex difference in the rate of task acquisition, collapsed across genotype, with female animals meeting criteria in shaping (at least 60 reinforcers for three consecutive or five non-consecutive sessions) and signal detection (70% accuracy for five consecutive or seven non-consecutive sessions) significantly more slowly than male animals. Presence of the HIV-1 transgene also had a significant effect on shaping and signal detection acquisition, with HIV-1 Tg animals displaying significant deficits in the rate of acquisition relative to control animals-deficits that were more prominent in female HIV-1 Tg animals. Once the animals' reached asymptotic performance in the signal detection task, female animals achieved a lower percent accuracy across test sessions and exhibited a decreased response rate relative to male animals, although there was no compelling evidence for any effect of transgene. Results indicate that the factor of biological sex may be a moderator of the influence of the HIV-1 transgene on signal detection. Understanding the impact of biological sex on neurocognitive deficits in HIV-1 is crucial for the development of sex-based therapeutics and cure strategies.

Keywords: HIV-1 transgenic rat; biological sex; neuroinflammation; sustained attention.

Figure 1

(A) The number of sessions required to meet criterion (at least 60 reinforcers for three consecutive or five non-consecutive days) in the shaping task is presented as a function of biological sex, representing both HIV-1 Tg and control animals by collapsing across genotype (±95% CI). A first-order polynomial provided a well-described fit for male animals (R²: 0.95), with all animals meeting criterion within 8 days. In sharp contrast, a one-phase association was the best fit for female animals (R²: 0.94). All female animals acquired the task within 43 days. (B) A global one-phase association was the best fit for male HIV-1 Tg and male control animals (R²: 0.98), suggesting no difference in the number of days to meet criterion between genotypes. (C) In female control animals, a one-phase association provided a well-described fit (R²: 0.96). However, in female HIV-1 Tg animals, shaping acquisition was best fit using a first-order polynomial (R²: 0.91). Presence of the HIV-1 transgene had a significant effect on the temporal acquisition of shaping in female, but not male, animals.

Figure 2

(A) The number of sessions required to meet criterion (70% accuracy for five consecutive or seven non-consecutive days) in the signal detection task is presented as a function of biological sex, representing both HIV-1 Tg and control animals by collapsing across genotype (±95% CI). A sigmoidal dose response was the best fit for both male (R²: 0.99) and female (R²: 0.99) animals, however, significant difference were observed in the fit of the function (F_(4,70) = 770.0, p ≤ 0.001). All male animals completed signal detection within 57 days. In stark contrast, female animals displayed a 27-day lag, relative to male animals, before beginning to acquire the task, with all animals meeting criterion within 113 days. (B) A first-order polynomial provided a well-described fit for male control animals (R²: 0.93), while a sigmoidal dose response curve was the best fit for male HIV-1 Tg (R²: 0.98). Male HIV-1 Tg animals exhibited a 14-day lag, relative to control animals, before beginning to acquire the signal detection task. (C) A sigmoidal dose response curve was the best fit for female control (R²: 0.99) animals. In sharp contrast, the number of test sessions required for female HIV-1 Tg animals to criterion were best fit with a plateau followed by a one-phase association (R²: 0.95). Overall, female control animals acquired the task at a significantly faster rate relative to female HIV-1 Tg animals.

Figure 3

Once animal’s met criterion in the final signal detection task, the factor of biological sex had a significant effect on percent accuracy and an animal’s response profile. (A) Both male and female animals, regardless of genotype, exhibited a linear increase in percent accuracy across test sessions. However, the rate of increase in percent accuracy occurred more slowly in female animals relative to male animals. (B) Female animals displayed a decreased response rate relative to male animals on all response types (i.e., false alarms, hits, misses and correct rejections). There was no compelling evidence for any effect of the HIV-1 transgene on either percent accuracy or an animal’s response profile once the task was well acquired.