High growth hormone serum partially protects mice against Trypanosoma cruzi infection

Chagas disease (CD) is one of the most devasting parasitic diseases in the Americas, affecting 7–8 million people worldwide. In vitro and in vivo experiments have demonstrated that growth hormone (GH) serum levels decrease as CD progresses. Interestingly, inactivating mutations in the GH receptor in humans result in Laron syndrome (LS), a clinical entity characterized by increased serum levels of GH and decreased insulin growth factor‐1 (IGF‐1). The largest cohort of LS subjects lives in the southern provinces of Ecuador. Remarkably, no clinical CD cases have been reported in these individuals despite living in highly endemic areas. In the current ex vivo study, we employed serum from GHR−/− mice, also known as LS mice (a model of GH resistance with high GH and low IGF‐1 levels), and serum from bovine GH (bGH) transgenic mice (high GH and IGF‐1), to test the effect on Trypanosoma cruzi infection. We infected mouse fibroblast L‐cells with T. cruzi (etiological CD infectious agent) and treated them with serum from each mouse type. Treatment with GHR−/− serum (LS mice) significantly decreased L‐cell infection by 28% compared with 48% from control wild‐type mouse serum (WT). Treatment with bGH mouse serum significantly decreased infection of cells by 41% compared with 54% from WT controls. Our results suggest that high GH and low IGF‐1 in blood circulation, as typically seen in LS individuals, confer partial protection against T. cruzi infection. This study is the first to report decreased T. cruzi infection using serum collected from two modified mouse lines with altered GH action (GHR−/− and bGH).

decreased T. cruzi infection using serum collected from two modified mouse lines with altered GH action (GHR À/À and bGH).
Chagas disease (CD) is a parasitic disease caused by the protozoan parasite Trypanosoma cruzi. Approximately 7-8 million people are currently infected worldwide, leading to~50 000 deaths per year [1]. Of those infected, 5 million are found in South American countries. Due to migration and globalization in recent decades, CD has spread globally to nonendemic areas such as Canada, the USA, Europe, Australia, and Japan [2]. Transmission of CD in endemic areas occurs mainly through contact with contaminated feces of triatomine insects, also known as kissing bugs [3]. Less frequent infection routes include oral transmission, contaminated food, or vertical transmission from mother to child during pregnancy and childbirth [3][4][5]. Clinical manifestations of CD infection involve an initial acute stage with high parasitemia and display no or mild symptoms such as fever and anorexia [6]. Subsequently, CD progresses to a chronic phase that may present with clinical abnormalities such as cardiomyopathy or nervous system abnormalities that can cause incapacity and even death [5,7]. Therapy for the acute phase of CD is limited to two oral antiparasitic drugs commercially known as nifurtimox and benznidazole [8]. Unfortunately, there is no effective treatment for the chronic stage of CD [7,9].
Growth hormone (GH) is a protein secreted from the anterior pituitary gland that regulates postnatal growth, metabolism, and organ development [10]. GH production and secretion are regulated by hypothalamic GH-releasing hormone, somatostatin, stomachderived ghrelin, and endocrine insulin growth factor-1 (IGF-1) [11]. Changes in GH activity have been associated with various diseases in humans. For instance, untreated oversecretion of GH by pituitary adenoma results in acromegaly (AC) in adults and gigantism in children. AC is a slowly progressive disease caused by chronic hypersecretion of GH with a concomitant increase in circulating IGF-1 produced primarily by the liver [11]. By contrast, decreased secretion of GH results in GH deficiency (GHD) and is associated with impeded growth and other abnormalities in children. An extreme condition known as Laron syndrome (LS) is caused by homozygous inactivating mutations in the growth hormone receptor (GHR) gene (GHR À/À ) and is characterized by GH insensitivity [12]. LS subjects are resistant to GH and have decreased serum levels of IGF-1 and elevated GH levels, have severely diminished stature, and are obese. In an apparent paradox, these subjects display enhanced insulin sensitivity due to the absence of the GH counter-regulatory effects on carbohydrate metabolism and a diminished incidence of cancer and diabetes [13,14]. LS subjects also display slower cognitive decline than their age and sexmatched relatives (GHR +/+ or GHR +/À ) [13,15]. The largest cohort of LS subjects live in the southern provinces of Ecuador [13,16], and despite living in highly CD endemic areas, no clinical cases of this parasitic infection have been reported (Jaime Guevara-Aguirre, personal communication). Interestingly, the absence of CD in LS patients from Ecuador resonates with a large cohort of adult GH deficiency patients (GHDdecreased IGF-1 serum levels) from Brazil, where no cases of CD were observed [17].
Emerging evidence suggests that GH influences the progression of T. cruzi infection [18][19][20] (Table 1). Moreover, T. cruzi infection directly promotes decreased GH and prolactin (PRL) production by the pituitary [18]. Notably, GH and PRL are known to inhibit parasitic infections by enhancing the immune response in the host by increasing the concentrations of tumor necrosis factor-alpha (TNF-a), interleukin 12 (IL-12), interferon-gamma (IFN-c), and nitric oxide (NO) production [17,[20][21][22]. For example, rats infected with T. cruzi and treated with GH resulted in decreased parasitemia in the blood leading to an improved immune response (increased TNF-a, NO, and IFN-c) compared with nontreated controls [23]. Accordingly, our previous in vitro studies showed that human HeLa and mouse fibroblast L-cells infected with T. cruzi and treated with relatively high GH concentrations have significantly less CD infection [24]. Moreover, the combination of high GH and low IGF-1 levels, simulating LS conditions in vitro, decreased T. cruzi infection by preventing parasitic cell invasion into the cells [24]. When human HeLa cells were treated with a GH receptor antagonist (Pegvisomant), the levels of infection were restored similarly to the control levels (PBS) [24]. These data strongly suggest that GH influences T. cruzi infection in vitro.
In the current study, we used serum collected from LS GHR À/À mice (elevated GH, decreased IGF-1) and AC bGH mice (elevated GH, elevated IGF-1) previously generated in our laboratory to assess the effect of on T. cruzi infection [25,26]. Results showed that elevated GH and diminished IGF-1 serum levels significantly protect against T. cruzi infection. This study is the first to explore the absence of T. cruzi infection in LS subjects using an ex vivo GH insensitivity mouse model. Our results suggest that serum from GHR À/À mice confer partial protection against T. cruzi infection.

Parasite maintenance
The parasite T. cruzi [strain Brazil (TcI)] life-stage epimastigotes were cultured in liver-infusion-tryptose broth and supplemented with inactivated 10% FBS (Gibco TM ; catalog number 1600044). Inactivation of FBS serum complement components is essential for parasite survival and to ensure infection of cells. Epimastigotes were starved for 15 days until metacyclic trypomastigotes (MT) formed spontaneously and then subsequently used for infection. Commercial horse serum (Fisher Scientific, Waltham, MA, USA; catalog 35030CV) was implemented to eliminate remaining epimastigotes in the media [28]. MTs were then collected, washed in PBS 19, suspended in DMEM supplemented with 2% FBS (DMEM2), and used to infect mouse L-cells as previously described [24,29]. After several rounds of replication, tissue-derived trypomastigotes were collected 4-5 days postinfection (PI) and used for further L-cells infection as described before [24].

Mouse lines
We have previously generated two mouse models in our laboratory: GHR À/À mice and bGH mice with a C57BL/6J genetic background [25,26]. The GHR À/À , also known as the LS mouse, exhibits increased serum GH and decreased IGF-1 and insulin concentrations [25]. Conversely, bovine GH transgenic (bGH) mice possess high GH and IGF-1 serum  [24] levels, resembling the characteristics of untreated AC patients [30]. Serums from control wild-type (WT) C57BL/6J mice, GHR À/À mice, and bGH mice were used in our experiments.

Serum collection
Serum was collected following the protocol approved by the Institutional Animal Care and Committee of Ohio University protocol #12H012. Sexually mature male mice at 3month-old were used in all experiments [31]. Three-monthold male mice were selected because the C57BL/6J mice at this stage mice had reached sexual maturity and appeared fully developed as young adults [31]. Before bleeding, mice were fasted for 6 h, then the blood sample was obtained by cutting 1 mm from the tip of the mouse tail and was collected using microvette CB300 tubes (Fisher Scientific) [32]. Collected blood was kept on ice for 20 min (min) at room temperature (RT), then spun at 6000 g for 15 min to remove the clot. Six bGH (strain C57BL/6J) and six WT (strain C57BL/ 6J) mice littermate controls and six GHR À/À (strain C57BL/ 6J) and six WT (strain C57BL/6J) mice littermate controls were bled every month for six consecutive months. Serum samples from each mouse were kept individually at À80°C for 6 months, thawed, and pooled immediately before experiments. In total, six serum samples from each mouse were used for the infection experiments. Next, serum was aliquoted for measurement of GH and IGF-1 levels via ELISA and simultaneously used for treating cells [32]. Serum glucose levels were determined using a glucose testing kit (Contour next strips, Contour next EZ Ò , Parsippany, NJ, USA).  [24]. For treatment of L-cells with (bGH + mIGF-1), DMEM2 was added every 24 h for 4 days (day 1, 2, 3, 4), followed by infection with T. cruzi (day 3) for 24 h. Then, the cells were washed once with PBS (2 mL), and parasites were removed (day 4). L-cells were infected with trypomastigotes using 1 9 10 6 parasites per cell for 24 h, as previously described [24]. Infection of cells was analyzed at 48 h PI on day 5 (Fig. 1B-D).

Treatment of cells
For the ex vivo experiments (Figs 2 and 3), mouse L-cells were treated with GHR À/À mouse serum (10%) + DMEM or bGH mouse serum (10%) + DMEM for four consecutive days (days 1, 2, 3, 4), with serum added every 24 h, followed by infection with T. cruzi (day 3). Infection proceeded for 24 h using 1 9 10 6 parasites per cell. After that time, cells were washed, and parasites were removed, as above. Infection of cells was analyzed at 48 h PI (day 5).

Statistical analysis
In total, GHR À/À (n = 6), WT (n = 6), bGH (n = 6), and WT (n = 6) serum sample treatments were used for experiments. Results were expressed as mean AE standard error (SE). An unpaired t-test was performed in GRAPHPAD PRISM version 9.1.2 (San Diego, CA, USA). A P-value of < 0.05 was considered statistically significant.

GH modulates T. cruzi infection via bioinformatic analysis and in vitro
Using previous RNA-seq data [37][38][39] from differentially expressed genes, the heat map showed that GHR gene expression is downregulated in human foreskin fibroblast cells infected with T. cruzi (Fig. 1A).
Importantly, these data indicate that different T. cruzi strains (Sylvio, CL Brenner), at different times of infections (96 and 72 hpi), consistently downregulate GHR expression. In normal conditions, GHR levels are downregulated as T. cruzi infection progresses. These data coincide with our previous in vitro studies that showed exogenous treatment of cells with high levels of GH decreased T. cruzi infection [24]. Together, these data imply that GH plays a modulatory role during T. cruzi infection and coincides with our hypothesis that high GH levels decreased T. cruzi infection [24]. We then simulated LS conditions in vitro and found that L-cells infected with T. cruzi and treated with high GH concentrations levels (200 ngÁmL) + low IGF-1 (50 ngÁmL) significantly decreased the number of the infected cells by 35% (P < 0.01) compared with 90% in control cells (2% FBS) (Fig. 1B-D). We also found that 10% FBS treatment significantly decreased the number of infected cells by 60% (P < 0.01) compared with 90% control (2% FBS), possibly by GH and growth factors included in the commercial 10% FBS composition (GH~39 ngÁmL, PRL~176 ngÁmL, IGF-1~111 ngÁmL, Insulin~10 ngÁmL). We did not find any significant changes in infection when L-cells were treated with high GH levels (200 ngÁmL) + high IGF-1 levels (900 ngÁmL) DMEM, simulating AC conditions in vitro.

Discussion
This study is the first to characterize the effect of serum derived from GHR À/À and bGH mice on T. cruzi infection. We observed a significant decrease in the number of infected cells after treatment with GHR À/À serum, providing further evidence of the role of GH during T. cruzi infection. These results agree with our previous in vitro studies showing that high GH levels and the combination of high GH and low IGF-1 (LS conditions) decrease T. cruzi infection in mammalian cells.
In this ex vivo study, both mice lines have increased serum GH levels but differ in corresponding serum IGF-1 concentrations. Treatment with GHR À/À mice serum (with elevated GH and low IGF-1) decreased T. cruzi-infected cells by 28% compared with 48% in WT controls. By contrast, treatment with bGH mouse serum (with elevated GH and IGF-1) decreased the number of infected cells by 4% compared with 54% in WT controls. Therefore, serum from mice with elevated GH and decreased IGF-1 levels appear to have a more substantial effect on the parasite's ability to infect mammalian cells than serum with elevated GH and IGF-1 levels. Thus, elevated IGF-1 levels may partially alter the inhibitory effect of GH on infection. We are aware that the effect of decreasing T. cruzi infection with serum from each strain of mice could also be mediated by cytokines, proteases, and other proteins present in the GHR À/À (increased G-CSF) and bGH (increased IL-1b) serum. For example, previous studies have shown that increased G-CSF in mice has been reported to promote the development, mobilization, and activation of neutrophils, leading to protection against infections [43]. By contrast, IL-1b is known to play a role in coordinating host immune and proinflammatory responses [44]. In our results, serum cytokines showed only a few differences between both mice models (increased TNF-a in GHR À/À and decreased levels in bGH mice compared with WT controls); thus, further in vivo experiments are needed to clarify this matter. Although this is the first ex vivo study using LS mice (GHR À/À ) to characterize the potential effects of GH on T. cruzi infection, these findings are concordant with previous in vitro and in vivo studies that indicate T. cruzi infection may lead to hypothalamic-pituitary-adrenal axis (HPA) imbalance (Table 1) [18]. In vivo mice models have also shown that during T. cruzi infection, modulation of pituitary hormones PRL and GH and adrenal glucocorticoids (GC) caused immune suppression and thymic atrophy by CD4 + CD8 + T-cell depletion. Moreover, data from previous RNA-Seq analysis [37][38][39] show that GHR gene expression is consistently downregulated during T. cruzi infection (Fig. 1A). These data indicate that GHR levels are altered as T. cruzi infection progresses, implying that the GH/IGF-1 axis might also play a detrimental role during infection. In support of these findings, our ex vivo results show that high GH levels in circulation are associated with protection against T. cruzi infection in mammalian cells.
Of note, GHR À/À mice have elevated serum GH and decreased IGF-1 and, despite being short and obese, display low serum insulin concentrations along with improved insulin sensitivity. By contrast, bGH mice with excess serum GH and IGF-1 levels have decreased adiposity, insulin resistance, and high susceptibility to diabetes [26]. The very high GH serum levels found in both mouse lines appear to protect against T. cruzi infection. Our findings correlate with the clinical observation that no clinical cases of T. cruzi infection were reported in Ecuadorian LS subjects. Remarkably, LS subjects are obese, display high insulin sensitivity, and have diminished incidence of cancer and insulin-resistant diabetes [14]. The absence of the GH counter-regulatory effects of GH on carbohydrate metabolism, despite the very high serum GH levels, as well as the low serum IGF-1 and insulin levels documented in LS subjects, have been proposed to explain the diminished incidence of these diseases [13,14]. Our previous in vitro findings and the present observations in this ex vivo report suggest that high circulating GH and low circulating IGF-1 levels might be, at the very least, partially protecting LS subjects from T. cruzi infection.
Interestingly, in a study from Barrios et al. [45], when isolated GHD macrophages from patients from Brazil [17] were treated with IGF-1 in vitro, there was an increased infection with the parasite Leishmania spp. (closely related to T. cruzi). These data correlate with our previous in vitro study, where high levels of IGF-1 also increased T. cruzi infection in vitro [24]. Additionally, the study of GHD patients from Brazil showed that GH deficiency is not associated with an increased frequency of infectious diseases such as CD, Leishmaniasis, HIV, hepatitis B, and C compared with controls [17]. Thus, altered GH action, as seen in our LS model, seems to play a protective role during infectious diseases that need further exploration.
In summary, we report decreased T. cruzi in vitro infection in the presence of serum collected from two modified mouse lines (GHR À/À and bGH) with altered GH action. Even though a direct and indirect influence of T. cruzi in endocrine homeostasis through HPA axis imbalance has been documented, the relationship between LS patients and resistance to T. cruzi infection has only recently been explored [24]. Our results suggest that the high circulating GH serum levels may confer partial protection against T. cruzi infection in humans. These data are consistent with our previous in vitro findings showing that high serum GH levels, as seen in LS patients, confer resistance to T. cruzi infection. This study also highlights the potential of using GH to decrease infectivity, an event worth considering when treating patients during the acute and chronic CD phases. Although additional studies are needed to fully understand the direct or indirect mechanisms of GH action during T. cruzi infection, our findings provide a potential mechanism for explaining the absence of clinical T. cruzi infection observed in LS individuals.