Human molecular genetics sheds light on the physiological significance of ribonuclease inhibitor (RNH1)

Ribonuclease

week 20 of gestation, and was not tested for the RNH1 mutation. The affected boy died at the age of 8 months while the girl survived with global developmental delay.
Taken together, RNH1 deficiency in humans seems to cause problems only in a homozygous state, whereas heterozygous carriers are typically healthy. RNH1 deficiency has a rather dramatic phenotype, manifesting shortly after birth or even already in utero, such as bilateral cataracts observed in one sibling. The affected tissues include neurons, muscles, eye lens, erythropoiesis, and the immune system. Based on the presented data, erythropoiesis at steady state is probably sufficient, but it deteriorates during infection, which hints to a defective stress-erythropoiesis [16].
It's worth noting that a previous study in mice provided some similar in vivo phenotypic evidences [14]. First, only homozygous Rnh1-deficient mice were affected, while heterozygous carriers remained healthy. Second, all affected mice died in utero, and therefore the onset was again early in their development. Third, erythropoiesis was reduced but still present in mouse embryos ( Fig. 1). However, the potential defects in neurons, muscles and eye lenses could not be examined in these E8.5 to E10 mouse embryos and may need further attention. In addition, Bombaci et al. reported a haematopoietic-specific conditional Rnh1 −/− mouse model that bypassed the vulnerable embryonic phase [2]. They are viable and do not show gross phenotypic defects. Interestingly, upon challenge with small doses of lipopolysaccharide (LPS), all conditional Rnh1-deficient mice died, whereas their wild-type littermates survived. This suggests that these mice struggle to cope with even mild immunological challenges. These findings are similar to those observed in the two reported human siblings who had up to five hospitalizations upon infection during their first year of life.
There were few interindividual as well as interspecies (between mouse and human) differences in RNH1 mutant phenotypes. The interindividual difference includes severity of disease where one of the affected siblings died at the age of 8 months while the other survived. This difference in severity was also observed between species: Constitutive Rnh1 knockout was completely lethal in utero for mice but two RNH1-deficient patients survived embryonic development. This variability in severity could potentially be attributed to a compensatory mechanism involving a truncated version of the RNH1 protein although such a protein was not detected by antibodies and in cell-type used by the authors. Moreover, presence of other disease-causing genetic mutations cannot be entirely ruled out that may have contributed to the observed differences. Despite these differences in severity, the shared phenotype described in patients is consistent with the one observed in Rnh1 knockout mice, strongly arguing for a causative role of RNH1 deficiency in the pathogenesis.
To explain the mechanism behind RNH1-mediated pathophysiology, the authors focused on the "Ribonuclease inhibitor" function of RNH1 [16]. Pancreatic-type RNases can enter cells and degrade RNAs leading to cell death or tissue damage [1]. Therefore, loss of RNH1 could increase RNase-mediated tissue damage. Supporting this Hedberg-Oldfors et al. showed increased cell death in RNH1 mutated patient-derived fibroblasts treated with increased concentration of RNase A compared to healthy control [16]. However, these results should be interpreted with caution because: (1) RNase A is a bovine counterpart for human RNase 1 and has~5 fold more activity than human RNase 1 [17]. (2) No cell death was observed in patient-derived fibroblasts when RNase A was added at lower concentrations or not added at all. At higher concentrations (about 200-1000 times higher than~0.5 ug/ml of RNase 1 usually present in blood and other body fluids [17,18]) of RNase A, only~15-20% of cell death was reported. This suggests that, under normal conditions, cells with a deficiency in RNH1 can still protect their RNAs and survive. (3) There is no data that RNase activity is increased in serum and cerebrospinal fluid of these patients. (4) RNase 1 is widely expressed in several tissues [17,19] but RNH1 mutated patients did not show multi-organ damage. Rather, they have specific phenotypes such as neurodevelopmental delay, myopathy, congenital cataracts and anemia. (5) RNA sequencing analysis of patient-derived fibroblasts, yolk-sac-derived cells from Rnh1 knockout mice, as well as Rnh1 knockout K562 cells, did not show large changes in the transcriptome compared to their respective controls [14,16] (6) RNase 1 knockout mice do not show an abnormal phenotype and have high extracellular RNA levels (eRNA) [20]. eRNA activates innate immune response [21], however these mice did not show any inflammatory phenotype despite increased eRNA levels. Although an increase in RNase activity in RNH1 mutant patients could potentially decrease eRNA levels and reduce inflammation, these patients, in fact, experience more inflammatory episodes. Collectively, it is less likely that RNasemediated toxic effects contribute to the disease phenotype in these patients. Rather, cell intrinsic functions of RNH1 might play an important role.
Further studies are still needed to check the extent to which RNases and mutated RNH1 contribute to the pathogenesis of the disease. For this, thorough characterization of the immune function and long-term follow-up of patients is needed. Additionally, generating RNH1 and RNases double knockout mice would be helpful to address these points.
Though there are several questions needed to address, this study by Hedberg-Oldfors et al. reveals an important role for RNH1 in human pathophysiology and motivates further studies on RNH1 [16]. Identifying more RNH1 mutations in humans and in combination with functional and genomic studies will unravel the precise role of RNH1 in human physiology. Constitutive Rnh1 knockout in mice is embryonically lethal, shows erythroid differentiation defects and anemia phenotype. Rnh1 deletion in myeloid cells increases inflammation by innate immune activators. Loss-of-function mutations in the human RNH1 gene affects brain, muscles, eyes and cause infection-induced psychomotor deterioration, seizures, and macrocytic anemia. RNH1 structure is taken from PDB.