Photoreceptor glucose metabolism determines normal retinal vascular growth

Abstract The neural cells and factors determining normal vascular growth are not well defined even though vision‐threatening neovessel growth, a major cause of blindness in retinopathy of prematurity (ROP) (and diabetic retinopathy), is driven by delayed normal vascular growth. We here examined whether hyperglycemia and low adiponectin (APN) levels delayed normal retinal vascularization, driven primarily by dysregulated photoreceptor metabolism. In premature infants, low APN levels correlated with hyperglycemia and delayed retinal vascular formation. Experimentally in a neonatal mouse model of postnatal hyperglycemia modeling early ROP, hyperglycemia caused photoreceptor dysfunction and delayed neurovascular maturation associated with changes in the APN pathway; recombinant mouse APN or APN receptor agonist AdipoRon treatment normalized vascular growth. APN deficiency decreased retinal mitochondrial metabolic enzyme levels particularly in photoreceptors, suppressed retinal vascular development, and decreased photoreceptor platelet‐derived growth factor (Pdgfb). APN pathway activation reversed these effects. Blockade of mitochondrial respiration abolished AdipoRon‐induced Pdgfb increase in photoreceptors. Photoreceptor knockdown of Pdgfb delayed retinal vascular formation. Stimulation of the APN pathway might prevent hyperglycemia‐associated retinal abnormalities and suppress phase I ROP in premature infants.

recombinant adiponectin or the agonist AdipoRon prevents retinopathy in an STZ-induced hyperglycemic mouse model. They further explore the mechanism connecting adiponectin /AdipoR to AMPK activation and mitochondrial function, which improves glucose metabolism in photoreceptor cells. Finally, they identified PDGFB as a responder effector of adiponectin function, mediating the generation of the vascular network. A timely and well performed set of studies. A few minor concerns: 1. Serum APN is generally reduced in established obesity/diabetes. In the STZ hyperglycemic retinopathy neonatal mice, both circulating adiponectin levels and AdipoR1 (the dominant receptor) expression in retinas are paradoxically higher, suggesting that the adiponectin pathway may be preserved and in fact enhanced under these conditions. What is the possible mechanism for this phenomenon in these STZ neonatal mice? Hyperglycemia is driving the retinopathy during the adiponectin /agonist treatment, is hyperglycemia corrected under these conditions? 2. The retinas are metabolically very active, as there is almost no reserve capacity in the vehicleretinas (Fig.4B). In vivo treatment with the agonist AdipoRon clearly increased maximal oxygen consumption. Are these measurements done in the retinas from hyperglycemic or eulglycemic mice? Does oxygen consumption decrease in hyperglycemic retinas, and is that improved by adiponectin? 3. In 661W cells, APMK is activated by AdipoRon treatment. Is the retinal AMPK activation required for adiponectin's effect in glucose metabolism and prevention of retinopathy? 4. Pdgfb is known to promote retinal vascular growth, as also shown in Fig 4, knockdown pdgfb in photoreceptor cells impairs the formation of the retinal vascular network. Does an intervention in the adiponectin receptor pathway reverse these defects? 5. Can the mice that have undergone retinopathy, upon activation of the adiponectin pathway, eventually develop normal eye function when grown up? 6. Is this strictly a glucotoxic phenomenon or is there a lipotoxic component involved as well? With STZ treatment, there is obviously a high degree of lipolysis ongoing, presumably also in these very young mice. What are the lipid levels under these conditions, and could this potentially involve the ceramide pathway?

Referee #3 (Remarks):
This is a very interesting and potentially important study that combines clinical and experimental investigations to elucidate a new mechanism for the development of retinopathy of prematurity (ROP), a sight-threatening disease in premature infants. Clinically, reduced levels of adiponectin (Apn) were found to be correlated with hyperglycemia and delayed retinal vessel formation in premature infants. Experimental studies using a newly developed mouse model of hyperglycemiaassociated retinopathy and 661w cell line demonstrate that depletion of Apn or activation of Apn pathway affects hyaloid vessel regression and retinal vascular formation in normal or hyperglycemic mice. Further experiments show that Apn increases retinal and photoreceptor metabolism and stimulates photoreceptor derived Pdgfb production. These findings clearly suggest a role of Apn in regulation of photoreceptor metabolism and vascular formation and therefore may serve as a potential target for ROP. Major points: 1. A major contribution of this manuscript would be the generation of a new model for study ROPrelated hyperglycemia-induced retinopathy. However, some important information appears to be missing. For example, what is the timing for the development of hyperglycemia and decrease of insulin production in this model. How about the survival rate of neonatal mice receiving STZ injection? In addition, injection of STZ from P1 -P9 only affects the deep layer but not superficial layer vascular formation. Can this phenomenon be explained by the timing for the development of hyperglycemia? 2. It is hypothesized that hyperglycemia suppresses photoreceptor metabolism resulting in delayed retinal vascularization. However, direct evidence for this is lacking (Apn has no effect on hyperglycemia). Perhaps it would be more convincing to demonstrate the role of hyperglycemia in inducing photoreceptor dysfunction by supplementation of insulin to the hyperglycemic neonatal mice. 3. While most data were derived from a large number of animals and appear to be very compelling, the quality of the deep layer vasculature seems to be suboptimal. Further, some representative images do not show the changes as indicated in the graph. For example, Figure 3C: the images on the left show fewer hyaloid vessels in Apn-/-retinas, while the graph shows an increase in these retinas. 4. Is AMPK activation altered by AdipoRon injection and in Apn-/-retinas in vivo? These data would be helpful in supporting the hypothesis ( Figure 5). 5. It is less clear how Apn regulates VEGFA expression, although Figure 5 indicates with Apn photoreceptors produce less VEGF (+PDGFB, -VEGF)? There is no data provided in this regard. 6. The use of "photoreceptors" instead of "661w cells" is confusing and misleading, since changes in photoreceptors generally refer to results from in vivo studies. Furthermore, 661w cells is a cone photoreceptor cell line. The results derived from the in vitro study should be carefully interpreted Minor points: 7. The definition of HAR should be consist in the manuscript. In the legends for Figure 2: hyperglycemia-associated retinopathy (HAR); Figure3: ...hyperglycemia-associated retinal vascularization (HAR) delay.... Which is correct? 8. Legend for Figure 4: Please specify the age of the mice used in each assay. (B): is the difference significant? (C): please clarify "In Apn-/-versus WT STZ" -were these mice hyperglycemic? 9. Figure EV3b, c shows that normal or hyperglycemic Apn-/-retinas display significant reduction in superficial layer vascular formation. This information is missing in the results section. The significance and mechanism for this defect should discussed. In contrast, there is only a modest decrease in the deep layer vascular formation in Apn-/-retinas ( Figure 3E). Is there a possible explanation for this? 10. Symbols for indications of significance are missing in some figures. Referee #2 (Remarks): Fu and colleagues report an interesting study focusing on adiponectin's function in hyperglycemia related "retinopathy of prematurity" (ROP). They report that low adiponectin correlated with delayed retinal vascularization in premature infants. They also found adiponectin deficiency delays hyaloid regression and retinal vascularization in neonatal mice. Activation of the APN pathway by recombinant adiponectin or the agonist AdipoRon prevents retinopathy in an STZ-induced hyperglycemic mouse model. They further explore the mechanism connecting adiponectin /AdipoR to AMPK activation and mitochondrial function, which improves glucose metabolism in photoreceptor cells. Finally, they identified PDGFB as a responder effector of adiponectin function, mediating the generation of the vascular network. A timely and well performed set of studies. A few minor concerns: 1. Serum APN is generally reduced in established obesity/diabetes. In the STZ hyperglycemic retinopathy neonatal mice, both circulating adiponectin levels and AdipoR1 (the dominant receptor) expression in retinas are paradoxically higher, suggesting that the adiponectin pathway may be preserved and in fact enhanced under these conditions. What is the possible mechanism for this phenomenon in these STZ neonatal mice? Hyperglycemia is driving the retinopathy during the adiponectin /agonist treatment, is hyperglycemia corrected under these conditions?
Responses: We thank the reviewer for the comments on the merit of our study and we understand the reviewer's concern.
In STZ-induced mouse neonates, serum insulin levels were decreased and the APN pathway was enhanced. Pharmaceutical activation of the APN pathway promoted and deficiency of APN further inhibited the deep retinal vascularization, suggesting that activation of the APN pathway may serves as a compensatory response with insulin deficiency. The phenomenon seen here is actually not unique. In type 1 diabetes (T1D), circulating APN levels are higher than non-diabetic controls (Pereira et al, 2012 Therefore, we speculate that the APN pathway is activated to protect against the cell metabolic stress with lack of insulin. However, under obese/T2D conditions, the production of APN and its receptors are affected and thus the modulatory effects of APN on metabolism are dampened. In neonatal STZ, although we did not observe a significant impact of APN on circulating blood glucose levels (mean blood glucose at P10 Apn -/-HAR mice: 206±8mg/dl; WT HAR mice: 197±6mg/dl), we did observe further increased circulating triglyceride levels with APN deficiency (as shown below).
In addition, we also found that APN deficiency decreased the local production of retinal metabolic enzymes involved in glucose pathway and mitochondrial activity ( Figure 4A), as well as a direct impact of APN treatment on retinal mitochondrial respiration ( Figure 4B). Taken together, we summarized that activation of the APN pathway compensated for the reduced circulating insulin levels, which in turn promoted retinal metabolism and neurovascular development in the HAR model.
2. The retinas are metabolically very active, as there is almost no reserve capacity in the vehicleretinas (Fig.4B). In vivo treatment with the agonist AdipoRon clearly increased maximal oxygen consumption. Are these measurements done in the retinas from hyperglycemic or neulglycemic mice? Does oxygen consumption decrease in hyperglycemic retinas, and is that improved by adiponectin?
Responses: We appreciate the reviewer's valuable comments. The measurement of AdipoRon on retinal oxygen consumption rate (OCR) was originally obtained from normal WT retinas when we tried to prove the concept that activation of the APN pathway increased retinal OCR.
We did observe an overall reduction in OCR in hyperglycemic versus normal control retinas ( Figure EV3, A). The mild changes seen possibly resulted from the fact that non-littermate controls were used. Unfortunately, we noticed that if we used mice in the same litter for controls, the STZinduced mouse pups were weaker than the vehicle (PBS)-treated pups, which led to less capability to fight for the nursing dam's milk and in turn they grew more slowly than the control pups. Therefore, in the HAR model, we treated all mice from same litter with either STZ or PBS for the phenotypic studies. The number of mouse pups was adjusted in PBS-treated group to achieve a similar range of body weights as mouse body weight may affect the development of retinopathy during developmental stage (Connor et al, 2009).
APN deficiency induced higher serum triglyceride (TG) levels in neonatal HAR model at P10. n=4-5 per group. Unpaired t test.
In the Seahorse analysis of retinal punches ex vivo, we noticed that oligomycin treatment had a negative impact on the induction of maximal respiration later, as previously reported (Kooragayala et al, 2015). Therefore, in our later studies, we did not treat retinas with oligomycin in order to achieve maximal respiration. We speculate that the lack of reserve metabolic capacity in normal WT retinas might due to the vehicle (DMSO, the diluting solution for AdipoRon) used. Therefore, we repeated the experiment with either msAPN or vehicle (PBS) treatment in WT HAR mice. Vehicletreated retinas showed increased respiration after FCCP and msAPN significantly increased retinal OCR (Fig. 4B). Responses: In our drug intervention study, we found that msAPN administration showed better retinal vascular formation than vehicle control (Fig. 3F). We now further measured local gene expression of pdgfb in retinas after msAPN administration. msAPN increased retinal pdgfb expression (Fig. 4H). APN deficiency led to reduced pdgfb expression, particularly in photoreceptors (Fig. 4I). Therefore, in summary, interventions affecting the APN pathway could increase retinal expression of pdgfb, which in turn could promote retinal vascular growth.

In 661W cells, AMPK is activated by
5. Can the mice that have undergone retinopathy, upon activation of the adiponectin pathway, eventually develop normal eye function when grown up?
Responses: Yes, msAPN treated mice (0.6ug/g) versus vehicle control (PBS) for up to one month in the HAR model had significantly better retinal function detected by electroretinography (ERG). In the HAR model, both rod ( Figure 5A) and cone ( Figure 5E) signals detected by ERG were significantly decreased. msAPN treatment significantly increased post-receptor b-wave amplitude; there was also a non-significant trend of increase in photoreceptor a-wave sensitivity and amplitude ( Figure 5D). As post-receptor responses positively correlated with the sum of changes in photoreceptors ( Figure 5B), our findings indicated that msAPN might rescue rod photoreceptor function in HAR. msAPN treated mice also demonstrated significantly better cone function (reflected by post-receptor response Vm) ( Figure 5E). Therefore, our findings suggested that activation of the APN pathway can improve eye function long term in the HAR model. 6. Is this strictly a glucotoxic phenomenon or is there a lipotoxic component involved as well? With STZ treatment, there is obviously a high degree of lipolysis ongoing, presumably also in these very young mice. What are the lipid levels under these conditions, and could this potentially involve the ceramide pathway?
Responses: Thanks for raising this important question and we agree with the reviewer that both glucose and lipid metabolic alterations might contribute to the disease progression.  In the HAR model, serum triglyceride levels were highly increased after the induction of hyperglycemia with STZ treatment (Figure 2D, Figure EV1, C). Our preliminary data also showed that retinal ceramide levels were higher in APN-deficient mice versus WT mice in HAR model (unpublished data). Ceramide induces photoreceptor apoptosis (German et al, 2006). A strong correlation between ceramide signaling and homeostasis of photoreceptor and retinal pigment epithelial cells is also suggested in retinal degeneration (Chen et al, 2012;Strettoi et al, 2010). APN receptor adipoR1and adipoR2 possess ceramidase activity to convert ceramide to anti-apoptotic metabolite sphingosine-1-phosphate (Holland et al, 2011). Therefore, we speculate that the protective effects of APN against HAR might also be mediated through the modulation of retinal ceramide levels. Further investigation is still ongoing to fully elucidate the contribution of lipid metabolic alterations and the ceramide pathway in HAR.

Referee #3 (Remarks):
This is a very interesting and potentially important study that combines clinical and experimental investigations to elucidate a new mechanism for the development of retinopathy of prematurity (ROP), a sight-threatening disease in premature infants. Clinically, reduced levels of adiponectin (Apn) were found to be correlated with hyperglycemia and delayed retinal vessel formation in premature infants. Experimental studies using a newly developed mouse model of hyperglycemiaassociated retinopathy and 661w cell line demonstrate that depletion of Apn or activation of Apn pathway affects hyaloid vessel regression and retinal vascular formation in normal or hyperglycemic mice. Further experiments show that Apn increases retinal and photoreceptor metabolism and stimulates photoreceptor derived Pdgfb production. These findings clearly suggest a role of Apn in regulation of photoreceptor metabolism and vascular formation and therefore may serve as a potential target for ROP.
Major points: 1. A major contribution of this manuscript would be the generation of a new model for study ROPrelated hyperglycemia-induced retinopathy. However, some important information appears to be missing. For example, what is the timing for the development of hyperglycemia and decrease of insulin production in this model. How about the survival rate of neonatal mice receiving STZ injection? In addition, injection of STZ from P1 -P9 only affects the deep layer but not superficial layer vascular formation. Can this phenomenon be explained by the timing for the development of hyperglycemia?
Responses: Thanks for the valuable comments. We have now examined the time course of hyperglycemia development in WT mice after STZ injections. We found that hyperglycemia was significantly induced at P8 and persisted at P10 (Figure EV1, C). We agree with the reviewer that the deep layer but not the superficial layer vascular formation could be partially explained by the timing of the onset of hyperglycemia, as in mouse, the superficial layer forms between P1-P10 and the deep layer starts forming at P8. At P10, when the retinal vasculature was examined, we found that the major metabolic changes were located in photoreceptors but not in other neuronal layers ( Figure 4C) nor in the deep blood vessels (Figure EV3, B). As retina (in particular photoreceptors) is the most metabolically demanding tissue in the body (Wong-Riley, 2010) and photoreceptors have the highest number of mitochondria of any cell (Hoang et al, 2002), therefore, we hypothesize that the disturbance of glucose metabolism observed at P8 may lead to photoreceptor metabolic dysfunction, which in turn affect deep retinal vascular development.
Unfortunately, we were not able to measure the serum insulin levels at very young ages as the mice are much too small to obtain 25µl of serum that is required according to the protocol. Figure EV1: (C) Blood glucose levels starting from P2 to P10 (n=7-19/group, ANOVA) and serum triglyceride levels at P8 (n=5-7/group, unpaired t test) in control and HAR mice.
Generally, we noticed that in the same litter, the STZ-induced mouse pups were weaker than the vehicle (PBS)-treated pups, which led to less capability to fight for the nursing dam's milk and in turn led to slower growth than the control pups. Therefore, we treated all mice in the same litter with either STZ or vehicle for phenotypic studies. The number of mouse pups was adjusted in PBStreated group to achieve the similar range of body weights as the mouse body weight may affect the development of retinopathy during developmental stage (Connor et al, 2009). The pup number in the STZ-injected litter was limited to six to achieve a similar body weight range among different litters. All six mice in a litter survived at P10 and the body weight was within the 4 to 5g range.
2. It is hypothesized that hyperglycemia suppresses photoreceptor metabolism resulting in delayed retinal vascularization. However, direct evidence for this is lacking (Apn has no effect on hyperglycemia). Perhaps it would be more convincing to demonstrate the role of hyperglycemia in inducing photoreceptor dysfunction by supplementation of insulin to the hyperglycemic neonatal mice.
Responses: We thank the reviewer for the suggestion. We have now treated WT HAR mice with insulin to examine if hyperglycemia is the major cause. To avoid any potential drug/drug interference, we injected STZ and insulin separately. As two intraperitoneal injections per day may cause too much stress and death in neonatal mice, we started the co-treatment from P7, a day before the deep retinal vascular layer begins to form. At P10, insulin treated mice showed better deep vascular formation versus the littermate vehicle-treated mice (Figure EV1, F).
3. While most data were derived from a large number of animals and appear to be very compelling, the quality of the deep layer vasculature seems to be suboptimal. Further, some representative images do not show the changes as indicated in the graph. For example, Figure 3C: the images on the left show fewer hyaloid vessels in Apn-/-retinas, while the graph shows an increase in these retinas.

Is AMPK activation altered by AdipoRon injection and in
Apn-/-retinas in vivo? These data would be helpful in supporting the hypothesis ( Figure 5).
Responses: Thank you for the comment. We tried to do Western blots of p-AMPK and AMPK in HAR retinas. Unfortunately, unlike in 661W cells, there were multiple non-specific bands and the  (Figure EV3, H). Therefore, AMPK activation is required for APN protection against hyperglycemia-induced retinal vascular changes. Figure 5 indicates with Apn photoreceptors produce less VEGF (+PDGFB, -VEGF)? There is no data provided in this regard.

It is less clear how Apn regulates VEGFA expression, although
Responses: Thanks for pointing out the missing data. We observed that although APN deficiency reduced Vegfa expression in the ONL layer ( Figure EV4A), knockdown of photoreceptor-derived VEGFA did not affect normal retinal vascularization in mice (Figure EV4, B). We speculate that the decrease in VEGFA with APN deficiency might not contribute substantially to the retinal vascularization in the deep layer in HAR model.
We also apologize for the confusion in this figure, we have now updated the figure as shown below.   6. The use of "photoreceptors" instead of "661w cells" is confusing and misleading, since changes in photoreceptors generally refer to results from in vivo studies. Furthermore, 661w cells is a cone photoreceptor cell line. The results derived from the in vitro study should be carefully interpreted Responses: Thanks for pointing this out. We now use the term "661W cells" instead of the term "photoreceptors" for in vitro studies. We also agree with the reviewer that 661W cells are a cone photoreceptor cell line. To examine if cone function could be improved with activation of the APN pathway, cone ERGs were obtained in WT HAR mice treated with msAPN or vehicle. In the HAR model, cone function (reflected by post-receptor response Vm, because the cone response is always small in mice) was decreased versus normal control mice. The msAPN treatment significantly improved cone responses (Figure 5E). Our findings suggested that the in vitro data correspond to the in vivo data.
Responses: HAR is now defined consistently as hyperglycemia-associated retinopathy. We have now updated the current manuscript.
8. Legend for Figure 4: Please specify the age of the mice used in each assay. (B): is the difference significant? (C): please clarify "In Apn-/-versus WT STZ" -were these mice hyperglycemic?
Responses: The mice were sacrificed and the retinas were isolated for studies at P10. In the current manuscript, we include this information. In B, we updated this figure with msAPN versus vehicle treatment in WT HAR mice, significantly increased maximal respiration was observed with msAPN treatment. (C) We have now updated it as "Apn -/versus WT HAR P10 retinas", these mice are hyperglycemic. 9. Figure EV3b, c shows that normal or hyperglycemic Apn-/-retinas display significant reduction in superficial layer vascular formation. This information is missing in the results section. The significance and mechanism for this defect should discussed. In contrast, there is only a modest decrease in the deep layer vascular formation in Apn-/-retinas ( Figure 3E). Is there a possible explanation for this?
Responses: We appreciate the reviewer's comment. We now state the results in the Result section as "APN deficiency inhibited superficial but not deep retinal vascularization in normal conditions at P10 (Fig. EV2B)".
We first speculated that this phenomenon might due to the persistent hyaloid vessels with APN deficiency seen at P8 (Fig. 3C, Apn -/-, control vs. WT, control). In both mice and human eyes, the hyaloid vessels are the first to form in utero and regress as the retinal vessels develop (Ito & Yoshioka, 1999;Stahl et al, 2010). Therefore persistent hyaloid vessels may keep supplying the nutrients and oxygen for the retina and thus decrease the need for superficial retinal vessels. In WT HAR vs normal mice, hyaloid vessels were persistent in HAR (Fig. 3C) but there were no significant differences in the superficial vascular network (Figure EV1, E). As noted, there is a sharp drop in the number counts of hyaloid vessels from P5 to P8 in mice (Rao et al, 2013). In HAR, hyperglycemia was induced from P6 to P8 (Figure EV1, C). The rise in blood glucose levels may interrupt the hyaloid vessel regression during this sensitive period, which leads to persistent hyaloid vessels. However, as the superficial retinal vessels cover more than 50% of the retinas at P5 and almost reach the most peripheral region at P8, the impact of hyperglycemia on the superficial vasculature may be mild. However, Apn -/mice, with absence of APN from conception, have a more profound change in hyaloid and even on superficial retinal vessels. Further investigation of the influence of the APN pathway on the systems controlling hyaloid vessel regression and superficial retinal vascular formation (e.g. retinal ganglion cells, RGCs (Rao et al, 2013)) are needed to better understand mechanisms. APN receptors were found in RGCs (Fig. 3B)  As noted above, complete loss of APN led to persistent hyaloid vessels and suppressed superficial retinal vessel growth. The relatively mild impact of APN deficiency on deep vascular layer versus on the superficial layer is possibly due to different controls for superficial and deep retinal vessel growth. As photoreceptors are highly metabolically active and require high levels oxygen/nutrients for energy production, the formation of deep vascular vessels may be more dependent on photoreceptor metabolic needs than the superficial vessels. We do not exclude the possibility that there are other factors to compensate for the loss of APN.
We now have included the discussion in the current manuscript.

Data
the data were obtained and processed according to the field's best practice and are presented to reflect the results of the experiments in an accurate and unbiased manner. figure panels include only data points, measurements or observations that can be compared to each other in a scientifically meaningful way. graphs include clearly labeled error bars for independent experiments and sample sizes. Unless justified, error bars should not be shown for technical replicates. if n< 5, the individual data points from each experiment should be plotted and any statistical test employed should be justified the exact sample size (n) for each experimental group/condition, given as a number, not a range; Each figure caption should contain the following information, for each panel where they are relevant:

Captions
The data shown in figures should satisfy the following conditions: Source Data should be included to report the data underlying graphs. Please follow the guidelines set out in the author ship guidelines on Data Presentation.
Please fill out these boxes ê (Do not worry if you cannot see all your text once you press return) a specification of the experimental system investigated (eg cell line, species name).

B--Statistics and general methods
the assay(s) and method(s) used to carry out the reported observations and measurements an explicit mention of the biological and chemical entity(ies) that are being measured. an explicit mention of the biological and chemical entity(ies) that are altered/varied/perturbed in a controlled manner.
a statement of how many times the experiment shown was independently replicated in the laboratory.
Any descriptions too long for the figure legend should be included in the methods section and/or with the source data.
In the pink boxes below, please ensure that the answers to the following questions are reported in the manuscript itself. Every question should be answered. If the question is not relevant to your research, please write NA (non applicable). We encourage you to include a specific subsection in the methods section for statistics, reagents, animal models and human subjects.
definitions of statistical methods and measures: a description of the sample collection allowing the reader to understand whether the samples represent technical or biological replicates (including how many animals, litters, cultures, etc.).