Phylogenetic analysis of SARS-CoV-2 in Boston highlights the impact of superspreading events

Analysis of 772 complete severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) genomes from early in the Boston-area epidemic revealed numerous introductions of the virus, a small number of which led to most cases. The data revealed two superspreading events. One, in a skilled nursing facility, led to rapid transmission and significant mortality in this vulnerable population but little broader spread, whereas other introductions into the facility had little effect. The second, at an international business conference, produced sustained community transmission and was exported, resulting in extensive regional, national, and international spread. The two events also differed substantially in the genetic variation they generated, suggesting varying transmission dynamics in superspreading events. Our results show how genomic epidemiology can help to understand the link between individual clusters and wider community spread.

Streamlined inactivation, amplification, and Cas13-based detection of SARS-CoV-2

The COVID-19 pandemic has highlighted that new diagnostic technologies are essential for controlling disease transmission. Here, we develop SHINE (Streamlined Highlighting of Infections to Navigate Epidemics), a sensitive and specific diagnostic tool that can detect SARS-CoV-2 RNA from unextracted samples. We identify the optimal conditions to allow RPA-based amplification and Cas13-based detection to occur in a single step, simplifying assay preparation and reducing run-time. We improve HUDSON to rapidly inactivate viruses in nasopharyngeal swabs and saliva in 10 min. SHINE's results can be visualized with an in-tube fluorescent readout - reducing contamination risk as amplification reaction tubes remain sealed - and interpreted by a companion smartphone application. We validate SHINE on 50 nasopharyngeal patient samples, demonstrating 90% sensitivity and 100% specificity compared to RT-qPCR with a sample-to-answer time of 50 min. SHINE has the potential to be used outside of hospitals and clinical laboratories, greatly enhancing diagnostic capabilities.

Development of a qualitative real-time RT-PCR assay for the detection of SARS-CoV-2: A guide and case study in setting up an emergency-use, laboratory-developed molecular assay

Developing and deploying new diagnostic tests is difficult, but the need to do so in response to a rapidly emerging pandemic such as COVID-19 is crucially important for an effective response. In the early stages of a pandemic, laboratories play a key role in helping health care providers and public health authorities detect active infection, a task most commonly achieved using nucleic acid-based assays. While the landscape of diagnostics is rapidly evolving, polymerase chain reaction (PCR) remains the gold-standard of nucleic acid-based diagnostic assays, in part due to its reliability, flexibility, and wide deployment. To address a critical local shortage of testing capacity persisting during the COVID-19 outbreak, our hospital set up a molecular based laboratory developed test (LDT) to accurately and safely diagnose SARS-CoV-2. We describe here the process of developing an emergency-use LDT, in the hope that our experience will be useful to other laboratories in future outbreaks and will help to lower barriers to fast and accurate diagnostic testing in crisis conditions.

Phylogenetic analysis of SARS-CoV-2 in the Boston area highlights the role of recurrent importation and superspreading events

SARS-CoV-2 has caused a severe, ongoing outbreak of COVID-19 in Massachusetts with 111,070 confirmed cases and 8,433 deaths as of August 1, 2020. To investigate the introduction, spread, and epidemiology of COVID-19 in the Boston area, we sequenced and analyzed 772 complete SARS-CoV-2 genomes from the region, including nearly all confirmed cases within the first week of the epidemic and hundreds of cases from major outbreaks at a conference, a nursing facility, and among homeless shelter guests and staff. The data reveal over 80 introductions into the Boston area, predominantly from elsewhere in the United States and Europe. We studied two superspreading events covered by the data, events that led to very different outcomes because of the timing and populations involved. One produced rapid spread in a vulnerable population but little onward transmission, while the other was a major contributor to sustained community transmission, including outbreaks in homeless populations, and was exported to several other domestic and international sites. The same two events differed significantly in the number of new mutations seen, raising the possibility that SARS-CoV-2 superspreading might encompass disparate transmission dynamics. Our results highlight the failure of measures to prevent importation into MA early in the outbreak, underscore the role of superspreading in amplifying an outbreak in a major urban area, and lay a foundation for contact tracing informed by genetic data.

Phylogenetic analysis of SARS-CoV-2 in the Boston area highlights the role of recurrent importation and superspreading events

SARS-CoV-2 has caused a severe, ongoing outbreak of COVID-19 in Massachusetts with 111,070 confirmed cases and 8,433 deaths as of August 1, 2020. To investigate the introduction, spread, and epidemiology of COVID-19 in the Boston area, we sequenced and analyzed 772 complete SARS-CoV-2 genomes from the region, including nearly all confirmed cases within the first week of the epidemic and hundreds of cases from major outbreaks at a conference, a nursing facility, and among homeless shelter guests and staff. The data reveal over 80 introductions into the Boston area, predominantly from elsewhere in the United States and Europe. We studied two superspreading events covered by the data, events that led to very different outcomes because of the timing and populations involved. One produced rapid spread in a vulnerable population but little onward transmission, while the other was a major contributor to sustained community transmission, including outbreaks in homeless populations, and was exported to several other domestic and international sites. The same two events differed significantly in the number of new mutations seen, raising the possibility that SARS-CoV-2 superspreading might encompass disparate transmission dynamics. Our results highlight the failure of measures to prevent importation into MA early in the outbreak, underscore the role of superspreading in amplifying an outbreak in a major urban area, and lay a foundation for contact tracing informed by genetic data.

Integrated sample inactivation, amplification, and Cas13-based detection of SARS-CoV-2

The COVID-19 pandemic has highlighted that new diagnostic technologies are essential for controlling disease transmission. Here, we develop SHINE (SHERLOCK and HUDSON Integration to Navigate Epidemics), a sensitive and specific integrated diagnostic tool that can detect SARS-CoV-2 RNA from unextracted samples. We combine the steps of SHERLOCK into a single-step reaction and optimize HUDSON to accelerate viral inactivation in nasopharyngeal swabs and saliva. SHINE's results can be visualized with an in-tube fluorescent readout - reducing contamination risk as amplification reaction tubes remain sealed - and interpreted by a companion smartphone application. We validate SHINE on 50 nasopharyngeal patient samples, demonstrating 90% sensitivity and 100% specificity compared to RT-PCR with a sample-to-answer time of 50 minutes. SHINE has the potential to be used outside of hospitals and clinical laboratories, greatly enhancing diagnostic capabilities.

123 Nerve growth factor-induced ovulation in llamas: Evidence of hypothalamic refractoriness to nerve growth factor during the declining phase of the luteinising hormone surge

Nerve growth factor (NGF) in semen is responsible for triggering ovulation after copulation in camelids. Interaction of NGF with its cognate receptors results in a preovulatory luteinising hormone (LH) surge that leads to ovulation, but the pharmacokinetics of NGF and the mechanism by which it mediates LH release are unknown. In an effort to elucidate the site and mechanism of action involved, the objective of this study was to determine whether the decline in the LH surge occurs as a result of pituitary depletion (i.e. diminished response to gonadotrophin-releasing hormone (GnRH)) or as an obtunded response at the level of the hypothalamus (diminished GnRH). Adult nonpregnant, nonlactating llamas (n=18) were synchronized by the administration of an ovulatory dose of a GnRH analogue (100μg IM; Fertiline, Vetoquinol). At 10 to 14 days later, the ovaries were examined to confirm the presence of a dominant follicle measuring ≥ 7mm, and a jugular catheter was put in place. The following day, a pretreatment blood sample was taken, and llamas were treated intravenously with 1mg of purified NGF from llama seminal plasma. Blood samples were taken every 30min for 7h from the time of NGF treatment. After the last blood sample was taken, llamas were treated with NGF (n=6; 1mg IV), GnRH (n=5; 100μg IV), or saline (Sal; IV; n=6), and blood samples were taken every 30min for another 7h. The ovaries were examined 48h after initial NGF treatment, via transrectal ultrasonography, to detect ovulation. Plasma was harvested and stored for analysis of LH concentration by radioimmunoassay. Data were compared using ANOVA for repeated measures, and single-point data were analysed using paired t-tests. As expected, most llamas ovulated in response to the initial NGF treatment (5 out of 5 in NGF-GnRH; 5 of 6 in NGF-NGF; 5 of 6 in NGF-Sal). Compared with pretreatment values, all llamas showed a 4-fold increase in plasma LH concentrations within 2h of the initial NGF treatment (P<0.05). Plasma LH concentrations peaked at 3h after initial NGF treatment and began to decline 4.5 to 5h after treatment (P<0.05). Plasma LH concentrations continued to decline following the second dose of NGF or Sal, whereas a transient elevation of LH was detected in llamas treated with GnRH (P<0.05). The LH concentration returned to basal levels (pretreatment) 8, 12, and 13h after NGF treatment in llamas treated with NGF-Sal, NGF-GnRH, and NGF-NGF, respectively. We conclude that the lack of LH response to the second dose of NGF is not because of pituitary depletion but rather due to diminished GnRH. The latter may be attributed to either a downregulation of NGF receptors within the hypothalamus or to temporary depletion of GnRH at the nerve terminals within the median eminence. Research was supported by the Natural Sciences and Engineering Research Council of Canada.

175 Use of equine chorionic gonadotrophin in a minimum-handling protocol for oocyte collection in bison

Wood bison (Bison bison athabascae) and plains bison (Bison bison bison) are threatened subspecies native to North America. The creation of a germplasm biobank will connect valuable and inaccessible genetics from geographically distant herds in a biosecure manner. Protocols that are feasible in field conditions are required for cumulus-oocyte complex (COC) collection for the purpose of invitro embryo production (IVP). The efficacy of a single dose of equine chorionic gonadotrophin (ECG) was tested in an effort to develop a minimum-handling ovarian superstimulation protocol for bison. The experimental design enabled comparison between ECG-treated and non-superstimulated bison. Transvaginal ultrasound-guided follicle ablation was performed in mature wood bison (n=24) during May (anovulatory season) to induce follicular wave emergence the following day. Immediately after ablation, the bison were assigned to one of three groups (n=8 per group) and treated intramuscularly with 5000IU of ECG (Folligon, Merck), 2500IU of ECG, or saline (control). Transvaginal COC collection was performed 5 days later. Follicular and COC data were recorded, and only grade 1 and 2 COC were used for IVP. The COC were matured invitro for 25-28h at 38.8°C, fertilised (2×106 spermmL−1), and co-incubated at 38.8°C in 5% O2, 5% CO2, and 90% N2 for 18h. Presumptive zygotes were denuded and cultured at 38.8°C in 5% O2, 5% CO2, and 90% N2. Nominal data were compared among groups using analysis of variance, and proportional data were compared using GLIMMIX. The total number of follicles ≥3mm on the day of COC collection was greater in the 5000-IU ECG group than in the 2500-IU ECG and control groups (37.5±6.9, 17.5±2.0, and 16.9±2.0, respectively; P<0.005). The number of follicles 5-8mm was also greater in the 5000-IU ECG group than in the 2500-IU ECG and control groups (12.5±2.1, 7.6±1.0, and 5.8±0.9, respectively; P<0.01), as was the number of follicles >8mm (21.1±5.4, 3.3±1.2, and 0.9±0.2, respectively; P<0.0005). The proportion of grade 1 and 2 COC/total COC recovered was greater in the 5000-IU ECG group than in the 2500-IU ECG and control groups (84/124 (68%), 39/76 (51%), and 37/75 (49%), respectively; P<0.05). The proportion of cleaved zygotes/number of COC matured, assessed at 56h after fertilisation, was similar among the 5000-IU ECG, 2500-IU ECG, and control groups (42/84 (50%), 27/39 (69%), and 21/37 (57%), respectively; P=0.20). The proportion of embryos of IETS grades 1-3/number of COC matured was also similar among the 5000-IU ECG, 2500-IU ECG, and control groups (15/84 (17.9%), 8/39 (20.5%), and 7/37 (18.9%), respectively; P=0.94), but the bison in the 5000-IU ECG group produced twice as many embryos than those in the other groups. In summary, a single dose of 5000IU of ECG increased the number and size of follicles available for COC collection, more than doubled the number of COC collected for IVP, and resulted in the production of more embryos than the other groups. A single dose of 5000IU of ECG is effective in a minimum-handling protocol for ovarian superstimulation and IVP in bison. This research was supported by NSERC.

Rapid Detection of Powassan Virus in a Patient With Encephalitis by Metagenomic Sequencing

We describe a patient with severe and progressive encephalitis of unknown etiology. We performed rapid metagenomic sequencing from cerebrospinal fluid and identified Powassan virus, an emerging tick-borne flavivirus that has been increasingly detected in the United States.

868. Prospective Pathogen Detection in Patients With Central Nervous System Inflammation Using Metagenomic Sequencing

Background

Metagenomic sequencing can identify pathogens in patients with central nervous system (CNS) inflammation, who often have no diagnosis achieved despite extensive clinical testing.

Methods

This prospective study enrolled patients with CNS inflammation at a tertiary hospital from 2016 to 2017. Total nucleic acid was extracted from cerebrospinal fluid (CSF). Libraries were constructed by random primer cDNA synthesis from RNA, and Nextera XT preparation from both cDNA and DNA. Sequencing was performed on an Illumina platform. Reads from human and environmental contaminants were removed. Metagenomic analysis was performed with Kraken and confirmed with viral-ngs. The Institutional Review Board approved the study, and informed consent was obtained.

Results

Of 68 subjects enrolled, 63% were men and 84% were white. The median age was 58 years. The median CSF pleocytosis was 80 cells/mm³ [IQR 17–132]. A median of 2.4 million RNA and 6.8 million DNA sequencing reads were generated per sample. Twenty-five subjects had no diagnosis achieved by routine clinical testing; metagenomic sequencing identified enterovirus in 2 of these subjects, and no pathogen in 23. Thirty-six subjects were clinically diagnosed with an infection. In 12 of these, pathogen nucleic acid was detected in CSF by clinical polymerase chain reaction (PCR); metagenomic sequencing detected the expected pathogen in 10 subjects (83%). The other 24 subjects were clinically diagnosed with infection by serology or PCR from blood. Among these, metagenomic sequencing detected the CSF presence of HIV and locally important tick-borne pathogens Powassan virus, Borrelia burgdorferi, and Anaplasma phagocytophilum. Four subjects with West Nile Virus (WNV) infection did not have WNV RNA detected in CSF by sequencing or clinical PCR testing. Among 7 subjects diagnosed with malignancy or autoimmune disease, no pathogens were detected by metagenomic sequencing.

Conclusion

When applied broadly to patients with CNS inflammation, metagenomic sequencing identified known and unexpected pathogens in CSF, including emerging tick-borne pathogens, highlighting its potential as a diagnostic tool. Patients in whom no pathogen nucleic acid was detected could have had an infection with low pathogen burden or short duration in CSF, or a noninfectious syndrome.

Disclosures

All authors: No reported disclosures.

202 ANTRAL FOLLICULAR COUNTS AND SUPERSTIMULATORY RESPONSE IN PREPUBERTAL CALVES

The number of follicles recruited in successive waves are consistent in postpubertal cattle (Singh et al. 2004 Theriogenology 62, 227), but ovarian response to gonadotropin stimulation is highly variable among animals. We tested the hypotheses that the number of follicles present at wave emergence are repeatable and are predictive of superstimulatory response in prepubertal calves; therefore, we expected that calves with higher antral follicular counts (AFC; follicles = 1mm) at wave emergence will result in a greater number of follicles available for oocyte collection after a conventional 4-day gonadotropin treatment. Hereford crossbreed calves (n = 52; 5.1 to 6.8 months of age) were ranked according to the number of follicles = 1 mm detected in transrectal ultrasound video recordings of both ovaries at the time of wave emergence (First AFC; range: 12 to 53 follicles). Calves in the bottom (Low AFC; <20 follicles; n = 6) and top (High AFC; >32 follicles; n = 5) quartiles were selected for ovarian superstimulation. Emergence of a new follicular wave (Day 0) was induced by transvaginal follicle ablation (14 to 57 days after first AFC; 5.7 to 7.1 months of age), AFC were performed again (Second AFC), and calves were given eight 12-hourly IM injections of 25 mg of pFSH (Folltropin-V®, Bioniche Animal Health Inc., Belleville, Canada) beginning on Day 0.5. All calves were given 12.5 mg of pLH (Lutropin-V®, Bioniche Animal Health Inc.) IM 12 h after the last FSH and number of follicles equal to 3, 3 to 5, 6 to 8, and equal to 6 and 9 were counted 24 h after LH treatment (at the time of oocyte collection). A t-test was used to compare the number of follicles and ovarian response (Low v. High AFC). Values of Pearson (0.8; P < 0.001) and Spearman (0.9; P < 0.001) correlation coefficients between First and Second AFC indicate strong repeatability of numbers of follicles present at the time of wave emergence. As expected, mean number of follicles were greater (P = 0.01) in the High- than Low-AFC group (24.2 ± 2.0 v. 15.7 ± 1.0) at the Second AFC. The High-AFC group had a greater number of follicles at oocyte collection than Low AFC for 6 to 8 mm (13.4 ± 2.1 v. 5.3 ± 1.7; P = 0.01), but not for 3 to 5 mm (9.4 ± 2.5 v. 5.3 ± 2.1; P = 0.2) or 9 mm (7.6 ± 2.9 v. 4.8 ± 2.0; P = 0.4) size categories. However, High AFC resulted in a greater total number of follicles 3 (30.4 ± 3.1 v. 15.5 ± 3.2; P = 0.009) and 6 mm (21.0 ± 4.1 v. 10.2 ± 2.9; P = 0.05). The number of 6-mm follicles at the end of superstimulation represented 80 and 60% of 1-mm follicles at wave emergence in the High- and Low-AFC groups (P = 0.3). In conclusion, the number of follicles at the beginning of a wave are predictive of follicles recruited into subsequent waves in 7-month-old calves, and higher AFC at wave emergence resulted in a greater number of 3- and 6-mm follicles available for oocyte collection. Research was supported by the Natural Science and Engineering Council of Canada (NSERC). Hormones provided by Vetoquinol Inc. ARTK funded by CNPq-Science Without Borders, Brazil.

121 THREE-DIMENSIONAL ASSESSMENT OF EARLY CORPUS LUTEUM VASCULARITY IN BUFFALO (BUBALUS BUBALIS)

The aim of the study was to develop an objective method to assess the vascular flow to the early corpus luteum (CL) in buffaloes using colour Doppler ultrasound data. Our hypothesis was that 3-dimensional (3D) volumetric analysis of vascularity would demonstrate lower variability between animals compared with conventional 2-dimensional (2D) analysis of single images. Wave emergence and ovulation was synchronized in buffalo (n = 16) using prostaglandin-GnRH based protocols. Colour Doppler ultrasonography (MyLab5, 7.5-MHz linear array, colour gain 65%) was performed daily from Day −2 to 4 (Day 0 = ovulation). Video clips of the ovaries (20 s at 18–28 frames per second, AVI) were recorded by slow and uniform free-hand movement of the transducer. Day 4 CL was used for analysis of vascular area and volume. For 2D vascularity assessment, 3 images (800 × 652 pixels, RGB, BMP) of each CL (at maximum apparent vascularity) were acquired through the clip image function on the ultrasound machine and analysed by ImageJ (Fiji) software (NIH, Bethesda, MD, USA). For 3D vascularity assessment, a portion of the video clip encompassing an entire ovary was identified and exported as a series of 2D TIFF images using Videomach software. The ultrasound scale bar was used to calculate the number of pixels per millimetre and to calibrate the X (horizontal) and Y (vertical) dimensions. For 2D analyses, the CL boundary was drawn using the free-hand manual selection tool in Fiji, the area of the CL (mm2) was recorded, and the border was then enlarged by 1.5 mm to include the peripheral vascular region of the CL. The colour threshold was adjusted to select the vascular region. The 2D vascularity score was calculated as the ratio of the coloured area to the enlarged luteal area. For 3D volumetric analyses, each series of TIFF images was imported as an image sequence in Fiji and colour thresholding (similar to 2D analysis) was applied to save a second TIFF series containing luteal vascular regions (coloured areas) only. The remaining volumetric analyses were completed in Imaris software using the ovarian volume (original TIFF series) and luteal vascular volume (second TIFF series) as separate channels. The Z-dimension thickness of each image was estimated by using the dimensions of a follicle within the same ovary (Z-axis diameter = mean diameter along X- and Y-axes). Similar to 2D analyses, the volume of the CL was obtained by drawing a border along the edge of the CL, the CL border was enlarged by 1.5 mm, and a 3D vascularity score was obtained by building a surface on the luteal vascular image and calculating the vascular to luteal volume ratio. The 2D vascularity score differed from 3D vascularity score (0.21 ± 0.02 v. 0.13 ± 0.02, paired t-test P < 0.01); however, variance did not differ (Bartlett’s test P = 0.32). Our initial results support the notion that the described technique of quantifying vascular volume of the corpus luteum may decrease the technical variability during image assessment and therefore better reflect the true vascularity compared with 2D image analyses. Research was supported by a grant from the Natural Sciences and Engineering Research Council of Canada.

Infection control link professionals' knowledge of Clostridium difficile

lostridium difficile is a common cause of diarrhoea in hospitalised patients. It can result in longer hospital stays and due to the need for strict isolation procedures can add significantly to nursing workload. Additionally, it can be very distressing for the patient and if patients are vulnerable to infection it can have serious health consequences. Cross-transmission can be limited by good infection prevention and control practices, however this relies on a sound knowledge base and support from the infection control team. This small-scale study reports on infection control link professionals' knowledge and assesses how they would utilise this knowledge in practice situations. Findings imply that the knowledge base concerning the microorganism was poor, but knowledge relating to general infection procedures was good.

The micro-architecture of human cancellous bone from fracture neck of femur patients in relation to the structural integrity and fracture toughness of the tissue

Osteoporosis is clinically assessed from bone mineral density measurements using dual energy X-ray absorption (DXA). However, these measurements do not always provide an accurate fracture prediction, arguably because DXA does not grapple with ‘bone quality’, which is a combined result of microarchitecture, texture, bone tissue properties, past loading history, material chemistry and bone physiology in reaction to disease. Studies addressing bone quality are comparatively few if one considers the potential importance of this factor. They suffer due to low number of human osteoporotic specimens, use of animal proxies and/or the lack of differentiation between confounding parameters such as gender and state of diseased bone. The present study considers bone samples donated from patients (n = 37) who suffered a femoral neck fracture and in this very well defined cohort we have produced in previous work fracture toughness measurements (FT) which quantify its ability to resist crack growth which reflects directly the structural integrity of the cancellous bone tissue. We investigated correlations between BV/TV and other microarchitectural parameters; we examined effects that may suggest differences in bone remodelling between males and females and compared the relationships with the FT properties. The data crucially has shown that TbTh, TbSp, SMI and TbN may provide a proxy or surrogate for BV/TV. Correlations between FT critical stress intensity values and microarchitecture parameters (BV/TV, BS/TV, TbN, BS/BV and SMI) for osteoporotic cancellous tissue were observed and are for the first time reported in this study. Overall, this study has not only highlighted that the fracture model based upon BMD could potentially be improved with inclusion of other microarchitecture parameters, but has also given us clear clues as to which of them are more influential in this role.

•

first time ever study to relate microarchitecture to the fracture toughness of cancellous bone from the femoral head of FNF victims

•

reduction in bone mass relates to a reduction in the number of trabeculae and trabecular thickness and an increase in trabeculae spacing

•

bone loss observed appears to be a consequence of thinning of the trabeculae in males and perforation of the trabeculae in females

•

study hints that TbTh, TbSp, SMI and TbN may provide a proxy or surrogate for BV/TV

•

fracture models can be improved by including microarchitecture, BMD and the bone mineral quality of osteoporotic cancellous bone

287 EFFECT OF FOLLICULAR AGING ON THE ATP CONTENT AND DISTRIBUTION OF MITOCHONDRIA IN BOVINE OOCYTES

Our objective was to determine how follicular aging affects the distribution and content of mitochondrial population and ATP in in vivo-matured bovine oocytes. We hypothesised that in vivo-matured bovine oocytes obtained from aged follicles (84 h of gonadotropin starvation) have altered mitochondrial distribution and decreased cytoplasmic ATP content compared to those obtained immediately at the end of a superstimulatory protocol (no starvation). Follicular waves were synchronized by ablation 5 to 8 d after ovulation and a CIDR device was given. Starting on the day of wave emergence (Day 0), short FSH and FSH starvation groups (n = 5 heifers each) were given 8 doses of FSH im over 4 d and the long FSH group (n = 4) was given 14 doses over 7 d. Two doses of PGF were given on Day 4 (short FSH) or Day 7 (FSH starvation and long FSH groups), the CIDR was removed, and LH was given 24 h after second PGF treatment. The ovaries were removed 24 h later by colpotomy and cumulus-oocyte-complexes (COC) were collected from follicles ≥8 mm. Denuded oocytes were either stained with Mitotracker Deep Red FM and imaged by confocal microscopy or processed for ATP assay. Mitochondria numbers were assessed by segmentation of 3D datasets. Proportions of COC within each grade were compared using Fischer's exact test, and ATP and mitochondrial data were compared by analysis of variance. Short and long FSH groups had a greater proportion of Grade 1 expanded COC than the FSH starvation group (P = 0.02). The ATP content of oocytes (from expanded COC) tended to be higher in the long FSH group than short FSH (P = 0.09), and the FSH starvation group was intermediate. The ATP content of oocytes from compact COC did not differ among groups (P = 0.49). The proportion of mitochondrial clusters was highest (P = 0.01) and the proportion of individual mitochondria was lowest (P = 0.01) in the FSH starvation group compared to short and long FSH groups. Mitochondria from the long FSH and FSH starvation groups had twice the relative intensity compared to the short FSH group (P < 0.01). In conclusion, follicular aging (FSH starvation) was associated with a decrease in oocyte morphologic grade and marked clustering of mitochondria, which may be a reflection of oxidative stress and atresia.