Benchmark of cellular deconvolution methods using a multi-assay reference dataset from postmortem human prefrontal cortex

Background

Cellular deconvolution of bulk RNA-sequencing (RNA-seq) data using single cell or nuclei RNA-seq (sc/snRNA-seq) reference data is an important strategy for estimating cell type composition in heterogeneous tissues, such as human brain. Computational methods for deconvolution have been developed and benchmarked against simulated data, pseudobulked sc/snRNA-seq data, or immunohistochemistry reference data. A major limitation in developing improved deconvolution algorithms has been the lack of integrated datasets with orthogonal measurements of gene expression and estimates of cell type proportions on the same tissue sample. Deconvolution algorithm performance has not yet been evaluated across different RNA extraction methods (cytosolic, nuclear, or whole cell RNA), different library preparation types (mRNA enrichment vs. ribosomal RNA depletion), or with matched single cell reference datasets.

Results

A rich multi-assay dataset was generated in postmortem human dorsolateral prefrontal cortex (DLPFC) from 22 tissue blocks. Assays included spatially-resolved transcriptomics, snRNA-seq, bulk RNA-seq (across six library/extraction RNA-seq combinations), and RNAScope/Immunofluorescence (RNAScope/IF) for six broad cell types. The Mean Ratio method, implemented in the DeconvoBuddies R package, was developed for selecting cell type marker genes. Six computational deconvolution algorithms were evaluated in DLPFC and predicted cell type proportions were compared to orthogonal RNAScope/IF measurements.

Conclusions

Bisque and hspe were the most accurate methods, were robust to differences in RNA library types and extractions. This multi-assay dataset showed that cell size differences, marker genes differentially quantified across RNA libraries, and cell composition variability in reference snRNA-seq impact the accuracy of current deconvolution methods.

Transcriptomic analysis of the human habenula in schizophrenia

Importance

Habenula (Hb) pathophysiology is involved in many neuropsychiatric disorders, including schizophrenia. Deep brain stimulation and pharmacological targeting of the Hb are emerging as promising therapeutic treatments. However, little is known about the cell type-specific transcriptomic organization of the human Hb or how it is altered in schizophrenia.

Objective

To define the molecular neuroanatomy of the human habenula and identify transcriptomic changes in individuals with schizophrenia compared to neurotypical controls.

Design Setting and Participants

This study utilized Hb-enriched postmortem human brain tissue. Single nucleus RNA-sequencing (snRNA-seq) and single molecule fluorescent in situ hybridization (smFISH) experiments were conducted to identify molecularly defined Hb cell types and map their spatial location (n=3-7 donors). Bulk RNA-sequencing and cell type deconvolution were used to investigate transcriptomic changes in Hb-enriched tissue from 35 individuals with schizophrenia and 33 neurotypical controls. Gene expression changes associated with schizophrenia in the Hb were compared to those previously identified in the dorsolateral prefrontal cortex (DLPFC), hippocampus, and caudate.

Main Outcomes and Measures

Semi-supervised snRNA-seq cell type clustering. Transcript visualization and quantification of smFISH probes. Bulk RNA-seq cell type deconvolution using reference snRNA-seq data. Schizophrenia-associated gene differential expression analysis adjusting for Hb and thalamus fractions, RNA degradation-associated quality surrogate variables, and other covariates. Cross-brain region schizophrenia-associated gene expression comparison.

Results

snRNA-seq identified 17 cell type clusters across 16,437 nuclei, including 3 medial and 7 lateral Hb populations. Cell types were conserved with those identified in a rodent model. smFISH for cell type marker genes validated snRNA-seq Hb cell types and depicted the spatial organization of subpopulations. Bulk RNA-seq analyses yielded 45 schizophrenia-associated differentially expressed genes (FDR < 0.05), with 32 (71%) unique to Hb-enriched tissue.

Conclusions

These results identify topographically organized cell types with distinct molecular signatures in the human Hb. They further demonstrate unique transcriptomic changes in the epithalamus associated with schizophrenia, thereby providing molecular insights into the role of Hb in neuropsychiatric disorders.

The gene expression landscape of the human locus coeruleus revealed by single-nucleus and spatially-resolved transcriptomics

Norepinephrine (NE) neurons in the locus coeruleus (LC) make long-range projections throughout the central nervous system, playing critical roles in arousal and mood, as well as various components of cognition including attention, learning, and memory. The LC-NE system is also implicated in multiple neurological and neuropsychiatric disorders. Importantly, LC-NE neurons are highly sensitive to degeneration in both Alzheimer's and Parkinson's disease. Despite the clinical importance of the brain region and the prominent role of LC-NE neurons in a variety of brain and behavioral functions, a detailed molecular characterization of the LC is lacking. Here, we used a combination of spatially-resolved transcriptomics and single-nucleus RNA-sequencing to characterize the molecular landscape of the LC region and the transcriptomic profile of LC-NE neurons in the human brain. We provide a freely accessible resource of these data in web-accessible and downloadable formats.

Data-driven identification of total RNA expression genes for estimation of RNA abundance in heterogeneous cell types highlighted in brain tissue

We define and identify a new class of control genes for next-generation sequencing called total RNA expression genes (TREGs), which correlate with total RNA abundance in cell types of different sizes and transcriptional activity. We provide a data-driven method to identify TREGs from single-cell RNA sequencing data, allowing the estimation of total amount of RNA when restricted to quantifying a limited number of genes. We demonstrate our method in postmortem human brain using multiplex single-molecule fluorescent in situ hybridization and compare candidate TREGs against classic housekeeping genes. We identify AKT3 as a top TREG across five brain regions.

Integrated single cell and unsupervised spatial transcriptomic analysis defines molecular anatomy of the human dorsolateral prefrontal cortex

Generation of a molecular neuroanatomical map of the human prefrontal cortex reveals novel spatial domains and cell-cell interactions relevant for psychiatric disease. The molecular organization of the human neocortex has been historically studied in the context of its histological layers. However, emerging spatial transcriptomic technologies have enabled unbiased identification of transcriptionally-defined spatial domains that move beyond classic cytoarchitecture. Here we used the Visium spatial gene expression platform to generate a data-driven molecular neuroanatomical atlas across the anterior-posterior axis of the human dorsolateral prefrontal cortex (DLPFC). Integration with paired single nucleus RNA-sequencing data revealed distinct cell type compositions and cell-cell interactions across spatial domains. Using PsychENCODE and publicly available data, we map the enrichment of cell types and genes associated with neuropsychiatric disorders to discrete spatial domains. Finally, we provide resources for the scientific community to explore these integrated spatial and single cell datasets at research.libd.org/spatialDLPFC/.

Circulating tumour cells demonstrate an altered response to hypoxia and an aggressive phenotype

Background:

Tumours contain hypoxic regions that select for an aggressive cell phenotype; tumour hypoxia induces metastasis-associated genes. Treatment refractory patients with metastatic cancer show increased numbers of circulating tumour cells (CTCs), which are also associated with disease progression. The aim of this study was to examine the as yet unknown relationship between hypoxia and CTCs.

Methods:

We generated human MDA-MB-231 orthotopic xenografts and, using a new technology, isolated viable human CTCs from murine blood. The CTCs and parental MDA-MB-231 cells were incubated at 21 and 0.2% (hypoxia) oxygen, respectively. Colony formation was assayed and levels of hypoxia- and anoxia-inducible factors were measured. Xenografts generated from CTCs and parental cells were compared.

Results:

MDA-MB-231 xenografts used to generate CTCs were hypoxic, expressing hypoxia factors: hypoxia-inducible factor1 alpha (HIF1α) and glucose transporter protein type 1 (GLUT1), and anoxia-induced factors: activating transcription factor 3 and 4 (ATF3 and ATF4). Parental MDA-MB-231 cells induced ATF3 in hypoxia, whereas CTCs expressed it constitutively. Asparagine synthetase (ASNS) expression was also higher in CTCs. Hypoxia induced ATF4 and the HIF1α target gene apelin in CTCs, but not in parental cells. Hypoxia induced lower levels of carbonic anhydrase IX (CAIX), GLUT1 and BCL2/adenovirus E1B 19-KD protein-interacting protein 3 (BNIP3) proteins in CTCs than in parental cells, supporting an altered hypoxia response. In chronic hypoxia, CTCs demonstrated greater colony formation than parental cells. Xenografts generated from CTCs were larger and heavier, and metastasised faster than MDA-MB-231 xenografts.

Conclusion:

CTCs show an altered hypoxia response and an enhanced aggressive phenotype in vitro and in vivo.

Discovery of molecular subtypes in leiomyosarcoma through integrative molecular profiling

Leiomyosarcoma (LMS) is a soft tissue tumor with a significant degree of morphologic and molecular heterogeneity. We employed integrative molecular profiling to discover and characterize molecular subtypes of LMS. Gene expression profiling was performed on 51 LMS samples. Unsupervised clustering demonstrated 3 reproducible LMS clusters. Array comparative genomic hybridization (aCGH) was performed on 20 LMS samples and demonstrated that the molecular subtypes defined by gene-expression showed distinct genomic changes. Tumors from the “muscle-enriched” cluster showed significantly increased copy number changes (p=0.04). Most muscle-enriched cases showed loss at 16q24 which contains FANCA, known to play an important role in DNA repair, and loss at 1p36 which contains PRDM16, whose loss promotes muscle differentiation. Immunohistochemistry was performed on LMS tissue microarrays (n=377) for five markers with high levels of mRNA in the muscle-enriched cluster (ACTG2, CASQ2, SLMAP,CFL2, MYLK) and demonstrated significantly correlated expression of the 5 proteins (all pairwise p < 0.005). Expression of the 5 markers was associated with improved disease-specific survival (DSS) in a multivariate Cox regression analysis (p < 0.04). In this analysis that combined gene expression profiling, aCGH and immunohistochemistry, we characterized distinct molecular LMS subtypes, provided insight into their pathogenesis, and identified prognostic biomarkers.

Plexiform fibrohistiocytic tumor with a clonal cytogenetic anomaly

Plexiform fibrohistiocytic tumors are rare lesions of proposed myofibroblastic origin occurring primarily in infants and children. There is a characteristic biphasic histology comprised of both fibroblastic and histiocyte-like components. These tumors tend to be locally aggressive with prognosis dependent on completeness of resection. A previous cytogenetic case report of this tumor described a stemline clone with a karyotype of 46,XY,-6,-8, del(4)(q25q31),del(20)(q11.2),+der(8)t(8;?) (p22;?),+mar. We report a different cytogenetic finding in another plexiform fibrohistiocytic tumor which demonstrated a simpler karyotype of 46,XY,t(4;15)(q21;q15). The implications of cytogenetic heterogeneity in fibroblastic tumors is briefly discussed.

Clinical and pathologic findings in hemochromatosis hip arthropathy

During a 9-year period, 15 patients with hemochromatosis hip arthropathy required 19 total hip arthroplasties for disabling hip pain. Preoperative presentation, hip function, pathologic evaluation of the femoral head, and radiographic findings were reviewed. Postoperative followup averaging 5.7 years (range, 2-11 years) was performed to assess hip pain and function after total hip arthroplasty. The average preoperative Hospital for Special Surgery hip score was 15 points (range, 4-24 points), and this improved to 30 points (range, 4-38 points) after total hip arthroplasty. Only one of 15 patients required revision surgery at 10 years for acetabular loosening. All other patients were pain free, with improved function at latest followup. Histologic evaluation of the resected femoral heads revealed evidence of primary or secondary osteonecrosis in seven of 19 (37%) specimens. Articular cartilage avulsion at the level of the tidemark was identified in eight of 19 (42%) specimens, and calcium pyrophosphate deposition was identified in five of 19 (26%) specimens. These pathologic findings suggest a predictable progression of the arthritic process in patients with hemochromatosis.

Practical management of venous thromboembolism following pelvic fractures

The management of thromboembolic complications remains one of the most controversial issues in the care of patients with pelvic and acetabular fractures. Recent studies have indicated that the incidence of proximal deep vein thrombosis is much higher than was previously believed. These patients should be managed with a formal institutional protocol that includes universal prophylaxis, supplemented in some cases by screening for deep vein thrombosis.

The detection and management of proximal deep venous thrombosis in patients with acute acetabular fractures: a follow-up report

OBJECTIVE

To determine the efficacy of a thromboprophylaxis protocol that included deep venous thrombosis (DVT) prophylaxis (subcutaneous heparin), preoperative screening with magnetic resonance venography (MRV), and therapeutic management (vena caval interruption preoperatively, therapeutic heparin anticoagulation postoperatively) when indicated.

DESIGN

Prospective, consecutive.

SETTING

Tertiary referral, teaching hospital in New York City.

PATIENTS

One hundred one patients with acutely displaced acetabular fractures.

MAIN OUTCOME MEASURE

Preoperative MRV was performed to assess vascular structures. Patients with proximal DVT received vena caval filter interruption preoperatively and therapeutic warfarin postoperatively. Patients without proximal DVT received only subcutaneous heparin preoperatively and low-dose warfarin postoperatively.

RESULTS

Forty-nine asymptomatic thrombi were identified in thirty-four of 101 patients (34 percent). Location of thrombi were in the popliteal vein in four of forty-nine patients (8 percent), superficial femoral vein in eight of forty-nine (16 percent), common femoral vein in thirteen of forty-nine (27 percent), external iliac vein in six of forty-nine (12 percent), internal iliac vein in fourteen of forty-nine (29 percent), and common iliac vein in four of forty-nine (8 percent). Thrombi were isolated to the injured extremity in twenty-six of thirty-four patients (76 percent), bilateral in four of thirty-four (12 percent), and isolated to the uninjured extremity in four of thirty-four (12 percent). Twenty-six of the thirty-four patients with proximal thrombi received preoperative vena caval filters. As a result of this protocol, only one patient (1 percent) developed a nonfatal pulmonary embolism.

CONCLUSION

MRV is a sensitive screening examination that allows the placement of inferior vena caval filters based on documented proximal thrombosis. We anticipate that preoperative DVT screening with MRV will significantly decrease the incidence of fatal pulmonary embolism in this high-risk population.

Thromboembolic complications in patients with pelvic trauma

Patients with pelvic trauma are known to be at increased risk for the development of thromboembolic complications. The incidence of deep venous thrombosis in patients with pelvic fractures is 35% to 60%. Proximal deep venous thrombosis, which is most likely to result in pulmonary embolism, occurs in 25% to 35% of these patients, and almost 1/2 of all proximal thrombi will be in the pelvic veins. The incidence of symptomatic pulmonary embolism in the pelvic trauma population is 2% to 10% whereas a greater proportion of patients will have clinically silent pulmonary embolism. Fatal pulmonary embolism occurs in 0.5% to 2% of patients with pelvic trauma. The cornerstone of effective management is prophylaxis and the most commonly used forms include low dose heparin, low molecular weight heparin, mechanical devices, and in some studies, inferior vena caval filters. Based on a critical review of the literature, in algorithm is proposed for the management of thromboprophylaxis in this trauma subgroup. This includes prophylaxis, screening, and treatment when proximal thrombosis is identified. Such a systematic approach to this potentially catastrophic problem may decrease the morbidity and mortality associated with thromboembolic complications in these patients.

Magnetic resonance venography to evaluate the deep venous system of the pelvis in patients who have an acetabular fracture

We performed a prospective, blinded study to assess and compare the values of preoperative contrast venography and magnetic resonance venography in the detection of deep venous thrombosis in the thigh and pelvis of forty-five consecutive patients who had a displaced acetabular fracture. The magnetic resonance venography and contrast venography were performed an average of seven days (range, one to twenty-nine days) after the injury. Twenty-four asymptomatic thrombi were identified with magnetic resonance venography in fifteen (33 percent) of the patients. Four of the thrombi were in the superficial femoral vein, nine were in the common femoral vein, one was in the external iliac vein, seven were in the internal iliac vein, and three were in the common iliac vein. Ten (42 percent) of the twenty-four thrombi were confirmed with contrast venography; nine of them were located in the thigh. The remaining fourteen thrombi (58 percent) that had been noted on magnetic resonance venography could not be seen with contrast venography because they were located either in the deep pelvic veins or in the uninjured extremity. The thrombi in the internal iliac vein were identified only with magnetic resonance venography. Twelve of the fifteen patients who had thrombi had a filter placed in the inferior vena cava preoperatively. In eight of these patients, the filter was placed because of the findings of magnetic resonance venography alone. Magnetic resonance venography resulted in a change in the therapeutic management of ten (22 per cent) of the forty-five patients. There were no pulmonary emboli. We concluded that magnetic resonance venography is superior to contrast venography for the preoperative evaluation of proximal deep venous thrombosis in patients who have an acetabular fracture. Magnetic resonance venography is non-invasive, does not require the use of contrast medium, images the proximal aspects of both lower extremities simultaneously, and, most importantly, allows for the identification of deep venous thrombosis in the pelvis.

Magnetic resonance imaging of the pelvis. New orthopaedic applications

A preliminary study of using magnetic resonance angiography to detect occult proximal thrombi in patients who had hip arthroplasty was done. Despite the presence of susceptibility artifact caused by metallic components, diagnostic visualization of thigh vessels was made in a preliminary series of 15 patients. Confirmation of all previously documented (by contrast venogram via dorsal foot vein cannulation or Doppler study) proximal thrombi was made in all 15 patients. One patient had a thrombus in the contralateral extremity that had been undetected by Doppler study; 4 additional pelvic thrombi occurred in 3 patients, which had been undocumented previously. Because magnetic resonance angiography is noninvasive, requiring no contrast agent, it has advantages over conventional venography to detect occult proximal thrombi. New fast spin echo sequences are discussed that enhance visualization of regional anatomic structures adjacent to metallic prosthetic components. Emphasis was placed on assessing the posterior soft tissue envelope in patients having recurrent dislocations after total hip arthroplasty, despite acceptable component alignment. Preliminary results show a consistent absence of a posterior pseudocapsule in patients having dislocations, as compared with control patients having no dislocations.

Genetic instability of chromosome 3 in HPV-immortalized and tumorigenic human keratinocytes

The HPV-1811 cell line is derived from primary human foreskin keratinocytes that have been transfected with human papilloma virus type 18. At late passage, these cells produce invasive squamous cell carcinomas when injected into nude mice. A striking, but unstable, aberration of chromosome 3 occurs very early after establishment of the culture; a consistent rearrangement is observed concomitant with tumorigenicity. Using molecular cytogenetic techniques, we characterized the complex development of this aberration. A whole chromosome probe to this chromosome was made by linker-adapter PCR amplification of a single flow-sorted chromosome. Hybridization of this probe to normal metaphase chromosomes revealed the der (3) to be composed of chromosome 3, distal 13q, and 21q22. Hybridization of a 3q subtelomeric probe and a glycoprotein V probe which maps to 3qter indicated that this locus is duplicated in the final form of the chromosome, but that much instability occurs prior to its establishment. The ETS2 oncogene, which maps to 21q22, is translocated to the der(3) when the cell line becomes tumorigenic, but not prior to this time. Early-passage cells which have been induced to become tumorigenic by exposure to the carcinogen nitrosomethylurea also have the localization of the ETS2 at 3qter.

MR imaging of acetabular fractures: value in detecting femoral head injury, intraarticular fragments, and sciatic nerve injury

OBJECTIVE

The purpose of this prospective study was to compare the value of MR imaging in the detection of injuries associated with acute acetabular fractures (femoral head fracture, free fragments within the hip joint, and injury to the sciatic nerve) with the value of preoperative CT examinations, intraoperative inspection, intraoperative somatosensory evoked potentials (SEP), and clinical neurologic examinations.

SUBJECTS AND METHODS

Coronal fat suppressed long TR/TE and unenhanced and contrast-enhanced T1-weighted MR images were obtained preoperatively in 37 patients with acetabular fractures. The sciatic nerve was assessed for injury and the femoral head was assessed for fracture, dislocation, and contusion. MR results were compared with CT findings for acetabular fractures and fractures of the femoral head. The appearance of the sciatic nerve on MR images was correlated with intraoperative changes in SEP and results of the clinical neurologic examination.

RESULTS

Although MR images showed acetabular fractures, intraarticular fragments were often obscured. Fragments were readily apparent on CT scans. MR images showed fracture of the femoral head in 10 (27%) of 37 cases. Nine of these fractures also were seen on CT scans. MR images showed subchondral contusion of the femoral head in 24 (65%) of 37 cases. The same areas appeared normal on CT scans. MR images of the sciatic nerve obtained after injection of contrast material showed intraneural or perineural enhancement in all patients with either changes in baseline SEP (n = 19) or preoperative neurologic deficit (n = 10). Although baseline changes in SEP were more common with intraneural enhancement, the difference in the prevalence of neurologic deficits was not significant. The preoperative enhancement pattern alone could not be used to predict a neurologic deficit.

CONCLUSIONS

MR imaging of acetabular fractures can be used to detect subclinical injury of the sciatic nerve and occult injuries of the femoral head not readily apparent on CT scans. However, intraarticular fragments may be obscured.