Club-Like Cells in Proliferative Inflammatory Atrophy of the Prostate - Karen Sfanos

August 24, 2023

Andrea Miyahira hosts a discussion with Karen Sfanos about her group's research on Proliferative Inflammatory Atrophy (PIA) of the prostate, published in the Journal of Pathology. The paper, titled "Club-Like Cells in Proliferative Inflammatory Atrophy of the Prostate," investigates the relationship between PIA and prostate cancer, focusing on the abnormal proliferation of PIA cells resembling lung club cells, a protective type involved in immunomodulation. Dr. Sfanos presents evidence linking club cells in PIA with inflammation and their potential oncogenic transformation. She also explores the connection between the prostate microbiome and club cells and suggests a link between urinary microbes and inflammation driving the club cell phenotype. The findings may have implications for understanding prostate cancer development, identifying early cancer risk factors, and guiding future research directions.


Karen Sandell Sfanos, PhD, MS, Johns Hopkins University School of Medicine, Baltimore, MD

Andrea K. Miyahira, PhD, Director of Global Research & Scientific Communications, The Prostate Cancer Foundation

Read the Full Video Transcript

Andrea Miyahira: Hi everyone. Thank you for joining us today. I'm Dr. Andrea Miyahira at the Prostate Cancer Foundation. I'm joined by Dr. Karen Sfanos, an associate professor at Johns Hopkins University, to discuss her group's recent paper, Club-Like Cells in Proliferative Inflammatory Atrophy of the Prostate, published in the Journal of Pathology. Thanks so much for joining me today, Dr. Sfanos.

Karen Sfanos: Thank you. So I'll go ahead and share my screen. Thanks very much for the opportunity to discuss our new research. And what I'm going to be talking about today is a prostate lesion that is pictured here called proliferative inflammatory atrophy or PIA. And as the name implies, this is an inflammation associated lesion and it's commonly observed when prostate tissue is being diagnosed after radical prostatectomy for prostate cancer.

Unlike normal prostate luminal epithelium that has a very low proliferative index, the cells in PIA are actively proliferating as shown here by KI-67, immunohistochemistry.

Studies dating back to 1968 have proposed that PIA serves as a risk factor lesion for prostate cancer. What is seen here is a PIA lesion that is potentially budding into micro adenocarcinoma, which is a term for a type of early cancer.

The luminal epithelial cells in PIA are very different from normal luminal epithelial cells in the prostate. I've already mentioned that they are highly proliferative, but they also have decreased expression of the tumor suppressor p27. They have suppressed apoptosis via increased Bcl-2 expression. And importantly these cells have altered cellular differentiation, meaning they make markedly lower amounts of normal prostate proteins such as NKX3.1, androgen receptor and PSA. And instead they express very high level inflammatory genes and stress response genes.

PIA cells also express keratins that are expressed in both luminal and basal epithelial cells, and have been previously given the name intermediate cells, which I'll refer to quite a bit.

PIA is a regenerative lesion. So that means that it functions to replace damaged epithelial cells that have presumably been damaged by inflammation or other insults in the prostate.

The other term that I need to introduce is club cells. Club cells are a specialized cell type that is found in the lung, and they play a protective role in immunomodulation and pathogen defense. One of the most abundant secreted proteins of club cells, is called secretoglobin family 1A member 1 or SCGB1A1.

Club cells are regarded as a second stem cell in the lungs because they function to regenerate damaged lung epithelium. So I think you can already start to sense that I'm drawing correlations between the function of lung club cells and prostate luminal epithelial cells in PIA. But I will also point out another phenotypic similarity between lung club cells and PIA, is that they are simple cuboidal epithelium. So whereas the normal luminal epithelial cells in the prostate are these tall columnar cells, the luminal epithelial cells in PIA are shortened cuboidal epithelium that are akin to club cells in the lung.

This great study by Doug Strand's group in 2018 identified cells in the prostate that expressed signature club cell genes including SCGB1A1. And importantly this study examined normal prostate epithelium in organ donor prostates. What they found is that SCGB1A1 expressing club cells were limited to being present in the prostatic urethra in the early collecting ducts of the prostate. But in this study, club cells were not identified in the peripheral zone of the prostate where most prostate cancers arise.

In contrast to this, a study by Franklin Huang's group published in 2022 examined radical prostatectomy tissues by single cell RNA sequencing, and these were from individuals undergoing treatment for prostate cancer. And they identified this other epithelial cell population that expressed a club cell gene signature including SCGB1A1, a gene called ceruloplasmin or CP, lactoferrin or LTF, matrix metallopeptidase 7 or MMP7, and polymeric immunoglobulin receptor or PIGR. It should be noted that all of these genes function in inflammation and immune modulation, and unlike the study in organ donor prostates, these club epithelial cells in Franklin's study were present in the peripheral zone of the prostate and were adjacent to prostate cancer.

So Franklin is a wonderful collaborator of mine and he reached out to ask if we could help him spatially locate and localize these club cells in the prostate. And that is basically summarizing how this study began.

So what we did is we conducted RNA in situ hybridization for the panel of club cell genes identified Franklin's study, including CP, LTF, and MMP7 and SCGB1A1. And the top row of this image shows expression of all of these club cell genes in the prostatic urethra. So this is confirmatory of the Doug Strand paper. But what we additionally found was that the expression of these club cell genes in the peripheral zone was associated with inflammation and was specifically in the epithelial cells of PIA. So it was localized only in these PIA lesions.

And so as shown here are examples of the club cell genes expressed in PIA in benign regions of the prostate as well as PIA that's admixed with areas of prostate cancer. I think you can appreciate that lactoferrin and PIGR are uniformly expressed in most PIA lesions. But the other club cell genes, CP, MMP7, and especially SCGB1A1 are only present in a subset of the epithelial cells in PIA.

We further quantitated the expression of PIGR as a surrogate marker for club cells in both organ donor and radical prostatectomy specimens. And what we found is that organ donor prostate tissues from younger males are largely devoid of club cells in the peripheral zone with the exception of the rare instance of inflammation that is present in organ donor prostates.

Quantitatively, PIGR was more highly expressed in benign regions of prostatectomy specimens over organ donor prostate or prostate cancer. And this higher expression in benign regions versus cancer is not unexpected because PIA is more commonly located in the benign regions than admixed in there with cancer.

Interestingly, PIGR was more highly expressed in both high grade cancer and high grade benign than organ donor prostate. And likewise, the higher expression of PIGR in benign regions versus cancer was maintained in both Black and white individuals.

I mentioned previously that the luminal epithelial cells in PIA have previously been referred to as intermediate cells because they express cytokeratins that are present in both luminal and basal cells in the prostate. And we further demonstrate that PIGR positive club cells in PIA are the same cells that are expressing the basal cell cytokeratins and that have markedly lower AR expression. And then in the single cell data from Franklin's study, we also show that the club cells that express PIGR co-express additional intermediate cell markers including CMET and BCL2 in the absence of CD38.

So importantly, over the years there have been multiple studies that have identified cell populations in the prostate that exhibit both regenerative potential and progenitor-like characteristics. And these cells have been given multiple different names depending on the study, including club cells in our study, intermediate cells that I've mentioned before, CD38 low luminal cells, club-like luminal cells, luminal 2 club cells. But importantly, all of these studies are likely describing the same cell type in the prostate. And what we further demonstrate is that what these cells are, are the luminal epithelial cells in PIA.

So finally, why are prostate club cells in PIA important? So we previously published a paper where we showed that PIA lesions can on very rare occasions, contain TMPRSS2-ERG gene fusions. And we demonstrated that ERG+ PIA can be observed directly budding into micro adenocarcinoma. We examined in our study here, the club cell genes, lactoferrin and PIGR, in these cases with ERG+ PIA. And what's shown here in graphical representation is what we found; normal luminal epithelial cells in the prostate do not express club cell genes. But in PIA lesions in the setting of inflammation, the luminal epithelial cells become shortened and cuboidal and express the club cell genes.

In the very rare instances where a TMPRSS2-ERG fusion occurs in PIA, PIGR expression is actually maintained, but lactoferrin expression is absent. We previously demonstrated that lactoferrin is silenced by promoter methylation in prostate cancer. So these findings suggest that lactoferrin methylation occurs very early in the carcinogenic process. And then in early invasive cancer, PIGR is actually variably maintained, but lactoferrin is also absent.

So in conclusion, the luminal epithelial cells in PIA express club cell genes. It's an inflammation associated lesion. So we submit that this is further evidence that inflammation drives lineage plasticity in the prostate. These club-like cells are not present in the peripheral zone in normal prostates. They're enriched in prostatectomy specimens, they're associated with inflammation, and we think that this means that they could possibly be associated with prostate cancer and/or aging in the prostate.

Club cells in PIA may be more prone to oncogenic transformation. And the reason that we think that is that club cells exist in the setting of inflammation where DNA damage may occur. They're likewise clearly undergoing lineage plasticity that includes reprogramming of the normal prostate gene signatures, and as I mentioned before, these cells are also proliferating. So any oncogenic insult to these cells may be propagated in this setting.

And then finally, club cells are enriched in cases with higher grade cancer. This could be reflective of a greater degree of inflammation in benign regions adjacent to high grade cancer, and that has been demonstrated other studies. But this also could be because inflammation could in fact be driving higher grade cancer or higher risk mutations. And this last point is very much to be determined.

So with that I just want to say thank you to all of the individuals that contributed to these studies, both at Johns Hopkins and UCSF. And then a huge thank you to the Prostate Cancer Foundation who funded this work via a Challenge Award 19CHAS03. So thank you so much.

Andrea Miyahira: Thank you so much for sharing that Dr. Sfanos. So it's really interesting that it seems like there's a possible role of club cells in prostate cancer initiation. Have you evaluated them for the presence of tumor mutations?

Karen Sfanos: Yes. So that's a great question and I would say that the one mutation that we have evaluated is TMPRSS2-ERG fusions. I touched upon that in the talk. So this can rarely occur, but we demonstrate that you can have ERG+ PIA. So it could be that something about inflammation driving the club cell phenotype might make them more prone to or at risk for getting a TMPRSS2-ERG fusion. But I think that it's entirely possible that there are other mutations that are associated with prostate cancer that could occur in these lesions and it's just something that we haven't looked at yet. But that's certainly a future direction of the study.

Andrea Miyahira: For others interested in studying club cells, are there markers that you suggest as being the most important to identify them in prostate cancer specimens?

Karen Sfanos: Yes, another great question. So throughout our panel that we looked at, there were several different club cell associated genes including SCGB1A1, which is like the classical club cell marker. But as you saw, that one was actually very heterogeneous in the club cells that exist in PIA in the peripheral zone.

So I would say that the best club cell markers in PIA, in the peripheral zone that we have so far are lactoferrin and PIGR. But it also brings up an important point that there's heterogeneity to the club cells in PIA. So they're not all the same. Not all the cells are the same and that might be important but also to be determined.

Andrea Miyahira: Thank you so much. What are your conclusions about the possible role of club cells in inflammation and in prostate cancer and lineage plasticity?

Karen Sfanos: I think what we are very interested in understanding is if... It's like a chicken or the egg thing, we don't know what comes first. But we think the inflammation happens first and it is driving the normal luminal epithelial cells in the prostate to become something very different both phenotypically and genotypically. So they are turning off all of the normal prostate differentiation genes including NKX3.1, including androgen receptor. Those are going way down and they're switching over to a more... They're like these inflammation hubs, they're like dumping out inflammatory cytokines.

But is there something about that, for lack of a better term, lineage plasticity or lineage switching that actually puts these cells at risk for getting an oncogenic hit in an area that normal luminal prostate epithelial cells would not? This is very much a hypothesis at this point, but something that we're very interested in pursuing.

Andrea Miyahira: Thank you. And then your lab is also highly interested in the prostate microbiome. Are there any findings from those studies which suggest interactions between prostate club cells and the prostate microbiome?

Karen Sfanos: Absolutely. And I appreciate the question that touches on the microbiome, which is of course another major focus of the lab.

So that prior study that I mentioned where we found ERG+ PIA was actually a study where we were looking at prostates that were infected by uropathogenic strains of E. coli that exist in the urinary tract. They're kind of hanging out in the urinary tract, but under certain situations might be able to get into the prostate where they shouldn't be and cause prostate infections. So I've been talking in my whole talk about inflammation, inflammation, inflammation, but you might be questioning, well, where is this inflammation coming from? What's causing it?

So we do think that one of the major sources of inflammation in the prostate is urinary microbes that exist in the urinary tract and maybe even these uropathogenic microbes that not only cause inflammation that drives this club cell phenotype, but also can carry additional things such as genotoxins that can cause DNA double-strand breaks. So there is possibly, maybe even probably a relationship between the microbiome and the development of club cells, and we're very, very, very interested in that topic.

Andrea Miyahira: Well, thank you again so much for coming on and sharing this with us today.

Karen Sfanos: Thank you so much for the opportunity. Really appreciate it.